JACC Volume 78, Issue 11 September

SEPTEMBER 14, 2021
VOLUME 78
NUMBER 11
Listen to this issue’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster.
SEPTEMBER 14, 2021
VOLUME 78
NUMBER 11
CONTENTS
ORIGINAL INVESTIGATIONS High-Sensitivity C-Reactive Protein Modifies the Cardiovascular Risk of Lipoprotein(a): 1083
Multi-Ethnic Study of Atherosclerosis
Wei Zhang, Jaime Lynn Speiser, Fan Ye, Michael Y. Tsai, Miguel Cainzos-Achirica,
Khurram Nasir, David M. Herrington, Michael D. Shapiro
The current study tested the hypothesis that lipoprotein(a) [Lp(a)]–associated atherosclerotic
cardiovascular disease (ASCVD) risk is modified by systemic inflammation in the general population.
A mean of 13.6 years follow-up in 4,679 participants in the MESA (Multi-Ethnic Study of
Atherosclerosis) cohort demonstrated that the risk of CVD was significantly associated with elevated
Lp(a) only in the setting of high-sensitivity C-reactive protein (hsCRP) $2 mg/L. Individuals with
concomitant increase in both Lp(a) and hsCRP had particularly higher ASCVD risk and all-cause
mortality. These findings suggest that individuals with concomitant presence of elevated Lp(a) and
systemic inflammation may merit closer surveillance and more aggressive ASCVD risk management.
SEE ADDITIONAL CONTENT ONLINE
n EDITORIAL COMMENT
Lifetime Risk Estimation in Atherosclerotic Cardiovascular Disease: 1095
Where Inflammation Meets Lipoprotein(a)
Xavier Rossello
Phenotypic Expression and Outcomes in Individuals With Rare Genetic Variants of 1097
Hypertrophic Cardiomyopathy
Antonio de Marvao, Kathryn A. McGurk, Sean L. Zheng, Marjola Thanaj, Wenjia Bai,
Jinming Duan, Carlo Biffi, Francesco Mazzarotto, Ben Statton, Timothy J.W. Dawes,
Nicolò Savioli, Brian P. Halliday, Xiao Xu, Rachel J. Buchan, A. John Baksi, Marina Quinlan,
Paweł Tokarczuk, Upasana Tayal, Catherine Francis, Nicola Whiffin, Pantazis I. Theotokis,
Xiaolei Zhang, Mikyung Jang, Alaine Berry, Antonis Pantazis, Paul J.R. Barton,
Daniel Rueckert, Sanjay K. Prasad, Roddy Walsh, Carolyn Y. Ho, Stuart A. Cook,
James S. Ware, Declan P. O’Regan
Hypertrophic cardiomyopathy (HCM) is an important cause of sudden death and is associated with
variants in sarcomere-encoding genes. Determining the significance of rare HCM-associated variants
outside of familial disease, and the significance of incidental unexplained hypertrophy, remain key
clinical challenges. In a community population of >200,000 adults, most individuals harboring
pathogenic variants did not have overt HCM; however, an increased risk of adverse outcomes was
observed, although the absolute event rate was low. Most individually rare sarcomeric variants,
which are collectively relatively common, appeared to be clinically benign. We also observed a low
yield of genetic testing in unselected adults with unexplained left ventricular hypertrophy.
n EDITORIAL COMMENT
Hypertrophic Cardiomyopathy in the General Population: Leveraging the UK Biobank 1111
Database and Machine Learning Phenotyping
Linnea M. Baudhuin
Articles have accompanying
audio accessible online at
www.onlineJACC.org.
CONTENTS
SEPTEMBER 14, 2021 VOLUME 78, NUMBER 11
Technology-Assisted Self-Selection of Candidates for Nonprescription Statin Therapy 1114

Steven E. Nissen, Howard G. Hutchinson, Tracy Y. Wang, Christie M. Ballantyne,
Sara Travis, Melanie Morris, William Miller, Jennifer Hynson, Kathy Wolski,
Paul M Ridker
Although statins reduce cardiovascular morbidity and mortality, only about one-half of eligible
patients receive treatment. A total of 500 participants, 83 with limited literacy, completed an at-
home Web-based application to assess appropriateness for treatment with rosuvastatin 5 mg. A
clinician, blinded to the information entered by the participant, performed an independent
assessment. Participant selection for statin therapy was concordant with clinician selection in 481
(96.2%) of 500 participants, of whom 23 (4.6%) were deemed appropriate and 458 (91.6%) were
deemed inappropriate for treatment. The use of a technology-assisted consumer self-selection for
statin therapy resulted in participant self-selection that showed substantial agreement with
clinician selection.
n EDITORIAL COMMENT
Should Cardiovascular Preventive Therapy Be Over-the-Counter? 1124
Neha J. Pagidipati, Eric D. Peterson
Innate Lymphoid Cells Promote Recovery of Ventricular Function After 1127

Myocardial Infarction
Xian Yu, Stephen A. Newland, Tian X. Zhao, Yuning Lu, Andrew S. Sage, Yanyi Sun,
Rouchelle S. Sriranjan, Marcella K.L. Ma, Brian Y.H. Lam, Meritxell Nus,
James E. Harrison, Simon J. Bond, Xiang Cheng, Jean-Sébastien Silvestre,
James H.F. Rudd, Joseph Cheriyan, Ziad Mallat
Innate lymphoid cells type 2 (ILC2s) reside in perivascular niches and limit the development of
atherosclerosis. Here, we found that ILC2s also reside in the pericardium and protect the heart
after ischemic injury in mice. Interleukin (IL)-2 played a critical upstream role in their activation
both in mice and in the circulating blood of patients with acute coronary syndromes. Low-dose IL-
2-induced ILC2 activation could constitute a novel therapeutic strategy to promote a reparative
response after myocardial infarction.
n EDITORIAL COMMENT
Innate Lymphoid Cells Participate in Myocardial Inflammation After Ischemia 1143
Matthias Nahrendorf
CONTENTS
THE PRESENT
AND FUTURE
JACC STATE-OF-THE- Medial Arterial Calcification: JACC State-of-the-Art Review 1145

ART REVIEW Peter Lanzer, Fadil M. Hannan, Jan D. Lanzer, Jan Janzen, Paolo Raggi, Dominic Furniss,
Mirjam Schuchardt, Rajesh Thakker, Pak-Wing Fok, Julio Saez-Rodriguez, Angel Millan,
Yu Sato, Roberto Ferraresi, Renu Virmani, Cynthia St. Hilaire
Medial arterial calcification (MAC) is a chronic systemic vascular disorder distinct from atherosclerosis
that is frequently associated with diabetes mellitus, chronic kidney disease, aging, and a large number
of less common diseases. The hallmarks of MAC include disseminated and progressive precipitation of
calcium phosphate within the medial layer and a prolonged and clinically silent course often leading
to chronic limb-threatening ischemia. MAC increases the risk of vascular interventions and mitigates
their outcomes. With the exception of rare monogenetic defects affecting adenosine triphosphate
metabolism, MAC pathogenesis remains unknown, and causal therapy is not available. The current
state of knowledge concerning MAC is provided, and future perspectives are outlined.
JACC STATE-OF-THE- Exercise Intolerance in Older Adults With Heart Failure With Preserved Ejection Fraction: 1166
ART REVIEW JACC State-of-the-Art Review
Ambarish Pandey, Sanjiv J. Shah, Javed Butler, Dean L. Kellogg, Jr., Gregory D. Lewis,
Daniel E. Forman, Robert J. Mentz, Barry A. Borlaug, Marc A. Simon, Julio A. Chirinos,
Roger A. Fielding, Elena Volpi, Anthony J.A. Molina, Mark J. Haykowsky, Flora Sam,
Bret H. Goodpaster, Alain G. Bertoni, Jamie N. Justice, James P. White, Jingzhone Ding,
Scott L. Hummel, Nathan K. LeBrasseur, George E. Taffet, Iraklis I. Pipinos,
Dalane Kitzman
Exercise intolerance (EI) is the primary manifestation of heart failure with preserved ejection fraction
(HFpEF) in older adults and is associated with poor outcomes. A better understanding of the
pathophysiology of EI in HFpEF is needed for its optimal management. This review provides a
comprehensive discussion of the evidence supporting key roles for extracardiac contributors to EI
in HFpEF and biological mechanisms of aging to the development and progression of HFpEF.
Future research leveraging a transdisciplinary approach that includes cardiovascular and
geroscience experts is needed for better phenotypic characterization of HFpEF and the
identification of key geroscience targets for treatment.

CONTENTS
FELLOWS-IN-TRAINING How Cardiovascular Disease Fellows Can Promote Diversity and Inclusion in Cardiology: 1188
& EARLY CAREER SECTION Doing Our Part
Joyce N. Njoroge, Quentin R. Youmans, Sarah Chuzi
n RESPONSE
Promoting Equity, Diversity, and Inclusion in Cardiology: 1191
Thinking Broadly, Acting Practically
Robert O. Roswell
LETTERS A Transatlantic Comparison of Patient-Reported Access to and 1193

Use of Aspirin in Contemporary Preventive Cardiology
Alan P. Jacobsen, Zi Lun Lim, Blair Chang, Kaleb D. Lambeth, Thomas M. Das,
Colin Gorry, Michael McCague, William Wijns, Patrick W.J.C. Serruys,
Roger S. Blumenthal, Seth S. Martin, John W. McEvoy
ONLINE FEATURE Calcium Score to Specify Assessment of Low-Flow Aortic e71

Stenosis Severity
Gudrun Lamm, Johann Auer
ONLINE FEATURE Reply

Guillaume Jean, Philippe Pibarot, Marie-Annick Clavel
THESE ARTICLES ONLY APPEAR IN THE ONLINE VERSION OF THE ISSUE.

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 78, NO. 11, 2021
ª 2021 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION
PUBLISHED BY ELSEVIER
ORIGINAL INVESTIGATIONS
High-Sensitivity C-Reactive Protein

Modifies the Cardiovascular Risk of
Lipoprotein(a)
Multi-Ethnic Study of Atherosclerosis
Wei Zhang, MD, PHD, MS,a Jaime Lynn Speiser, MSC, PHD,b Fan Ye, MD, PHD,a Michael Y. Tsai, PHD,c
Miguel Cainzos-Achirica, MD, MPH, PHD,d Khurram Nasir, MD, MPH, MSC,d David M. Herrington, MD, MHS,a
Michael D. Shapiro, DO, MCRa
ABSTRACT
BACKGROUND Little is known about the relationship between lipoprotein (a) [Lp(a)] and high-sensitivity C-reactive
protein (hsCRP) and their joint association with atherosclerotic cardiovascular disease (ASCVD).
OBJECTIVES The purpose of this study was to assess whether Lp(a)-associated ASCVD risk is modified by hsCRP in the
context of primary prevention.
METHODS The current study included 4,679 participants from the MESA (Multi-Ethnic Study of Atherosclerosis)
Apolipoprotein ancillary data set. Cox proportional hazards models and Kaplan-Meier curves were used to assess the
association among Lp(a), hsCRP, and time to cardiovascular disease (CVD) events.
RESULTS During a mean follow-up of 13.6 years, 684 CVD events occurred. A significant interaction was observed
between Lp(a) and hsCRP (P ¼ 0.04). With hsCRP <2 mg/L, no significant CVD risk was observed at any level of Lp(a)
from <50 mg/dL to >100 mg/dL. However, with hsCRP $2 mg/L, a significant CVD risk was observed with Lp(a) of
50-99.9 mg/dL (HR: 1.36; 95% CI: 1.02-1.81) and Lp(a) $100 mg/dL (HR: 2.09; 95% CI: 1.40-3.13). Isolated elevations
of either Lp(a) or hsCRP were not associated with increased CVD risk. In contrast, the combination of elevated Lp(a)
($50 mg/dL) and hsCRP ($2 mg/L) was independently associated with significant CVD risk (HR: 1.62; 95% CI: 1.25-2.10)
and all-cause mortality (HR: 1.39; 95% CI: 1.12-1.72).
CONCLUSIONS Lp(a)-associated ASCVD risk is observed only with concomitant elevation of hsCRP. Individuals with
concomitant presence of elevated Lp(a) and systemic inflammation have greater ASCVD risk and all-cause mortality, and
thus may merit closer surveillance and more aggressive ASCVD risk management.
(J Am Coll Cardiol 2021;78:1083–1094) © 2021 by the American College of Cardiology Foundation.
Listen to this manuscript’s

audio summary by From the aCenter for Prevention of Cardiovascular Disease, Section on Cardiovascular Medicine, Department of Internal Medicine,
Editor-in-Chief Wake Forest University Baptist Medical Center, Winston-Salem, North Carolina, USA; bThe Department of Biostatistics and Data
Dr. Valentin Fuster on Sciences, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA; cDepartment of Laboratory Medicine
JACC.org. and Pathology, University of Minnesota, Minneapolis, Minnesota, USA; and the dDivision of Cardiovascular Prevention and
Wellness, Department of Cardiology, DeBakey Heart and Vascular Center, Houston Methodist Hospital, Houston, Texas, USA.
The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’
institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information,
visit the Author Center.
Manuscript received June 7, 2021; revised manuscript received June 30, 2021, accepted July 2, 2021.
ISSN 0735-1097/$36.00 https://doi.org/10.1016/j.jacc.2021.07.016

1084 Zhang et al. JACC VOL. 78, NO. 11, 2021
hsCRP Modifies Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094
S METHODS
ABBREVIATIONS ignificant declines in atherosclerotic
AND ACRONYMS cardiovascular disease (ASCVD) mor-
tality have occurred over the last 4 de- STUDY POPULATION. The study design of MESA
ACC = American College of
Cardiology
cades (1). Despite this progress, (Multi-Ethnic Study of Atherosclerosis) has been
cardiovascular disease remains the leading previously described by Bild et al (19). More protocol
AHA = American Heart
Association cause of death and disability globally, ac- information about the MESA study can be found on-
ASCVD = atherosclerotic counting for about 17.6 million deaths world- line (20). Briefly, MESA recruited 6,814 asymptomatic
cardiovascular disease wide in 2016 (1). Residual risk of men and women aged 45-84 years without overt
CVD = cardiovascular disease atherosclerotic events remains high (2). Addi- clinical cardiovascular disease (CVD) from 6 commu-
hsCRP = high sensitivity tional targets of therapy may be needed to nities in the United States. Informed consent and
C-reactive protein further improve cardiovascular health and institutional review board approval was completed at
LDL-C = low-density outcomes (3). Lipoprotein(a) [Lp(a)] is one all MESA participating sites. A total of 6 examinations
lipoprotein cholesterol
such putative target (4). Lp(a) is a low- have been completed since 2000.
Lp(a) = lipoprotein(a) density lipoprotein (LDL)–like particle with The current study included 4,679 participants from
an additional glycoprotein, apolipoprotein(a), cova- the MESA Apolipoprotein ancillary dataset with
lently bound to its apolipoprotein B moiety (5). There- the following race/ethnic distribution: Caucasian
fore, Lp(a) is an atherogenic, proinflammatory, and (n ¼ 1,709; 36.6%), Chinese American (n ¼ 560; 12%),
prothrombotic lipoprotein particle. Epidemiologic, African American (AA) (n ¼ 1,346; 28.7%), and His-
experimental, and genetic studies suggest a causal panic (n ¼ 1,064; 22.7%). Of this study population,
relationship between elevated Lp(a) and risk for 4,654 participants had measurements of both Lp(a)
myocardial infarction (MI), stroke, and calcific aortic and hsCRP and were included in the analysis for this
valve stenosis (6-10). A novel antisense oligonucleo- paper.
tide therapy has been developed and showed potent DEMOGRAPHICS AND BASELINE CHARACTERISTICS.
Lp(a) lowering efficacy in a phase II trial (11,12). Age, sex, race/ethnicity, smoking status, education,
medical history, family history, and medication use
SEE PAGE 1095
data were collected by standard questionnaires.
Inflammation is another emerging target of ther- Diabetes mellitus was defined as fasting
apy to potentially reduce residual ASCVD risk. Re- glucose $126 mg/dL or hypoglycemic medication use.
sults of recent large randomized controlled trials Body mass index (BMI) was calculated as weight (in
demonstrated improved cardiovascular outcomes kg) divided by height (in m2 ). Resting blood pressure
with specific anti-inflammatory therapies (13,14). was repeated 3 times after 5 minutes in the seated
Among the numerous inflammatory biomarkers, position using a Dinamap automated oscillometric
high-sensitivity C-reactive protein (hsCRP) has the sphygmomanometer (Critikon), and the average of
most extensive predictive validation with regard to the second and third readings was used in analyses.
ASCVD outcomes (15). Lp(a) and hsCRP are both LABORATORY MEASUREMENTS. Lp(a) mass was
listed as risk enhancers in the 2018 American Heart determined in plasma using a latex-enhanced turbi-
Association (AHA)/American College of Cardiology dimetric immunoassay (Denka Seiken) at Health Di-
(ACC) cholesterol guideline (16) and the 2019 ACC/ agnostics Laboratory. Total cholesterol, high-density
AHA guideline on primary prevention of cardiovas- lipoprotein cholesterol (HDL-C), and triglycerides
cular disease (17). A recent post hoc analysis of the were measured using blood samples collected after
ACCELERATE (Assessment of Clinical Effects of overnight fasting. Low-density lipoprotein choles-
Cholesteryl Ester Transfer Protein Inhibition with terol (LDL-C) was calculated using the Friedewald
Evacetrapib in Patients at a High Risk for Vascular equation in specimens with triglycerides <400 mg/dL
Outcomes) trial reported that Lp(a) was associated (21); 43 participants with triglyceride >400 mg/dL
with greater risk of major adverse cardiovascular were assigned missing values. Serum creatinine was
events only when hsCRP was $2 mg/L (18). It is not measured by rate reflectance spectrophotometry us-
known whether the Lp(a)-associated ASCVD risk is ing thin film adaptation of the creatine amidinohy-
modified by inflammation as measured by plasma drolase method on the Vitros analyzer (Johnson &
hsCRP in primary prevention. The purpose of the Johnson Clinical Diagnostics, Inc) and estimated
current study was to examine the relationship glomerular filtration rate was calculated using the
among Lp(a), subclinical inflammation, and their creatinine-based CKD-EPI (Chronic Kidney Disease
joint association with ASCVD in a population free of Epidemiology Collaboration) equation. hsCRP was
clinical ASCVD at baseline. measured by a BNII nephelometer using a particle-
JACC VOL. 78, NO. 11, 2021 Zhang et al. 1085
SEPTEMBER 14, 2021:1083–1094 hsCRP Modifies Lp(a)-Associated ASCVD Risk
enhanced immunonephelometric assay (N High

T A B L E 1 Baseline Characteristics of the Study Population Classified by hsCRP
Sensitivity CRP, Dade Behring Inc).
hsCRP <2 mg/L hsCRP $2 mg/L Total
CLINICAL ENDPOINTS. Incident cardiovascular (n ¼ 2,388) (n ¼ 2,273) (N ¼ 4,661) P Value
events were recorded from baseline (2000) through Age, y 61.7 10.7 62.2 10.0 62 10.4 0.10
calendar year 2017. Participants were followed for a Female 43.2 62.4 52.5 <0.001
mean of 13.6 4.43 years. The cardiovascular events BMI, kg/m2 26.3 4.4 30.2 5.8 28.2 5.5 <0.001
analyzed in this study include myocardial infarction, Ethnicity <0.001
Caucasian 38.8 34.2 36.6
fatal and nonfatal coronary heart disease (CHD),
Chinese American 18.1 5.6 12.0
definite angina, and probable angina if followed by
African American 23.7 34.0 28.7
revascularization, resuscitated cardiac arrest, fatal
Hispanic 19.4 26.3 22.7
and nonfatal stroke, and other atherosclerotic or CVD hsCRP, mg/L 0.88 (0.53-1.33) 4.44 (3.02-7.93) 1.94 (0.86-4.32) <0.001
death. Telephone interviews were conducted at 9- to Lipids
12-month intervals for each participant to obtain data Lp(a), mg/dL 16 (7.7-37.2) 20 (8.5-42.5) 17.85 (8.2-40.0) 0.001
regarding interim hospitalizations, outpatient car- LogLp(a), U 1.19 0.50 1.25 0.53 1.22 0.52 <0.001
diovascular procedures and diagnoses, and deaths. LDL-C, mg/dL 119 31 120 32 120 31 0.20
HDL-C, mg/dL 52 15.2 50 14.9 51 15.1 0.33
Additional medical encounters were identified
Triglycerides, mg/dL 101 (17-147) 116 (82-169) 109 (76-158) <0.001
through cohort clinic visits, call-ins, medical record
Total cholesterol, mg/dL 195.0 34.0 198.0 36.0 196.0 35.5 0.01
abstractions, or obituaries. Copies of all death certif-
HTN 34.9 48.8 41.7 <0.001
icates and medical records of hospitalizations and HTN therapy 27.4 39.1 33.1 <0.001
selected outpatient cardiovascular procedures and Lipid-lowering therapy 0.13 0.09 0.10 0.69
diagnoses were collected to verify self-reported di- Diabetes 9.8 14.0 11.9 <0.001
agnoses (19). Current smoker 10.9 15.2 13.0 <0.001
STATISTICAL ANALYSIS. Continuous normally eGFR, mL/min/1.73 m2 75.6 15.1 74.7 16.6 75 15.8 0.08
distributed variables were reported as mean SD.

Values are mean SD, %, or median (interquartile range).
Variables with skewed distributions were reported as BMI ¼ body mass index; eGFR ¼ estimated glomerular filtration rate; HDL-C ¼ high-density lipoprotein
cholesterol; hsCRP ¼ high-sensitivity C-reactive protein; HTN ¼ hypertension; LDL-C ¼ low-density lipoprotein;
median (interquartile range). Categorical variables
LogLp(a) ¼ log transformed lipoprotein(a); Lp(a) ¼ lipoprotein(a).
were reported as frequency and percentage (%). Log
transformed values of Lp(a) and hsCRP were used
when they were included as continuous variables in
adjusting to corrected LDL-C [total LDL 30% Lp(a)].
statistical models (18,22). Elevated hsCRP was defined
The proportional hazard assumption was assessed by
as $2 mg/L, consistent with the current ACC/AHA
the ASSESS statement with the Ph option and also by
Cholesterol guideline (17). Lp(a) was also examined as
the Schoenfeld residuals method.
a categorical variable by commonly used clinical cut
points (50 and 100 mg/dL) (23) or quartiles. Risk of
CVD was also compared among 4 risk groups classi- T A B L E 2 Lipoprotein(a)-Associated Cardiovascular Risk According to hsCRP Threshold
fied by elevation of none, 1, or both risk factors [Lp(a)
Groups Events/Total (%) HR (95% CI) P Value a
and hsCRP]. Differences in baseline characteristics
In entire cohort
were compared using the chi-square test for cate-
LogLp(a), per unit 676/4,611 (14.7) 1.18 (1.00-1.41) 0.05
gorical variables and an unpaired Student’s t-test or Lp(a) <50 mg/dL 528/3,764 (14) 1.00 (reference) NA
Mann-Whitney U test for continuous variables. The Lp(a) $50 mg/dL 156/889 (17.5) 1.36 (1.10-1.65) 0.001
time-to-ASCVD across Lp(a) categories by hsCRP hsCRP <2 mg/L
threshold was evaluated using Kaplan-Meier curves. LogLp(a), per unit 335/2,379 (14) 1.02 (0.81-1.27) 0.88
The cumulative incidence of events during the Lp(a) <50 mg/dL 268/1,960 (13.7) 1.00 (reference) NA
Lp(a) $50 mg/dL 66/417 (15.8) 1.19 (0.89-1.58) 0.23
follow-up period was computed. Multivariable Cox
hsCRP $2 mg/L
proportional hazards models were used to assess the
LogLp(a), per unit 348/2,263 (15.4) 1.32 (1.05-1.65) 0.016
relationship between Lp(a) and endpoints. Formal
Lp(a) <50 mg/dL 256/1,788 (14.2) 1.00 (reference) NA
interaction test between Lp(a) and hsCRP was per- Lp(a) $50 mg/dL 90/470 (19.1) 1.52 (1.18-1.95) 0.001
formed. Model adjustments were made for age, sex,
ethnicity, hypertension, use of antihypertensive Elevation of Lp(a), either by log unit increase or above the cutoff of 50 mg/dL, was associated with significant
CVD risk only in the setting of hsCRP $2 mg/L. Multivariable Cox proportional hazards regression model was
medication, diabetes, smoking status (never, former, adjusted for age, sex, ethnicity, hypertension, use of hypertension medications, diabetes, smoking status, high-
density lipoprotein cholesterol, triglycerides, total cholesterol, and renal function (estimated glomerular filtra-
current), HDL-C, triglycerides, total cholesterol, and
tion rate). aP value for interaction: LogLp(a) LogCRP ¼ 0.04; LogLp(a) hsCRP(<2 or $2) ¼ 0.08.
renal function as assessed by estimated glomerular NA ¼ not available; other abbreviations as in Table 1.
filtration rate. Sensitivity analysis was performed by

F I G U R E 1 Cardiovascular Risk Across Clinical Strata of Lp(a)
A hsCRP <2 mg/L

Lp(a) Strata Events/Total (%) HR (95% CI) P Value
<50 mg/dL 268/1,960 (13.7) 1.00 (reference) NA

50–99.9 mg/dL 55/332 (16.5) 1.21 (0.89–1.64) 0.21
≥100 mg/dL 11/85 (12.9) 1.10 (0.58–2.08) 0.77
hsCRP ≥2 mg/L
Lp(a) Strata Events/Total (%) HR (95% CI) P Value
<50 mg/dL 256/1,788 (14.3) 1.00 (reference) NA
50–99.9 mg/dL 62/359 (17.3) 1.36 (1.02–1.81) 0.03
≥100 mg/dL 28/111 (25.2) 2.09 (1.40–3.13) < 0.001
0.5 1 2 4
(A) HRs for cardiovascular events. Cox proportional hazards model was adjusted for age, sex, race/ethnicity, hypertension, use of hypertension medications, diabetes,
smoking status, high-density lipoprotein cholesterol, triglycerides, total cholesterol, and renal function (estimated glomerular filtration rate). Point estimates of HRs
are represented by the blue dots and 95% CI by the horizontal lines. (B) Kaplan-Meier curves of cumulative incidence of cardiovascular events. Lp(a) <50 mg/L was
used as the reference group. CVD ¼ cardiovascular disease; hsCRP ¼ high-sensitivity C-reactive protein; Lp(a) ¼ lipoprotein(a).
Continued on the next page
Given the observation that median levels and dis- cholesterol, and BMI. Additionally, there was a
tributions of Lp(a) differ significantly between AA and higher prevalence of female sex, hypertension, use
other ethnic groups, a subgroup analysis was per- of antihypertensive medications, diabetes, and
formed to compare the relationship of Lp(a) and current smoking among those with hsCRP $2 mg/L.
hsCRP to CVD in AA and non-AA. A subgroup analysis AA participants manifested a significantly higher
was also conducted to explore this relationship across median Lp(a) level of 35 mg/dL compared with
sex. Statistical analysis was performed using SAS median values of approximately 13 mg/dL in other
version 9.4 (SAS Institute Inc). Statistical significance ethnicities, consistent with prior reports (22)
was defined as a 2-tailed P value <0 .05, except for (Supplemental Figure 1).
interaction testing where a P value <0.2 was consid-
ASSOCIATION BETWEEN CARDIOVASCULAR
ered significant. We prespecified 0.2 as the cutoff to
EVENTS AND LP(A) ACCORDING TO hsCRP
be conservative in terms of determining if there was
THRESHOLDS. Table 2 displays the results of the Cox
an interaction between Lp(a) and hsCRP and justi-
proportional hazard models evaluating the associa-
fying more detailed inspection on its potentially
tion between risk of CVD events and Lp(a) by log-
important impact on clinical outcomes.
transformed [LogLp(a)] or 50 mg/dL threshold. In
RESULTS the total study population, Lp(a) was significantly
associated with CVD assessed by either per unit in-
BASELINE CHARACTERISTICS OF STUDY POPULATION. crease of LogLp(a) (HR: 1.18; 95% CI: 1.00-1.41;
Baseline characteristics were presented by hsCRP P ¼ 0.05), or categorical threshold of 50 mg/dL (HR:
thresholds (Table 1). The mean age of participants in 1.36; 95% CI: 1.10-1.65; P ¼ 0.001). A significant
this analysis was 62 years (52.5% women). Of the interaction was observed between Lp(a) and hsCRP
study population, 36.6% were Caucasian, 28.7% tested as log-transformed (LogCRP) (P for
were AA, 22.7% were Hispanic, and 12% were Asian. interaction ¼ 0.04, and 0.08 when hsCRP was tested
Median Lp(a) and hsCRP were 17.85 mg/dL and by 2 mg/L threshold) (Table 2). In those with hsCRP
1.94 mg/L, respectively. Elevated hsCRP was asso- <2 mg/L, there was not a significant association be-
ciated with higher levels of Lp(a), triglycerides, total tween risk of CVD events and Lp(a) when evaluated
F I G U R E 1 Continued
B hsCRP <2 mg/L

35
Log-Rank P = 0.41
Cumulative Incidence of CVD (%)

30
25
20
15
10
0
0 2 4 6 8 10 12 14 16 18
Time to First Event (Years)
No. at risk
Lp(a) <50 mg/dL 1,960 1,887 1,812 1,714 1,622 1,529 1,427 1,286 723
Lp(a) 50–99.9 mg/dL 332 324 308 294 277 257 245 212 125
Lp(a) ≥100 mg/dL 85 81 78 74 71 66 57 54 32
hsCRP ≥2 mg/L
35
Log-Rank P = 0.001
30
25
20
15
10
0
0 2 4 6 8 10 12 14 16 18
No. at risk
Lp(a) <50 mg/dL 1,788 1,711 1,617 1,528 1,416 1,321 1,236 1,102 626
Lp(a) 50–99.9 mg/dL 359 340 328 300 283 258 237 213 114
Lp(a) ≥100 mg/dL 111 103 94 84 76 71 64 56 28
either by LogLp(a) (HR: 1.02; 95% CI: 0.81-1.27) or consistent, as shown in Supplemental Table 1. When
categorical threshold of 50 mg/dL (HR: 1.19; 95% CI: coronary heart disease and stroke were analyzed as
0.89-1.58). On the other hand, in those with separate endpoints, greater HRs were consistently
hsCRP $2 mg/L, elevation of Lp(a), either by log unit observed in the group with Lp(a) $50 mg/dL and
increase (HR: 1.32; 95% CI: 1.05-1.65; P ¼ 0.016) or hsCRP $2 mg/L (Supplemental Tables 2A and 2B,
categorical threshold of 50 mg/dL (HR: 1.52; 95% CI: Supplemental Figure 3). Additionally, dual elevation
1.18-1.95; P ¼ 0.001) was significantly associated with of Lp(a) and hsCRP was associated with significantly
risk of CVD events. higher all-cause mortality (HR: 1.39; 95% CI: 1.12-1.72;
RISK OF CARDIOVASCULAR EVENTS ACROSS LP(A) P < 0.01) (Supplemental Table 2C).
STRATA ACCORDING TO hsCRP THRESHOLDS.
Figure 1A shows the results of Cox proportional haz-
EVENTS AND LP(A) ACCORDING TO hsCRP IN AA
ard model exploring the risk of CVD events across
AND NON-AA. Given the observation that median
Lp(a) strata (<50, 50-99.9, and $100 mg/dL). Again, in
levels and distributions of Lp(a) differ significantly
the setting of hsCRP <2 mg/L, no significant associa-
between AA and other ethnic groups (Supplemental
tion between Lp(a) and CVD risk was observed in any
Table 3, Supplemental Figure 1), a subgroup analysis
of the Lp(a) categories. However, in the setting of
was conducted by categorizing the study population
elevated hsCRP ($2 mg/L), a significant risk of CVD
to AA and non-AA (Table 3, Figure 3). The P values for
events was observed with increased levels of Lp(a)
interaction were 0.03 and 0.2 in AA and non-AA,
50-99.9 mg/dL (HR: 1.36; 95% CI: 1.02-1.81; P ¼ 0.03)
respectively. P values for interaction were 0.42 and
and Lp(a) $100 mg/dL (HR: 2.09; 95% CI: 1.40-3.13;
0.37 when Lp(a) was tested by the 50 mg/dL threshold
P < 0.001).
and hsCRP by the 2 mg/L threshold in AA and non-AA,
Figure 1B displays the cumulative incidence of CVD
respectively. In both AA and non-AA, elevation of
events over a mean follow-up of 13.6 years across
either Lp(a) ($50 mg/dL) or hsCRP ($2 mg/L) alone
Lp(a) strata according to hsCRP thresholds. In those
was not associated with a higher risk of CVD events
with hsCRP <2 mg/L, increasing Lp(a) was not asso-
compared with the reference group [Lp(a) <50 mg/dL
ciated with a higher cumulative incidence of CVD
and hsCRP <2 mg/L). The greatest risk of CVD events
(P ¼ 0.41). However, in those with hsCRP $2 mg/L,
was observed in those with concomitant elevation in
cumulative incidence of CVD was significantly higher
Lp(a) ($50 mg/dL) and hsCRP ($2 mg/L) (HR: 1.63;
in the Lp(a) 50-99.9 mg/dL group, and highest in
95% CI: 1.09-2.44; P ¼ 0.016 and HR: 1.66; 95% CI:
those with Lp(a) $100 mg/dL (P ¼ 0.001). Similar re-
1.17-2.35; P ¼ 0.005 for AA and non-AA, respectively).
sults were found while Lp(a) was assessed by quar-
tiles (Supplemental Figure 2).
RISK OF CARDIOVASCULAR EVENTS AND EVENTS AND LP(A) ACCORDING TO hsCRP BY SEX.
ALL-CAUSE MORTALITY IN DIFFERENT RISK Supplemental Table 4 demonstrates that men had
GROUPS CLASSIFIED BY LP(A) AND hsCRP. Risk of relatively lower median levels of Lp(a) and hsCRP
CVD was then compared by classifying participants to than women. Table 4 and the Central Illustration
4 risk groups: elevation of none, 1, or both risk factors display the results of the Cox proportional hazard
[Lp(a) and hsCRP]. Figure 2 displays the results of Cox models and Kaplan-Meier curves of the cumulative
proportional hazard models and Kaplan-Meier curves incidence of CVD events by sex. The P values for
of the cumulative incidence of CVD events over a interaction were 0.04 and 0.47 in women and men,
mean of 13.6 years follow-up, grouped by Lp(a) and respectively. The P values for interaction were 0.09
hsCRP thresholds. Compared with the reference group and 0.85 when Lp(a) was tested by the 50 mg/dL
[Lp(a) <50 mg/dL and hsCRP <2 mg/L], an isolated threshold and hsCRP by the 2 mg/L threshold,
elevation of either Lp(a) ($50 mg/dL) or hsCRP respectively. Compared with the reference group
($2 mg/L) was not associated with a higher risk of CVD [Lp(a) <50 mg/dL and hsCRP <2 mg/L], the only
events. However, concomitant elevation in both Lp(a) significantly increased risk of CVD events was
($50 mg/dL) and hsCRP ($2 mg/L) was associated observed among participants with dual elevations of
with a significantly increased risk of CVD events (HR: Lp(a) and hsCRP in both women and men (HR: 1.52;
1.62; 95% CI: 1.25-2.10; P < 0.001). Given the inherent 95% CI: 1.06-2.18; P ¼ 0.02 and HR: 1.61; 95% CI: 1.12-
relationship between Lp(a), LDL-C, and CVD risk, a 2.31; P ¼ 0.01, respectively). Isolated elevation of
sensitivity analysis was performed by adjusting to either Lp(a) or hsCRP had higher cumulative inci-
corrected LDL-C [total LDL – 30% Lp(a)] in the Cox dence of CVD events in men, but this association was
proportional hazard model, and the results were not statistically significant.
F I G U R E 2 Cardiovascular Risk Classified by Lipoprotein(a) and hsCRP
A Risk Groups Events/Total (%) HR (95% CI) P Value
Lp(a) <50 mg/dL & hsCRP <2 mg/L 268/1,960 (13.7) 1.00 (reference) NA
Lp(a) ≥50 mg/dL & hsCRP <2 mg/L 66/417 (15.8) 1.19 (0.90–1.57) 0.23
Lp(a) <50 mg/dL & hsCRP ≥2 mg/L 256/1,788 (14.3) 1.06 (0.89–1.27) 0.54
Lp(a) ≥50 mg/dL & hsCRP ≥2 mg/L 90/470 (19.1) 1.62 (1.25–2.10) < 0.001
B 25
Log-Rank P = 0.006
20
15
10
0
0 2 4 6 8 10 12 14 16 18
No. at risk
Lp(a) <50 mg/dL & hsCRP <2 mg/L 1,960 1,887 1,812 1,714 1,622 1,529 1,427 1,286 723
Lp(a) ≥50 mg/dL & hsCRP <2 mg/L 417 404 386 368 347 323 301 265 156
Lp(a) <50 mg/dL & hsCRP ≥2 mg/L 1,788 1,711 1,617 1,528 1,416 1,321 1,236 1,102 626
Lp(a) ≥50 mg/dL & hsCRP ≥2 mg/L 470 443 421 384 358 330 301 268 140
(A) HRs for cardiovascular events. Cox proportional hazards model was adjusted for age, sex, race/ethnicity, hypertension, use of hypertension medications, diabetes,
smoking status, high-density lipoprotein cholesterol, triglycerides, total cholesterol, and renal function (estimated glomerular filtration rate). (B) Kaplan-Meier curves
of cumulative incidence of cardiovascular events. Lp(a) <50 mg/dL and hsCRP <2 mg/L was used as the reference group. P value for interaction between Lp(a) and
hsCRP ¼ 0.04 (as continuous variable) or 0.17 (as a dichotomous variable). Abbreviations as in Figure 1.
DISCUSSION A recent post hoc analysis suggested that hsCRP

may modulate Lp(a)-associated CVD risk in the
Despite optimal management of traditional risk fac- ACCELERATE trial, which is essentially in a secondary
tors, including aggressive LDL-C–lowering therapy, prevention population (18). However, it is not known
significant residual cardiovascular risk remains. whether the ASCVD risk associated with elevated
There are many drivers of residual risk, including Lp(a) is modified by the presence of inflammation as
inflammatory, pro-thrombotic, and metabolic path- measured by plasma hsCRP in the general population.
ways (3). Lp(a) and hsCRP are established biomarkers The purpose of the current study was to determine
that can facilitate ASCVD risk estimation in primary whether subclinical inflammation modifies Lp(a)-
prevention according to the 2018 AHA/ACC choles- associated cardiovascular risk in a population
terol guideline (16) and the 2019 ACC/AHA guideline without established clinical ASCVD at baseline.
on primary prevention of cardiovascular disease (17). Several key features of MESA, including large sample
T A B L E 3 HR for Cardiovascular Events by Race
African American Non-African American
Risk Groups Events/Total (%) HR (95% CI) P Value Events/Total (%) HR (95% CI) P Value
Lp(a) <50 mg/dL and hsCRP <2 mg/L 49/388 (12.6) 1.00 (reference) NA 219/1,572 (13.9) 1.00 (reference) NA
Lp(a) $50 mg/dL and hsCRP <2 mg/L 31/172 (19) 1.26 (0.80-2.00) 0.32 35/245 (14.3) 1.15 (0.80-1.66) 0.44
Lp(a) <50 mg/dL and hsCRP $2 mg/L 61/501 (12.2) 0.99 (0.68-1.45) 0.97 195/1,287 (15.2) 1.13 (0.92-1.37) 0.24
Lp(a) $50 mg/dL and hsCRP $2 mg/L 50/266 (18.8) 1.63 (1.09-2.44) 0.016 40/204 (19.6) 1.66 (1.17-2.35) 0.005
In both African Americans and Non-African Americans, elevated Lp(a) $50 mg/dL was associated with greater cardiovascular risk only with concomitant elevation of hsCRP
$2 mg/L. Cox proportional hazards model was adjusted for age, sex, hypertension, use of hypertension medications, diabetes, smoking status, high-density lipoprotein
cholesterol, triglycerides, total cholesterol and renal function (estimated glomerular filtration rate). P for interaction between Lp(a) and hsCRP ¼ 0.03 in African Americans and
0.2 in Non-African Americans.
Abbreviations as in Tables 1 and 2.
size, participation among multiple ethnic groups, and in the Lp(a) HORIZON trial (Assessing the Impact of
long follow-up duration allow this question to be Lipoprotein(a) Lowering with TQJ230 on Major Car-
addressed in a robust fashion. diovascular Events in Patients with CVD)
In this analysis of over 4,600 MESA participants, a (NCT04023552).
significant interaction was observed between Lp(a) Systemic inflammation, frequently found as a
and hsCRP. Lp(a) was associated with incident CVD pathophysiological feature of metabolic syndrome,
events over a mean of 13.6-year follow-up only in has been recognized as a major component of re-
those with elevated hsCRP ($2 mg/L). In participants sidual cardiovascular risk (3,25). Recent trials on the
without elevation in hsCRP (<2 mg/L), no significant proprotein convertase subtilisin/kexin type 9 in-
risk of CVD events was observed across any level of hibitors also demonstrated persistent residual in-
Lp(a), from <50 mg/dL to >100 mg/dL. Compared flammatory CVD risk among patients with stable
with those who had neither elevated Lp(a) nor ASCVD despite treatment with both high-intensity
elevated hsCRP, the presence of either elevated Lp(a) statin therapy and proprotein convertase subtilisin/
or hsCRP was not significantly associated with an kexin type 9 inhibition (26,27). Among the
increased risk of CVD events. However, concomitant numerous inflammatory biomarkers, hsCRP has the
elevation of both Lp(a) and hsCRP was associated greatest validation (15). In a review of multiple
with a significantly higher risk of CVD events. Similar clinical trials and registry data, the proportion of
impacts of dual elevations in hsCRP and Lp(a) were atherosclerotic patients on statin therapy who have
observed in both AA and non-AA and in both sexes. residual inflammatory risk (defined as an on-
Finally, dual elevation of Lp(a) and hsCRP was asso- treatment LDL-C <70 mg/dL and hsCRP $2 mg/L)
ciated with significantly higher risk of all-cause is more than twice the proportion of patients with
mortality. residual cholesterol risk (defined as an on-treatment
Epidemiologic, genetic, and post hoc analyses of LDL-C $70 mg/dL and hsCRP <2 mg/L) (28). Results
clinical trials suggest that Lp(a) plays a causal role in of recent large randomized controlled trials testing
the development of MI, stroke, and calcific aortic anti-inflammatory therapies targeting the NLRP3
valve stenosis (6-9). A meta-analysis of multiple inflammasome (canakinumab and colchicine) in
statin trials including 95,576 person-years at risk and participants with established ASCVD demonstrated
5,751 events revealed a linear association between improved cardiovascular outcomes (13,14), although
elevated Lp(a) level and ASCVD, with risk being the cardiovascular benefits of canakinumab were
particularly high among patients with Lp(a) levels not observed in those who had on-treatment
>50 mg/dL (8). Plasma Lp(a) levels are largely deter- hsCRP $2 mg/L (13). The connection between Lp(a)
mined by the LPA gene locus with minimal effects and inflammation was suggested by an early in vivo
from lifestyle or environmental factors (24). Recently, study by Dangas et al (29), where a strong correla-
an antisense oligonucleotide, pelacarsen, was devel- tion between plaque KP-1 and Lp(a) was found in
oped to silence the expression of the apolipopro- atherectomy specimens from patients with unstable
tein(a) gene, LPA (11). In a phase IIB trial, pelacarsen rest angina. Along those lines, a recent post hoc
reduced plasma Lp(a) by up to 80% (12). Whether analysis of participants from the ACCELERATE trial,
Lp(a) lowering with this strategy will improve car- a secondary prevention study, reported that
diovascular outcomes in patients with established elevated Lp(a) was only significantly associated
CVD and elevated Lp(a) is currently being evaluated with major adverse cardiovascular events when
F I G U R E 3 Cumulative Incidence of Cardiovascular Events in African Americans and Non-African Americans
African American
25
Log-Rank P = 0.018

20
15
10
0
0 2 4 6 8 10 12 14 16 18
No. at risk
Lp(a) <50 mg/dL & hsCRP <2 mg/L 388 371 357 338 313 288 268 243 117
Lp(a) <50 mg/dL & hsCRP ≥2 mg/L 501 482 450 427 395 367 346 309 145
Non-African American
25
Log-Rank P = 0.12
20
15
10
0
0 2 4 6 8 10 12 14 16 18
No. at risk
Lp(a) <50 mg/dL & hsCRP <2 mg/L 1,572 1,516 1,455 1,376 1,309 1,242 1,159 1,044 607
Lp(a) <50 mg/dL & hsCRP ≥2 mg/L 1,287 1,230 1,167 1,101 1,022 954 890 794 482
Kaplan-Meier curves of the cumulative incidence of CVD events are shown. In both African Americans and Non-African Americans, greater cumulative risk
of CVD was observed only in individuals with concomitant elevation of Lp(a) ($50 mg/dL) and hsCRP ($2 mg/L). Abbreviations as in Figure 1.
T A B L E 4 Hazard Ratio for Cardiovascular Events by Sex
Women Men
Risk Groups Events/Total (%) HR (95% CI) P Value Events/Total (%) HR (95% CI) P Value
Lp(a) <50 mg/dL and hsCRP <2 mg/L 81/820 (9.9) 1.00 (reference) NA 187/1,140 (16.4) 1.00 (reference) NA
Lp(a) $50 mg/dL and hsCRP <2 mg/L 21/204 (10.3) 0.91 (0.55-1.50) 0.71 45/213 (21.1) 1.35 (0.97-1.87) 0.08
Lp(a) <50 mg/dL and hsCRP $2 mg/L 115/1,078 (10.7) 0.95 (0.71-1.27) 0.71 141/710 (19.9) 1.12 (0.90-1.40) 0.31
Lp(a) $50 mg/dL and hsCRP $2 mg/L 53/333 (15.9) 1.52 (1.06-2.18) 0.02 37/137 (27.0) 1.61 (1.12-2.31) 0.01
In both women and men, elevated Lp(a) $50 mg/dL was associated with greater cardiovascular risk only with concomitant elevation of hsCRP $2 mg/L. Cox proportional
hazards model was adjusted for age, race/ethnicity, hypertension, use of hypertension medications, diabetes, smoking status, high-density lipoprotein cholesterol, triglycerides,
total cholesterol and renal function (estimated glomerular filtration rate). P value for interaction between Lp(a) and hsCRP ¼ 0.04 and 0.47 in women and men, respectively.
Abbreviations as in Table 1.
hsCRP was $2 mg/L (18). The results of our present In the 2018 AHA/ACC cholesterol guideline and the
analysis in primary prevention are consistent with 2019 ACC/AHA primary prevention guideline, hsCRP
the findings from ACCELERATE. Thus, data from and Lp(a) are both recommended as a risk-enhancers
both primary and secondary prevention studies to help inform clinical decision-making about initia-
suggest that elevated Lp(a) associates with ASCVD tion of statin therapy in individuals with intermediate
risk only in the presence of systemic inflammation, ASCVD risk (16). According to the current study,
as measured by elevated hsCRP. Subgroups analysis hsCRP levels may further refine Lp(a)-associated
further confirmed the presence of systemic inflam- cardiovascular risk in the setting of primary preven-
mation potentiated Lp(a)-associated ASCVD risk in tion. Individuals with elevated levels of both Lp(a)
both AAs and non-AAs as well as in both sexes. and hsCRP appear to be at particularly higher ASCVD
C E NT R AL IL L U STR AT IO N Cumulative Incidence of Cardiovascular Events in Women and Men
Female Male
25 40
Log-Rank P < 0.001

20 Log-Rank P = 0.01
30
15
20
10
10
5
0 0
0 2 4 6 8 10 12 14 16 18 0 2 4 6 8 10 12 14 16 18
Time to First Event (Years) Time to First Event (Years)
Lp(a) ≥50 mg/dL & hsCRP ≥2 mg/L (n = 333) Lp(a) ≥50 mg/dL & hsCRP ≥2 mg/L (n = 137)
Lp(a) <50 mg/dL & hsCRP ≥2 mg/L (n = 1,078) Lp(a) <50 mg/dL & hsCRP ≥2 mg/L (n = 710)
Lp(a) ≥50 mg/dL & hsCRP <2 mg/L (n = 204) Lp(a) ≥50 mg/dL & hsCRP <2 mg/L (n = 213)
Lp(a) <50 mg/dL & hsCRP <2 mg/L (n = 820) Lp(a) <50 mg/dL & hsCRP <2 mg/L (n = 1,140)
Zhang, W. et al. J Am Coll Cardiol. 2021;78(11):1083–1094.
Kaplan-Meier curves of the cumulative incidence of cardiovascular disease events are shown. In both women and men, greater cumulative risk of cardiovascular disease
was observed only in individuals with concomitant elevation of lipoprotein(a) ($50 mg/dL) and high-sensitivity C-reactive protein ($2 mg/L).
risk and all-cause mortality, and thus may need to be needed to confirm whether the observed interaction
more closely monitored and aggressively treated. A between Lp(a) and hsCRP plays a causal role in ASCVD.
common dilemma in the assessment and manage-
ment of individuals without known ASCVD but CONCLUSIONS
elevated Lp(a) is whether to obtain additional diag-
nostic tests and/or to start preventive medical ther- In the setting of primary prevention, Lp(a)-associated
apies. Based on the data from this analysis, ASCVD risk is observed only with concomitant
measurement of subclinical inflammation with hsCRP elevation of hsCRP. Measurement of hsCRP may
may help to guide further decisions, although this further refine Lp(a)-associated ASCVD risk. In-
would need to be confirmed in trials designed to dividuals with elevated levels of both hsCRP and
specifically answer this question. Lp(a) may merit closer surveillance and more
STUDY LIMITATIONS. Baseline levels of Lp(a), hsCRP, aggressive ASCVD risk management strategies.
and LDL-C were used in the analysis, which precluded
the ability to evaluate the impact of the dynamic FUNDING SUPPORT AND AUTHOR DISCLOSURES
changes of these markers over time. Although very few
MESA is supported by contracts HHSN268201500003I, N01-HC-95159,
participants were on lipid-lowering therapy at base- N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-
line, use of lipid-lowering medications during the 95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168,
follow-up period may represent a source of con- and N01-HC-95169 from the National Heart, Lung, and Blood Insti-
tute; and by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-
founding. We evaluated lipid-lowering therapy at
001420 from National Center for Advancing Translational Sciences.
follow-up visit 5. As summarized in Supplemental The MESA Air ancillary study was supported by a grant from the US
Tables 5 and 6, the group with Lp(a) $50 mg/dL Environmental Protection Agency’s Science to Achieve Results
demonstrated a larger proportion of statin or any lipid- (STAR) program. Assistance Agreement number RD831697 awarded
by the U.S. Environmental Protection to the University of Washing-
lowering therapy compared with those with
ton. The authors have reported that they have no relationships
Lp(a) <50 mg/dL. Additionally, the group with relevant to the contents of this paper to disclose.
concomitant elevation of Lp(a) and hsCRP had the
highest proportion on lipid-lowering therapy. These ADDRESS FOR CORRESPONDENCE: Dr Michael D.
findings are consistent with the primary hypothesis Shapiro, Section on Cardiovascular Medicine,
that this is the group with highest CVD risk and thus Department of Internal Medicine, Wake Forest Uni-
had a higher percentage of patients on lipid therapy at versity Baptist Medical Center, Medical Center
follow-up. Moreover, if there is any bias, it would be Boulevard, Winston Salem, North Carolina 27157,
toward the null hypothesis and thus would not USA. E-mail: mdshapir@wakehealth.edu. OR Dr Wei
dramatically change the conclusion of the paper. Zhang, Section on Cardiovascular Medicine, Depart-
Although we consistently observed significant inter- ment of Internal Medicine, Wake Forest University
action between Lp(a) and hsCRP with or without Baptist Medical Center, Medical Center Boulevard,
including BMI in the initial model, the final models did Winston Salem, North Carolina 27157, USA. E-mail:
not include BMI as a covariate given that BMI is not weiwakeheart@gmail.com. Twitter: @endoweiwei.
included in the Pooled Cohort Equations and was not
included in prior studies on Lp(a) (22). All ASCVD PERSPECTIVES
events was used as the primary outcome measure,
because pinpointing which specific cardiovascular COMPETENCY IN MEDICAL KNOWLEDGE: Patients with
disease was more or less affected by the Lp(a)-hsCRP elevated Lp(a) and systemic inflammation face greater risk of
interaction was not the primary focus of the current ischemic events and mortality.
study. Future studies with larger sample sizes may be
needed to address this question. Finally, results from TRANSLATIONAL OUTLOOK: Future studies of
the study suggest an association among Lp(a), hsCRP, Lp(a)-lowering agents should assess risk reduction according to
and CVD events. However, given that this is an anal- baseline levels of hsCRP.
ysis of a population-based observational cohort,
future experimental and randomized studies are
REFERENCES
1. Benjamin EJ, Muntner P, Alonso A, et al. Heart randomised, double-blind, placebo-controlled, protein cholesterol in plasma, without use of the
disease and stroke statistics-2019 update: a report dose-ranging trials. Lancet. 2016;388:2239–2253. preparative ultracentrifuge. Clin Chem. 1972;18:
from the American Heart Association. Circulation. 12. Tsimikas S, Karwatowska-Prokopczuk E, 499–502.
2019;139:e56–e528. Gouni-Berthold I, et al. Lipoprotein(a) Reduction 22. Guan W, Cao J, Steffen BT, et al. Race is a
2. Patel KV, Pandey A, de Lemos JA. Conceptual in Persons with Cardiovascular Disease. N Engl J key variable in assigning lipoprotein(a) cutoff
framework for addressing residual atheroscle- Med. 2020;382:244–255. values for coronary heart disease risk assess-
rotic cardiovascular disease risk in the era of 13. Ridker PM, MacFadyen JG, Everett BM, et al. ment: the Multi-Ethnic Study of Atheroscle-
precision medicine. Circulation. 2018;137:2551– Relationship of C-reactive protein reduction to rosis. Arterioscler Thromb Vasc Biol. 2015;35:
2553. cardiovascular event reduction following treat- 996–1001.
3. Dhindsa DS, Sandesara PB, Shapiro MD, ment with canakinumab: a secondary analysis from 23. Verbeek R, Sandhu MS, Hovingh GK, et al.
Wong ND. The evolving understanding and the CANTOS randomised controlled trial. Lancet. Lipoprotein(a) improves cardiovascular risk pre-
approach to residual cardiovascular risk manage- 2018;391:319–328. diction based on established risk algorithms. J Am
ment. Front Cardiovasc Med. 2020;7:88. 14. Tardif JC, Kouz S, Waters DD, et al. Efficacy Coll Cardiol. 2017;69:1513–1515.
4. Mayr M, Gerszten R, Kiechl S. Cardiovascular and safety of low-dose colchicine after myocardial 24. Kronenberg F, Utermann G. Lipoprotein(a):
risk beyond low-density lipoprotein cholesterol. infarction. N Engl J Med. 2019;381:2497–2505. resurrected by genetics. J Intern Med. 2013;273:6–
J Am Coll Cardiol. 2018;71:633–635. 15. Ridker PM. From C-reactive protein to 30.
interleukin-6 to interleukin-1: moving upstream to 25. Dandona P, Aljada A, Chaudhuri A, Mohanty P,
5. Tsimikas S. A Test in Context: Lipoprotein(a):
identify novel targets for atheroprotection. Circ Garg R. Metabolic syndrome: a comprehensive
diagnosis, prognosis, controversies, and emerging
Res. 2016;118:145–156. perspective based on interactions between
therapies. J Am Coll Cardiol. 2017;69:692–711.
16. Grundy SM, Stone NJ, Bailey AL, et al. 2018 obesity, diabetes, and inflammation. Circulation.
6. Kamstrup PR, Tybjaerg-Hansen A, Steffensen R, AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/ 2005;111:1448–1454.
Nordestgaard BG. Genetically elevated lip- APhA/ASPC/NLA/PCNA guideline on the manage- 26. Pradhan AD, Aday AW, Rose LM, Ridker PM.
oprotein(a) and increased risk of myocardial ment of blood cholesterol: a report of the Amer- Residual inflammatory risk on treatment with
infarction. JAMA. 2009;301:2331–2339. ican College of Cardiology/American Heart pcsk9 inhibition and statin therapy. Circulation.
Association Task Force on Clinical Practice Guide- 2018;138:141–149.
7. Clarke R, Peden JF, Hopewell JC, et al. Genetic
lines. J Am Coll Cardiol. 2019;73:e285–e350.
variants associated with Lp(a) lipoprotein level 27. Bohula EA, Giugliano RP, Leiter LA, et al. In-
and coronary disease. N Engl J Med. 2009;361: 17. Arnett DK, Blumenthal RS, Albert MA, et al. flammatory and cholesterol risk in the FOURIER
2518–2528. 2019 ACC/AHA guideline on the primary pre- Trial. Circulation. 2018;138:131–140.
vention of cardiovascular disease: a report of the
8. Willeit P, Ridker PM, Nestel PJ, et al. Baseline 28. Ridker PM. How common is residual inflam-
American College of Cardiology/American Heart
and on-statin treatment lipoprotein(a) levels for matory risk? Circ Res. 2017;120:617–619.
Association Task Force on Clinical Practice
prediction of cardiovascular events: individual
Guidelines. J Am Coll Cardiol. 2019;74(10):e177– 29. Dangas G, Mehran R, Harpel PC, et al. Lip-
patient-data meta-analysis of statin outcome tri-
e232. oprotein(a) and inflammation in human coronary
als. Lancet. 2018;392:1311–1320.
atheroma: association with the severity of clinical
18. Puri R, Nissen SE, Arsenault BJ, et al. Effect of
9. Emerging Risk Factors C, Erqou S, Kaptoge S, presentation. J Am Coll Cardiol. 1998;32:2035–
C-reactive protein on lipoprotein(a)-associated
et al. Lipoprotein(a) concentration and the risk of 2042.
cardiovascular risk in optimally treated patients
coronary heart disease, stroke, and nonvascular
with high-risk vascular disease: a prespecified
mortality. JAMA. 2009;302:412–423.
secondary analysis of the ACCELERATE Trial.
10. Tsimikas S, Hall JL. Lipoprotein(a) as a po- JAMA Cardiol. 2020;5(10):1136–1143. KEY WORDS ASCVD, cardiovascular risk,
tential causal genetic risk factor of cardiovascular high-sensitivity C-reactive protein (hsCRP),
19. Bild DE, Bluemke DA, Burke GL, et al. Multi-
disease: a rationale for increased efforts to un- inflammation, lipoprotein(a), Multi-Ethic
Ethnic Study of Atherosclerosis: objectives and
derstand its pathophysiology and develop Study of Atherosclerosis (MESA)
design. Am J Epidemiol. 2002;156:871–881.
targeted therapies. J Am Coll Cardiol. 2012;60:
716–721. 20. MESA Website. Accessed July 21, 2021.
https://www.mesa-nhlbi.org
11. Viney NJ, van Capelleveen JC, Geary RS, et al. A PPE NDI X For supplemental tables and
Antisense oligonucleotides targeting apolipopro- 21. Friedewald WT, Levy RI, Fredrickson DS. Esti- figures, please see the online version of this
tein(a) in people with raised lipoprotein(a): two mation of the concentration of low-density lipo- paper.
EDITORIAL COMMENT
Lifetime Risk Estimation in

Atherosclerotic Cardiovascular Disease
Where Inflammation Meets Lipoprotein(a)*
Xavier Rossello, MD, PHD
D espite significant advances in the diagnosis

and management of atherosclerotic cardio-
vascular disease (ASCVD), the incidences
of myocardial infarction, stroke, and peripheral arte-
consensus, Lp(a) is believed to cause ASCVD through a
combination of proatherogenic, proinflammatory, and
prothrombotic mechanisms (5). In addition to Lp(a),
systemic inflammation, frequently found as a patho-
rial disease still remain unacceptably high (1). The ad- physiologic feature of metabolic syndrome (6), has
vances in identifying modifiable cardiovascular risk been also recognized as a major component of residual
factors (CVRFs) for ASCVD, including smoking, hyper- cardiovascular risk. Among the numerous inflamma-
tension, dyslipidemias, diabetes mellitus, and tory biomarkers, hsCRP is the most widely used to
obesity, have contributed to decrease ASCVD mortal- predict ASCVD outcomes.
ity (2). Likewise, the implementation of early inter-
SEE PAGE 1083
ventions to reduce lifetime exposure to CVRFs has
been key to reduce the expected burden of ASCVD. In this issue of the Journal, Zhang et al (7) describe
However, despite optimal management of traditional their use of a cohort from the Multi-Ethnic Study of
CVRFs, significant residual cardiovascular risk re- Atherosclerosis (MESA) to describe an interaction
mains to be identified and adequately addressed. between Lp(a) and hsCRP in individuals without
Many drivers of residual risk have been described, established clinical ASCVD. In this study of >4,600
including the inflammatory, prothrombotic, and participants, Lp(a) was associated with incident
metabolic pathways (3). Among them, lipoprotein(a) ASCVD events during a mean of 13.6 years of follow-
(Lp[a]) and high-sensitivity C-reactive protein up only in those with elevated hsCRP ($2 mg/L). In
(hsCRP) deserve special attention. participants without elevated hsCRP (<2 mg/L), no
Lp(a) is a low-density lipoprotein (LDL) particle significant risk of CVD events was observed, regard-
with an Apo(a) moiety covalently bound to its ApoB less of Lp(a) levels. In comparison with neither
component (4). Given its small size (diameter <70 nm), elevated Lp(a) nor elevated hsCRP, the presence of
it can freely flux across the endothelial barrier, become either elevated Lp(a) or hsCRP was not associated
retained within the arterial wall, and thus increase the with an increased risk of ASCVD events.
risk of ASCVD. Although there is not a general Lp(a) is not a young CVRF anymore. There is a body
of evidence showing a link between Lp(a) and ASCVD
events, including myocardial infarction and ischemic
stroke, in both the primary and the secondary pre-
*Editorials published in the Journal of the American College of Cardiology
reflect the views of the authors and do not necessarily represent the vention settings (8). Likewise, inflammation is an old
views of JACC or the American College of Cardiology. foe. We have already some treatments available: an
From the Centro Nacional de Investigaciones Cardiovasculares, Madrid, anti-inflammatory therapy (canakinumab) has been
Spain; Cardiology Department, Institut d’Investigació Sanitària Illes shown to reduce ASCVD events (9). Lp(a) and hsCRP
Balears, Hospital Universitari Son Espases, Palma, Spain; Facultad de
are both considered risk enhancing factors in Amer-
Medicina, Universitat de les Illes Balears, Palma, Spain; and the Medical
Statistics Department, London School of Hygiene and Tropical Medicine,
ican guidelines (10,11). The novelty of this report is the
London, United Kingdom. finding suggesting that inflammation may modulate
The author attests they are in compliance with human studies commit- Lp(a)-associated ASCVD risk and that the concomitant
tees and animal welfare regulations of the author’s institution and Food
elevation of both Lp(a) and hsCRP has a synergistic
and Drug Administration guidelines, including patient consent where
appropriate. For more information, visit the Author Center.
impact on ASCVD events. Because it is already known

1096 Rossello JACC VOL. 78, NO. 11, 2021
Lifetime Risk Estimation in ASCVD SEPTEMBER 14, 2021:1095–1096
that no pharmacologic therapies are currently lifetime risk. Instead of 10-year prediction tools,
approved to reduce Lp(a) concentrations and ASCVD lifetime risk and treatment benefit predictions, pre-
risk (PCSK-9 monoclonal antibodies reduce Lp[a], but sented in lifetime cardiovascular risk and in cardio-
they are not approved for individuals who meet LDL vascular free life years gained from (combinations of)
cholesterol targets), and other therapies are under preventive measures, respectively, might be
development (12), the immediate key question with more appropriate to estimate risk in young in-
clinical implications is whether measuring Lp(a) en- dividuals (15).
hances risk prediction at the bedside. This question Current risk prediction tools in ASCVD prevention
was partly addressed by the Bruneck study >20 years are designed to make estimations using cross-
ago (13). In study participants at intermediate risk sectional data (eg, LDL levels at the time of the
based on Framingham risk score and Reynolds outpatient visit), although we know that lifetime
risk score variables, the addition of Lp(a) to these risk exposure is what really matters. Perhaps it is time to
scores allowed reclassification of w40% of individuals incorporate into our clinical practice longitudinal
into either lower or higher risk categories, depending measurements (LDL levels at several timepoints,
on their Lp(a) levels. estimating an area under the curve for lifetime
Accumulated evidence shows that lifetime cumu- exposure), stable predictors not fluctuating over a
lative exposure to lipids has an impact on the risk of lifetime (eg, Lp[a]), inflammatory biomarkers that
ASCVD events (14). Given that w90% of circulating might influence other CVRFs (eg, hsCRP), and lifetime
Lp(a) levels are genetically determined and that risk estimation tools.
plasma levels do not fluctuate substantially around a
preset baseline over a lifetime (5), Lp(a) should be FUNDING SUPPORT AND AUTHOR DISCLOSURES
easy to incorporate into lifetime risk estimation tools.
Dr Rossello has reported that he has no relationships relevant to the
Currently, the most widely used prediction tools, contents of this paper to disclose.
such as the race- and sex-specific pooled cohort
equations and the SCORE (Systematic Coronary Risk
Estimation), make predictions of 10-year risk (15). ADDRESS FOR CORRESPONDENCE: Dr Xavier Ros-
Because age is the most important predictor of 10- sello, Centro Nacional de Investigaciones Car-
year risk, standard risk algorithms yield inaccurate diovasculares Carlos III, c/Melchor Fernández
estimates to identify younger individuals aged <50 Almagro 3, 28029, Madrid, Spain. E-mail: fjrossello@
years at very high relative risk who may have high cnic.es. Twitter: @RosselloXavier.
REFERENCES
1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart lipoprotein(a): Multi-Ethnic Study of Atheroscle- JACC focus seminar 2/4. J Am Coll Cardiol.
disease and stroke statistics–2015 update: a report rosis. J Am Coll Cardiol. 2021;78(11):1083–1094. 2021;77(12):1576–1589.
from the American Heart Association. Circulation. 8. Miksenas H, Januzzi Jr JL, Natarajan P. Lip- 13. Kronenberg F, Kronenberg MF, Kiechl S, et al.
2015;131(4):e29–e322. oprotein(a) and cardiovascular diseases. JAMA. Role of lipoprotein(a) and apolipoprotein(a)
2. Ibanez B, Fernández-Ortiz A, Fernández- 2021;326(4):352–353. phenotype in atherogenesis: prospective results
Friera L, García-Lunar I, Andrés V, Fuster V. Pro- 9. Ridker PM, Everett BM, Thuren T, et al. Anti- from the Bruneck study. Circulation. 1999;100(11):
gression of early subclinical atherosclerosis (PESA) inflammatory therapy with canakinumab for 1154–1160.
study: JACC focus seminar 7/8. J Am Coll Cardiol. atherosclerotic disease. N Engl J Med.
2021;78(2):156–179. 14. Raposeiras-Roubin S, Rosselló X, Oliva B, et al.
2017;377(12):1119–1131.
Triglycerides and residual atherosclerotic risk.
3. Rossello X, Raposeiras-Roubin S, Oliva B, et al. 10. Grundy SM, Stone NJ, Bailey AL, et al. 2018 J Am Coll Cardiol. 2021;77(24):3031–3041.
Glycated hemoglobin and subclinical atheroscle- AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/
rosis in people without diabetes. J Am Coll Cardiol. APhA/ASPC/NLA/PCNA Guideline on the manage- 15. Rossello X, Dorresteijn JA, Janssen A, et al.
2021;77(22):2777–2791. ment of blood cholesterol: executive summary: a Risk prediction tools in cardiovascular disease
prevention: a report from the ESC prevention of
4. Mehta A, Virani SS, Ayers CR, et al. Lipoprotein(a) report of the American College of Cardiology/
American Heart Association task force on clinical CVD programme led by the European Association
and family history predict cardiovascular disease
practice guidelines. J Am Coll Cardiol. of Preventive Cardiology (EAPC) in collaboration
risk. J Am Coll Cardiol. 2020;76(7):781–793.
2019;73(24):3168–3209. with the Acute Cardiovascular Care Association
5. Shah NP, Pajidipati NJ, McGarrah RW, et al. Li-
(ACCA) and the Association of Cardiovascular
poprotein (a): an update on a marker of residual 11. Arnett DK, Blumenthal RS, Albert MA, et al. Nursing and Allied Professions (ACNAP). Eur Heart
risk and associated clinical manifestations. Am J 2019 ACC/AHA guideline on the primary pre- Journal Acute Cardiovasc Care. 2020;9(5):522–
Cardiol. 2020;126:94–102. vention of cardiovascular disease: a report of 532.
6. Rossello X, Fuster V, Oliva B, et al. Association the American College of Cardiology/American
between body size phenotypes and subclinical Heart Association task force on clinical practice
atherosclerosis. J Clin Endocrinol Metab. guidelines. J Am Coll Cardiol. 2019;74:e177–
KEY WORDS ASCVD, cardiovascular risk, high-
2020;105(12):3734–3744. e232.
sensitivity C-reactive protein (hsCRP),
7. Zhang W, Speiser JL, Ye F, et al. High-sensitivity 12. Tsimikas S, Moriarty PM, Stroes ES. Emerging inflammation, lipoprotein(a), Multi-Ethic Study of
C-reactive protein modifies the cardiovascular risk of RNA therapeutics to lower blood levels of Lp(a): Atherosclerosis (MESA)
ª 2021 THE AUTHORS. PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN
COLLEGE OF CARDIOLOGY FOUNDATION. THIS IS AN OPEN ACCESS ARTICLE UNDER
THE CC BY LICENSE (http://creativecommons.org/licenses/by/4.0/).
Phenotypic Expression and Outcomes in

Individuals With Rare Genetic Variants of
Hypertrophic Cardiomyopathy
Antonio de Marvao, MBCHB, PHD,a Kathryn A. McGurk, PHD,b Sean L. Zheng, BMBCH,b,c Marjola Thanaj, PHD,a
Wenjia Bai, PHD,d,e Jinming Duan, PHD,a,d,f Carlo Biffi, PHD,a,d Francesco Mazzarotto, PHD,b,c,g,h Ben Statton, MSC,a
Timothy J.W. Dawes, MB BCHIR, PHD,a,b Nicolò Savioli, PHD,a Brian P. Halliday, MBCHB, PHD,b,c Xiao Xu, PHD,b,c
Rachel J. Buchan, MSC,b,c A. John Baksi, MBBS, PHD,b,c Marina Quinlan, MSC,a Paweł Tokarczuk, PHD,a
Upasana Tayal, BMBCH, PHD,b,c Catherine Francis, BMBCH, PHD,b,c Nicola Whiffin, PHD,b,c,i
Pantazis I. Theotokis, MSC,a Xiaolei Zhang, PHD,b Mikyung Jang, PHD,b,c Alaine Berry, MSC,a Antonis Pantazis, MD,b,c
Paul J.R. Barton, PHD,a,b,c Daniel Rueckert, PHD,d,j Sanjay K. Prasad, MD,b,c Roddy Walsh, PHD,k Carolyn Y. Ho, MD,l
Stuart A. Cook, MBBS, PHD,a,b,c,m,n James S. Ware, MB BCHIR, PHD,a,b,c,* Declan P. O’Regan, MBBS, PHDa,*
ABSTRACT
BACKGROUND Hypertrophic cardiomyopathy (HCM) is caused by rare variants in sarcomere-encoding genes, but little
is known about the clinical significance of these variants in the general population.
OBJECTIVES The goal of this study was to compare lifetime outcomes and cardiovascular phenotypes according to the
presence of rare variants in sarcomere-encoding genes among middle-aged adults.
METHODS This study analyzed whole exome sequencing and cardiac magnetic resonance imaging in UK Biobank
participants stratified according to sarcomere-encoding variant status.
RESULTS The prevalence of rare variants (allele frequency <0.00004) in HCM-associated sarcomere-encoding genes in
200,584 participants was 2.9% (n ¼ 5,712; 1 in 35), and the prevalence of variants pathogenic or likely pathogenic for HCM
(SARC-HCM-P/LP) was 0.25% (n ¼ 493; 1 in 407). SARC-HCM-P/LP variants were associated with an increased risk of death
or major adverse cardiac events compared with controls (hazard ratio: 1.69; 95% confidence interval [CI]: 1.38-2.07;
P < 0.001), mainly due to heart failure endpoints (hazard ratio: 4.23; 95% CI: 3.07-5.83; P < 0.001). In 21,322 participants
with both cardiac magnetic resonance imaging and whole exome sequencing, SARC-HCM-P/LP variants were associated
with an asymmetric increase in left ventricular maximum wall thickness (10.9 2.7 mm vs 9.4 1.6 mm; P < 0.001), but
hypertrophy ($13 mm) was only present in 18.4% (n ¼ 9 of 49; 95% CI: 9%-32%). SARC-HCM-P/LP variants were still
associated with heart failure after adjustment for wall thickness (hazard ratio: 6.74; 95% CI: 2.43-18.7; P < 0.001).
CONCLUSIONS In this population of middle-aged adults, SARC-HCM-P/LP variants have low aggregate penetrance
for overt HCM but are associated with an increased risk of adverse cardiovascular outcomes and an attenuated
cardiomyopathic phenotype. Although absolute event rates are low, identification of these variants may enhance
risk stratification beyond familial disease. (J Am Coll Cardiol 2021;78:1097–1110) © 2021 The Authors. Published by
Elsevier on behalf of the American College of Cardiology Foundation. This is an open access article under the CC BY
license (http://creativecommons.org/licenses/by/4.0/).
From the aMRC London Institute of Medical Sciences, Imperial College London, Hammersmith Hospital Campus, London, United
Kingdom; bNational Heart and Lung Institute, Imperial College London, London, United Kingdom; cCardiovascular Research
audio summary by
Centre at Royal Brompton and Harefield Hospitals, London, United Kingdom; dBiomedical Image Analysis Group, Department of
Editor-in-Chief
Computing, Imperial College London, London, United Kingdom; eDepartment of Brain Sciences, Imperial College London, Lon-
Dr. Valentin Fuster on
don, United Kingdom; fSchool of Computer Science, University of Birmingham, Birmingham, United Kingdom; gDepartment of
JACC.org.
Experimental and Clinical Medicine, University of Florence, Florence, Italy; hCardiomyopathy Unit, Careggi University Hospital,
Florence, Italy; iWellcome Centre for Human Genetics, University of Oxford, Oxford, United Kingdom; jFaculty of Informatics and
Medicine, Klinikum Rechts der Isar, TU Munich, Munich, Germany; kDepartment of Experimental Cardiology, Amsterdam UMC,
AMC Heart Centre, Amsterdam, the Netherlands; lCardiovascular Division, Brigham and Women’s Hospital, Boston, Massachu-
setts, USA; mNational Heart Centre Singapore, Singapore; and the nDuke-NUS Graduate Medical School, Singapore. *Drs Ware and
O’Regan are joint senior authors.
ISSN 0735-1097 https://doi.org/10.1016/j.jacc.2021.07.017

1098 de Marvao et al. JACC VOL. 78, NO. 11, 2021
Impact of HCM-Associated Variants on the Adult Population SEPTEMBER 14, 2021:1097–1110
H
ABBREVIATIONS ypertrophic cardiomyopathy 2 groups of variants found in 8 genes with definitive
AND ACRONYMS (HCM) is characterized by clinical evidence for HCM (3): sarcomeric variants P/LP spe-
and genetic heterogeneity, incom- cifically for HCM (SARC-HCM-P/LP) and rare sarco-
ACMG = American College of
Medical Genetics and Genomics
plete and age-dependent penetrance, and meric variants of indeterminate significance (SARC-
variable expressivity (1). Most individuals IND) with the potential to cause HCM, dilated
CMR = cardiac magnetic
resonance imaging with HCM have a normal life expectancy cardiomyopathy (DCM), or have little impact on car-
DCM = dilated cardiomyopathy but are at increased risk of adverse outcomes diomyopathy risk. Using high-precision, deep
HCM = hypertrophic
such as heart failure, atrial fibrillation, learning phenotyping of cardiac magnetic resonance
cardiomyopathy stroke, or sudden cardiac death (2). imaging (CMR), we also characterized phenotypic
LVH = left ventricular SEE PAGE 1111
manifestations and estimated the prevalence of
hypertrophy penetrant disease. Lastly, we determined the preva-
MACE = major adverse
A recent expert-led assessment of the lence and genetic yield of sequencing in unexplained
cardiovascular events validity of reported gene associations with left ventricular hypertrophy (LVH) among this adult
P/LP = pathogenic or likely HCM identified 8 sarcomeric genes with population.
pathogenic definitive evidence for disease causation (3),
SARC-HCM-P/LP = including MYH7 and MYBPC3, that account METHODS
pathogenic or likely pathogenic
for the majority of genetically explained
variants for hypertrophic
disease (4). The American College of Medical STUDY COHORTS. The UKBB recruited 500,000 par-
cardiomyopathy in sarcomere-
encoding genes Genetics and Genomics (ACMG) includes this ticipants aged 40 to 69 years across the United
SARC-IND = indeterminate set of genes among those for which specific Kingdom between 2006 and 2010 (National Research
variants in hypertrophic
variants are known to be causative of disease Ethics Service, 11/NW/0382) (9). This study was
cardiomyopathy–associated
phenotypes and are clinically actionable (5). conducted under terms of access approval number
sarcomere-encoding genes
(rare variants that do not meet The ACMG recommends that these genes 40616. In each case, written informed consent was
criteria for pathogenic/likely
be analyzed whenever clinical exome provided. The results are reported in accordance with
pathogenic annotation)
sequencing is undertaken and that patho- the Strengthening the Reporting of Observational
SARC-NEG = genotype Studies in Epidemiology guidelines (10) and the
negative/no rare variants in
genic or likely pathogenic (P/LP) variants
sarcomere-encoding genes should be proactively reported to patients checklist provided in Supplemental Table 1.
UKBB = UK Biobank as secondary findings. With increasing CARDIAC PHENOTYPING USING MACHINE LEARNING.
WES = whole exome

availability of whole exome sequencing Participants in the imaging substudy were randomly
sequencing (WES), both in wider clinical settings and as invited, and the response rates with exclusion criteria
WT = wall thickness direct-to-consumer asymptomatic testing, have been previously reported (11). Each underwent
this raises questions regarding potential CMR at 1.5-T (12). Segmentation of the cine images
benefit as well as downstream risks (6). Evidence is was performed by using a deep learning neural
not currently available that would allow a critical network algorithm developed in-house and opti-
evaluation of genomic screening at the population mized on the UKBB cohort. The performance of image
level (7). Specifically, it is unclear what risk annotation using this algorithm is equivalent to a
cardiomyopathy-associated variants confer in the consensus of expert human readers and achieves
general adult population and their phenotypic subpixel accuracy for cardiac segmentation (13).
expression outside families with penetrant disease. Myocardial wall thickness (WT) was measured along
Current evidence is based on aggregating data from radial line segments connecting the endocardial and
small and often underpowered studies (8), using epicardial surfaces perpendicular to the myocardial
different variant classifications and relying on center-line and excluding trabeculae (Figure 1), an
inconsistent phenotyping. approach that also exceeds the reliability of human
Here, we sought to determine the population experts (14). Chamber volumes and mass were
prevalence of rare sarcomeric variants in a prospec- calculated from the segmentations according to
tively recruited cohort of >200,000 middle-age standard post-processing guidelines (15). Myocardial
participants drawn from the UK Biobank (UKBB) and strain analysis was performed by using nonrigid
to assess lifetime risk of adverse events. We analyzed free-form deformation image registration (16,17).
Manuscript received May 7, 2021; revised manuscript received July 1, 2021, accepted July 6, 2021.
JACC VOL. 78, NO. 11, 2021 de Marvao et al. 1099
SEPTEMBER 14, 2021:1097–1110 Impact of HCM-Associated Variants on the Adult Population
F I G U R E 1 Cardiac Image Analysis in the UK Biobank
A Cardiac Image Analysis
3-Dimensional cardiac segmentation
Wall Thickness/mm
100 10
Radial Strain %
80 8
60
40 6
20 4
0 2
−20 3-Dimensional
−40 wall thickness model 0
Cardiac strain analysis
B Distribution of Maximum Left Ventricular Wall Thickness (n = 21,272)
1,500
1,000
Count
500
5 10 15 20
Maximum LV Wall Thickness (mm)
Male Female
(A) Machine learning segmentation of the heart from cardiac magnetic resonance imaging (right atrium: light blue; right ventricle: dark blue; left atrium: yellow; left
ventricle: red; left ventricular myocardium: green). Motion analysis was used to derive strain and strain rates (radial strain in diastole and systole shown). Regional
analysis of left ventricular (LV) wall thickness was performed by using 3-dimensional modeling. Mean wall thickness for 21,322 UK Biobank participants is mapped onto
the LV surface; the right ventricle is shown as a mesh. (B) Histogram of maximum LV wall thickness according to sex.
Trabecular traits were quantified by using fractal Appendix). Individuals harboring variants that did
dimension analysis in which a higher value indicates not fit into these 3 categories were removed from
more complex trabeculation (18). Parametric analyses, including those with rare variants in genes
3-dimensional analysis of the geometry of the left associated with HCM genocopies, those with
ventricle was performed to map regional patterns intermediate-frequency variants (0.00004 < allele
of remodeling and quantify the association with frequency <0.001), those with variant classes not
genetic and environmental predictors (16,19-21). robustly established as disease-causing (eg, trun-
Further details on phenotyping are given in the cating variants in MYH7), and those identified as P/LP
Supplemental Appendix. for DCM (n ¼ 7) (3). Details on the variant curation
pipelines are presented in the Supplemental
SEQUENCING AND VARIANT CATEGORIZATION
Appendix.
PIPELINE. UKBB participants underwent WES as
previously described (22). They were divided into 3 OUTCOME MEASURES. The effect of genotype strata
genetic strata. Individuals were classified as genotype on clinical outcomes was assessed by using lifetime
negative (SARC-NEG) if they had no rare protein- risk. The UKBB reports the date of first occurrence of
altering genetic variation (minor allele frequency a diagnosis, identified from self-reporting, primary
<0.001 in the UKBB and the Genome Aggregation care, hospital in-patient, and death register records.
Database) (23) in any of 25 genes that may cause or This permitted the identification of events preceding
mimic HCM. These 25 genes represent an inclusive recruitment to the UKBB. The primary clinical
list of 8 sarcomere-encoding genes with definitive outcome was a composite of all-cause mortality or
evidence of an association with HCM, 3 moderate- major adverse cardiovascular events (MACE) defined
evidence sarcomere-encoding genes, and 14 genes as a diagnostic code for heart failure (including car-
associated with syndromic phenotypes that can diomyopathy), arrhythmia, stroke, or cardiac arrest
include LVH (3). This SARC-NEG group was compared events. Secondary clinical outcomes were the indi-
with individuals with rare variants in 8 sarcomere- vidual components of the primary clinical outcome. A
encoding genes definitively associated with HCM full list of endpoint definitions and data fields used
(MYBPC3, MYH7, MYL2, MYL3, TNNI3, TNNT2, TPM1, from the UKBB database is presented in the
and ACTC1). Analysis was restricted to robustly Supplemental Appendix and Supplemental Table 2.
disease-associated variant classes for each gene and STATISTICAL ANALYSIS. Statistical analysis was
to variants sufficiently rare to cause penetrant performed with R version 3.6.0 (R Foundation for
disease (filtering allele frequency <0.00004) (24). Statistical Computing) and RStudio Server version
Variants were classified as pathogenic/likely path- 1.043, unless otherwise stated. Variables are
ogenic (SARC-HCM-P/LP) if reported as P/LP for HCM expressed as percentages if categorical, mean SD if
in ClinVar and confirmed by manual review (n ¼ 81), continuous and normal, and median (interquartile
or if annotated as P/LP according to ACMG criteria, range) if continuous and non-normal. Baseline
using the semi-automated CardioClassifier decision anthropometric data were compared by using
support tool (24) (n ¼ 19); the curation is depicted in Kruskal-Wallis tests and, if differences were identi-
Supplemental Figure 1. Other variants that were fied, a Wilcoxon test was used for pairwise compari-
consistent with known disease mechanisms and suf- sons with Benjamini-Hochberg adjustment for
ficiently rare, but not recorded in ClinVar and without multiple testing. Imaging parameters in 2 or more
other computationally available data to robustly groups were compared by using analysis of covari-
classify as P/LP, were defined as indeterminate ance, adjusted for relevant clinical covariates. When
sarcomeric variants (SARC-IND) (25,26). Despite being differences were significant, a Tukey post hoc test
curated for HCM, given that 5 of the 8 genes (MYH7, was applied for pairwise multiple comparisons.
ACTC1, TNNT2, TNNI3, and TPM1) that have defini- Three-dimensional phenotypic regression modeling
tive evidence for HCM have also been implicated in applied threshold-free cluster enhancement and
DCM (27), it is possible that some SARC-IND variants permutation testing to derive the P values associated
have the potential to cause DCM. The SARC-IND with each regression coefficient following adjustment
strata differs from those classically termed variants to control the false discovery rate, as previously
of unknown significance as it likely contains addi- described (16,20).
tional variants that would be reported as P/LP if Clinical outcomes were analyzed in participants
subject to full manual curation, which is not feasible with WES stratified according to genotype categories.
in >5,000 individuals (further subgroup in- Cox proportional hazards were calculated for lifetime
vestigations are presented in the Supplemental risk of clinical events. For the primary outcome,
F I G U R E 2 Flowchart of UKBB Participants
200,584 participants with WES data 39,551 participants with CMR data
157,922 SARC-NEG 5,219 SARC-IND 493 SARC-HCM-P/LP 38,348 WT <13 mm 763 WT = 13-15 mm 163 WT ≥15 mm
36,950 excludeda 277 excludedb
21,322 UKBB participants with CMR and WES data
16,844 SARC-NEG 513 SARC-IND 49 SARC-HCM-P/LP
3,916 excludeda
We included 200,548 participants with whole exome sequencing (WES) in the UK Biobank (UKBB) and stratified them according to variant pathogenicity for outcome
analysis. Machine learning was also used to characterize left ventricular traits in 39,551 participants, of whom 21,322 also had sequencing. aIndividuals excluded if
carriers of: 1) rare variants in genes associated with HCM genocopies or LV phenotype; 2) intermediate frequency variants (0.00004 < AF < 0.001); 3) variant classes
not robustly established as disease causing. bCMRs excluded from WT measure due to nondiagnostic imaging, incomplete sequences, and other technical reasons.
AF ¼ allele frequency; CMR ¼ cardiac magnetic resonance imaging; HCM ¼ hypertrophic cardiomyopathy; SARC-HCM-P/LP ¼ pathogenic or likely pathogenic variants
for hypertrophic cardiomyopathy in sarcomere-encoding genes; SARC-IND ¼ indeterminate variants in hypertrophic cardiomyopathy–associated sarcomere-encoding
genes (rare variants that do not meet criteria for pathogenic/likely pathogenic annotation); SARC-NEG ¼ genotype negative; WT ¼ wall thickness.
competing risk analysis was performed by using the (0.25%) heterozygotes for 100 SARC-HCM-P/LP vari-
cause-specific survival method (28). Secondary anal- ants, including 16 compound heterozygotes of a
ysis of incident clinical events from recruitment SARC-HCM-P/LP variant and a SARC-IND variant, and
excluded individuals with events preceding enroll- 5,219 (2.6%) heterozygotes with SARC-IND variants,
ment and was performed by using Cox proportional including 112 compound heterozygotes where neither
hazards adjusted for age at recruitment. Time-to- variant met criteria to be classified as SARC-HCM-
event was censored at first event for each outcome, P/LP (Supplemental Table 3).
death, or last recorded follow-up. The relationship LEFT VENTRICULAR WALL THICKNESS IN THE
between genotype or CMR phenotypes and outcomes GENERAL POPULATION. Of the 39,274 subjects who
was assessed with multivariable Cox proportional underwent CMR analysis, 763 (1.9% [1 in 51]) had mild
hazards models and robust SE estimates. Sex was hypertrophy (WT 13-15 mm) and 163 (0.4% [1 in 241])
included as a covariate in all models. Hazard pro- had at least moderate hypertrophy (WT $15 mm).
portionality assumption was tested by using Participants with WT $13 mm were older (66.3 7.4
Schoenfeld residuals. Sex was found to be in violation years vs 63.6 7.6 years; P < 0.001), more often male
of these assumptions, and therefore a sex-stratified (88.4% vs 46.9%; P < 0.001), and with higher systolic
analysis was conducted with interaction co- blood pressure (149.5.2 18.8 mm Hg vs 138
efficients. Outcomes are reported as hazard ratios 18.2 mm Hg; P < 0.001) and body surface area (2.07
(HRs) with 95% confidence intervals (CIs) and pre- 0.21 m 2 vs 1.86 0.21 m 2; P < 0.001). There were no
sented graphically as cumulative hazards and Cox differences in terms of race. The prevalence of
proportional hazards curves. phenotypic HCM, defined as WT $15 mm in the
RESULTS absence of hypertension and valve disease, was
0.19% (n ¼ 76 [1 in 517]).
PARTICIPANTS. We analyzed WES data from 200,584 PENETRANCE AND EXPRESSIVITY OF RARE
participants and CMR from 39,551 subjects. A total of SARCOMERIC VARIANTS. Analyses were restricted
21,322 had both CMR and WES data available to UKBB participants with both CMR and WES
(Figure 2, Table 1). (n ¼ 21,322). In this subset, 16,844 were denoted as
PREVALENCE OF RARE SARCOMERIC VARIANTS. SARC-NEG, 513 as SARC-IND, and 49 as SARC-HCM-P/
There were 157,922 SARC-NEG subjects, 493 LP. CMR-derived cardiac measurements by genotype
T A B L E 1 Subject Characteristics and CMR-Derived Cardiac Measurements According to Genotype
SARC-NEG SARC-IND SARC-HCM-P/LP SARC-NEG vs SARC-NEG vs SARC-IND vs

(n ¼ 157,922) (n ¼ 5,219) (n ¼ 493) SARC-HCM-P/LP SARC-IND SARC-HCM-P/LP
Female 86,710 (54.9) 2,866 (54.9) 276 (56.0)

Age, y 56.5 8.1 56.3 8.2 56.2 8.1
White 150,403 (95.2) 4,756 (91.1) 461 (93.5) 0.11 <0.001 0.11
SBP, mm Hg 139.7 19.6 139.5 19.8 139.2 20.2
DBP, mm Hg 82.2 10.7 82.2 10.5 81.4 10.7
BSA, m2 1.9 0.22 1.9 0.22 1.9 0.22
eGFR, mL/min/1.73 m2 90.7 13.4 91.1 13.3 90.1 13.3
Hypercholesterolemia 29,137 (18.5) 950 (18.2) 94 (19.1)
Hypertension 52,356 (33.2) 1,723 (33.0) 168 (34.1)
Diabetes 8,429 (5.3) 309 (5.9) 25 (5.1)
Aortic valve disease 1,567 (1.0) 37 (0.7) 8 (1.6)
Known HCM 109 (0.07) 9 (0.17) 18 (3.65) <0.001 0.006 <0.001
Known DCM 265 (0.17) 9 (0.17) 1 (0.2)
On medication for cholesterol, 18,537 (11.7) 599 (11.5) 64 (13.0)
blood pressure, diabetes
SARC-NEG SARC-IND SARC-HCM-P/LP

(n ¼ 16,844) (n ¼ 513) (n ¼ 49)
LVEDV, mL 147.9 33.5 150.1 34.6 139.4 30.3 0.18 0.005 0.02
LVESV, mL 60.4 19 61.4 19.7 54.7 15.9 0.07 0.04 0.01
LVEF, % 59.5 6.1 59.5 6.1 61.1 5.9
LVM, g 86 22.1 87.1 23.6 86.2 26.6 0.43 0.001 0.99
LV maximum WT, mm 9.4 1.6 9.5 1.7 10.9 2.7 <0.001 0.008 <0.001
LV concentricity, g/mL 0.58 0.09 0.58 0.09 0.62 0.12 0.001 0.95 0.002
LV global longitudinal strain, % –18.5 2.8 –18.4 2.8 –19 2.5
LV global radial strain, % 45 8.3 44.8 8.1 46.7 9.3
LV global circumferential strain, % –22.3 3.4 –22.2 3.3 –22.8 3.4
LV longitudinal PDSR 1.7 0.6 1.6 0.6 1.7 0.6
LV radial PDSR –5.7 2.0 –5.7 1.9 –5.9 2.0
LAEDV, mL 72.7 23.1 73.6 27.8 81.8 26.8 0.004 0.29 0.02
LAESV, mL 29.2 15.3 30.2 20.8 34.4 15.4 0.03 0.18 0.12
LAEF, % 61.3 9.2 60.9 9.8 59.2 8.2
RVEDV, mL 156.6 37.4 157.7 37.6 138.6 32.2 <0.001 0.08 <0.001
RVESV, mL 67.6 21.3 68.1 21 55 15.3 <0.001 0.26 <0.001
RVEF, % 57.3 6.2 57.2 5.9 60.6 5.4 <0.001 0.99 <0.001
RAEDV, mL 86 27.7 89.1 32.9 80.6 27.9 0.41 0.007 0.08
RAESV, mL 46.1 19.1 47.9 24.4 42.2 19.4 0.33 0.03 0.08
RAEF, % 47.1 9.4 47.5 9.8 48.9 9.9
Mean apical FD 1.14 0.04 1.14 0.04 1.16 0.05 0.006 0.88 0.02
Mean basal FD 1.19 0.03 1.19 0.03 1.2 0.04
Mean global FD 1.17 0.03 1.17 0.03 1.18 0.03 0.002 0.96 0.003
Values are n (%) or mean SD. Table includes data for genotype-negative individuals (SARC-NEG), individuals with pathogenic or likely pathogenic sarcomeric variants (SARC-
HCM-P/LP), or those with other rare indeterminate sarcomeric variants (SARC-IND). Bold P values are statistically significant. Kruskal-Wallis tests were conducted for each
variable to determine whether differences in participants’ characteristics between genotype group were significant; if so, a Wilcoxon test was used to perform pairwise
comparisons between groups, with Benjamini-Hochberg adjustment for multiple testing, and those adjusted P values are shown. For the cardiac magnetic resonance imaging
(CMR)-derived cardiac parameters, analysis was adjusted for age, sex, race, systolic blood pressure (SBP), and body surface area (BSA) using an analysis of covariance; when
differences between genotype groups were significant, a Tukey post hoc test was applied for pairwise multiple comparisons, and those P values are shown.
concentricity ¼ (left ventricular mass/left ventricular end-diastolic volume); DBP ¼ diastolic blood pressure; DCM ¼ dilated cardiomyopathy; eGFR ¼ estimated glomerular
filtration rate; FD ¼ fractal dimension; HCM ¼ hypertrophic cardiomyopathy; LAEDV ¼ left atrial end-diastolic volume; LAEF ¼ left atrial ejection fraction; LAESV ¼ left atrial
end-systolic volume; LV ¼ left ventricular; LVEDV ¼ left ventricular end-diastolic volume; LVEF ¼ left ventricular ejection fraction; LVESV ¼ left ventricular end-systolic volume;
LVM ¼ left ventricular mass; PDSR ¼ peak diastolic strain rate; RAEF ¼ right atrial ejection fraction; RVEDV ¼ right ventricular end-diastolic volume; RVESV ¼ right ventricular
end-systolic volume; WT ¼ wall thickness.
are summarized in Table 1. Wall thickness was greater showed evidence of concentric remodeling, smaller
in SARC-HCM-P/LP versus SARC-NEG (Figure 3) right ventricular volumes, higher left atrial
and marginally greater in SARC-IND after adjustment volume, and increased trabeculation. Independent,
for relevant clinical variables. Compared with 3-dimensional analysis of cardiac geometry showed
SARC-NEG, those harboring SARC-HCM-P/LP variants that SARC-IND variants were positively associated
F I G U R E 3 Relationship Between Rare Variants in HCM-Associated Genes and WT
A ***
****
ns
20
Maximum LV Wall Thickness (mm)
15 mm
15
13 mm
10
SARC-NEG (n = 16,844) SARC-IND (n = 513) SARC-HCM-P/LP (n = 49)

Genotype
0.2
Beta Coefficients
0.1
0
SARC-IND
−.01
−.02
C
SARC-HCM-P/LP
(A) Dot and boxplots of maximum wall thickness according to genotype. (B and C) 3-dimensional modeling of LV geometry with standardized
beta-coefficients showing the strength of association between genotype and regional WT. Contour lines indicate significant regions
(P < 0.05) after correction for multiple testing. LV projections are septal (left) and anterior (right). ***P # 0.001; ****P # 0.0001. ns ¼ not
significant; other abbreviations as in Figures 1 and 2.
with an asymmetric increase in WT (n ¼ 508; mean 95 (19.3%) primary clinical outcome events (all-cause
b ¼ 0.08; significant area ¼ 70.1%), predominantly mortality or MACE), and by age 70 years there were
across the mid-basal anterior septum in continuity 14,168 (cumulative incidence: 12%; 95% CI: 12%-12%),
with the anterior wall. There was a much stronger 461 (cumulative incidence: 12%; 95% CI: 11%-13%),
positive association between SARC-HCM-P/LP vari- and 80 events (cumulative incidence: 21%; 95%
ants and increased WT (n ¼ 48; mean b ¼ 0.31; signifi- CI: 16%-25%), respectively.
cant area ¼ 47%) in an asymmetric pattern involving Examining lifetime risks, SARC-HCM-P/LP variants
most of the septum, anterior wall, and apex. were associated with an increased risk of death or
In the 49 subjects harboring SARC-HCM-P/LP vari- MACE (HR: 1.69; 95% CI: 1.38-2.07; P < 0.001), heart
ants, LVH $13 mm was present in 9, corresponding to a failure (HR: 4.23; 95% CI: 3.07-5.83; P < 0.001), and
penetrance of 18.4% (95% CI: 9%-32%) for this arrhythmia (HR: 1.59; 95% CI: 1.22-2.09; P < 0.001)
phenotype. In 3 of these individuals, LVH was $15 mm (Figure 4, Supplemental Table 5). There was no dif-
(6.1%; 95% CI: 1.3%-16.8%). Of the 513 with SARC-IND ference in risk in any of the primary or secondary
variants, 15 individuals (2.9%; 95% CI: 1.6%-4.8%) clinical outcomes when comparing SARC-IND and
had WT $13 mm. In 390 of these 562 individuals with a SARC-NEG. Sex was independently associated with
SARC-IND or SARC-HCM-P/LP variant, there was no all clinical outcomes, with men having an increased
concomitant hypertension or valve disease (69.4%). risk. Although male subjects have a higher overall
Among these 390 individuals, the penetrance of risk of adverse outcomes (HR: 1.76; 95% CI: 1.71-1.81;
LVH $13 mm was 10.8% for SARC-HCM-P/LP (4 of 37) P < 0.001), the incremental genetic risk from SARC-
and 0.57% for SARC-IND (2 of 353). Details of variants HCM-P/LP mutations compared with SARC-NEG is
found in these 6 individuals with otherwise greater in female subjects (HR for female subjects:
unexplained LVH are shown in Supplemental Table 4. 2.24 [95% CI: 1.71-2.94; P < 0.001]; HR for male sub-
Among the 173 individuals with a rare sarcomere- jects: 1.31 [95% CI: 0.97-1.76; P ¼ 0.08]; HR for sex*-
encoding variant and hypertension or valve disease, SARC-HCM-P/LP interaction: 1.72 [95% CI: 1.15-2.58;
the penetrance of LVH was 41.7% for SARC-HCM-P/LP P ¼ 0.009]) (Supplemental Table 6, Supplemental
(5 of 12) and 8.1% for SARC-IND (13 of 161). Only 1 in- Figure 2). Sensitivity analyses of lifetime risk of
dividual of the 39 with a wall thickness $15 mm (2.6%), death or MACE, excluding participants with any car-
not accounted for by hypertension or valve disease, diomyopathy and excluding cardiomyopathy events
harbored a SARC-HCM-P/LP variant. There was no as an outcome, yielded concordant results to the
difference in the prevalence of left ventricular ejection primary analysis comparing SARC-HCM-P/LP with
fraction <50% (P ¼ 0.12) between subjects harboring SARC-NEG (excluding participants with cardiomyop-
SARC-HCM-P/LP variants (1 of 49 [2%]; 95% CI: 0.0%- athy HR: 1.29 [95% CI: 1.02-1.63; P ¼ 0.03]; excluding
10.9%), subjects with a SARC-IND variant (29 of 513 cardiomyopathy diagnosis from the HF composite
[5.7%]; 95% CI: 3.9%-8.0%), and those identified as outcome HR: 1.60 [95% CI: 1.25-2.04; P < 0.001])
SARC-NEG (888 of 16,844 [5.3%]; 95% CI: 4.9%-5.6%). (Supplemental Table 7), confirming that previous
The prevalence of left ventricular dilatation (defined findings were not solely driven by individuals with
by using the UKBB-derived normal ranges of left ven- known disease, nor only by a diagnostic label of
tricular end-diastolic volume [abnormal high >232 mL cardiomyopathy as an endpoint. Examining incident
in male subjects and >175 mL in female subjects]) (29) risks, primary clinical events (HR: 1.57; 95% CI: 1.2-
was also not different in SARC-NEG individuals (530 of 2.04; P < 0.001), including heart failure (HR: 3.15;
16,844 [3.15%]; 95% CI: 2.9%-3.4%) and those with 95% CI: 2.03-4.89; P < 0.001), were increased in
SARC-IND (16 of 513 [3.12%]; 95% CI: 1.8%-5.0%) or SARC-HCM-P/LP but not in SARC-IND (Supplemental
SARC-HCM-P/LP (0 of 49 [0%]; 95% CI: 0.0%-7.3%) Table 8, Supplemental Figure 3).
variants. In the 39,551 individuals who underwent CMR, the
CLINICAL OUTCOMES ASSOCIATED WITH RARE number of primary clinical events was 2,568 (6.7%),
SARCOMERIC VARIANTS. Clinical outcomes for 112 (14.7%), and 34 (20.9%) in the WT <13 mm,
163,634 participants were analyzed. Median age at 13-15 mm, and $15 mm subgroups, respectively.
recruitment was 58 years (interquartile range: 50-63 Increased WT was associated with an increased risk of
years), and participants were followed up for a me- death or MACE (per millimeter increase HR: 1.15;
dian of 10.8 years (interquartile range: 9.9-11.6 years) 95% CI: 1.11-1.20; P < 0.001) and each individual
with a total of 19,504 primary clinical events endpoint (Supplemental Table 9, Supplemental
reported. Among SARC-NEG (n ¼ 157,922), SARC-IND Figure 4), with similar findings in the imaging and
(n ¼ 5,219), and SARC-HCM-P/LP (n ¼ 493) in- WES subgroup (n ¼ 17,447) (Supplemental Table 10,
dividuals, there were 18,793 (11.9%), 616 (11.8%), and Supplemental Figure 5). Increased trabecular
F I G U R E 4 Outcomes Stratified According to Variant Pathogenicity
A Death and MACE
0.40
1.0
0.35
0.9 0.30
0.25
Cumulative Hazards of Event
0.8 0.20
0.15
0.7 0.10
0.05
0.6
0.00
0.5 0 10 20 30 40 50 60 70 80
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80
Years
Number at risk
Genotype
SARC-NEG 157,922 157,888 157,845 157,702 157,260 155,565 123,305 70,044 7,248
SARC-IND 5,219 5,219 5,216 5,213 5,199 5,140 4,005 2,264 208
SARC-HCM-P/LP 493 493 493 492 491 480 364 200 17
0 10 20 30 40 50 60 70 80
Genotype SARC-NEG SARC-IND SARC-HCM-P/LP
Cumulative hazard curves with zoomed plots for lifetime risk of: (A) death and major adverse cardiovascular events (MACE), consisting of heart failure, arrhythmia,
stroke, and cardiac arrest events, or (B) heart failure, stratified according to genotype, consisting of SARC-NEG, SARC-IND, or SARC-HCM-P/LP. (C) Forest plot of
comparative lifetime risk of clinical endpoints (Cox proportional hazards models adjusted for sex) according to genotype. Sex refers to male subjects compared with
female subjects. Abbreviations as in Figure 2.
complexity, as measured by fractal dimension anal- end-diastolic volume (HR: 8.05; 95% CI: 2.46-26.3;
ysis, was associated with adverse clinical outcomes P < 0.001) (Supplemental Tables 13 and 14).
(HR per 0.1 unit increase in mean global fractal
dimension: 1.04; 95% CI: 1.01-1.08; P ¼ 0.01) after DISCUSSION
adjustment for systolic blood pressure and WT
(Supplemental Table 11, Supplemental Figure 6). In this study of >200,000 adults, we found the preva-
Among 17,406 individuals in the 3 genotype strata lence of rare variants in HCM-associated sarcomere-
with imaging and WES data, there were 1,177 (7.0%), 44 encoding genes to be 2.9%, with the prevalence of SARC-
(8.3%), and 87 (16.0%) primary clinical events in SARC- HCM-P/LP variants conservatively measured at 0.25%,
NEG, SARC-IND, and SARC-HCM-P/LP, respectively. both less frequent than reported in smaller cohorts (30),
The risk of heart failure was higher with SARC-HCM-P/ prior to contemporary variant classification guidelines
LP variants despite adjustment for WT (HR: 6.74; (25) and a consensus emerging on genes with definitive
95% CI: 2.43-18.7; P < 0.001) (Supplemental Table 12) evidence for disease causation in HCM (3). By examining
and also when accounting for WT, left atrial volume, lifetime risks alongside precision cardiac phenotyping,
left ventricular ejection fraction, and left ventricular we provide a critical evaluation of large-scale genomic
B Heart Failure
0.20
1.0
0.9 0.15
Cumulative Hazards of Event
0.8 0.10
0.7 0.05
0.6
0.00
0.5 0 10 20 30 40 50 60 70 80
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80
Years
Number at risk
Genotype
SARC-NEG 157,922 157,918 157,917 157,914 157,888 157,572 127,633 76,261 8,537
SARC-IND 5,219 5,219 5,219 5,218 5,216 5,203 4,125 2,468 253
SARC-HCM-P/LP 493 493 493 492 492 483 384 222 19
0 10 20 30 40 50 60 70 80
Genotype SARC-NEG SARC-IND SARC-HCM-P/LP
screening for sarcomeric variants outside of familial dis- SARC-HCM-P/LP variants (32). SARC-HCM-P/LP vari-
ease (Central Illustration). ants were associated with a greater increase in life-
We found that SARC-HCM-P/LP variants were time risk of death and MACE in female subjects than
associated with an increased lifetime risk of adverse in male subjects. This supports findings in patients
cardiovascular events, predominantly due to heart with HCM, in whom, despite apparent reduced dis-
failure and arrhythmia, which was partially inde- ease penetrance (33), women have lower survival
pendent of left ventricular wall thickness and not regardless of genotype (34,35). In contrast, SARC-IND
restricted to individuals with cardiomyopathy or variants (which include a mixture of variants with
reaching cardiomyopathy endpoints. The fact that potential to cause HCM, DCM, or have little impact on
heart failure risk is only partially explained by cardiomyopathy risk) express a minimal phenotype in
observable phenotypic expressivity is consistent with aggregate and have a generally benign clinical course.
the profibrotic state that precedes the development of Recent study shows that the same genetic pathways
LVH in those with HCM variants (31). This finding may lead to distinct disorders through opposing ge-
highlights the potential of genomic testing for sarco- netic effects in the general population (36).
meric variants to modify cardiovascular risk assess- Adults harboring SARC-HCM-P/LP variants had an
ment even in the absence of image phenotyping. Such attenuated HCM-related phenotype characterized by
variants are also associated with greater risk in HCM increased wall thickness in an asymmetric pattern
patient cohorts, with adverse outcomes reported to involving the anteroseptum and apex, concentric
be twice as likely in patients with HCM harboring ventricular remodeling, increased left atrial volume,
C
1.76 (1.71-1.81)
Death or MACE 1.69 (1.38-2.07)
1.00 (0.92-1.09)
1.73 (1.64-1.82)
Death 1.45 (0.99-2.13)
0.9 (0.77-1.05)
1.81 (1.75-1.87)
MACE 1.83 (1.47-2.27)
1.00 (0.92-1.10)
Endpoint
2.30 (2.13-2.47)
Heart Failure 4.23 (3.07-5.83)
1.08 (0.89-1.31)
1.80 (1.73-1.87)
Arrhythmia 1.59 (1.22-2.09)
1.03 (0.93-1.14)
1.80 (1.68-1.92)
Stroke 1.54 (0.96-2.48)
0.86 (0.71-1.05)
2.71 (2.25-3.26)
Cardiac Arrest 2.52 (0.94-6.74)
0.65 (0.36-1.19)
0.5 1.0 2.0 4.0

Hazard Ratio
Sex SARC-HCM-P/LP vs SARC-NEG SARC-IND vs SARC-NEG
C E NT R AL IL L U STR AT IO N Outcomes and Expression of Rare Variants in Hypertrophic

Cardiomyopathy–Associated Genes
de Marvao, A. et al. J Am Coll Cardiol. 2021;78(11):1097–1110.
In 200,000 adults, the prevalence of variants pathogenic or likely pathogenic for hypertrophic cardiomyopathy (SARC-HCM-P/LP) was 1 in 407, whereas the
aggregate prevalence of indeterminate sarcomeric variants was 1 in 38. The SARC-HCM-P/LP variants were associated with increased risk of death and major adverse
cardiovascular events. We found associations with hypertrophic cardiomyopathy–like imaging phenotypes although the prevalence of overt cardiomyopathy was low.
SARC-IND ¼ indeterminate variants in hypertrophic cardiomyopathy–associated sarcomere-encoding genes (rare variants that do not meet criteria for pathogenic/
likely pathogenic annotation); SARC-NEG ¼ genotype negative.
and greater trabeculation. In subjects harboring cohorts, in which even patients without family his-
SARC-HCM-P/LP variants, hypertrophy ($13 mm) was tory have a comparatively high 30% yield of sarco-
present in 18.4%, with 3 individuals presenting with mere variants (39), suggesting that patients enrolled
hypertrophy $15 mm (6.1%). Penetrance estimates into cohorts and/or referred for diagnostic
from family studies vary, but a recent study of rela- sequencing represent a skewed subset, likely
tives of HCM probands observed an HCM phenotype enriched by those with clinical risk factors (40,41).
in 37% of genotype-positive relatives at first Although the natural history of disease and family
screening (37). Although UKBB participants have a pedigrees are not known for UKBB participants with
higher degree of kinship than would be expected in a unexplained LVH, it is plausible that multifactorial
random sample (38), these results suggest that the sarcomere-negative hypertrophy is the predominant
penetrance of familial HCM is additionally driven by etiology in the community (42,43). Nonetheless,
other genetic or environmental disease–modifying increasing WT was associated with higher risk of
characteristics, shared within families but that over- death and MACE (and each individual component),
lap less commonly in the community. Conversely, which persisted despite adjustment for sex, sarco-
only 2.6% of individuals with unexplained LVH meric genotype, and systolic blood pressure. These
(WT $15 mm) harbored SARC-HCM-P/LP variants. findings illustrate the additive role of clinical and
This contrasts with the diagnostic yield in HCM genetic assessment in risk stratifying patients with
unexplained LVH and managing the appropriate FUNDING SUPPORT AND AUTHOR DISCLOSURES
screening of relatives (2).
This study was supported by the Medical Research Council, UK (MC-
STUDY LIMITATIONS. UKBB is a large-cross sectional
A658-5QEB0); the National Institute for Health Research Imperial
study that is subject to selection bias, excluding College Biomedical Research Centre; the National Institute for Health
younger and potentially more severe cases (44); Research Royal Brompton Cardiovascular Biomedical Research Unit;
the British Heart Foundation (NH/17/1/32725, RG/19/6/34387, RE/18/
however, risk factor associations seem to be broadly
4/34215); Fondation Leducq (16 CVD 03); Wellcome Trust (107469/Z/
generalizable (45). The population is predominantly 15/Z, 200990/A/16/Z); the National Heart and Lung Institute Foun-
European, and further work is required to explore dation; the Royston Centre for Cardiomyopathy Research; Rosetrees
traits and outcomes in people of diverse ancestries. and CORDA (Dr Prasad); Academy of Medical Sciences (SGL015/1006;
Dr de Marvao); Mason Medical Research Trust grant (Dr de Marvao);
Although we included outcome data before enroll-
SmartHeart EPSRC Programme Grant (EP/P001009/1; Dr Bai and Dr
ment in the UKBB, we do not know the natural history Rueckert); and a Rosetrees and Stoneygate Imperial College Research
of disease in the community or what factors may in- Fellowship (Dr Whiffin). Dr Ware has consulted for MyoKardia, Inc.
fluence conversion to penetrant disease; in addition, and Foresite Labs. Dr Cook holds shares in Enleofen Bio Pte. Ltd. Dr
O’Regan has consulted for Bayer AG. All other authors have reported
co-segregation data were not available. Although
that they have no relationships relevant to the contents of this paper
there is genetic overlap between HCM and DCM, we to disclose.
found that DCM traits do not drive clinical outcomes
in individuals with SARC-IND or SARC-HCM-P/ ADDRESS FOR CORRESPONDENCE: Dr James S.
LP variants. Ware, Imperial College London, MRC London Institute
CONCLUSIONS of Medical Sciences, Hammersmith Hospital Campus,
Du Cane Road, London W12 0HS, United Kingdom.
We found that SARC-HCM-P/LP variants are present E-mail: j.ware@imperial.ac.uk. OR Dr Declan P. O’Re-
in 1 in 407 adults. Although the presence of SARC- gan, Imperial College London, MRC London Institute
HCM-P/LP variants in individuals in the community of Medical Sciences, Hammersmith Hospital Campus,
was rarely associated with the degree of unexplained Du Cane Road, London W12 0HS, United Kingdom.
hypertrophy required for a diagnosis of HCM, they are E-mail: declan.oregan@imperial.ac.uk.
associated with an attenuated phenotype and an
increased risk of adverse cardiovascular events even
PERSPECTIVES
in the absence of cardiomyopathy. SARC-HCM-P/LP
variants are likely to be harbored by >18 million
COMPETENCY IN MEDICAL KNOWLEDGE: Most individuals
people worldwide, and an improved understanding of
with pathogenic sarcomeric variants do not have overt HCM, but
their clinical significance is crucially important,
a subclinical phenotype is associated with an increased risk of
especially in light of recommendations to actively
adverse cardiovascular events.
screen for such variants as secondary findings during
clinical sequencing (46). Although penetrance for
TRANSLATIONAL OUTLOOK: Further genomic analyses are
overt HCM is modest and absolute event rates are
needed to characterize the specific risks levels and types of
low, identification of these variants may enhance risk
complications associated with various inherited forms of HCM.
stratification beyond familial disease, even when
cardiomyopathy is not manifest.
REFERENCES
1. Elliott PM, Anastasakis A, Borger MA, et al. 3. Ingles J, Goldstein J, Thaxton C, et al. Evalu- Genomics (ACMG). Genet Med. 2021;23:1391–
2014 ESC guidelines on diagnosis and manage- ating the clinical validity of hypertrophic cardio- 1398.
ment of hypertrophic cardiomyopathy: the Task myopathy genes. Circ Genom Precis Med. 2019;12:
6. Green RC, Berg JS, Grody WW, et al. ACMG
Force for the Diagnosis and Management of Hy- e002460.
recommendations for reporting of incidental
pertrophic Cardiomyopathy of the European So-
findings in clinical exome and genome sequencing.
ciety of Cardiology (ESC). Eur Heart J. 2014;35: 4. Maron BJ, Maron MS, Semsarian C. Genetics of
Genet Med. 2013;15:565–574.
2733–2779. hypertrophic cardiomyopathy after 20 years:
clinical perspectives. J Am Coll Cardiol. 2012;60: 7. ACMG response to Nussbaum et al. Letter on
2. Ommen SR, Mital S, Burke MA, et al. 2020 AHA/
705–715. ACMG policy statement: the use of secondary
ACC guideline for the diagnosis and treatment of
findings recommendations for general population
patients with hypertrophic cardiomyopathy: a 5. Miller DT, Lee K, Gordon AS, et al. Recom-
screening: a policy statement of the ACMG. Genet
report of the American College of Cardiology/ mendations for reporting of secondary findings
Med. 2019;21:2838–2839.
American Heart Association Joint Committee on in clinical exome and genome sequencing,
Clinical Practice Guidelines. J Am Coll Cardiol. 2021 update: a policy statement of the 8. Semsarian C, Ingles J, Maron MS, Maron BJ.
2020;76:e159–e240. American College of Medical Genetics and New perspectives on the prevalence of
hypertrophic cardiomyopathy. J Am Coll Cardiol. aortic stenosis are revealed by 3D model machine and clinical outcomes in hypertrophic cardiomy-
2015;65:1249–1254. learning. Eur Heart J Cardiovasc Imaging. 2020;21: opathy. Circ Genom Precis Med. 2021;14(1):
417–427. e003062.
9. Sudlow C, Gallacher J, Allen N, et al. UK Bio-
bank: an open access resource for identifying the 22. Szustakowski JD, Balasubramanian S, Sasson A, 36. Tadros R, Francis C, Xu X, et al. Shared genetic
causes of a wide range of complex diseases of et al. Advancing human genetics research and drug pathways contribute to risk of hypertrophic and
middle and old age. PLoS Med. 2015;12:e1001779. discovery through exome sequencing of the UK dilated cardiomyopathies with opposite directions
Biobank. Nat Genet. 2021;53(7):942–948. of effect. Nat Genet. 2021;53:128–134.
10. von Elm E, Altman DG, Egger M, Pocock SJ,
Gøtzsche PC, Vandenbroucke JP. The Strength- 23. Karczewski KJ, Francioli LC, Tiao G, et al. The 37. van Velzen HG, Schinkel AFL, Baart SJ, et al.
ening the Reporting of Observational Studies in mutational constraint spectrum quantified from Outcomes of contemporary family screening in
Epidemiology (STROBE) statement: guidelines for variation in 141,456 humans. Nature. 2020;581: hypertrophic cardiomyopathy. Circ Genom Precis
reporting observational studies. Lancet. 434–443. Med. 2018;11:e001896.
2007;370:1453–1457.
24. Whiffin N, Walsh R, Govind R, et al. Car- 38. Bycroft C, Freeman C, Petkova D, et al. The UK
11. Littlejohns TJ, Holliday J, Gibson LM, et al. The dioClassifier: disease- and gene-specific compu- Biobank resource with deep phenotyping and
UK Biobank imaging enhancement of 100,000 tational decision support for clinical genome genomic data. Nature. 2018;562:203–209.
participants: rationale, data collection, manage- interpretation. Genet Med. 2018;20:1246–1254.
39. Ingles J, Burns C, Bagnall RD, et al. Nonfa-
ment and future directions. Nat Commun. 2020;11:
25. Richards S, Aziz N, Bale S, et al. Standards and milial hypertrophic cardiomyopathy: prevalence,
2624.
guidelines for the interpretation of sequence var- natural history, and clinical implications. Circ Car-
12. Petersen SE, Matthews PM, Francis JM, et al. iants: a joint consensus recommendation of the diovasc Genet. 2017;10:e001620.
UK Biobank’s cardiovascular magnetic resonance American College of Medical Genetics and Geno- 40. Maron BJ, Spirito P. Impact of patient selec-
protocol. J Cardiovasc Magn Reson. 2016;18:8. mics and the Association for Molecular Pathology. tion biases on the perception of hypertrophic
Genet Med. 2015;17:405–424. cardiomyopathy and its natural history. Am J
13. Bai W, Sinclair M, Tarroni G, et al. Automated
cardiovascular magnetic resonance image analysis 26. Walsh R, Mazzarotto F, Whiffin N, et al. Cardiol. 1993;72:970–972.
with fully convolutional networks. J Cardiovasc Quantitative approaches to variant classification 41. Maron BJ, Rowin EJ, Casey SA, Maron MS. How
Magn Reson. 2018;20:65. increase the yield and precision of genetic testing hypertrophic cardiomyopathy became a contem-
in Mendelian diseases: the case of hypertrophic porary treatable genetic disease with low mortal-
14. Augusto JB, Davies RH, Bhuva AN, et al.
cardiomyopathy. Genome Med. 2019;11:5. ity: shaped by 50 years of clinical research and
Diagnosis and risk stratification in hypertrophic
cardiomyopathy using machine learning wall 27. Jordan E, Peterson L, Ai T, et al. An evidence- practice. JAMA Cardiol. 2016;1:98–105.
thickness measurement: a comparison with human based assessment of genes in dilated cardiomy- 42. Butters A, Bagnall RD, Ingles J. Revisiting the
test-retest performance. Lancet Digit Health. opathy. Circulation. 2021;144(1):7–19. diagnostic yield of hypertrophic cardiomyopathy
2020;3:e20–e28. 28. Austin PC, Lee DS, Fine JP. Introduction to the genetic testing. Circ Genom Precis Med. 2020;13:
analysis of survival data in the presence of e002930.
15. Schulz-Menger J, Bluemke DA, Bremerich J,
et al. Standardized image interpretation and post- competing risks. Circulation. 2016;133:601–609. 43. Harper AR, Goel A, Grace C, et al. Common
processing in cardiovascular magnetic resonance— 29. Petersen SE, Aung N, Sanghvi MM, et al. genetic variants and modifiable risk factors un-
2020 update: Society for Cardiovascular Magnetic Reference ranges for cardiac structure and function derpin hypertrophic cardiomyopathy susceptibility
Resonance (SCMR): Board of Trustees Task Force using cardiovascular magnetic resonance (CMR) in and expressivity. Nat Genet. 2021;53:135–142.
on Standardized Post-Processing. J Cardiovasc Caucasians from the UK Biobank population cohort. 44. Munafò MR, Tilling K, Taylor AE, Evans DM,
Magn Reson. 2020;22:19. J Cardiovasc Magn Reson. 2017;19:18. Davey Smith G. Collider scope: when selection bias
16. Biffi C, de Marvao A, Attard MI, et al. Three- 30. Bick AG, Flannick J, Ito K, et al. Burden of rare can substantially influence observed associations.
dimensional cardiovascular imaging-genetics: a sarcomere gene variants in the Framingham and Int J Epidemiol. 2018;47:226–235.
mass univariate framework. Bioinformatics. Jackson Heart Study cohorts. Am J Hum Genet. 45. Batty GD, Gale CR, Kivimäki M, Deary IJ, Bell S.
2018;34:97–103. 2012;91:513–519. Comparison of risk factor associations in UK Bio-
17. Duan J, Bello G, Schlemper J, et al. Automatic 3D 31. Ho CY, López B, Coelho-Filho OR, et al. bank against representative, general population
Bi-ventricular segmentation of cardiac images by a Myocardial fibrosis as an early manifestation of based studies with conventional response rates:
shape-refined multi-task deep learning approach. hypertrophic cardiomyopathy. N Engl J Med. prospective cohort study and individual partici-
IEEE Trans Med Imaging. 2019;38:2151–2164. 2010;363:552–563. pant meta-analysis. BMJ. 2020;368:m131.
18. Meyer HV, Dawes TJW, Serrani M, et al. Ge- 32. Ho CY, Day SM, Ashley EA, et al. Genotype and 46. Kalia SS, Adelman K, Bale SJ, et al. Recom-
netic and functional insights into the fractal lifetime burden of disease in hypertrophic cardio- mendations for reporting of secondary findings in
structure of the heart. Nature. 2020;584:589– myopathy: insights from the Sarcomeric Human clinical exome and genome sequencing, 2016 up-
594. Cardiomyopathy Registry (SHaRe). Circulation. date (ACMG SF v2.0): a policy statement of the
2018;138:1387–1398. American College of Medical Genetics and Geno-
19. de Marvao A, Dawes TJ, Shi W, et al. Pre-
mics. Genet Med. 2017;19:249–255.
cursors of hypertensive heart phenotype develop 33. Lorenzini M, Norrish G, Field E, et al. Pene-
in healthy adults: a high-resolution 3D MRI study. trance of hypertrophic cardiomyopathy in sarco-
J Am Coll Cardiol Img. 2015;8:1260–1269. mere protein mutation carriers. J Am Coll Cardiol. KEY WORDS cardiovascular magnetic
2020;76:550–559. resonance, deep learning, genetics,
20. Schafer S, de Marvao A, Adami E, et al. Titin-
truncating variants affect heart function in disease 34. Geske JB, Ong KC, Siontis KC, et al. Women hypertrophic cardiomyopathy, penetrance
cohorts and the general population. Nat Genet. with hypertrophic cardiomyopathy have worse
2017;49:46–53. survival. Eur Heart J. 2017;38:3434–3440. A PPE NDI X For supplemental figures and
21. Bhuva AN, Treibel TA, De Marvao A, et al. Sex 35. Lakdawala NK, Olivotto I, Day SM, et al. As- tables, please see the online version of this
and regional differences in myocardial plasticity in sociations between female sex, sarcomere variants paper.
EDITORIAL COMMENT
Hypertrophic Cardiomyopathy in the

General Population
Leveraging the UK Biobank Database and
Machine Learning Phenotyping*
Linnea M. Baudhuin, PHD
I n the past 2 decades, the field of genomics has

undergone tremendous growth, resulting in a
transformative impact on the practice of medi-
cine. This is due in large part to advances in technol-
defined as a wall thickness (WT) $15 mm, with a
lower threshold (WT of 13-14 mm) in the setting of
genetic predisposition (2,3). The clinical manifesta-
tions of HCM are highly variable and can include
ogy, especially as it relates to next-generation asymptomatic LVH, arrhythmias, and refractory heart
sequencing (NGS) and bioinformatics, whereby more failure, with variable interfamilial and intrafamilial
rapid and comprehensive testing can be performed expressivity. HCM is primarily an autosomal domi-
for rare, heterogeneous genetic disorders. The cardio- nant disorder with 8 main sarcomeric genes (MYBPC2,
myopathies, which are caused by numerous genes, MYH7, MYL2, MYL3, TNNI3, TNNT2, TPM1, ACTC1)
are no stranger to the diagnostic benefits derived that contribute to >92% of hereditary HCM (4).
from NGS testing. However, genetic testing of cardio- Numerous additional genes have been associated
myopathies oftentimes reveals novel genetic variants with HCM as well, albeit to a much lesser extent,
caused by a high rate of private (familial) variants. and other genetic influences (eg, genetic modifiers)
This, combined with the phenotypic heterogeneity are suspected to affect disease phenotype and
(incomplete penetrance and variable expressivity) of penetrance.
cardiomyopathies, can create challenges in the classi- Clinical laboratories have widely adopted genetic
fication of novel variants detected in families with variant classification guidelines from the American
cardiomyopathy. These challenges are further ampli- College of Medical Genetics and Genomics/Associa-
fied in presumably healthy individuals who choose to tion for Molecular Pathology (5). However, these
undergo predictive exome or genome sequencing, guidelines are conservative in nature, and combined
which subsequently uncovers the presence of cardio- with the genetic heterogeneity, novel variants, and
myopathy gene variants. incomplete penetrance/variable expressivity of HCM,
Hypertrophic cardiomyopathy (HCM) is an inheri- clinical laboratories tend to undercall pathogenic/
ted disorder that most commonly leads to isolated left likely pathogenic (P/LP) variants in sarcomeric genes
ventricular hypertrophy (LVH) (1). American and Eu- (6). A Clinical Genome Resource (ClinGen) Expert
ropean society guidelines state that HCM is clinically Panel has recently recommended a set of clinical pa-
rameters that should be collected on individuals
tested for HCM to enhance and refine genetic variant
interpretation in HCM (7). However, in the absence of
reflect the views of the authors and do not necessarily represent the
HCM (or with unknown HCM), such as what might be
views of JACC or the American College of Cardiology. encountered during population screening, classifica-
From the Department of Laboratory Medicine and Pathology, Mayo tion of sarcomeric variants becomes even more chal-
Clinic, Rochester, Minnesota, USA. lenging, and relatively little is understood about the
The author attests they are in compliance with human studies commit- clinical significance of detected variants. This is part
tees and animal welfare regulations of the author’s institution and Food
of the reason why ACMG guidelines for reporting
appropriate. For more information, visit the Author Center. secondary findings in clinical exome and genome

1112 Baudhuin JACC VOL. 78, NO. 11, 2021
Population Screening for Hypertrophic Cardiomyopathy SEPTEMBER 14, 2021:1111–1113
sequencing note that their list of 73 genes, which in- reduced to 10.8% for SARC-HCM-P/LP and 0.57% for
cludes the 8 sarcomeric HCM genes, is not validated SARC-IND. This low-level penetrance contrasts with
for population screening (8). other studies that have demonstrated 37%-50%
The United Kingdom Biobank (UKBB) is a long- penetrance in genotype-positive families with HCM
term, prospective, population-based genomic medi- (11,12). These results suggest that intrafamilial modi-
cine initiative involving 500,000 adults aged 40-69 fiers for penetrance are potentially genetically driven,
years from the United Kingdom. It is the largest open- in contrast to the general population where genetic
access exome sequencing resource with extensive underpinnings are not necessarily shared. In addition
health information currently available. As such, the to genetic modifiers, a number of other secondary
UKBB is ripe with opportunity for discoveries that can factors, such as sex differences and comorbidities,
affect the practice of genomic medicine, including likely influence penetrance of primary sarcomeric
variant frequency utilization, ascertainment of pop- P/LP variants in HCM. Future investigations could be
ulation disease risk, and novel disease gene discovery done to help determine why some variants are more
penetrant, eg, based on the type of variant present
SEE PAGE 1097
and its location in the gene/protein and whether
(9). Along these lines, in this issue of the Journal, de other more common potential modifier variants are
Marvao et al (10) took a deep dive into trying to better present, and then marry this data with phenotypic
understand population-based HCM through the factors.
analysis of data from >200,000 UKBB participants, in In considering subjects with unexplained
which they report on the prevalence of rare sarco- LVH $15 mm, only 2.6% harbored SARC-HCM-P/LP
meric gene variants, machine learning-derived variants, which is in contrast to other studies where
phenotype, and the associated lifetime risk of an approximately 32% diagnostic yield has been re-
adverse events in genotype-positive individuals. ported (13). This discrepancy could be due in part to
They observed a 2.9% prevalence of rare (filtering the involvement of other genes, either known or
allele frequency <0.00004) sarcomeric variants, with novel. Some genes, eg, those associated with Fabry
a conservative 0.25% (1 in 407) prevalence of P/LP disease (GLA), PRKAG2 cardiomyopathy, Noonan
(“SARC-HCM-P/LP”) variants. SARC-HCM-P/LP vari- syndrome (PTPN11), and transthyretin amyloidosis
ants were associated with increased risk of death (TTR), can cause conditions that mimic HCM with
or major adverse cardiovascular events (MACE) isolated, or seemingly isolated, LV hypertrophy (14).
(HR: 1.69; 95% CI: 1.38-2.07), heart failure (HR: 4.23; There is also an opportunity to potentially discover
95% CI: 3.07-5.83), and arrhythmia (HR: 1.59; 95% novel genes involved in HCM/LVH. Second, sarco-
CI: 1.22-2.09). Cardiac magnetic resonance images mere gene-negative HCM has been suggested to be a
from a substudy of approximately 39,000 participants complex, polygenic trait (15), with a low probability of
were subjected to machine learning phenotyping via clinically relevant HCM in relatives. Thus, additional
a deep learning neural network algorithm to charac- analyses (eg, with genotyping arrays and other
terize phenotypic manifestations as well as estimate methods) of this cohort might yield information per-
the prevalence of penetrant disease. In this substudy taining to additional genetic and environmental
cohort, 1.9% (1 of 52) had mild hypertrophy (WT contributions to HCM. As noted in the previous text,
13-15 mm) and 0.4% (1 of 241) had at least moderate there are many complexities and challenges to accu-
hypertrophy (WT $15 mm). The prevalence of rate classification of sarcomeric variants, which has
phenotypic HCM (WT $15mm in the absence of hy- been demonstrated to lead to undercalling P/LP var-
pertension and valve disease) was 0.19% (1 of 517), iants in HCM. Along these lines, the authors per-
which is in line with previous reports. formed a supplemental substudy whereby they
Penetrance of rare sarcomeric variants was further stratified the SARC-HCM-IND variants and
analyzed in 21,322 UKBB participants who had both determined that there were some groupings of in-
cardiac magnetic resonance imaging and exome dividuals for which the variants could be bumped up
sequencing. There were 49 individuals who harbored to “SARC-IND high probability.” Additional genomic
P/LP variants, and 513 had rare variants of indeter- analyses could potentially be performed to further
minate significance (“SARC-IND”), with the majority characterize specific risk levels of the various variants
of variants occurring in MYBPC3 and MYH7. Pene- detected.
trance for HCM (LVH $13 mm) in SARC-HCM-P/LP This study sheds light on the prevalence of rare
and SARC-IND was 18.4% (9 of 49) and 2.9% (15 of sarcomeric variants and unexplained LVH in the
513), respectively. In those without concomitant general population, as well as penetrance and clinical
hypertension or valve disease, the penetrance was outcomes in those who carry P/LP variants, and
JACC VOL. 78, NO. 11, 2021 Baudhuin 1113
SEPTEMBER 14, 2021:1111–1113 Population Screening for Hypertrophic Cardiomyopathy
highlights the importance of recognizing subclinical population working groups established by ACMG (8),
phenotypes with adverse cardiovascular events. The and this study paves the way for similar analyses to
low yield of P/LP variants in unexplained LVH pro- be performed on other disease states using the UKBB
vides a great reminder that there are potentially other database. Finally, the study by de Marvao et al (10)
genetic causes that could explain phenotype, such as details a unique, machine-learning algorithm for
genes that cause conditions that mimic HCM. Addi- cardiac phenotyping that was optimized on the UKBB
tionally, although P/LP variants had low penetrance, cohort and can potentially be extended to other
the attenuated phenotype and adverse cardiovascular studies.
events in the presence of some SARC-HCM-P/LP var-
iants provides important implications for counseling
and risk stratification when these variants are detec- ADDRESS FOR CORRESPONDENCE: Dr Linnea M.
ted as secondary findings. Additionally, because the Baudhuin, Department of Laboratory Medicine and
ACMG secondary findings gene list is not validated for Pathology, Mayo Clinic, Rochester, Minnesota 55905,
population screening, the results from this study can USA. E-mail: baudhuin.linnea@mayo.edu. Twitter:
provide helpful insights to the asymptomatic/ @LinneaBaudhuin.
REFERENCES
1. Maron BJ, Ommen SR, Semsarian C, Spirito P, American College of Medical Genetics and Geo- with rare genetic variants of hypertrophic cardio-
Olivotto I, Maron MS. Hypertrophic cardiomyopa- mics and the Association for Molecular Pathology. myopathy. J Am Coll Cardiol. 2021;78:1097–1110.
thy: present and future, with translation into Genet Med. 2015;17:405–423.
11. Lorenzini M, Norrish G, Field E, et al. Pene-
contemporary cardiovascular medicine. J Am Coll
6. Walsh R, Thomson KL, Ware HS, et al. trance of hypertrophic cardiomyopathy in sarco-
Cardiol. 2014;64:83–99.
Reassessment of Mendelian gene pathogenicity mere protein mutation carriers. J Am Coll Cardiol.
2. Gersh BJ, Maron BJ, Bonow RO, et al. 2011 using 7,855 cardiomyopathy cases and 60,706 2020;76:550–559.
ACCF/AHA guideline for the diagnosis and treat- reference samples. Genet Med. 2017;19:192– 12. Van velzen HG, Schinkel AFL, Baart SJ, et al.
ment of hypertrophic cardiomyopathy: executive 203. Outcomes of contemporary family screening in
summary: a report of the American College of
7. Morales A, Ing A, Antolik C, et al. Harmo- hypertrophic cardiomyopathy. Circ Genom Precis
Cardiology Foundation/American Heart Associa-
nizing the collection of clinical data on genetic Med. 2018;11:e001896.
tion Task Force on Practice Guidelines. J Am Coll
testing requisition forms to enhance variant 13. Butters A, Bagnall RD, Ingles J. Revisiting the
Cardiol. 2011;58:2703–2738.
interpretation in hypertrophic cardiomyopathy diagnostic yield of hypertrophic cardiomyopathy
3. Elliott PM, Anastasakis A, Borger MA, et al. 2014 (HCM): a study from the ClinGen Cardiomyopa- genetic testing. Circ Genom Precis Med. 2020;13:
ESC guidelines on diagnosis and management of thy Variant Curation Expert Panel. J Mol Diag. e002930.
hypertrophic cardiomyopathy: the Task Force for 2021;23:589–598.
the Diagnosis and Management of Hypertrophic 14. Hoss S, Habib M, Silver J, et al. Genetic testing
Cardiomyopathy of the European Society of Cardi- 8. Miller DT, Lee K, Gordon AS, et al. Recom- for diagnosis of hypertrophic cardiomyopathy
ology (ESC). Eur Heart J. 2014;35:2733–2779. mendations for reporting secondary findings in mimics. Yield and clinical significance. Circ Genom
clinical exome and genome sequencing, 2021 up- Precis Med. 2020;13:e002748.
4. Ommen SR, Mital S, Buke MA, et al. 2020 AHA/
date: a policy statement of the American College
ACC guideline for the diagnosis and treatment of 15. Harper AR, Goel A, Grace C, et al. Common
of Medical Genetics and Genomics. Genet Med.
patients with hypertrophic cardiomyopathy: a genetic variants and modifiable risk factors un-
2021;23:1391–1398.
report of the American College of Cardiology/ derpin hypertrophic cardiomyopathy susceptibility
American Heart Association Joint Committee on 9. Alaimo JT, Saunders CJ. Big data strikes again: and expressivity. Nat Genet. 2021;53:135–142.
Clinical Practice Guidelines. J Am Coll Cardiol. future utilization of the UK Biobank as a resource
2020;76:e159–e240. for clinical laboratories. Clin Chem. 2021;67:932–
934.
5. Richards S, Aziz N, Bale S, et al. Standards and KEY WORDS cardiovascular magnetic
guidelines for the interpretation of sequence var- 10. de Marvao A, McGurk KA, Zheng SL, et al. resonance, deep learning, genetics, hypertrophic
iants: a joint consensus recommendation of the Phenotypic expression and outcomes in individuals cardiomyopathy, penetrance
Technology-Assisted Self-Selection of
Candidates for Nonprescription
Statin Therapy
Steven E. Nissen, MD,a,b Howard G. Hutchinson, MD,c Tracy Y. Wang, MD, MHS, MSC,d Christie M. Ballantyne, MD,e
Sara Travis, BS,f Melanie Morris, BA,c William Miller, MD,f Jennifer Hynson, BA,f Kathy Wolski, MPH,a,b
Paul M Ridker, MDg
ABSTRACT
BACKGROUND Although statins reduce cardiovascular morbidity and mortality, only about one-half of eligible patients
receive treatment. Safe and appropriate consumer access to statins could have a significant positive public health impact.
OBJECTIVES This study compares the concordance between a participant and clinician assessment of eligibility for
statin therapy using a technology-assisted approach.
METHODS A total of 500 participants, 83 with limited literacy, completed an at-home Web-based application to assess
appropriateness for treatment with rosuvastatin 5 mg. The Web application is designed to assess eligibility for a
moderate-intensity statin based on current guidelines and deny access to individuals with contraindications to rosu-
vastatin. Subsequently, participants visited a research site where clinicians, blinded to the information the participant
entered, performed an independent Web application assessment. The Web application is programmed for 1 of 3 rosu-
vastatin treatment outcomes: “OK to use,” “not right for you,” or “ask a doctor.” The primary endpoint was the percent of
participants whose self-selected eligibility for nonprescription rosuvastatin was concordant with clinician assessment.
RESULTS For the primary endpoint, participant selection for statin therapy was concordant with clinician selection in
481 (96.2%) of 500 participants (95% confidence interval: 94.1%-97.7%), of whom 23 (4.6%) were deemed appropriate
and 458 (91.6%) were deemed inappropriate for treatment. Discordance was due to incorrect self-selection (“OK to use”)
in 3 cases, incorrect rejection (“not right for you”) in 14 cases and an incorrect “ask a doctor” outcome in 2 cases.
CONCLUSIONS The use of a technology-assisted approach to consumer self-selection for statin therapy resulted in
participant self-selection that showed substantial agreement with clinician selection.
(J Am Coll Cardiol 2021;78:1114–1123) © 2021 by the American College of Cardiology Foundation.
A dministration of statins to reduce low-

density lipoprotein cholesterol (LDL-C) plays
a pivotal role in public health efforts to
improve cardiovascular outcomes in primary and
some communities. Recognition of the problem of
undertreatment with statins has resulted in efforts
to make these drugs available without a prescription.
Five prior efforts to achieve regulatory approval of
secondary prevention populations. However, studies nonprescription statins failed due to multiple issues,
demonstrate that only about one-half of patients particularly the inability to demonstrate that eligible
eligible to receive statins are actually treated (1,2). consumers would receive treatment and consumers
Barriers to effective treatment are complex but for whom statins were inappropriate or unsafe would
include reluctance of patients to seek regular medical not be able to access over-the-counter (OTC) treat-
care and lack of access to health care resources in ment (3-7).

audio summary by From the aDepartment of Cardiovascular Medicine, Cleveland Clinic, Cleveland Ohio, USA; bCleveland Clinic Coordinating Center
Editor-in-Chief for Clinical Research, Cleveland Clinic, Cleveland Ohio, USA; cAstraZeneca BioPharmaceuticals, Wilmington, Delaware, USA;
Dr. Valentin Fuster on d
Duke Clinical Research Institute, Duke University, Durham, North Carolina, USA; eBaylor College of Medicine, Houston, Texas,
JACC.org. USA; fConcentrics Research, Indianapolis, Indiana, USA; and the gCenter for Cardiovascular Disease Prevention, Harvard Medical
School, Boston, Massachusetts, USA.
Manuscript received April 2, 2021; revised manuscript received May 20, 2021, accepted June 10, 2021.

JACC VOL. 78, NO. 11, 2021 Nissen et al. 1115
SEPTEMBER 14, 2021:1114–1123 Nonprescription Statins
To address this problem, we used a novel approach household were a health care practitioner or ABBREVIATIONS
termed technology assisted self-selection designed to were employed by a health care practice, AND ACRONYMS
qualify consumers for treatment with rosuvastatin a manufacturer of medicines, a consumer

BP = blood pressure
5 mg daily, while excluding those for whom treatment health company, a health care application-
CI = confidence interval
is not appropriate. The self-selection tool was incor- development company, or a marketing com-
HDL-C = high-density
porated into a Web-based application that was pany. From a planned enrollment of 500 par-
lipoprotein cholesterol
developed as Software as a Medical Device based on ticipants, 80 were required to have limited
LDL-C = low-density
current regulatory guidance (8). The Web application literacy assessed by the Rapid Estimate of lipoprotein cholesterol
was programmed to replicate the current guidelines Adult Literacy in Medicine (REALM) test.
OTC = over the counter
for treatment with a moderate intensity statin and the Limited literacy was defined as a REALM
REALM = Rapid Estimate of
warnings and precautions for rosuvastatin. The score <61, which corresponds to a seventh- or Adult Literacy in Medicine
CREST (Consumer Rosuvastatin Electronic Self- eighth-grade education or lower (10). To test TC = total cholesterol
Selection Trial) was designed to test whether tech- the overall utility of the Web application, the
TG = triglyceride
nology can help consumers make appropriate self- study design precluded enrollment of an
selection decisions for rosuvastatin 5 mg by excessive number of young and old participants who
comparing concordance between participant and would be rejected strictly due to age. The enrollment of
clinician assessments of treatment eligibility using women between age 20 and 49 years was capped at #50
this Web-based application. participants, and men aged <40 years or those aged
>75 years were limited to 30 participants each.
SEE PAGE 1124
ONLINE PARTICIPANT SELF-ASSESSMENT PROCEDURE.

METHODS Participants were provided with a brief study over-
view and a link to the Web-based application for use
STUDY ORGANIZATION AND OVERSIGHT. The trial at home or any location outside of a research facility.
was designed by the sponsor and Concentrics Upon starting the online self-assessment, participants
Research, a contract research organization that spe- acknowledged a privacy consent. They subsequently
cializes in developing evidence to support approval of responded to questions related to medical history,
nonprescription medications. This research program medication use, total cholesterol (TC), high-density
was guided by an academic advisory board consisting lipoprotein-cholesterol (HDL-C), LDL-C, triglycerides
of experts in lipids and public health. At the conclu- (TG), blood pressure (BP), and if needed for their
sion of the trial, all data were provided to the Cleve- assessment, waist circumference, high-sensitivity
land Clinic Coordinating Center for Clinical Research C-reactive protein, and coronary artery calcium
for confirmation of analyses. The study design was score (Central Illustration). The participant responded
reviewed by the U.S. Food and Drug Administration to questions until 1 of 3 possible self-selection out-
and approved by a central Institutional Review Board comes was obtained for treatment with rosuvastatin
(Advarra) prior to commencing patient enrollment. 5 mg (“OK to use,” “not right for you” or “ask a doc-
This type of noninterventional behavior study is tor”). The Web application used the 2018 Cholesterol
exempt from requirements for registration (9). Treatment Guidelines and a proposed drug facts label
STUDY POPULATION SELECTION. The study protocol for nonprescription rosuvastatin to assess eligibility
is available in the Supplemental Appendix. for treatment (11). Participants were deemed ineli-
Participants were recruited via digital advertising gible if their risk was <5% or $20%, if they had a risk
(eg, paid search, social media, display advertising), $5% and <7.5% without a risk-enhancing factor, or if
traditional media (eg, radio, television, print, and they had a contraindication to rosuvastatin. Ques-
community outreach), and supplemental methods (eg, tions were suspended once a self-selection outcome
local sites and flyers). The advertising was approved by was achieved. The technology-assisted outcome was
the central Institutional Review Board. The adver- not revealed to participants.
tising asked participants if they were interested in CLINICIAN ASSESSMENT OF STATIN APPROPRIATENESS.
cholesterol and heart health and invited them to phone After completion of the online self-assessment,
a study call center to assess eligibility to participate in a participants were directed to a research site for a
study. Potential participants provided written scheduled visit and instructed to bring verification of
informed consent prior to first study visit. Those the laboratory values and BP they entered into the
eligible were any sex, were aged $20 years, and were online health assessment. At the research site, partic-
able to read, speak, and understand English. Partici- ipants signed an informed consent and staff collected
pants were excluded if they or a member of their the verification documents provided and ensured that
1116 Nissen et al. JACC VOL. 78, NO. 11, 2021
Nonprescription Statins SEPTEMBER 14, 2021:1114–1123
C E N T R A L IL LU ST R A T I O N Examples of Web Application Screens Presented to Trial Participants
Nissen, S.E. et al. J Am Coll Cardiol. 2021;78(11):1114–1123.
Four examples of computer screens presented to participants as part of the self-selection application. The top left screen informs the
consumer that recent lipid values and blood pressure are required. The top right screen shows an inquiry about medications that would make
the patient ineligible for nonprescription statin treatment. The bottom left screen allows participants to enter lipid values. The bottom right
screen asks about medical history that would define participants as secondary prevention patients who require high-intensity statin treat-
ment. HDL ¼ high-density lipoprotein; LDL ¼ low-density lipoprotein.
F I G U R E 1 Flow of Participants Through the Study
1,563 Assessed for Eligibility
500 Excluded
- 162 Past participation
- 81 Health care professional
- 48 Employment security
- 5 Decline to disclose age
- 204 Patient declined
- 13 Not worth time/travel
- 48 Research site too far
- 143 Other
1,063 sent link to technology-assisted self-selection
563 Did not complete studya

- 331 Lost to follow-up
- 43 Patient withdrew
- 38 Ineligible
- 3 Investigator decision
- 148 Otherb
500 Completed study
Reasons for exclusion of participants during the study. From 1,563 participants initially assessed, 500 consumers met all eligibility criteria and
eventually completed the study. a198 of these participants completed the health assessment but never scheduled a visit to complete the
study. b140 enrollment completed (no more participants required for study), 7 age quota full, 1 past participation in a research study.
these were obtained within 12 months of the visit. independent medical evaluation and used this infor-
Verification could consist of written results from a mation, along with the verified laboratory and BP
doctor’s office, laboratory, or community testing site, data, to perform a technology-assisted assessment
or online viewing of health portal results. If the using the same Web application used by the
participant did not have a verified source, they could participant.
reschedule the visit and bring the verified source, or TRIAL ENDPOINTS. The primary analysis compared
the site staff could obtain a fingerstick blood sample the concordance between self-assessed and clinician-
for TC, LDL-C, HDL-C, and TG and measure BP and, if assessed appropriateness for treatment with 5 mg of
applicable, waist circumference. Research sites did not rosuvastatin daily. The primary endpoint was the
have a point-of-care test for high-sensitivity C-reac- percentage of participants whose technology-assisted
tive protein, so if a participant provided this informa- self-assessment yielded an identical rosuvastatin
tion within the online self-assessment and did not treatment outcome (“OK to use,” “not right for you,”
provide a verified source to the site visit, these data or “ask a doctor”) to the clinician’s technology-
were classified as unverified. The coronary calcium assisted assessment. Discordant outcomes were
score was treated similarly. A urine pregnancy test was evaluated to understand the root cause for the
conducted on women of child-bearing potential. The differences. Because an individual could incorrectly
REALM test was administered (12). answer a question or multiple questions in the online
The information collected at the site was provided health assessment and still arrive at a correct overall
to a telemedicine clinician. The clinician was unaware outcome, the secondary endpoint evaluated the
of the information the participant entered into the percentage of participants with correct entries for
Web-based application. The clinician conducted an all self-selection questions to determine the potential
potential, history of atherosclerotic cardiovascular

T A B L E 1 Summary of Characteristics for All Participants and by Literacy Category
Obtained by Research Sites
disease, history of taking cholesterol- or TG-lowering
medicines, diabetes, and age.
All Normal Limited Did Not
Participants Literacy Literacy Complete
PRESPECIFIED MITIGATION PLAN. The primary
(N ¼ 500) (n ¼ 417) (n ¼ 83) Study (n ¼ 563) endpoint was adjusted based on prespecified criteria
Sex at birth that corrected for minor differences between the
Female 311 (62.2) 270 (64.7) 41 (49.4) 337 (59.9) patient and clinician assessment that did not repre-
Age, y 59.5 12.7 60.4 12.4 54.7 13.3 NA
sent a clinically meaningful problem, such as a patient
<45 y 54 (10.8) 37 (8.9) 17 (20.5) 88 (15.6)
having a birthday between self-assessment and the
45-54 y 75 (15.0) 62 (14.9) 13 (15.7) 89 (15.8)
site visit, resulting in an age difference. Because of the
55-64 y 172 (34.4) 136 (32.6) 36 (43.4) 190 (33.7)
>65 y 199 (39.8) 182 (43.6) 17 (20.5) 196 (34.8) temporal variability in measurement of BP and lipids,
Race the values obtained during the site visit were invari-
White 305 (61.0) 289 (69.3) 16 (19.3) 296 (52.6) ably different from those reported by the participant.
Black 166 (33.2) 103 (24.7) 63 (75.9) 212 (37.7) Such variability may lead to differences in the risk
Asian 5 (1.0) 5 (1.0) 0 (0.0) 8 (1.4) scores between the participant and clinician assess-
Other 24 (4.8) 20 (4.8) 4 (4.8) 47 (8.3)
ments, resulting in different selection outcomes. To
Ethnicity
account for these expected differences, a participant
Hispanic 28 (5.6) 21 (5.0) 7 (8.4) 24 (6.0)
Non-Hispanic 471 (94.2) 395 (94.7) 76 (91.6) 527 (93.6)
who had a home-based cardiovascular risk score be-
Refused 1 (0.2) 1 (0.2) 0 (0.0) 2 (0.4) tween 5.0% and 20.0% was classified as within range
Education if the on-site laboratory and BP measurements resul-
Graduated college/technical school 224 (44.8) 206 (49.4) 18 (21.7) 199 (35.3) ted in a risk score between 4.0% and 21.0%. Partici-
Some college/technical school 156 (31.2) 135 (32.4) 21 (25.3) 199 (35.3) pants who had a home-based risk score <5.0% or
Completed high school/GED 106 (21.2) 70 (16.8) 36 (43.4) 131 (23.3)
$20.0% were classified as outside the range if their
Less than high school 14 (2.8) 6 (1.4) 8 (9.6) 34 (6.0)
cholesterol and BP measurements on-site yielded a
Employment status
risk score <6.0% or $19.0%.
Full-time 127 (25.4) 112 (26.9) 15 (18.1) 119 (21.1)
Part-time 62 (12.4) 49 (11.8) 13 (15.7) 84 (14.9) Data entered into the online health assessment
Unemployed 58 (11.6) 38 (9.1) 20 (24.1) 84 (14.9) application were used to determine if a risk-
Retired 205 (41.0) 186 (44.6) 19 (22.9) 199 (35.3) enhancing factor was present as described in the
Othera 48 (9.6) 32 (7.7) 16 (19.2) 77 (13.7) 2018 Cholesterol Treatment Guidelines and whether
Prior cardiovascular disease 69 (13.8) 60 (14.4) 9 (10.8) NA the laboratory values fell within an acceptable range
for treatment with a moderate-intensity statin (11).
Values are n (%) or mean SD. aIncludes students or those who refused to answer the question.
NA ¼ not available. The allowed variability in measurements between the
participant’s values and the on-site values was based
on the National Cholesterol Education Program
for errors on individual questions and better under- analytical performance goals for TC (9%), TG
stand the questions most likely to lead to errors. (15%), LDL-C (12%), and HDL-C (13%) (13). No
Subgroup analyses were prespecified, including established standard exists for acceptable BP
literacy (normal, limited), women of childbearing variability. Because the normal circadian variability
in BP is 10% to 20%, the acceptable variability for BP
T A B L E 2 Verified Laboratory and BP Data for the Study Population by Literacy was set at 15% for both systolic and diastolic pres-
sures (14).
All Participants Normal Literacy Limited Literacy
(N ¼ 500) (n ¼ 417) (n ¼ 83) STATISTICAL ANALYSIS. The sample size of 500
Total cholesterol, mg/dL 498 186.9 50.7 415 188.8 51.1 83 177.5 48.2 provided >84% power to test the hypothesis for the
HDL-C, mg/dL 500 56.4 17.6 417 56.4 17.6 83 56.8 17.7 primary endpoint (H0: P ¼ 85% vs H1: P > 85%). The
LDL-C, mg/dL 492 105.7 42.9 410 107.4 43.3 82 97.2 39.7 null hypothesis would be rejected in favor of the
Triglycerides, mg/dL 497 107.0 414 107.0 83 114.0
(77.0-154.0) (77.0-153.0) (75.0-166.0)
alternative if the lower bound of the 2-sided 95% exact
Systolic BP, mm Hg 500 127.2 16.6 417 126.2 16.6 83 132.3 16.0 binomial (ie, Clopper-Pearson) confidence interval (CI)
Diastolic BP, mm Hg 500 76.9 10.2 417 76.1 10.0 83 81.0 9.9 was >85%. This sample size calculation was performed
Waist circumference, 500 41.1 7.0 417 40.9 6.9 83 42.2 7.4 using SAS version 9.4 PROC Power (SAS Institute, Inc)
inches
for a 1-sample, 2-sided a ¼ 0.05 exact test for binomial
Values are n, mean SD, or median (interquartile range). Verified data include data provided by the participant
proportion. Agreement between participants and
with documentation or obtained by the research site. clinicians was also assessed using a weighted kappa
BP ¼ blood pressure; HDL-C ¼ high-density lipoprotein cholesterol; LDL-C ¼ low-density lipoprotein
cholesterol.
statistic. Demographic and baseline variables are
reported using descriptive statistics.
T A B L E 3 Primary Endpoint: Overall Technology-Assisted T A B L E 4 Overall Technology-Assisted Self Selection Outcome by Literacy
Self-Selection Outcome With and Without Prespecified Subgroups With Prespecified Correction
Mitigation Plan (n ¼ 500)
Normal Literacy Limited Literacy
With Prespecified Without (n ¼ 417) (n ¼ 83)
Correction Correction
Outcome n (%) 95% CI (%) n (%) 95% CI (%)
Outcome n (%) 95% CI (%) n (%) 95% CI (%) Overall concordant outcome 401 (96.2) 93.8-97.8 80 (96.4) 89.8-99.2
Overall concordant 481 94.1-97.7 480 93.9-97.5 Concordant “OK to use” 17 (4.1) 6 (7.2)
outcome (96.2) (96)
Concordant “Ask a doctor” 0 (0) 0 (0)
Concordant “OK to use” 23 (4.6) 23 (4.6)
Concordant “Not right for you” 384 (92.0) 74 (89.2)
Concordant “Ask a doctor” 0 (0) 0 (0)
Concordant “Not right for 458 457 Values represent the numbers and percentages of the normal and limited literacy participants
you” (91.6) (91.4) whose self-selection outcome was concordant with the outcome determined by the clinician.
CI ¼ confidence interval.
Values represent the numbers and percentages of the 500 participants whose self-
selection outcome was concordant with the outcome determined by the clinician.
CI ¼ confidence interval.
(95% CI: 93.9%-97.5%). Table 4 shows outcomes for
both the normal and limited literacy subgroups. In the
RESULTS limited literacy subgroup, 80 (96.4%) of 83 partici-
pants achieved a concordant outcome (95% CI: 89.8%-
STUDY POPULATION. Figure 1 shows the flow of 99.2%) with an “OK to use” result in 7.2%.
participants through the study. From 1,563 partici- DETAILS: CONCORDANCE AND DISCORDANCE.
pants assessed for eligibility, 1,063 were sent a link to Greater than 90% of the outcomes for both partici-
the Web-based health assessment. Of these participants and clinicians was “not right for you.” The most
pants, 563 did not schedule an on-site visit, resulting
T A B L E 5 In-Home and Center-Derived Technology-Assisted Outcomes
in 500 completers defined as those with both a
participant and a clinician technology-assisted health In-Home Outcome In-Center Outcome
(n ¼ 500) (n ¼ 500)
assessment. Table 1 shows the characteristics of these
“OK to use” 26 (5.2) 36 (7.2)
500 participants obtained at the clinical research
“Ask a doctor” 2 (0.4) 1 (0.2)
centers and the data available for 563 participants “Not right for you” 472 (94.4) 463 (92.6)
who did not complete the self-selection process. The Taking cholesterol- or triglyceride- 206 (43.6) 212 (45.8)
mean age was 59.2 12.7 years, 62.2% were female, lowering medications
61.0% were White, 33.2% were Black, and 44.8% had Low risk score 62 (13.1) 67 (14.5)
Female aged <50 y 51 (10.8) 50 (10.8)
graduated college or a technical school. These char-
Age: males aged <20 y or either sex 30 (6.4) 30 (6.5)
acteristics were similar in the 563 participants who aged >75 y
did not complete the self-selection process. No risk enhancing factor 37 (7.8) 30 (6.5)
BP AND LIPIDS. Table 2 shows the verified laboratory No family history, males aged 20-39 y 15 (3.2) 13 (2.8)
and BP data for all participants and those in the Low LDL-C 13 (2.8) 22 (4.8)
Combination of prohibited medications 9 (1.9) 3 (0.6)
normal and limited literacy categories. The mean
High LDL-C 8 (1.7) 8 (1.7)
LDL-C was 106 mg/dL and the mean systolic BP was
Low HDL-C 7 (1.5) 0 (0.0)
127 mm Hg. Limited literacy participants had similar
Risk score too high 6 (1.3) 10 (2.2)
values with slightly lower LDL-C (97 mg/dL) and Diabetes not aged 40-75 y 5 (1.1) 4 (0.9)
slightly higher systolic BP (132 mm Hg). Documenta- High HDL-C 5 (1.1) 4 (0.9)
tion for the lipid data entered into the Web applica- Cardiac risk 3 (0.6) 2 (0.4)
tion were provided by 73.3% of participants and Procedure 3 (0.6) 3 (0.6)
documented BP data were provided by 32.6%. High triglycerides 2 (0.4) 0 (0.0)

Males aged 20-39 y with 2 (0.4) 3 (0.6)
PRIMARY ENDPOINT: OVERALL CONCORDANCE
LDL <160 mg/dL
BETWEEN PARTICIPANTS AND CLINICIANS. Table 3 Prior stroke 2 (0.4) 1 (0.2)
shows overall results for technology-assisted self- Warfarin 2 (0.4) 0 (0.0)
selection with and without the application of the pre- Liver disease 1 (0.2) 1 (0.2)
specified mitigation plan. Concordance between self- Pregnancy 1 (0.2) 0 (0.0)
selection and clinician assessment occurred in 481 Systolic BP high 1 (0.2) 0 (0.0)
Systolic BP low 1 (0.2) 0 (0.0)
(96.2%) of the 500 participants (95% CI: 94.1%-97.7%),
of whom 23 (4.6%) were deemed appropriate for Values are n (%). Data are results of the technology-assisted outcomes for the participants and
nonprescription statin treatment and 458 (91.6%) were clinicians. This table also presents the reasons for a “not right for you” outcome from both the
participant and clinician assessments.
deemed inappropriate. A concordant outcome without Abbreviations as in Table 2.
mitigation was achieved in 480 (96.0%) participants
assessment and on-site assessment. One participant

T A B L E 6 Comparison of Technology-Assisted Self-Selection Outcomes
Obtained by Participants and Clinicians at Research Sites
answered “yes” to a health assessment question
regarding whether they had a procedure on the heart
Technology-Assisted Self-Selection by Participant
and cardiac catheterization was prespecified as an
“OK to use” “Ask a doctor” “Not right for you” Total
acceptable answer. The most common errors included
Clinician-verified outcome
21 participants who incorrectly answered the ques-
“OK to use” 23 (88.5) 2 (100) 14 (3.0) 39
tion about taking another statin, 19 who had incorrect
“Ask a doctor” 0 0 1 (0.2) 1
“Not right for you” 3 (11.5) 0 457 (96.8) 460
LDL-C entries, 16 who had incorrect TG entries, and
Total 26 2 472 500 14 who had incorrect reporting of a family history of
cardiovascular disease.
Values are n (%) or n. The weighted kappa coefficient is 0.70 (95% confidence interval: 0.58-
Table 8 presents the secondary endpoint that
0.83), indicating substantial agreement.
evaluated the percentage of participants with
concordant entries for all self-selection questions.
common reasons for not qualifying included use of Overall, 405 (81.0%) participants (95% CI: 77.3-84.3)
cholesterol- or TG-lowering medications, a low answered all of the questions concordantly with
cardiovascular risk score or low-risk demographic clinicians after corrections were applied according to
characteristics, such as young women (Table 5). the prespecified mitigation plan.
Table 6 shows details for results obtained by partici-
pants compared with results obtained by clinicians for DISCUSSION
all possible outcomes. For the 3 of 26 participants who
received an “OK to use” result, clinician assessment Expanded access to statin could have a significant
resulted in a “not right for you” result. Two (0.4%) of positive public health impact given the benefits of
these 3 participants self-selected for nonprescription these drugs in primary prevention and the observa-
rosuvastatin therapy but were deemed ineligible by tion that as many as one-half of patients who are
the clinician due to incorrect self-evaluation of family guideline-eligible for statin therapy are not treated
history. An additional participant self-selected for (1,2,15-17). To expand access to statins, several
statin therapy with self-entered data yielding an attempts to provide OTC availability of these drugs
estimated 10-year risk of 15%, which was discordant have been undertaken, but all failed to achieve reg-
with clinician assessment that resulted in an esti- ulatory approval due to the inability to demonstrate
mated 10-year risk of 24%, indicating the need for a that consumers could appropriately self-select for
higher dose of rosuvastatin. The discrepancy in risk treatment. In the current study, a novel strategy us-
score was due to a systolic BP that was higher at the ing a Web application for self-selection was highly
research site than at home (132 mm Hg vs 100 mm Hg). successful with >95% concordance for the primary
In 14 (3.0%) cases, participant entries resulted in endpoint that compared results obtained by partici-
rejection that was discordant with clinician assess- pants and an independent clinician assessment
ment. These discordant rejections resulted from er- (Table 3). Using technology assistance, limited liter-
rors such as laboratory input mistakes or laboratory acy consumers also achieved a high concordance with
data above the acceptable limits of variability, incor- the clinician’s assessments (Table 4). This outcome is
rectly identifying as taking a statin, or failing to potentially important because limited literacy con-
recognize a family history of heart disease. A post hoc sumers may have less access to health care providers
analysis in the 431 participants without a history of and may benefit from enhanced efforts to deliver
cardiovascular disease (ie, primary prevention group) medications directly.
demonstrated that 88.0% were correct selectors and More than 90% of participants were deemed not
96.2% were correct rejecters. suitable for nonprescription rosuvastatin treatment,
ACCURACY OF PARTICIPANTS’ RESPONSES TO most commonly because they were already taking a
QUESTIONS. Table 7 shows summary results for the cholesterol-lowering medication or had low cardio-
overall accuracy of participants in responses to vascular risk (Table 5). The fact that the majority of
questions. There were a total of 108 participant participants were ineligible for treatment was ex-
entries that did not agree with the clinician assess- pected because the study was designed to enroll a
ment. Thirteen errors were corrected based on the wide range of participants, including those with
prespecified mitigation plan. Of these, 12 were due to cardiovascular disease. Such a study is pivotal in
a 1-year difference in age between participant and development, as inability to prevent treatment of
clinician because the participant had a birthday be- ineligible participants was a key failing of prior ef-
tween the date that they completed the online forts. Discordance was observed between trial
participants and clinicians for 19 potential rosuvas-

T A B L E 7 Data From the 500 Participants With Incorrect Self-Selection
tatin consumers with the largest number, 14 trial Entry Results by Question
participants, due to incorrect rejection (Table 6).
Total Incorrect Home Entries Incorrect
These discordant rejections (“not right for you”) or Home Entries Corrected Using Home Entries
“ask a doctor” outcomes resulted from participants Compared With Prespecified Compared With
Clinician at Research Mitigation Clinician Not
incorrectly reporting that they were taking a statin, Site (n ¼ 108) Plan (n ¼ 13) Corrected (n ¼ 95)
failing to recognize a family history of heart disease, Age 15 (3.0) 12 (2.4) 3 (0.6)
or laboratory input mistakes (Table 7). In the context Family history 14 (2.8) 0 (0.0) 14 (2.8)
of a Web application that will be used for access to a Risk score 7 (1.4) 0 (0.0) 7 (1.4)
nonprescription medicine, incorrect rejections BP medications 1 (0.2) 0 (0.0) 1 (0.2)
represent a lost opportunity for treating appropriate Cholesterol medication reaction 2 (0.4) 0 (0.0) 2 (0.4)
hs-CRP class ($2.0 or <2.0) 1 (0.2) 0 (0.0) 1 (0.2)
consumers but pose no inherent risks from treat-
Cyclosporine 4 (0.8) 0 (0.0) 4 (0.8)
ment. Similarly, incorrect “ask a doctor” outcomes
Diabetes 1 (0.2) 0 (0.0) 1 (0.2)
pose no inherent risk because access to medication Sex 1 (0.2) 0 (0.0) 1 (0.2)
would be denied without a physician discussion. HDL-C 10 (2.0) 0 (0.0) 10 (2.0)
Conversely, incorrect self-selection could raise Prior heart attack 1 (0.2) 0 (0.0) 1 (0.2)
safety concerns by allowing access to drug by con- Kidney disease 1 (0.2) 0 (0.0) 1 (0.2)
sumers who are unlikely to benefit or who may be at LDL-C 19 (3.8) 0 (0.0) 19 (3.8)
Other statins 21 (4.2) 0 (0.0) 21 (4.2)
risk for adverse events. In the current study, no in-
Pregnant 1 (0.2) 0 (0.0) 1 (0.2)
dividuals at increased risk for adverse events self-
Procedure 1 (0.2) 1 (0.2) 0 (0.0)
selected for access to rosuvastatin 5 mg using the
Race 14 (2.8) 0 (0.0) 14 (2.8)
Web application. The 3 individuals who could have Smoker 1 (0.2) 0 (0.0) 1 (0.2)
accessed drug through the self-selection, but were Systolic BP 2 (0.4) 0 (0.0) 2 (0.4)
deemed ineligible by the clinician, would have actu- Total cholesterol 7 (1.4) 0 (0.0) 7 (1.4)
ally benefited from treatment. Two trial participants Triglycerides 16 (3.2) 0 (0.0) 16 (3.2)
who received an “OK to use” outcome had 10-year Warfarin 4 (0.8) 0 (0.0) 4 (0.8)
risk scores of 6% and 13%, respectively. The discor-

Values are n (%). All percentages are calculated based on the 500 enrolled participants.
dance with the clinician outcomes occurred because hs-CRP ¼ high-sensitivity C-reactive protein; other abbreviations as in Table 2.
they incorrectly identified as having a family history
of cardiovascular disease. The third participant was
discordant because the BP at home was lower than self-selection ranged from 52%-72% (2,6,7). The 20 mg
on-site, resulting in a risk score of 15% at home and lovastatin program was designed to demonstrate a
24% on-site. point estimate for self-selection >80% without
Previous efforts have attempted to achieve but defining a lower 95% bound of the CI (4). Because the
were unsuccessful in achieving regulatory approval current trial used technology-assisted self-selection
for OTC statins, including pravastatin 10 mg, lova- to potentially improve results, the current study was
statin 10 mg and 20 mg, and atorvastatin 10 mg (3-7). powered to rule out a substantially more stringent
The pravastatin and lovastatin programs were eval- 85% lower bound of the 95% CI and actually achieved
uated and rejected by U.S. Food and Drug Adminis- a lower bound of 94.1%.
tration Advisory Committees (3,6). For pravastatin, The failure of previous OTC statin programs can
the concerns centered on sufficient efficacy and the be attributed to 2 critical factors. The first is the use
ability of consumers to make appropriate self- of the drug facts label alone to guide consumers on
selection determinations. For lovastatin, a 10-mg
dose was rejected due to concerns about sufficient T A B L E 8 Secondary Endpoint: Overall Identical Data Entry Comparing
Participants With Physicians With and Without Prespecified
efficacy, the potential for drug-drug interactions, and
Mitigations (N ¼ 500)
appropriate patient self-selection (6). A second
attempt sought approval for a 20-mg dose but was Total Correct Total Correct
(After Mitigation) (Prior to Mitigation)
also rejected because of continued concerns about
Outcome n (%) 95% CI (%) n (%) 95% CI (%)
drug-drug interactions and appropriate self-selection
Overall identical 405 (81.0)a 77.3-84.3 392 (78.4)a 74.5-81.9
(18). The atorvastatin program was abandoned self-selection entry
following an actual use study that demonstrated poor
a
compliance with LDL-C testing requirements and Number and percent of the 500 participants whose data entry exactly agreed with clinicians for
all questions (detailed breakdown of incorrect entries shown in Table 6).
adherence to warnings in the drug facts label (7). In CI ¼ confidence interval.
the lovastatin and pravastatin OTC programs, correct
self-selection decisions and use. Studies suggest that medication. For prescription use, adherence to statin
the majority of consumers do not read the drug facts treatment is relatively poor, with less than one-half of
labels for OTC products prior to initiating treatment patients still receiving therapy 1 year after initiation
or during actual use (19-21). Additionally, while the (17). This current approach provides continuing con-
drug facts label was designed to help consumers use tact with the consumer to encourage compliance and
OTC drugs appropriately, many struggle to correctly enable ongoing confirmation of eligibility.
apply the instructions. The second issue for previous STUDY LIMITATIONS. The study was, by design,
statin programs was the choice of statin for nonpre- intended to study a broad population that included
scription use. The 5-mg dose of rosuvastatin was few patients who were eligible for nonprescription
chosen for this development program because it is statin treatment. Although the current study helps
more efficacious in lowering LDL-C than other statin determine whether ineligible consumers can be
entry doses with safety demonstrated in a clinical successfully excluded from treatment, it does not
trial that used a 4-fold higher dose (22). Studies have provide a robust assessment of technology-assisted
demonstrated a >40% reduction in LDL-C with the self-selection in identifying patients most likely to
5-mg dose and a low incidence of adverse effects benefit from nonprescription statin therapy. Future
(23,24). The projected >40% LDL-C reduction is studies will be required that enroll participants
important given the potential for patients to occa- specifically selected for eligibility to receive nonpre-
sionally miss doses. Studies have demonstrated a scription rosuvastatin to determine whether tech-
meaningful reduction in cholesterol even when the nology can improve accurate selection of these
drug is taken only 3 days per week (25). The 5-mg consumers and assess their adherence to treatment.
dose has a safety profile appropriate for use in a
nonprescription setting with a low potential for CONCLUSIONS
drug-drug interactions (24).
The current Web-based application was developed A novel approach using technology-assisted self-
using the regulatory guidance for Software as a selection was successful in overcoming a significant
Medical Device. Such an approach was designed to barrier to the development of a nonprescription statin
overcome the problem of poor compliance with the by ensuring that a high percentage of ineligible con-
drug facts label that undermined prior efforts at reg- sumers were denied access and that only those with
ulatory approval for OTC statins. The self-selection an appropriate level of risk were deemed eligible to
component of the Web application incorporates 3 access this medication.
critical elements to determine eligibility for nonpre- ACKNOWLEDGMENTS The authors thank Julie Aker,
scription rosuvastatin. The first component is use of BS (Concentrics Research); Brooke Purvis, MS, and
the American College of Cardiology/American Heart Lori Christman, PhD (StatKing); and James Eudicone,
Association risk calculator pooled cohort equations to PhD (AstraZeneca).
calculate cardiovascular risk (26). The second is use of
FUNDING SUPPORT AND AUTHOR DISCLOSURES
the 2018 Cholesterol Treatment Guidelines, and the
final component is the proposed drug facts label (11). This work was funded by AstraZeneca Pharmaceuticals. The sponsor
The consumer navigates a series of screens until the was centrally involved in the study design and reviewed the manu-
software generates an eligibility outcome. Several script, but the final decisions on content were made by the lead
author after consultation with the other academic coauthors (Dr
examples of screens are shown in the Central
Nissen). Dr Nissen reports that the Cleveland Clinic Center for Clinical
Illustration. Research has received funding to perform clinical trials from AbbVie,
For this development program, participants who AstraZeneca, Amgen, Eli Lilly, Esperion, Medtronic, MyoKardia,
meet the 2018 Cholesterol Treatment Guidelines for Novartis, Pfizer, and Silence Therapeutics—he is involved in these
clinical trials, but receives no personal remuneration for his partici-
use of a moderate-intensity statin without a contrain-
pation; and has served as a consultant for many pharmaceutical
dication would be permitted to make an online pur- companies but requires them to donate all honoraria or consulting
chase with rosuvastatin 5 mg shipped directly to them. fees directly to charity so that he receives neither income nor a tax
The medication would not be available on the shelves deduction. Dr Hutchinson and Ms Morris are employees of AstraZe-
neca. Dr Wang has received research grants to the Duke Clinical
of pharmacies or stores where consumers have the
Research Institute from Abbott, AstraZeneca, Bristol Myers Squibb,
potential to make incorrect self-selection decisions. Boston Scientific, CryoLife, Chiesi, Merck, Portola, and Regeneron;
Renewal reminders are provided by email prior to and has received consulting honoraria from AstraZeneca, Bristol
Myers Squibb, CryoLife, and Novartis. Dr Ballantyne has received
depletion of drug supply with consumers directed to
grants to his institution from Akcea, Amgen, Esperion, Ionis, Novar-
the Web application to take a short health reassess- tis, Regeneron, and the American Heart Association; and has served
ment to confirm they are still eligible for treatment. as a consultant for Althera, Amarin, Amgen, Arrowhead, AstraZeneca,
Those who remain eligible are allowed to reorder Corvidia, Esperion, Genentech, Gilead, Matinas BioPharma Inc., New
Amsterdam, Novartis, Novo Nordisk, Pfizer, Regeneron, and Sanofi-

PERSPECTIVES
Synthelabo. Ms Travis, Dr Miller, and Ms Hynson are employees of
Concentrics Research. Dr Ridker has reported that during the course
of this project he received unrelated investigator-initiated research
COMPETENCY IN SYSTEMS-BASED PRACTICE: Consumers
grant support from Novartis, Kowa, Amarin, Pfizer, and the National
Heart, Lung, and Blood Institute; and has served as a consultant to
using a Web-based application to determine eligibility for
Corvidia, Novartis, Flame, Agepha, Inflazome, AstraZeneca, Janssen, nonprescription rosuvastatin drew conclusions concordant with
Civi Biopharm, SOCAR, Novo Nordisk, Uptton, Omeicos, and Boeh- those made by clinicians.
ringer Ingelheim. Ms Wolski has reported that she has no relation-
ships relevant to the contents of this paper to disclose.
TRANSLATIONAL OUTLOOK: Future studies should assess
the utility of technology-assisted self-selection in populations
ADDRESS FOR CORRESPONDENCE: Dr Steven E.
with a high likelihood of eligibility for nonprescription statin
Nissen, Cleveland Clinic Heart and Vascular Institute,
therapy, including long-term clinical outcomes.
9500 Euclid Avenue, Cleveland, Ohio 44195, USA.
E-mail: nissens@ccf.org.
REFERENCES
1. Navar AM, Wang TY, Li S, Robinson JG, et al. 9. National Library of Medicine. FDAAA 801 and archive-it.org/7993/20170405052057/https://
Lipid management in contemporary community the Final Rule. Accessed July 30, 2021. https:// www.fda.gov/ohrms/dockets/ac/07/briefing/2
practice: results from the Provider Assessment of www.clinicaltrials.gov/ct2/manage-recs/fdaaa 007-4331b1-01-FDA.pdf
Lipid Management (PALM) Registry. Am Heart J. 19. King JP, Davis TC, Bailey SC, et al. Developing
10. Davis TC, Crouch MA, Long SW, et al. Rapid
2017;193:84–92. consumer-centered, nonprescription drug label-
assessment of literacy levels of adult primary care
patients. Fam Med. 1991;23:433–435. ing: a study in acetaminophen. Am J Prev Med.
2. Bradley CK, Wang TY, Li S, et al. Patient-re-
2011;40:593–598.
ported reasons for declining or discontinuing 11. Grundy SM, Stone NJ, Bailey AL, et al. 2018
statin therapy: insights from the PALM registry. 20. Cryer B, Barnett MA, Wagner J, Wilcox CM.
AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/
J Am Heart Assoc. 2019;8:e011765. Overuse and misperceptions of nonsteroidal anti-
APhA/ASPC/NLA/PCNA guideline on the manage-
inflammatory drugs in the United States. Am J
3. Food and Drug Administration. Joint meeting of ment of blood cholesterol: a report on the Amer-
Med Sci. 2016;352:472–480.
the Nonprescription DFLs Advisory Committee and ican College of Cardiology/American Heart
Association Task Force on Clinical Practice Guide- 21. Caitlin JR, Brass EP. The effectiveness of
Endocrinological and Metabolic Drugs Advisory
lines. J Am Coll Cardiol. 2019;73:3168–3209. nonprescription drug labels in the united states:
Committee, Pravachol NDA 21-198, pravastatin
insights from recent research and
sodium [PowerPoint slides]. 2000. Accessed July 12. Dumenci L, Matsuyama RK, Kuhn L, Perera RA,
opportunities for the future. Pharmacy (Basel).
30, 2021. https://wayback.archive-it.org/7993/2 Siminoff LA. On the validity of the Rapid Estimate
2018;6:119.
0170404111314/https://www.fda.gov/ohrms/ of Adult Literacy in Medicine (REALM) scale as a
dockets/ac/00/slides/3622s2.htm measure of health literacy. Commun Methods 22. Ridker PM, Danielson E, Fonseca FA, et al,
Meas. 2013;7:134–143. JUPITER Study Group. Rosuvastatin to prevent
4. Melin JM, Stuble WE, Tipping RW, et al. A
vascular events in men and women with elevated
Consumer Use Study of Over-the-Counter Lova- 13. Contois JH, Warnick GR, Sniderman AD. Reliability
C-reactive protein. N Engl J Med. 2008;359:2195–
statin (CUSTOM). Am J Cardiol. 2004;94:1243– of low-density lipoprotein cholesterol, non-high-
2207.
1248. density lipoprotein cholesterol, and apolipoprotein
B measurement. J Clin Lipidol. 2011;5:264–272. 23. Blasetto JW, Stein EA, Brown WV, Chitra R,
5. Brass EP, Vassil T, Replogle A, et al.
Raza A. Efficacy of rosuvastatin compared with
Can consumers self-select for appropriate use 14. Staessen JA, Bieniaszewski L, O’Brien E, et al. other statins at selected starting doses in hyper-
of an over-the-counter statin? The Self Evalu- Nocturnal blood pressure fall on ambulatory cholesterolemic patients and in special population
ation of Lovastatin to Enhance Cholesterol monitoring in a large international database. Hy- groups. Am J Cardiol. 2003;91:3C–10C.
Treatment Study. Am J Cardiol. 2008;101:1448– pertension. 1997;29:30–39.
1455. 24. Shepherd J, Hunninghake DB, Stein EA, et al.
15. Virani SS, Alonso A, Benjamin EJ, et al. Heart The safety of rosuvastatin. J Cardiol. 2004;94:
6. Food and Drug Administration, Merck Research Disease and Stroke Statistics-2020 Update: a 882–888.
Labs. Joint meeting of the Nonprescription Drugs report from the American Heart Association. Cir-
25. Mampuya WM, Frid D, Rocco M, et al. Treat-
Advisory Committee and Endocrinological and Meta- culation. 2020;141:e139–e596.
ment strategies in patients with statin intolerance:
bolic Drugs Advisory Committee, Mevacor NDA 21-
16. Cholesterol Treatment Trialists’ (CTT) Collab- the Cleveland Clinic experience. Am Heart J.
213, lovastatin. 2000; 2000. Accessed July 30,
orators, Mihaylova B, Emberson J, et al. The ef- 2013;166:597–603.
2021. https://wayback.archive-it.org/7993/2
fects of lowering LDL cholesterol with statin
0170404111228/https://www.fda.gov/ohrms/ 26. Goff Jr DC, Lloyd-Jones DM, Bennett G, et al.
therapy in people at low risk of vascular disease:
dockets/ac/00/slides/3622s1.htm 2013 ACC/AHA guideline on the assessment of
meta-analysis of individual data from 27 rando-
cardiovascular risk: a report of the American Col-
7. Actual Use Trial of Atorvastatin Calcium 10 mg. mised trials. Lancet. 2012;380:581–590.
lege of Cardiology/American Heart Association
A multicenter, actual use trial in a simulated over- 17. Deshpande S, Quek RG, Forbes CA, et al. Task Force on Practice Guidelines. J Am Coll Car-
the-counter environment of atorvastatin calcium A systematic review to assess adherence and diol. 2014;63:2935–2959.
10 mg. ClinicalTrials.gov identifier: NCT01964326. persistence with statins. Curr Med Res Opin.
Accessed July 30, 2021. https://clinicaltrials.gov/ 2017;33:769–778.
ct2/show/NCT01964326
18. Joint session of the Nonprescription Drugs
KEY WORDS consumer self-selection,
8. Policy for Device Software Functions and Mo- Advisory Committee and the Endocrinology and
nonprescription statins, Web application
bile Medical Applications: Guidance for Industry Metabolic Drugs Advisory Committee for NDA 21-
and Food and Drug Administration Staff. Accessed 213 proposing over-the-counter Mevacor TM
July 30, 2021. https://www.fda.gov/media/8 (lovastatin) 20 mg. [Advisory Committee Briefing A PP END IX For the study protocol, please
0958/download Book]. Accessed July 30, 2021. https://wayback. see the online version of this paper.
EDITORIAL COMMENT
Should Cardiovascular Preventive

Therapy Be Over-the-Counter?*
Neha J. Pagidipati, MD, MPH,a Eric D. Peterson, MD, MPHb
C ardiovascular disease (CVD) remains the

leading cause of mortality worldwide (1).
Although there are multiple CVD preventive
therapies available that can significantly lower CV
informed decisions about CV preventive therapies
without the direct involvement of a health care
professional.
SEE PAGES 1114 AND 1193

risk, routine use of these in clinical practice is far
from ideal. Only one-third of persons recommended In this issue of the Journal, 2 studies address this
for primary prevention statin therapy are on an crucial question. Jacobsen et al (4) performed a sur-
appropriately-dosed statin (2). There are also wide so- vey of patients with or at high risk for CVD in the
cioeconomic and racial disparities in the prescription United States and in Ireland to assess their under-
of CVD prevention therapies (2). Clearly, the current standing of the risks, benefits, and indications for
approach of relying on a fragmented health care sys- aspirin therapy. In the United States, where aspirin is
tem to deliver consistent CV preventive care is not already an OTC medication, 46% of participants re-
working for most patients. ported taking aspirin regularly. In contrast, in
A potential alternative to this clinician-centric Ireland, where aspirin is largely a prescribed therapy,
model would be to directly empower patients to get only 26% of participants were on aspirin. Although
CV preventive therapies without a physician’s pre- this correlation supports the notion that making a
scription. The concept of making certain CV preven- medication OTC can promote higher use, the authors
tive therapies “over-the-counter” (OTC) has been also raise concern about whether these choices were
tried. Aspirin is OTC in several countries, and some “informed” and safe. For example, only one-half
have proposed similar consideration for statin ther- (53%) of the surveyed participants understood that
apy (3). An OTC prevention approach has several po- aspirin was not routinely indicated for primary pre-
tential advantages: 1) reducing the burden and cost of vention. Additionally, those surveyed had a very low
obtaining these medications; 2) limiting disparities ability to accurately assess their own perceived CVD
resulting from differential access to care; and 3) risk vs actual measured risk (Kappa statistic 0.10).
increasing self-efficacy and therapeutic adherence. Based on these data, the authors conclude that
On the other hand, an OTC approach may result in widespread availability of OTC aspirin therapy in the
overuse or misuse of medication and lead to safety United States may be unsafe.
concerns. This raises a key health policy question of Readers of this study should note that the survey
whether lay individuals can make appropriate, was limited in sample size (n ¼ 300) and came from
only 2 hospitals. Thus, generalization of these find-
ings to the whole of the U.S. or European populations
should be done with caution. Further, the study did
reflect the views of the authors and do not necessarily represent the not provide analysis of responses among those who
views of JACC or the American College of Cardiology. were or were not on aspirin therapy. Thus, we do not
From the aDuke Clinical Research Institute, Durham, North Carolina, know whether the patients’ estimate of their CVD risks
USA; and the bUniversity of Texas Southwestern, Dallas, Texas, USA. or understanding of the risks of therapy influenced
The authors attest they are in compliance with human studies commit-
their selection. Finally, the study did not determine
tees and animal welfare regulations of the authors’ institutions and Food
whether patients who obtained aspirin OTC did so
appropriate. For more information, visit the Author Center. after discussion with their clinicians (ie, advised

JACC VOL. 78, NO. 11, 2021 Pagidipati and Peterson 1125
SEPTEMBER 14, 2021:1124–1126 Safety of Self-Prescribed CV Medications
vs self-directed). These limitations notwithstanding, Together, these 2 studies demonstrate both the
this study by Jacobsen et al (4) shows that OTC op- problems with and the potential for OTC therapies.
tions increase access, but it also raises an important The study by Jacobsen et al (4) clearly indicates that
concern about whether lay individuals have an ac- without assistance, patients’ innate assessment of
curate idea of their own CVD risk and/or whether their CVD risk is poor, putting them at risk for making
they understand the indications for therapy. inappropriate therapeutic decisions. Yet, the study by
In another paper in this issue of the Journal, Nissen Nissen et al (5) demonstrates that patients can accu-
et al (5) suggest that technology may be an exciting rately report their medical histories and can use a
option to support patient-led decision making about self-directed app to reach recommendations that are
preventive therapies. Rather than expecting patients concurrent with those of clinicians.
to have existing knowledge about statins, the authors Where do we go from here? First and foremost, it is
evaluated whether a technology-assisted decision aid clear that new innovations to improve cardiovascular
could be used to support patients’ determination of disease prevention are desperately needed. Digitally-
their need for statin therapy. The application first aided “self-service” has become an integral part of
asked patients a series of health questions related to many aspects of our modern-day lives. It is now
their medical history, medication use, and test re- possible to shop, invest, do taxes, and select insur-
sults. Based on this information, the application ance using self-guided computer-aided programs.
determined whether patients met eligibility for statin Relative to interacting with humans, these online
therapy (specifically rosuvastatin 5 mg). The in- experiences may lose a bit in “personalization,” but
vestigators then evaluated the concordance between are a win in turns of convenience, cost, objectivity,
the self-directed decision tool and decisions made by and scalability.
an online physician assessment. The study found that In a similar fashion, health care too is becoming
of the 500 participants studied, the self-directed de- more virtual and self-directed. Patients now have
cision-aid matched clinician assessments in 96.2% wider online access to their medical records as well as
(95% CI: 94.1%-97.7%) of cases. In only 3 patients, the options to visit their clinicians virtually. Patients can
app suggested statin therapy when clinicians deemed receive a prescription for many medications after a
it not to meet guidelines; in these, none put the pa- brief review from a third-party online clinician. Is it
tient at increased risk for adverse events. The study that much of a stretch to imagine that patients could
also demonstrated that patients were reasonably ac- further access cardiovascular preventive therapies
curate in recalling their medical and medication his- after guidance from an online decision aid? To date,
tories (with 81% of self-reported responses matching medicine has paternalistically controlled
those determined by the clinician online interview). prescription-writing by clinicians as one of the last
Importantly, the authors also found that the high human hold-outs against technology. Yet, the cost of
concordance between the self-directed tool and the clinicians’ control over their patients’ medication
clinician assessment did not differ by patients’ base- access has been an overall failure to deliver high-
line health literacy. quality, preventive care at scale.
This study by Nissen et al (5) is notable for evalu- Before we leap into this new world of technology-
ating an innovative technology in a diverse popula- aided self-directed care, however, we need both
tion. It also carefully analyzed overall decision optimized digital decision-aids and larger imple-
concordance, reasons for discordance, and mitigation mentation studies. On the technology side, although
strategies. Readers should, however, note that the Nissen et al (5) found that patients were reasonable at
study population was self-selected and may differ reporting their health histories, the accuracy,
from the general population. Additionally, the study completeness, and ease of collection of heath infor-
cohort was relatively young and few had actual in- mation could be improved if the tools had been digi-
dications for primary prevention statin therapy. The tally integrated with the participants’ electronic
study tool also considered only a single statin medical record. Similarly, the “output” from decision-
strength and did not consider the complex algorithms support needs to be evaluated. Patient assessment of
of other lipid therapy options beyond statins. Last, their own risk can vary markedly depending on what
because statins are not currently OTC, the study information is provided and how these data are dis-
could not assess the impact of the tool on actual pa- played (6). Additionally, although the study by Nissen
tient initiation of or adherence to a statin. Despite et al (5) indicated that even those with low health lit-
these limitations, the study does support the poten- eracy can use their tool, there is always a concern that
tial of technology-assisted decision aids to facilitate some will be left behind in the digital divide. Thus,
accurate patient assessments for OTC statin therapy. widespread user acceptability testing is needed.
1126 Pagidipati and Peterson JACC VOL. 78, NO. 11, 2021
Safety of Self-Prescribed CV Medications SEPTEMBER 14, 2021:1124–1126
Finally, and most importantly, we need larger real- resources, the potential upside gained from better
world studies. Regulators require huge randomized cardiovascular prevention could be enormous. The
clinical trials evaluating the safety and efficacy of time has come for us to learn how to safely empower
drugs to prevent CVD before they can be marketed. patients to self-prescribe preventive therapies.
Yet, most implementation studies (like the 2 current
ones) that evaluate how to get these therapies used
appropriately in practice tend to be small, select, and Dr Pagidipati has received research support from Amgen, AstraZe-
short-lived. Implementation studies in the future neca, Boehringer Ingelheim, Eggland’s Best, Eli Lilly, Novartis,
need to be part of the development process for new Novo Nordisk, Sanofi, and Verily Life Sciences; and has received
consulting support from AstraZeneca, Boehringer Ingelheim, Eli Lilly,
therapies and conducted in large and diverse pop-
and Novo Nordisk. Dr Peterson has received research support from
ulations. These studies also need to assess outcomes: Amgen, Janssen, Bristol Myers Squibb, and Esperion; and has
not only whether the decision-aid successfully guides received consulting support from Janssen, Boehringer Ingelheim, and
the right patients to initiate guideline-based thera- Cerner.
pies, but also whether such engagement helps keep

patients on therapy over time. These future imple- ADDRESS FOR CORRESPONDENCE: Dr Neha J.
mentation studies should also assess for adverse ef- Pagidipati, Duke Clinical Research Institute, 300
fects resulting from using self-guided patient West Morgan Street, Durham, North Carolina 27701,
decision aids (such as overuse, safety concerns, and USA. E-mail: neha.pagidipati@duke.edu. Twitter:
patient anxiety). Although such studies will take @ericpetersonMD.
REFERENCES
1. Roth GA, Mensah GA, Johnson CO, et al. Global the-counter statin? The Self Evaluation of Lova- for nonprescription statin therapy. J Am Coll Car-
burden of cardiovascular diseases and risk factors, statin to Enhance Cholesterol Treatment Study. diol. 2021;78:1114–1123.
1990-2019: update from the GBD 2019 Study. Am J Cardiol. 2008;101:1448–1455.
6. Navar AM, Wang TY, Mi X, et al. Influence of
J Am Coll Cardiol. 2020;76:2982–3021.
4. Jacobsen AP, Lim ZL, Chang B, et al. cardiovascular risk communication tools and pre-
2. Navar AM, Wang TY, Li S, et al. Lipid manage- A transatlantic comparison of patient-reported sentation formats on patient perceptions and
ment in contemporary community practice: results access to and use of aspirin in contemporary preferences. JAMA Cardiol. 2018;3:1192–1199.
from the Provider Assessment of Lipid Management preventive cardiology. J Am Coll Cardiol. 2021;78:
(PALM) Registry. Am Heart J. 2017;193:84–92. 1193–1195.
3. Brass EP, Vassil T, Replogle A, et al. Can con- 5. Nissen SE, Hutchinson HG, Wang TY, et al. KEY WORDS consumer self-selection,
sumers self-select for appropriate use of an over- Technology-assisted self-selection of candidates nonprescription statins, Web application
ª 2021 THE AUTHORS. PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN
COLLEGE OF CARDIOLOGY FOUNDATION. THIS IS AN OPEN ACCESS ARTICLE UNDER
THE CC BY LICENSE (http://creativecommons.org/licenses/by/4.0/).
Innate Lymphoid Cells Promote Recovery

of Ventricular Function After
Myocardial Infarction
Xian Yu, MD, PHD,a,b,* Stephen A. Newland, PHD,a,* Tian X. Zhao, MD, PHD,a Yuning Lu, MD, PHD,a
Andrew S. Sage, PHD,a Yanyi Sun, MD, PHD,c Rouchelle S. Sriranjan, PHD,a Marcella K.L. Ma, PHD,d
Brian Y.H. Lam, PHD,d Meritxell Nus, PHD,a James E. Harrison, BSC,a Simon J. Bond, PHD,e Xiang Cheng, MD, PHD,b
Jean-Sébastien Silvestre, PHD,c James H.F. Rudd, MD, PHD,a Joseph Cheriyan, MBCHB, MA,e,f Ziad Mallat, MD, PHDa,c
ABSTRACT
BACKGROUND Innate lymphoid cells type 2 (ILC2s) play critical homeostatic functions in peripheral tissues. ILC2s
reside in perivascular niches and limit atherosclerosis development.
OBJECTIVES ILC2s also reside in the pericardium but their role in postischemic injury is unknown.
METHODS We examined the role of ILC2 in a mouse model of myocardial infarction (MI), and compared mice with or
without genetic deletion of ILC2. We determined infarct size using histology and heart function using echocardiography.
We assessed cardiac ILC2 using flow cytometry and RNA sequencing. Based on these data, we devised a therapeutic
strategy to activate ILC2 in mice with acute MI, using exogenous interleukin (IL)-2. We also assessed the ability of low-
dose IL-2 to activate ILC2 in a double-blind randomized clinical trial of patients with acute coronary syndromes (ACS).
RESULTS We found that ILC2 levels were increased in pericardial adipose tissue after experimental MI, and genetic
ablation of ILC2 impeded the recovery of heart function. RNA sequencing revealed distinct transcript signatures in ILC2,
and pointed to IL-2 axis as a major upstream regulator. Treatment of T-cell–deficient mice with IL-2 (to activate ILC2)
significantly improved the recovery of heart function post-MI. Administration of low-dose IL-2 to patients with ACS led to
activation of circulating ILC2, with significant increase in circulating IL-5, a prototypic ILC2-derived cytokine.
CONCLUSIONS ILC2s promote cardiac healing and improve the recovery of heart function after MI in mice.
Activation of ILC2 using low-dose IL-2 could be a novel therapeutic strategy to promote a reparative response after MI.
(J Am Coll Cardiol 2021;78:1127–1142) © 2021 The Authors. Published by Elsevier on behalf of the American College of Car-
diology Foundation. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
T he mechanisms that lead to heart failure

post-myocardial infarction (MI) are initiated
very early after ischemic injury, triggering
waves of immune responses that play critical roles
current
remodeling
therapies
directly
aimed
target
at
the
improving
immune
around the time of ischemic injury (2). Therefore,
important advances in therapeutic management
cardiac
system
in tissue healing and remodeling (1). None of the could be expected from a better understanding
From the aDepartment of Medicine, Division of Cardiovascular Medicine, University of Cambridge, Cambridge, United Kingdom;
Listen to this manuscript’s b
Department of Cardiology, Union Hospital, Tongji, Medical College, Huazhong University of Science and Technology, Wuhan,
audio summary by China; cUniversité de Paris, PARCC, INSERM, F-75015 Paris, France; dThe Wellcome Trust-MRC Institute of Metabolic Science-
Editor-in-Chief Metabolic Research Laboratories, University of Cambridge, Cambridge, United Kingdom; eDivision of Experimental Medicine
Dr. Valentin Fuster on and Immunotherapeutics, University of Cambridge, Cambridge, United Kingdom; and the fCambridge Clinical Trials Unit, Cam-
JACC.org. bridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom. *Drs Yu and Newland contributed equally to this
work.
Manuscript received May 14, 2021; revised manuscript received July 6, 2021, accepted July 13, 2021.
ISSN 0735-1097 https://doi.org/10.1016/j.jacc.2021.07.018

1128 Yu et al. JACC VOL. 78, NO. 11, 2021
ILC2 Promotes Healing After Myocardial Infarction SEPTEMBER 14, 2021:1127–1142
ABBREVIATIONS and early targeting of the post-ischemic im- regulation of the immune and healing response post-
AND ACRONYMS mune response. MI remains unexplored.
During the initial inflammatory phase after
ACS = acute coronary
ischemic injury, the rapid infiltration of METHODS
syndrome
neutrophils to the infarct area begins the
Arg1 = Arginase 1
process of clearing the necrotic tissue. This is EXPERIMENTAL MICE. All work was conducted under
IL = interleukin
aided by Ly6C hi
monocytes recruited through UK Home Office project license regulations after
ILC2 = type 2 innate lymphoid
secretion of chemokines such as CCL2 and approval by the Ethical Review Committee of the
cells
CCL7, which in turn potentially differentiate University of Cambridge. The following mouse
LAD = left anterior descending
into activated macrophages. The macro- strains were used: C57BL/6 (in house), Rag2 /(Jax),
MI = myocardial infarction
phages are further conditioned during the Rorafl/fl Cd127 cre Ldlr / (ILC2 KO), and Rora fl/fl Cd127WT
PcAT = peri-cardial adipose
subsequent regulatory phase (3), shifting the Ldlr / (ILC2 WT ) (14,17,18). All experimental mice
tissue
balance from proinflammatory classical acti- were females.
RNAseq = RNA sequencing
vation toward the alternative activation LEFT ANTERIOR DESCENDING CORONARY ARTERY
Treg = regulatory T cells
phenotype, which is more conducive for tis- LIGATION MODEL. Heart failure post-MI is mostly due
sue repair (4-7). Circulating CD4þ T cells infiltrate the to abnormal cardiac healing of large infarcts. There-
wounded heart early on and their presence helps to fore, we used the permanent left anterior descending
slew the macrophage phenotype toward the alterna- (LAD) ligation model (19). Expansion of ILC2s after MI
tive activation phenotype (8). During the resolution was achieved using IL-2/jes6-1 (14,20) injected intra-
phase, the activation of regulatory T cells (Tregs) is peritoneally at 1 m g per mouse 3 times a week for
beneficial, reducing effector T-cell activation, and 4 weeks. Soluble ST2 was a kind gift from Suzanne
promoting alternative macrophage activation and Cohen. Mice received 10 mg/kg soluble ST2 (sST2)
tissue repair (9,10). intravenously on day 0 (D0) and, D1 to D3 post-MI.
SEE PAGE 1143 ECHOCARDIOGRAPHY. Transthoracic echocardiog-
raphy was performed using Vevo 3100 with an MX400
Type 2 innate lymphoid cells (ILC2s) form a cluster
linear array transducer (VisualSonics) at 30 MHz.
of cells that secrete large amounts of type 2 cytokines,
Cardiac function was measured on M-mode images.
such as interleukin (IL)-5 and IL-13. Morphologically,
they are very similar to T cells but lack the ability to HISTOLOGICAL ANALYSIS. Heart cryostat sections
express recombined surface antigen receptors, and were stained with Masson trichrome to determine
thus function in an antigen-independent manner. scar size. Scar size (in %) was calculated as total
Activation of these cells is primarily through the cy- infarct circumference divided by total left ventricular
tokines IL-25 and IL-33, further supported by circumference. Collagen deposition was determined
lymphocyte-derived IL-2 (11). Although ILC2s are using Sirius red staining under polarized light (14)
present in secondary lymphoid tissues, they are 10- and quantified using Fiji (21) as % total area.
fold more prevalent in barrier tissues (12,13). In our
FLOW CYTOMETRY. Single-cell suspension for
previous work, we found that the adventitia of ar-
spleen, mesenteric lymph nodes, and pericardial and
teries and fat-associated lymphoid clusters in peri-
peri-gonadal white adipose tissue was generated (14).
vascular adipose tissue contain ILC2s, which expand
Cells were washed in phosphate-buffered saline (PBS)
in response to atherosclerosis (14). Further work (15)
1% fetal calf serum before labeling with desired
showed that adventitial stromal cells provide a sup-
antibody cocktails (Supplemental Table 1). Intracel-
portive niche in these tissues, supplying them with
lular antigens were detected using the IC Fix/perm kit
IL-33 and thymic stromal lymphopoietin. Perivascular
(BD Biosciences) as per the manufacturer’s in-
ILC2s are critical regulators of macrophage behavior,
structions. Cells were analyzed on an LSR-Fortessa
promoting alternative macrophage activation,
(BD Biosciences). Subsequent data were analyzed
thereby limiting the development of unstable
with FloJo X analysis software (FreeStar). Examples
atherosclerotic plaques, directly through the secre-
of each population gating hierarchy are provided in
tion of IL-13 (14). ILC2s also accumulate in pericardial
Supplemental Figure 1.
fat-associated lymphoid clusters and mediastinum,
and a population of IL-33-dependent ILC2s present in LOW-INPUT RNA SEQUENCING. A total of 100
human and mouse hearts (16) has been shown to ILC2s were sorted directly into single-cell lysis
expand in settings of myocarditis, pericarditis, and buffer (Takara Bio) and stored at 70o C until library
MI. However, the role of endogenous ILC2s in the preparation. The lysate was subjected to direct
JACC VOL. 78, NO. 11, 2021 Yu et al. 1129
SEPTEMBER 14, 2021:1127–1142 ILC2 Promotes Healing After Myocardial Infarction
oligo-dT-based reverse transcription and comple- tissue (GWAT) by flow cytometry (Supplemental
mentary DNA (cDNA) amplification using SMART-Seq Figure 1A). ILC2 populations in heart and PcAT
v4 Ultra Low Input RNA Kit (Takara Bio); 150 pg of (Supplemental Figure 2A) were proportionally higher
amplified cDNA was then used to generate than normally found in peripheral lymph nodes or
sequencing library using Illumina Nextera XT DNA spleen and observations of ST2 expression suggests
Library Preparation Kit, validated by an Agilent 2100 cardiac-infiltrating ILC2s are similar to tissue resident
Bioanalyzer and pooled at equal molar concentrations. ILC2s found in GWAT, whereas the lower ST2
Sequencing was performed on an Illumina HiSeq4000 expression in PcAT more closely resembles those
instrument (Single-end 50). An average of approxi- found in the periphery (Supplemental Figure 2B).
mately 15 million reads were obtained per sample. In the PcAT, the ILC2 population expanded 2-fold on
SEQUENCING BIOINFORMATICS. Raw sequence D3 before returning to presurgery levels (Figure 1A)
reads were aligned to the mouse GRCm38 genome while heart resident ILC2s were largely unchanged.
and gene level counts generated using STAR During this period, ST2 expression within the PcAT
(v2.5.0a). Differential expression was performed us- compartment increased stepwise with the
ing DeSeq2. Genes were considered differentially expansion of ILC2s, before regressing to its initial
expressed when logFC $1 and false discovery rate expression level by D7 (Figure 1A). Intracellular stain-
was <0.05. Canonical pathways enrichment predicted ing for proliferation marker Ki67 in PcAT-resident
upstream regulatory components and network anal- ILC2s showed that there was a 3-fold increase in the
ysis was performed using QIAGEN Ingenuity pathway proliferating population at the D3 time point
analysis software (QIAGEN IPA). RNA sequencing (Figure 1B), suggesting that the increase in cell number
(RNAseq) data will be available at the Gene Expres- at this time point was due in part to local proliferation.
sion Omnibus. In the context of MI, it is possible that release of IL-
33 from damaged myocardium was driving the pro-
SERUM CYTOKINE QUANTIFICATION. Mouse IL-5
liferation of ST2þ ILC2s in the inflamed tissues. To
and IL-13 were detected by enhanced sensitivity CBA
address this, mice were treated with 10 mg/kg sST2 or
flex set (R and D systems), following the manufac-
isotype control the day before MI surgery (D0), as well
turer’s instructions. Human cytokine analysis was
as D1 to D3, and killed on D3. Flow cytometric analysis
performed using a MesoScale Diagnostics
showed that in PcAT, there was a significant
Sector Imager 6000.
(P ¼ 0.026) decrease in the abundance of ILC2 present
CLINICAL TRIAL DATA. Samples from the LILACS in the group receiving sST2 compared with immuno-
(Low-dose interleukin-2 in patients with stable globulin G control, with a similar trend in heart-
ischemic heart disease and acute coronary syndrome; infiltrating cells, indicating that IL-33 likely drives
NCT03113773) clinical trial (22) permitted ILC2 the expansion of ILC2s after MI. These data taken
fluorescence-activated cell sorting analysis on banked together provide evidence that ILC2s are part of the
peripheral blood mononuclear cells (PBMCs) using a orchestrated response to cardiac tissue damage.
LSRII Fortessa cytometer. The trial received a favor-
able evaluation from the Greater Manchester Central GENETIC ABLATION OF ILC2 IMPEDES THE RECOVERY
Research Ethics Committee, UK (17/NW/0012). Circu- OF HEART FUNCTION AFTER MI. Previously, we have
lating eosinophil count was performed using a Sys- shown that genetic ablation of ILC2 in Rora fl/fl
mex automated analyzer. Cd127 cre mice was sufficient to increase atheroscle-
rotic burden during high-fat feeding (14). Using these
STATISTICAL ANALYSIS. Statistical analysis was
mice in the LAD ligation MI model, we observed a
performed using GraphPad Prism 7 (Graph Pad Soft-
trend toward a difference in mortality between ILC2
ware). Mann-Whitney U test was used for nonpara-
replete (ILC2 WT ) and ILC2 knockout (ILC2 KO) cohorts
metric datasets or 2-way analysis of variance where
(Figure 2A). The surviving mice were followed for
appropriate. P values are indicated in individual
4 weeks with echocardiography (Figure 2B), and dur-
figure legends. Error bars represent SEM.
ing this period, ILC2 KO mice had consistently poorer
RESULTS heart function compared with ILC2WT mice. By D14
post-MI, cardiac function (% ejection fraction) had
ILC2s EXPAND IN PERICARDIAL ADIPOSE TISSUE fallen by a mean of 41.04% vs a decrease of 30.03% in
AFTER MI IN AN ST2-DEPENDENT MANNER. On D0, ILC2 KO and ILC2 WT mice, respectively.
D1, D3, D5, and D7 post-MI, mice were killed and the Four weeks after MI, scar size was significantly
proportions of ILC2 were measured in heart, pericar- larger in in ILC2KO mice compared with controls
dial adipose tissue (PcAT), and perigonadal adipose (Figure 2C). Sirius red staining of the sections revealed
F I G U R E 1 ILC2s Proliferate in PcAT After MI
Heart
A Sham MI Day 3 MI Day 7 0.25
105 105 105 0.20
0.15
ILC2%
4 4
10 104 10
Heart
0.10
103 103 103
0.05
0 0 0 0.00
Sham D1 D3 D5 D7
Lineage FITC
0 103 104 105 0 103 104 105 0 103 104 105

Post-MI
105 105 105 PcAT

0.4
104 104 104 P = 0.021
0.3
PcAT
103
ILC2%
103 103
0.2
102
0 0
0 0.1
0 103 104 105 0 103 104 105 0 103 104 105 0.0
Sham D1 D3 D5 D7
ICOS APC Post-MI
PcAT ST2
105 105 105 1,500
P = 0.0022
104 104 104
KLRG1 PeCy7
1,000
ST2 MFI
103
PcAT
103 103
102 0 0 500
0 –10 3
–10 3
–103 0 103 104 105 –103 0 103 104 105 –103 0 103 104 105 0
Sham D1 D3 D5 D7
ST2 e710 Post-MI
B 105 105 105 80

ILC2 Proliferation
P = 0.0031
104 104 104 60
% Ki-67+
Ki67 Bv450
3
PcAT
10 3 10 103 40
0 0 0 20
–103 –103 –103
0
0 103 104 105 0 103 104 105 0 103 104 105 Sham D1 D3 D5 D7
KLRG1 Pe-Cy7 Post-MI
Flow cytometric analysis detects the expansion of ILC2 3 days post-MI (A) in heart (top) and pericardial adipose tissue accompanied with elevated ST2 surface
expression (bottom), and increased proliferation (Ki67) (B). ILC2 proliferation in the PcAT was inhibited by injections of sST2, less pronounced in the heart (C).
Representative images shown. Mann-Whitney U test. ILC2 ¼ type 2 innate lymphoid cells; MI ¼ myocardial infarction; PcAT ¼ pericardial adipose tissue; sST2 ¼ soluble
ST2.

C sST2 IgG
Heart
105 10 5
0.25
P = 0.17
104 104 0.20
Heart
ILC2 %
0.15
103 103
0.10
0 0 0.05
Lineage FITC
0.00
–103 0 103 104 105 –103 0 103 104 105
SST2 IgG
105 105 PcAT

0.10
104 P = 0.026
104 0.08
PcAT
ILC2 %
103 10 3 0.06
0.04
0 0
0.02
0 10 3
104
105
0 10 3
10 4
10 5 0.00
SST2 IgG
ICOS APC
that although the scars were larger, the per unit area of these cells receive during the period of inflammation
collagen in each section (Figure 2D) was significantly may shed light on cardioprotective regulatory path-
lower in ILC2 KO mice. Together, these findings suggest ways promoted by ILC2. We performed RNAseq
that ILC2s play a positive role in the early response to analysis of PcAT ILC2 cells sorted from D3 post-MI or
MI, slewing scar formation toward less intrusive from sham mice. Within the sequence data, tran-
remodeling in the proceeding recovery period. scripts typically high in ILC2 (eg, Rora , Gata3,
We performed detailed phenotypic analysis on the Supplemental Figure 4A) were well represented,
infiltrating cells by flow cytometry at D5 post-MI. In the whereas those that would be highly expressed in
absence of ILC2, circulating eosinophils and Tregs potential “contaminating” cell types (eg, Cd19,
were suppressed while inflammatory Ly6CHi mono- Cd11b, Supplemental Figure 4B) were relatively rare,
cytes were more abundant (Figure 3A). Within the PcAT confirming sort purity. Approximately 800 differen-
(Figure 3B), the populations are largely unchanged, tially expressed transcripts were detected with an
although in this tissue the macrophages express more adjusted P value of <0.05, of those 309 were of
surface CD206. Analysis of the heart-infiltrating known, characterized genes (Supplemental Figure 5).
myeloid cells (Figure 3C) showed there were no dif- Prominent ILC2 genes (Il5, Il13, St2, and Arg1) were
ferences in eosinophil or macrophage representation maintained or even increased after the onset of
in ILC2-deficient mice; however, the number of infil- inflammation (Figure 4A), and several genes of in-
trating Ly6C hi inflammatory monocytes were terest were also identified (Figure 4B). Changes to
increased and cardiac macrophages expressed less the IL-18 pathway through decreased expression of
CD206, reflective of a more inflammatory phenotype. Il18r1 and enhanced Il18bp (IL-18 decoy protein)
Heart-infiltrating T cells and Tregs were fewer in might suggest reinforcement of the protective type 2
number, although the proportion of Tregs within the pathway. Other genes involved in lymphocyte traf-
population was constant (Supplemental Figure 3). ficking signals during MI and remodeling, Cxcr4 and
Cxcl1, were increased. Subsequent in silico pathway
IL-2 AXIS AS A MAJOR UPSTREAM REGULATOR OF analysis identified the most likely pathways modi-
ILC2s IN POST-MI. ILC2s are more abundant in fied in ILC2 revolve around cell cycle, proliferation,
pericardial adipose tissue, which may in this and survival (Figure 4C). Most useful from this
instance be functioning as a reservoir for responding analysis was identifying genes linked to certain
cells that then traffic to the heart. The conditioning activation pathways, so-called upstream regulators
F I G U R E 2 Diminished Cardiac Function in ILC2 KO Mice After MI
A 100
75
% Survival
50
25
0
0 5 10 15 20 25 30
Time (Days)
ILC2 WT ILC2 KO
B ILC2 WT ILC2 KO
M-mode
80 P = 0.0025
Echocardiography
4 mm P = 0.0006
60
EF (%)
0.1s 40
20
0
B-mode WT KO WT KO WT KO
Baseline D14 D28
3 mm
C Masson’s trichrome
100
P = 0.036
ILC2 WT
80
(% Midline)
Scar Size
60
40
20
0
ILC2 KO
KO
W
C2
C2
IL
IL
D Collagen Sirius Red

1,000,000
P = 0.0104
ILC2 WT
800,000
Area/PM2
600,000
400,000
200,000
ILC2 KO
0
T
KO
W
C2
C2
IL
IL

F I G U R E 3 Alterations to the Myeloid Compartment Following MI
P = 0.0022 P = 0.0455
A 8 3 15 24 8 P = 0.0087
% Macrophage
%Ly6CHi Mono
% Eosinophil
%CD3+ CD4+
18 6
2 10
Blood
% Treg
4 12 4
1 5
2 6 2
0 0 0 0 0
KO
KO
T
T
KO
KO
KO
W
W
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
B 600 1,500 2,500 20,000
P = 0.008
600 8
Ly6Chi Mono / mg
Macrophage / mg
Mac+ CD206 MFI

Eosinophil / mg
Treg Cells / mg
2,000
T Cells / mg
15,000 6
400 1,000 400
PcAT
1,500
10,000 4
1,000
200 500 200
500 5,000 2
0 0 0 0 0 0
T
T
KO
KO
KO
T
KO
KO
KO
W
W
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
P = 0.026 P = 0.0043 P = 0.0238 P = 0.0303
C 100 900 800 3,600 80 4
Ly6Chi Mono / mg
Macrophage / mg
Mac+ CD206 MFI

Eosinophil / mg
Treg Cells / mg
80 3,000
T Cells / mg
600 60 3
600 2,400
Heart
60
400 1,800 40 2
40
300 1,200
20 200 20 1
600
0 0 0 0 0 0
T
T
KO
KO
KO
T
KO
T
KO
KO
W
W
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
C2
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
Cells from heart (A), pericardial adipose tissue (PcAT) (B), and blood (C). Eosinophils, monocytes, macrophage, and Treg populations were characterized by flow
cytometry at Day 5 post-MI. Mann-Whitney U test. Treg ¼ T regulatory cell; other abbreviation as in Figure 1.
(Figure 4D). IL-2 was the most well represented IL-2 AXIS IS CENTRAL TO ILC2 FUNCTION AND MI
9
pathway (P ¼ 2.44 10 ), key to ILC2 survival, and OUTCOME IN MICE. IL-2 was chosen for its high
its receptor CD25 is a defining molecule for the ILC2 score, high relevance to ILC2 function, and its po-
population. Others were the MYC pathway tential as a therapeutic modulator of immune re-
7
(P ¼ 1.28 10 ) is a very well characterized cell sponses in patients with acute coronary syndrome
7
survival/growth regulator, TREM1 (P ¼ 6.51 10 ) is (ACS) (22,26). We therefore subjected T-cell–deficient
an activatory receptor usually found on the surface Rag2 / mice to LAD ligation and treated them for
of myeloid cells that can drive activation through 3 days with the IL-2/Jes6-1 complex (27) or PBS
ERK phosphorylation and Ca þþ mobilization (23); vehicle (Figure 5). Serum cytokine secretion was
and EGLN1 (P¼ 2.42 106 ) is a prolyl hydroxylase enhanced following IL-2/Jes6-1 injection and
that suppresses the HIF1 a response through both elevated IL-5 was observed (Figure 5A), although at
proteolytic degradation and suppression of tran- this time point no change in IL-13 was detected
scriptional activity (24,25). (Supplemental Figure 6A). Complementing this, there
(A) Survival curve for knockout (KO: Rorafl/fl Cd127Cre/þ, n ¼ 5/14 ruptured) compared with wild type (WT: Rorafl/fl Cd127wt, n ¼ 1/9 ruptured) mice after experimentally
induced MI (P ¼ 0.2). (B) A significant impact on cardiac function (% ejection fraction, EF) in KO mice at Day 14 post-MI (P ¼ 0.0025) (representative images shown)
and recovery was severely impacted up to 4 weeks post-surgery (P ¼ 0.0006). (C) WT mice maintained more cardiac muscle compared with KO and had significantly
smaller scars. (D) Further, Sirius red labeling demonstrates decreased collagen deposition in the scar tissue of KO mice. Mann-Whitney U test. Abbreviations as in
Figure 1.
was an expansion of peripheral ILC2s in the spleen to secrete IL-5 (Figure 7B), indicating an activa-
(Figure 5C), and IL-5 expression in ILC2 was enhanced tion status.
(Figures 5B and 5C), a pattern also observed in PcAT The LILACS trial was designed to evaluate the
and in the heart (Figure 5C). Proportions of splenic safety and pharmacodynamics of low-dose IL-2 in
macrophages were unchanged; however, the expres- patients with ACS. After recruitment, individuals
sion of the Type II marker Arginase 1 (Arg1) was received repeated blinded daily doses of IL-2 (1.5
decreased. Conversely, within the PcAT macrophage, 106 IU, n ¼ 6 patients or 2.5 106 IU, n ¼ 6 patients) or
Arg1 expression was elevated (coinciding with more placebo (n ¼ 4 patients) over a period of 5 days, with
CD206 in this case, Supplemental Figure 6B) slewing follow-up sampling on D6 and D13 (Figure 7C). Flow
the population toward a wound-healing phenotype. cytometry analysis of PBMCs from this trial showed
Macrophages were also detected within the heart that although placebo had no effect on ILC2 pop-
tissue in greater abundance, and this was accompa- ulations, patients receiving 1.5 MIU/d of IL-2 for 5
nied by a greater proportion of them expressing Arg1 consecutive days showed a trend toward a decrease in
(Figures 5B and 5C). ILC2 canonical markers CRTH2 and CD127 immedi-
Assessing the impact of these phenotypic changes ately after the treatment course (follow-up 1), which
on long-term MI outcome, Rag2 / mice received then recovered after 1 week (follow-up 2) (Figure 7D).
3 weekly injections of IL-2/Jes6-1 or PBS vehicle for Patients receiving the higher 2.5 MIU/d dose for 5
4 weeks (Figure 6A) accompanied by regular assess- consecutive days showed a significant decrease in
ment of cardiac function. There were no immediate ILC2 canonical markers immediately after the treat-
differences between the IL-2 and PBS cohorts; how- ment course, which remained low 1 week later
ever, distinct improvement in cardiac function was (Figure 7D). These features are indicative of ILC2
observed in the IL-2-treated cohort after 7 days activation on treatment with low-dose IL-2 (28).
(Figure 6B). This improvement was maintained for the Serum IL-5 titers and peripheral eosinophil counts
duration of the experiment, increasing to 45% by were elevated in both treatment groups after
week 4 (P ¼ 0.0012) compared with 28% in the PBS low-dose IL-2 in a dose-dependent manner and both
cohort (Figure 6B), despite no difference in infarct subsequently returned to basal levels after 1 week
size (Figure 6C) or collagen deposition (Figure 6D). No (Figures 7E and 7F), further supporting an IL-2-driven
alterations to macrophage infiltrate or apoptosis (via activation of circulating ILC2s.
TUNEL staining) were observed at this time (data not
shown). DISCUSSION
LOW-DOSE IL-2 INFLUENCES ILC2 BEHAVIOR IN Here, we have shown that ILC2s have a direct and
PATIENTS WITH ACS. So far, we have shown that in protective role in recovery from experimental MI.
mouse models of MI, an ILC2/IL-2 axis is beneficial to ILC2s were detected in the heart and pericardial adi-
improved recovery of cardiac function. Repurposing pose tissue in steady state and expanded during the
recombinant IL-2 (aldesleukin) for treating human initial inflammatory phase of MI, trafficking from the
cardiovascular disease is of interest, although blood under some conditions. The expansion within
currently its licensed indication is limited to meta- the inflamed tissues is rapid but not sustained and falls
static melanoma and metastatic renal cell carcinoma. away within the first week of recovery. This comple-
In humans, IL-2 has been shown to activate ILC2 ments earlier observations that there is a cardiac resi-
in vitro, which is associated with reduced expression dent population of ILC2s that are present in the heart
of canonical markers CRTH2 and CD127 (28), poten- during periods of pericarditis and MI (16), and we show
tially masking the presence and importance of this that cardiac remodeling and wound stability is
population. To address this, we studied the effect of severely compromised in their absence. As we have
IL-2 treatment on whole PBMCs and how it modified seen in other models of ILC2 response to tissue damage
the ILC2 population. PBMCs isolated from donor blood (eg, atherosclerosis), the types of wounds that develop
was cultured with 10 ng/mL IL-2, IL-33, or in combi- in the absence of ILC2s tend to be larger, have less
nation for 24 to 96 hours, and any changes to ILC2s collagen deposition, and are overall more fragile than
were monitored by flow cytometry. Here, ILC2s ILC2-replete mice (14); as a consequence, the trend
rapidly lost their surface phenotype during treatment toward higher mortality rate seen in ILC2 KO mice dur-
(Figure 7A) in association with increased capacity ing experimental MI was due to cardiac rupture.
F I G U R E 4 ILC2 Transcriptome After MI Highlight a Potentially Protective IL-2 Stimulus
A B
Sorting ILC2 from –2 0 2 PcAT ILC2 RNASeq
Row Z-Score
LAD ligation pericardial adipose Rab32
Idh2
tissues Fnbp1l
Thsd4
Rhobtb3
Phldb1
Myo1c
E2f3
Gamt
Epb41l1
Tpmt
Tkfc
D0 D3 Mastl
Ldha.ps
Syngr1
Flrt3
Nfxl1
Sfmbt2
Cep68
II18bp
Pik3cb
Cdk5r1
Mtmr10
IL-5 IL-13 ST2 Nrp2
Vegfa
600,000 2,000 50,000 Cebpa
Cd86
P = 0.34 P = 0.14 Ly6g
II1rl2
40,000
Counts/FPKM
Counts/FPKM
Counts/FPKM
Tmem2
1,500 Pcnx
400,000 Abca1
Snx8
30,000 Lclat1
1,000 Tcaim
II6ra
20,000 Cxcl1
Rgs6
200,000 Foxn2
500 10,000
Bcar1
Urb1
Gtf3c3
Erc1
Alg12
0 0 0 Adgrg3
Cables1
Cfap53
am
am
am
I
M
M
Nkx2.3
Elac1
Sh
Sh
Sh
Zfp143
Sytl4
Slc4a8
Gene
Arg1 IL-18R1 IL-18bp
Spock1
Zfp462
Slc8a3
60,000 4,000 600 Pkp1
P = 0.4 P = 0.057 Ppie
Cd79b
Rdm1
Counts/FPKM
Counts/FPKM
Counts/FPKM
II18r1
3,000 Olfr95
40,000 400 Cbx4
Matk
Irx5
2,000 Hs6st2
Xk
Cep76
20,000 200 Pgpep1
Dscam
1,000 Gemin8
Fancf
Fzd3
Zgrf1
0 0 0 Cenpu
Spdya
Srr
Hook1
am
am
am
I
M
Rapsn
Sh
Sh
Sh
Ezh1
Ifi206
Trmt61b
Zfat
Dtd2
IL-6ra CXCR4 CXCL1 Asb17
Armc6
400 8,000 20,000 Lzts1
P = 0.0286 P = 0.0286 P = 0.0286

Arv1
Dscc1
Intu
Dnajc17
Counts/FPKM
Counts/FPKM
Counts/FPKM
300 6,000 15,000 Bid

ApoI7b
Slc2a3
Msh3
Oxnad1
200 4,000 10,000 Gcm1
Ttc38
Xlr3c
Cd3d
100 2,000 5,000 Otud1
Cldn5
Ccdc61
C1qtnf4
Snai3
0 0 0 Tprn
SHAM.1
SHAM.2
Sham.3
SHAM.4
MI.1
MI.2
MI.3
MI.4
am
am
am
I
M
M
Sh
Sh
Sh
Surgery Condition
C Modified Pathways in Post-MI ILC2 D Upstream Regulators in Post-MI ILC2
Cell Cycle IL-2

Cellular Movement MYC
Cellular Assembly and Organization TREM1
Cell Death and Survival EGLN1
Cellular Function and Maintenance
IFNg
–8
–7
–6
–5
–4
–9
–8
–7
–6
–5
–4
10
10
10
10
10
10
10
10
10
10
10
P Value
P Value
RNAseq analysis of ILC2s sorted from the PcAT on Day 3 post-MI. (A) Initial indications of alterations in signature ILC2 transcripts in MI mice as well as significant
differences in expression of inflammatory and trafficking molecules. (B) Heatmap of in-depth pathway and predictive gene cluster analysis highlights changes in
expression of novel genes induced by infarction, which may be central around modulation of cell proliferation and survival pathways (C). (D) Predictive upstream
regulatory pathways identified an IL-2-driven survival fingerprint. IL ¼ interleukin; RNAseq ¼ RNA sequencing; other abbreviations as in Figure 1.
F I G U R E 5 IL-2 Improves ILC2 Function During Acute MI in Rag2 / Mice
A 50 P = 0.0043 B FM1 FM1

40
IL-5 pg/mL
30 PBS PBS
20
IL-2 IL-2
10
0
–103 0 103 104 105 –103 0 103 104
S
-2
PB
IL-5 PE Arg1 PE
IL
C
0.03 P = 0.0498 4,000 15 15
P = 0.002 P = 0.3 P = 0.002
% Macrophage
3,000
0.02 10 10
% Arg1+
IL-5 MFI
Spleen
% ILC2
2,000
0.01 5 5
1,000
0.00 0 0 0
S
-2
-2
-2
S
-2
PB
PB
PB
PB
IL
IL
IL
IL
P = 0.13
0.3 P = 0.58 2,500 40 15
P = 0.02 P = 0.043
% Macrophage
2,000 30
0.2 10
MFI IL-5
% Arg1+
% ILC2
PcAT
1,500
20
1,000
0.1 5
500 10
0.0 0 0 0
S
-2
-2
-2
-2
PB
PB
PB
PB
IL
IL
IL
0.15 400
P = 0.04
60 P = 0.0152 3 IL
P = 0.013
% Macrophage
300
0.10 40 2
% Arg1+
MFI IL-5
% ILC2
Heart
200
0.05 20 1
100
0.00 0 0 0
S
-2
-2
-2
-2
PB
PB
PB
PB
IL
IL
IL
IL
IL-2/Jes6-1 treatment enhances serum cytokine secretion of IL-5 at Day 3 post-MI (A). Intracellular detection of IL-5 (B, black histogram FM1, blue PBS,
and red IL-2 treated) complements this. Splenic ILC2s (C) were expanded and expressed more IL-5 per cell. Splenic macrophages slew toward Arg1þ
population (M2-like). PcAT ILC2s (C) were not expanded but expressed more IL-5 with biased macrophage population toward Arg1þ expression. Heart
resident ILC2s (C) were not expanded but expressed more IL-5 per cell. Subsequent heart-infiltrating macrophages slew toward the Arg1þ population
(M2-like). Mann-Whitney U test. Arg1þ ¼ Arginase1þ; PBS ¼ phosphate-buffered saline; other abbreviations as in Figures 1 and 4.
Transcript analysis highlighted 2 main signatures heart through interaction with CXCL12 (29) (known to
of ILC2 activation during MI in the PcAT: proliferation increase during ischemia [30]) or MIF (31). The sup-
and trafficking. The expression of CXCR4 would allow pression of the IL-18 axis through decreased expres-
the ILC2 to be recruited to the inflamed tissue in the sion of IL18R and increased IL18BP (which sequesters
F I G U R E 6 IL-2 Treatment Improves the Recovery of Heart Function After MI
A PBS or
1 Pg IL-2/Jes6-1 complex
PBS/ PBS/ PBS/ PBS/
IL-2 IL-2 IL-2 IL-2
Day –1 0 1 7 14 21 28
Rag2–/–
ho
lig cho
Ec n
mouse
o
ol o
y
io
ist ch
og
at
Ec
Ec
Ec
E
E
D
H
LA
B
80 P = 0.010 P < 0.0001
PBS
P = 0.8 P = 0.007 P = 0.0019

60
%EF
40
20
0
IL-2
S
-2
S
-2
S
-2
S
-2
S
-2
S
-2
PB
PB
PB
PB
PB
PB
IL
IL
IL
IL
IL
IL
Baseline D1 D7 D14 D21 D28
C 100 P = 0.0871 D 2,000,000 P = 0.8
80 1,500,000
(%midline)
Area/PM2
Scar size
60
1,000,000
40
20 500,000
0 0
S
-2
-2
PB
PB
IL
IL
(A) Rag2/ mice (n ¼ 9 per group) were subjected to LAD ligation and treated intraperitoneally 3 times a week with the IL-2/Jes6-1 complex or PBS vehicle alone (see
Methods). IL-2 treatment substantially improves cardiac recovery after MI as shown by echocardiography (B) and the significant difference in % ejection fraction (%EF)
between the 2 groups of mice by the primary endpoint at Day 28, despite no difference in infarct size (C) or collagen deposition (D). Mann-Whitney U test or 2-way
analysis of variance where appropriate. Abbreviations as in Figures 1 and 4.
IL-18 [32,33]) may reinforce the polarization of ILC2 Pathway analysis of the RNAseq dataset provided a
and potentially nearby T cells, reducing the oppor- useful insight of what may be triggering the ILC2
tunity for plasticity. This is consistent with the trend response, and identification of IL-2 and other novel
toward reduced accumulation of Tregs in the hearts upstream regulatory pathways could also provide
of ILC2-deficient mice, but more work is needed to useful mechanisms to modify ILC2 behavior during
fully describe the mechanisms. the acute phase of MI.
F I G U R E 7 Low-Dose IL-2 Activates ILC2s in Patients With ACS
A 24 h 96 h
0.006 0.006
P = 0.0286 P = 0.0286
% CRTH2+ CD127+ ILC2

0.004 0.004
0.002 0.002
0.000 0.000
ia
-2
ia
-2
3
-3
-3
-3
-3
ed
ed
IL
IL
IL
IL
IL
IL
M
M
+
+
-2
-2
IL
IL
B 24 h
25
96 h
40
P = 0.0286 P = 0.0286
20
30
IL-5 ng/mL
IL-5 ng/mL
15
20
10
10
5
0 0
ia
-2
ia
-2
3
-3
-3
-3
-3
ed
ed
IL
IL
IL
IL
IL
IL
M
M
+
+
-2
-2
IL
IL
C Day-7 to –1 Day 1 to 5 Day 6 Day 13
Screening Treatment period: 5 Days Follow-up Follow-up

2 placebo, 6 IL-2 per group 1 2
Randomization*
Visit 1 Visit 2 Visit 3 Visit 4 Visit 5 Visit 6 Visit 7 Visit 8
PBMC culture with IL-2 (10 ng/mL) show rapid downregulation of ILC2 surface markers following IL-2, IL-33, and in combination (A). Secretion
of IL-5 was increased (B). Patients with ACS enrolled in LILACS (C) were sampled at Visit 2 (Baseline) before receiving 5 daily consecutive IL-2
doses (1.5 U or 2.5 U) or placebo and follow-up before blood sampling on Visit 7 (Follow-up 1, Day 6) and Visit 8 (Follow-up 2, Day 13). (D)
Flow cytometric analysis shows a dose-dependent decrease in canonical markers of peripheral blood ILC2 (CRTH2þCD161þ cells within
LinCD127þ cells) over the course of the treatment, suggesting activation. Corresponding amounts of serum IL-5 (E) and blood eosinophil
count (F) were both elevated at follow-up 1 and subsequently decreased after a further 6 days. Error bars show SEM. Multiple comparisons
were made using the Kruskal-Wallis test followed by Mann-Whitney U test. ACS ¼ acute coronary syndrome; PBMC ¼ peripheral blood
mononuclear cell; other abbreviations as in Figures 1 and 4.

D Flow Cytometry E Serum Cytokine

0.015 1,000
P = 0.002
100 P = 0.002
IL-5 pg/mL
0.010 10
P = 0.02
1
0.005
0.1
Placebo 1.5 MIU 2.5 MIU
Baseline Follow-Up 1 Follow-Up 2
0.000
F Eosinophils
1.5
P = 0.0087
P = 0.0216
P = 0.06
Count (u109/L)
1.0
0.5
0.0
The results observed during the direct targeting of effect will require further investigations. We used
ILC2 with IL-2/Jes6-1 complexes were promising, not a permanent coronary artery ligation model to
only preventing a marked decrease in cardiac func- create large infarcts. The role of ILC2 in the
tion during the initial phase but also improving response to ischemia-reperfusion injury will require
recovery over the subsequent 4 weeks. further studies.
Data from our LILACS human clinical trial provide H u m a n c l i n i c a l r e s u l t s . Our data show increased
translational relevance of our preclinical findings. We serum IL-5 and blood eosinophil counts after low-
demonstrate that low-dose IL-2 activates ILC2s in pa- dose IL-2, in association with signs of ILC2 activa-
tients presenting with ACS, with increases in IL-5 and tion; however, the potential protective role of such
eosinophil levels as a consequence of ILC2 activation changes will require further investigation.
(34). Interestingly, recent data in mice have shown that
CONCLUSIONS
eosinophils may play a significant role in MI, and their
deficiency promoted adverse cardiac remodeling (35).
We have presented data that highlight a critical role for
STUDY LIMITATIONS. B a s i c and preclinical ILC2 during the response to MI. Expanding early to
r e s u l t s . Our data suggest that the protective effects condition the immune response, their absence in-
of ILC2 after MI could be mediated through the creases the accumulation of inflammatory monocytes
production of the type 2 cytokine IL-5; however, the and macrophages and severely impacts cardiac func-
direct role of ILC2-derived IL-5 (and downstream tion after MI. We have shown that expansion of ILC2
eosinophil activation) in mediating this protective with exogenous IL-2 is beneficial, driving heart
C E NT R AL IL L U STR AT IO N Type 2 Innate Lymphoid Cells Orchestrate a Cardioprotective Response

After Myocardial Infarction
Yu, X. et al. J Am Coll Cardiol. 2021;78(11):1127–1142.
In this report, we demonstrate that type 2 innate lymphoid cells (ILC2s) are resident in the heart and pericardium, expanding during experimental myocardial infarction.
In mouse models deficient in ILC2s, cardiac recovery and repair are significantly retarded; the recovering scar contains less collagen, fewer T regulatory cells (Tregs)
and more inflammatory monocyte/macrophage populations. Supplementing the mice with interleukin (IL)-2 increases ILC2 activation and promotes the secretion of
cardioprotective cytokines (eg, IL-5) and ultimately improves recovery. We observe a similar phenotypic shift in ILC2s in patients with acute coronary syndromes after
administration of low-dose IL-2, coupled with increased IL-5 secretion and associated eosinophilia.
macrophages toward an alternatively activated and support from European Research Area Network on Cardiovascular
Diseases Joint Transnational Call (ERA-CVD JTC) 2017-
potentially reparative phenotype and improving car-
PLAQUEFIGHT). Dr Sun has been funded by the Federation Fran-
diac output for the duration of the trial (Central çaise de Cardiologie. The LILACS trial was funded by the Medical
Illustration). Future work should address the precise Research Council, grant number MR/N028015/1, the British Heart
mechanisms through which ILC2s control the inflam- Foundation Cambridge Centre of Excellence (RCAG/521). Mr Cheriyan
has received funding support from the NIHR Cambridge Biomedical
matory and cardiac remodeling response after MI. Early
Research Centre (BRC-1215-20014). The views expressed are those of
translational data of low-dose IL-2 in patients with ACS the authors and not necessarily those of the NIHR or the Department
suggest that this may promote ILC2 activation, poten- of Health and Social Care. Dr Rudd has been part-supported by the
tially impacting on long-term cardiac dysfunction. NIHR Cambridge Biomedical Research Centre, the British Heart
Foundation, HEFCE, the EPSRC, and the Wellcome Trust. This NGS
Furthermore, acute MI may drive the progression of
work was performed with the Genomics and Transcriptomics Core,
atherosclerosis (36), which clinically results in which is funded by the UK Medical Research Council (MRC) Metabolic
increased rate of further atherosclerotic cardiovascular Disease Unit (MRC_MC_UU_00014/5) and a Wellcome Trust Major
events in the intervening period. Therefore, one can Award (208363/Z/17/Z). All authors have reported that they have no
relationships relevant to the contents of this paper to disclose.
hypothesize that targeting ILC2s may be beneficial not
only in cardiac remodeling post-MI but potentially also
ADDRESS FOR CORRESPONDENCE: Dr Ziad Mallat,
retards progression of atherosclerosis.
Department of Medicine, Division of Cardiovascular
ACKNOWLEDGMENTS The authors acknowledge Giles
Medicine, University of Cambridge, CB2 0QQ Cam-
Yeo and The Wellcome Trust-MRC Institute of Metabolic
bridge, United Kingdom. E-mail: zm255@medschl.
Science-Metabolic Research Laboratories for practical
cam.ac.uk. Twitter: @ZMallat, @mallat_lab.
and bioinformatics support; Suzanne Cohen, AstraZeneca,
Cambridge, UK, for the kind gift of sST2 reagent; Hans-
Reimer Rodewald, Division of Cellular Immunology,
PERSPECTIVES
German Cancer Research Center, Heidelberg, Germany, for
providing the Cd127Cre mouse strain; and Andrew N.J.
McKenzie, Medical Research Council Laboratory of Molec- COMPETENCY IN MEDICAL KNOWLEDGE: Inflammation
ular Biology, Cambridge, UK, for providing the Rora fl/fl impedes cardiac remodeling and recovery of ventricular function
mouse strain. after MI. IL-2 plays a critical role in ILC2 activation and myocar-
dial repair.
TRANSLATIONAL OUTLOOK: Clinical trials of low-dose IL-2
This research was supported by the Cambridge NIHR BRC Cell Phe- are needed to evaluate therapeutic efficacy in patients with
notyping Hub. Dr Newland has received funding from the British
acute MI.
Heart Foundation (PG/18/20/33595). Dr Yu has been funded by a Royal
Society Newton Advanced fellowship (NA140277). The work received
REFERENCES
1. Zlatanova I, Pinto C, Silvestre JS. Immune repair after infarction. Circ Res. 2013;113:1004– macrophage differentiation. Circ Res. 2014;115:
modulation of cardiac repair and regeneration: the 1012. 55–67.
art of mending broken hearts. Front Cardiovasc
6. Howangyin KY, Zlatanova I, Pinto C, et al. 10. Tang TT, Yuan J, Zhu ZF, et al. Regulatory T
Med. 2016;3:40.
Myeloid-epithelial-reproductive receptor tyrosine cells ameliorate cardiac remodeling after
2. Bulluck H, Yellon DM, Hausenloy DJ. Reducing kinase and milk fat globule epidermal growth myocardial infarction. Basic Res Cardiol.
myocardial infarct size: challenges and future op- factor 8 coordinately improve remodeling after 2012;107:232.
portunities. Heart. 2016;102:341–348. myocardial infarction via local delivery of vascular
endothelial growth factor. Circulation. 2016;133: 11. Roediger B, Kyle R, Tay SS, et al. IL-2 is a
3. Nahrendorf M, Swirski FK, Aikawa E, et al. The critical regulator of group 2 innate lymphoid cell
826–839.
healing myocardium sequentially mobilizes two function during pulmonary inflammation. J Allergy
monocyte subsets with divergent and comple- 7. Zlatanova I, Pinto C, Bonnin P, et al. Iron
Clin Immunol. 2015;136:1653–1663.e7.
mentary functions. J Exp Med. 2007;204:3037– regulator hepcidin impairs macrophage-
3047. dependent cardiac repair after injury. Circulation. 12. Bando JK, Gilfillan S, Di Luccia B, et al. ILC2s
2019;139:1530–1547. are the predominant source of intestinal ILC-
4. Hilgendorf I, Gerhardt LM, Tan TC, et al. Ly-
derived IL-10. J Exp Med. 2020;217:e20191520.
6Chigh monocytes depend on Nr4a1 to balance 8. Hofmann U, Beyersdorf N, Weirather J, et al.
both inflammatory and reparative phases in the Activation of CD4þ T lymphocytes improves 13. Nussbaum JC, Van Dyken SJ, von Moltke J,
infarcted myocardium. Circ Res. 2014;114:1611– wound healing and survival after experimental et al. Type 2 innate lymphoid cells control eosin-
1622. myocardial infarction in mice. Circulation. ophil homeostasis. Nature. 2013;502:245–248.
2012;125:1652–1663.
5. Wan E, Yeap XY, Dehn S, et al. Enhanced 14. Newland SA, Mohanta S, Clement M, et al.
efferocytosis of apoptotic cardiomyocytes through 9. Weirather J, Hofmann UD, Beyersdorf N, et al. Type-2 innate lymphoid cells control the devel-
myeloid-epithelial-reproductive tyrosine kinase Foxp3þ CD4þ T cells improve healing after opment of atherosclerosis in mice. Nat Commun.
links acute inflammation resolution to cardiac myocardial infarction by modulating monocyte/ 2017;8:15781.
15. Dahlgren MW, Jones SW, Cautivo KM, et al. 23. Bouchon A, Dietrich J, Colonna M. Cutting dial ischemia/reperfusion injury: role of the cardiac
Adventitial stromal cells define group 2 innate edge: inflammatory responses can be triggered by stromal cell derived factor-1 alpha CXCR4 axis.
lymphoid cell tissue niches. Immunity. 2019;50: TREM-1, a novel receptor expressed on neutro- Circulation. 2007;116:654–663.
707–722.e6. phils and monocytes. J Immunol. 2000;164:4991–
31. Doring Y, Pawig L, Weber C, Noels H. The
4995.
16. Bracamonte-Baran W, Chen G, Hou X, et al. CXCL12/CXCR4 chemokine ligand/receptor axis in
Non-cytotoxic cardiac innate lymphoid cells are a 24. To KK, Huang LE. Suppression of hypoxia- cardiovascular disease. Front Physiol. 2014;5:212.
resident and quiescent type 2-commited popula- inducible factor 1alpha (HIF-1alpha) transcrip- 32. Rao A, Strauss O, Kokkinou E, et al. Cytokines
tion. Front Immunol. 2019;10:634. tional activity by the HIF prolyl hydroxylase regulate the antigen-presenting characteristics of
EGLN1. J Biol Chem. 2005;280:38102–38107. human circulating and tissue-resident intestinal
17. Schlenner SM, Madan V, Busch K, et al. Fate
mapping reveals separate origins of T cells and 25. Zhou Y, Ouyang N, Liu L, Tian J, Huang X, Lu T. ILCs. Nat Commun. 2020;11:2049.
myeloid lineages in the thymus. Immunity. An EGLN1 mutation may regulate hypoxic 33. Novick D, Kim SH, Fantuzzi G, Reznikov LL,
2010;32:426–436. response in cyanotic congenital heart disease Dinarello CA, Rubinstein M. Interleukin-18 binding
through the PHD2/HIF-1A pathway. Genes Dis. protein: a novel modulator of the Th1 cytokine
18. Oliphant CJ, Hwang YY, Walker JA, et al.
2019;6:35–42. response. Immunity. 1999;10:127–136.
MHCII-mediated dialog between group 2 innate
lymphoid cells and CD4(þ) T cells potentiates 26. Zhao TX, Kostapanos M, Griffiths C, et al. Low- 34. Van Gool F, Molofsky AB, Morar MM, et al.
type 2 immunity and promotes parasitic helminth dose interleukin-2 in patients with stable ischaemic Interleukin-5-producing group 2 innate lymphoid
expulsion. Immunity. 2014;41:283–295. heart disease and acute coronary syndromes (LI- cells control eosinophilia induced by interleukin-2
LACS): protocol and study rationale for a rando- therapy. Blood. 2014;124:3572–3576.
19. Zouggari Y, Ait-Oufella H, Bonnin P, et al.
mised, double-blind, placebo-controlled, phase I/II
B lymphocytes trigger monocyte mobilization and 35. Toor IS, Ruckerl D, Mair I, et al. Eosinophil
clinical trial. BMJ Open. 2018;8:e022452.
impair heart function after acute myocardial deficiency promotes aberrant repair and adverse
infarction. Nat Med. 2013;19:1273–1280. 27. Boyman O, Kovar M, Rubinstein MP, Surh CD, remodeling following acute myocardial infarction.
Sprent J. Selective stimulation of T cell subsets J Am Coll Cardiol Basic Trans Sci. 2020;5:665–681.
20. Polhill T, Zhang GY, Hu M, et al. IL-2/IL-2Ab
with antibody-cytokine immune complexes. Sci-
complexes induce regulatory T cell expansion and 36. Dutta P, Courties G, Wei Y, et al. Myocardial
ence. 2006;311:1924–1927.
protect against proteinuric CKD. J Am Soc Nephrol. infarction accelerates atherosclerosis. Nature.
2012;23:1303–1308. 28. Camelo A, Rosignoli G, Ohne Y, et al. IL-33, IL- 2012;487:325–329.
25, and TSLP induce a distinct phenotypic and
21. Schindelin J, Arganda-Carreras I, Frise E,
activation profile in human type 2 innate lymphoid
et al. Fiji: an open-source platform for
cells. Blood Adv. 2017;1:577–589. KEY WORDS cytokines, heart failure,
biological-image analysis. Nat Methods. 2012;9:
innate lymphoid cells, interleukin-2,
676–682. 29. Thackeray JT, Derlin T, Haghikia A, et al. Mo-
lymphocytes, myocardial infarction
lecular imaging of the chemokine receptor CXCR4
22. Zhao TX, Newland SA, Mallat Z. 2019 ATVB
after acute myocardial infarction. J Am Coll Cardiol
plenary lecture: interleukin-2 therapy in cardio-
Img. 2015;8:1417–1426.
vascular disease: the potential to regulate innate A PPE NDI X For supplemental figures and a
and adaptive immunity. Arterioscler Thromb Vasc 30. Hu X, Dai S, Wu WJ, et al. Stromal cell derived table, please see the online version of this
Biol. 2020;40:853–864. factor-1 alpha confers protection against myocar- paper.
EDITORIAL COMMENT
Innate Lymphoid Cells Participate in

Myocardial Inflammation After Ischemia*
Matthias Nahrendorf, MD, PHD
C ardiac ischemia triggered by atherothrom-

botic events remains a highly prevalent
clinical challenge. Although the mortality
of myocardial infarction (MI) has plummeted, its
macrophages (1) and their subtypes (2), discovering
the role of Ccl2/Ccr2 after MI (3), that heart resident
macrophages die in the ischemic myocardium (4),
the involvement of the spleen (5), the bone marrow
incidence continues to top disease statistics, and (6), and the epicardial space (7) as purveyors of
numerous patients with MI develop heart failure or various leukocyte subtypes, the contribution of
reinfarction. Re-establishing blood flow rapidly and lymphocytes to the infarct microenvironment (8)
managing atherosclerosis risk factors have improved and elucidation of specific leukocyte functions during
outcomes, but the size of the persisting problem chal- heart development (9), the myocardial steady state
lenges the field to search for orthogonal therapeutic (10,11), and after injury.
options. The success of immunomodulatory therapy
SEE PAGE 1127
in cancer and autoimmune diseases indicates that
the immune response to MI may be one area ripe for In this issue of the Journal, Yu et al (12) focus on a
a breakthrough. That MI triggers leukocytosis, which previously less-studied cell type in acute MI: innate
associates with worse outcomes, has been known lymphoid cells type 2 (ILC2). These cells are known
for decades; however, only the recently expanding for secreting cytokines such as interleukin (IL)-5 after
insight into the diverse actions of leukocytes after parasitic and viral infections and in autoimmune
cardiac injury suggests that these cells are causally disorders. Yu et al (12) now report that ILC2 reside in
involved in shaping outcomes after ischemia. This pericardial fat of mice and double their numbers after
knowledge expansion sheds light on the participation induction of MI by coronary ligation due to their
of new classes and subsets of innate and adaptive leu- increased proliferation in this tissue compartment
kocytes, the different compartments they derive with emerging importance for cardiac inflammation.
from, and these cells’ functions, including various cy- Inhibiting ST2 signaling reduced the post-MI ILC2
tokines that regulate myocardial tissue homeostasis expansion. A central finding of this work is that Ror-
and repair. Milestones on this path included, for afl/fl Cd127 cre mice, which lack ILC2, had worse post-
instance, the description of resident heart MI outcomes. Profiling of leukocytes in infarcts of
Rora fl/fl Cd127 cre mice revealed a more inflammatory
monocyte and macrophage repertoire and reduced
T-regulatory lymphocytes, both of which have pre-
reflect the views of the author and do not necessarily represent the views viously been connected to impaired infarct healing
of JACC or the American College of Cardiology. (13,14). RNA sequencing of ILC2 suggested that the
From the Center for Systems Biology, Massachusetts General Hospital cytokine IL-2 may be among the factors that promote
Research Institute and Harvard Medical School, Boston, Massachusetts, the observed effects after MI. Indeed, IL-2 supple-
USA; Department of Radiology, Massachusetts General Hospital, Boston,
mentation in mice with MI improved outcomes, sup-
Massachusetts, USA; Cardiovascular Research Center, Massachusetts
General Hospital and Harvard Medical School, Boston, Massachusetts,
porting the notion that this may be a relevant effector
USA; and the Department of Internal Medicine I, University Hospital cytokine, although cell-specific deletion studies are
Wuerzburg, Wuerzburg, Germany. warranted to gauge whether ILC2 are indeed the
The author attests they are in compliance with human studies
critical cellular source of IL-2 after MI. In particular T
committees and animal welfare regulations of the author’s institution
and Food and Drug Administration guidelines, including patient consent
lymphocytes, which also expand after MI, may be an
where appropriate. For more information, visit the Author Center. alternative source for the cytokine. Noteworthy,

1144 Nahrendorf JACC VOL. 78, NO. 11, 2021
Innate Lymphoid Cells in Myocardial Infarction SEPTEMBER 14, 2021:1143–1144
injecting IL-2 (a drug called aldesleukin) into 12 pa- the destructive forces of immune cells while avoiding
tients with MI triggered a dose-dependent ILC2 an increased risk of infections. A successful solution
phenotype shift and increased IL-5 production, which will likely focus on specific immune functions or
paralleled the authors’ observations in mice. Alde- overwhelmingly harmful subpopulations of immune
sleukin is currently used as immunomodulatory cells. In addition, identification of the appropriate
therapy in patients with renal cell cancer, where it is patient population that will benefit from immuno-
thought to expand and activate lymphocytes. The modulation (the inflammatory activity in infarcts is
precise mechanisms how IL-2 or ILC2 improves post- likely as heterogeneous as comorbidities and the in-
MI recovery remain to be determined. Such actions dividual past medical history) and appropriate
may be multifactorial and leveraged through other outcome measures (perhaps initially circulating and
leukocytes, for instance T cells or the more abundant imaging biomarkers of inflammation) are also pre-
macrophages, which the current study found to be requisites for successful clinical studies.
affected in Rora fl/fl Cd127cre mice.
This work adds another cellular component to our
expanding knowledge on immune cell subsets’ roles Dr Nahrendorf has received funds or material research support
in the recovery from organ ischemia. It also empha- from Lilly, Alnylam, Biotronik, CSL Behring, GlycoMimetics,
sizes that local adipose tissue may act as an immune GlaxoSmithKline, Medtronic, Novartis, and Pfizer; and has received
consulting fees from Biogen, Gimv, IFM Therapeutics, Molecular
cell reservoir, and provides a potential path for
Imaging, Sigilon, and Verseau Therapeutics.
immunomodulatory interventions. There are many
different potential ways of modulating inflammation
after MI. Finding the optimal therapeutic path will ADDRESS FOR CORRESPONDENCE: Dr Matthias
require a better understanding of the interaction of Nahrendorf, Center for Systems Biology, MGH
immune, hematopoietic, and cardiovascular systems. Research Institute, 185 Cambridge Street, Boston,
Only then will we be able to prioritize clinical testing Massachusetts 02114, USA. E-mail: mnahrendorf@
of targeted strategies with the best potential to curb mgh.harvard.edu. Twitter: @MatthiasNahrend.
REFERENCES
1. Pinto AR, Paolicelli R, Salimova E, et al. An their deployment to inflammatory sites. Science. 11. Nicolás-Ávila JA, Lechuga-Vieco AV, Esteban-
abundant tissue macrophage population in the 2009;325:612–616. Martínez L, et al. A network of macrophages
adult murine heart with a distinct alternatively- supports mitochondrial homeostasis in the heart.
6. Dutta P, Sager HB, Stengel KR, et al. Myocardial
activated macrophage profile. PLoS One. 2012;7: Cell. 2020;183:94–109.e23.
infarction activates CCR2(þ) hematopoietic stem
e36814.
and progenitor cells. Cell Stem Cell. 2015;16:477– 12. Yu X, Newland SA, Zhao TX, et al. Innate
2. Epelman S, Lavine KJ, Beaudin AE, et al. Em- 487. lymphoid cells promote recovery of ventricular
bryonic and adult-derived resident cardiac mac- function after myocardial infarction. J Am Coll
7. Deniset JF, Belke D, Lee WY, et al. Gata6þ Peri-
rophages are maintained through distinct Cardiol. 2021;78:1127–1142.
cardial cavity macrophages relocate to the injured
mechanisms at steady state and during inflam- 13. Nahrendorf M, Swirski FK, Aikawa E, et al.
heart and prevent cardiac fibrosis. Immunity.
mation. Immunity. 2014;40:91–104. The healing myocardium sequentially mobilizes
2019;51:131–140.e5.
two monocyte subsets with divergent and com-
3. Frangogiannis NG, Dewald O, Xia Y, et al. Crit- 8. Hofmann U, Beyersdorf N, Weirather J, et al. plementary functions. J Exp Med. 2007;204:
ical role of monocyte chemoattractant protein-1/ Activation of CD4þ T lymphocytes improves wound 3037–3047.
CC chemokine ligand 2 in the pathogenesis of healing and survival after experimental myocardial
ischemic cardiomyopathy. Circulation. 2007;115: 14. Weirather J, Hofmann UD, Beyersdorf N, et al.
infarction in mice. Circulation. 2012;125:1652–1663.
584–592. Foxp3þ CD4þ T cells improve healing after
9. Leid J, Carrelha J, Boukarabila H, Epelman S, myocardial infarction by modulating monocyte/
4. Leuschner F, Rauch PJ, Ueno T, et al. Rapid Jacobsen SE, Lavine KJ. Primitive embryonic mac- macrophage differentiation. Circ Res. 2014;115:55–
monocyte kinetics in acute myocardial infarction rophages are required for coronary development 67.
are sustained by extramedullary monocytopoiesis. and maturation. Circ Res. 2016;118:1498–1511.
J Exp Med. 2012;209:123–137.
10. Hulsmans M, Clauss S, Xiao L, et al. Macro-
5. Swirski FK, Nahrendorf M, Etzrodt M, et al. phages facilitate electrical conduction in the heart. KEY WORDS inflammation, innate lymphoid
Identification of splenic reservoir monocytes and Cell. 2017;169:510–522.e20. cells, myocardial infarction
ª 2021 THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION.
PUBLISHED BY ELSEVIER. ALL RIGHTS RESERVED.
THE PRESENT AND FUTURE
JACC STATE-OF-THE-ART REVIEW
Medial Arterial Calcification

JACC State-of-the-Art Review
Peter Lanzer, MD,a Fadil M. Hannan, DPHIL,b Jan D. Lanzer, MD,c,d,e Jan Janzen, MD,f Paolo Raggi, MD,g
Dominic Furniss, DM, MBBCH,h Mirjam Schuchardt, PHD,i Rajesh Thakker, SCD,j Pak-Wing Fok, PHD,k
Julio Saez-Rodriguez, PHD,c Angel Millan, PHD,l Yu Sato, MD,m Roberto Ferraresi, MD,n Renu Virmani, MD,m
Cynthia St. Hilaire, PHDo,p,q
ABSTRACT
Medial arterial calcification (MAC) is a chronic systemic vascular disorder distinct from atherosclerosis that is frequently
but not always associated with diabetes mellitus, chronic kidney disease, and aging. MAC is also a part of more complex
phenotypes in numerous less common diseases. The hallmarks of MAC include disseminated and progressive precipitation
of calcium phosphate within the medial layer, a prolonged and clinically silent course, and compromise of hemodynamics
associated with chronic limb-threatening ischemia. MAC increases the risk of complications during vascular interventions
and mitigates their outcomes. With the exception of rare monogenetic defects affecting adenosine triphosphate meta-
bolism, MAC pathogenesis remains unknown, and causal therapy is not available. Implementation of genetics and omics-
based approaches in research recognizing the critical importance of calcium phosphate thermodynamics holds promise to
unravel MAC molecular pathogenesis and to provide guidance for therapy. The current state of knowledge concerning
MAC is reviewed, and future perspectives are outlined. (J Am Coll Cardiol 2021;78:1145–1165) © 2021 the American
College of Cardiology Foundation. Published by Elsevier. All rights reserved.
M edial arterial calcification (MAC) is a sys-

temic vascular disorder distinct from
atherosclerosis that is associated with
the following: an increase in arterial stiffness (1); dia-
(3); and chronic limb-threatening ischemia (CLTI)
(4-7). MAC is frequently, but not always (8), associ-
ated with diabetes mellitus (DM) (9,10), chronic kid-
ney disease (CKD) (11,12), and aging (13).
stolic heart failure (2); impaired perfusion of a high Furthermore, MAC is a part of more complex pheno-
blood flow organs such as brain, kidney, and liver types in a large number of less common disorders
From the aMiddle German Heart Center-Bitterfeld, Bitterfeld-Wolfen Health Care Center, Bitterfeld, Germany; bNuffield Depart-
ment of Women’s & Reproductive Health, University of Oxford, Oxford, United Kingdom; cInstitute for Computational
Biomedicine, Bioquant, Faculty of Medicine, Heidelberg University, Heidelberg, Germany; dDepartment of Internal Medicine II,
Heidelberg University Hospital, Heidelberg, Germany; eFaculty of Biosciences, Heidelberg University, Heidelberg, Heidelberg,
Germany; fVascPath Switzerland, Bern, Switzerland; gDivision of Cardiology, Department of Medicine, University of Alberta,
Edmonton, Alberta, Canada; hBotnar Research Centre, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal
Sciences, University of Oxford, Oxford, United Kingdom; iDepartment of Nephrology and Medical Intensive Care, Charité–Uni-
versitätsmedizin Berlin, Corporate Member of Freie Universität and Humboldt Universität Berlin, Campus Benjamin Franklin,
Listen to this manuscript’s Berlin, Germany; jAcademic Endocrine Unit, Radcliffe Department of Medicine, University of Oxford, Oxford, United Kingdom;
audio summary by k
Department of Mathematical Sciences, University of Delaware, Newark, Delaware, USA; lInstitute of Materials Science, University
Editor-in-Chief of Zaragoza, Zaragoza, Spain; m
CVPath Institute, Gaithersburg, Maryland, USA; nCardiovascular Department, Humanitas Gav-
Dr. Valentin Fuster on azzeni, Bergamo, Italy; oDivision of Cardiology, Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA;
JACC.org. p
Pittsburgh Heart, Lung, and Blood Vascular Medicine Institute, University of Pittsburgh, Pittsburgh, Pennsylvania, USA; and the
q
Department of Bioengineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA.
Manuscript received April 12, 2021; revised manuscript received June 23, 2021, accepted June 28, 2021.

1146 Lanzer et al. JACC VOL. 78, NO. 11, 2021
Medial Arterial Calcification SEPTEMBER 14, 2021:1145–1165
ABBREVIATIONS (Supplemental Appendix A). MAC interferes

AND ACRONYMS HIGHLIGHTS
with revascularization procedures and miti-
gates their outcomes (14-16). MAC is a systemic vascular disorder
AC = arterial calcification
MAC is characterized by accumulation of distinct from atherosclerosis that results
ATP = adenosine triphosphate
calcium phosphate (CaP) with formation of in progressive calcification of the medial
CaP = calcium phosphate
hydroxyapatite (HAP) crystals (17), resulting layer of the arterial wall.
CaSR = calcium-sensing in progressive petrification of the medial
receptor MAC is frequently associated with dia-
layer of the arterial wall. In some cases
CKD = chronic kidney disease betes mellitus, chronic kidney disease,
ectopic vascular osteogenesis may also
CLTI = chronic limb- and aging, and it can result in severe limb
develop (18).
threatening ischemia ischemia.
The discovery of bone morphogenetic
CT = computed tomography
protein in samples of calcified human Genetic and proteomic investigations are
DM = diabetes mellitus
atherosclerotic lesions launched the biologic needed to clarify the pathogenesis of
ECM = extracellular matrix
hypothesis of arterial calcifications (ACs) (19). MAC and develop specific therapeutic
GWAS = genome-wide However, to date, with the exception of the approaches.
association study
monogenic disorders (20-23), the molecular
HAP = hydroxyapatite
pathogenesis of AC is poorly understood (24), dissection of the thoracic aorta, MAC was found in
HDAC9 = histone deacetylase 9
and no causal treatment is available (25). The 22% and 52%, respectively (34).
IAC = intimal arterial slow progress in unraveling the pathogenesis
calcification
stems from confounding distinct AC entities, CLINICAL IMPLICATIONS
MAC = medial arterial
a lack of experimental models replicating
calcification
MAC (26), and a deficient appreciation of the In the past, MAC was considered an innocent
MV = matrix vesicle
thermodynamic principles governing CaP bystander. However, studies now demonstrate that
PAD = peripheral artery
disease
precipitation (27). MAC can be considered the silent killer of the car-
Here, we review the clinical and research diovascular system.
T2DM = type 2 diabetes
mellitus signatures of MAC, address some of the In a 10-year follow-up study of 133 patients with
VC = vascular calcification perceived current shortcomings, and suggest T2DM, MAC was found to be a powerful and inde-
VSMC = vascular smooth
directions for future research. pendent risk factor for cardiovascular mortality and
muscle cell substantially stronger than the impact of IAC (9). MAC
EPIDEMIOLOGY
was found to be a strong and independent predictor
of total cardiovascular mortality, and it predicts the
The true prevalence of MAC is not known. MAC
risk of future coronary events in diabetic patients
frequently represents incidental findings, or it is
(10). Similarly, MAC is closely associated with the
misdiagnosed as atherosclerosis and may therefore
duration of hemodialysis and CaP disorders (11). In
escape clinical attention. The prevalence of MAC, on
patients with T2DM, the finding of MAC in foot ar-
the basis of an ankle-brachial index (ABI) >1.3, has
teries has been associated with below-the-knee artery
been estimated to be w0.5% of adults, with a male-to-
disease and foot ulcers (35,36). A recent meta-analysis
female ratio of 3:2 (28); however, this is not an ac-
of below-the-knee MAC demonstrated a strong asso-
curate metric for diagnosis (29). MAC is found in 17%
ciation between infrapopliteal MAC and lower limb
to 42% of type 2 DM (T2DM) patients (9,10), in 27% to
amputation risk (37).
40% of patients with advanced CKD (11,12) and in up
These clinical data provide definite proof that MAC
to 72% in patients with CLTI (30,31).
is an independent cause of peripheral artery disease
The prevalence of MAC in the coronary arteries is
(PAD) shadowing the role of atherosclerotic PAD in
unknown. However, intravascular ultrasound imag-
predisposed patients, and they clearly justify sys-
ing revealed isolated deep calcifications and deep
tematic studies targeting detection and clinical rele-
combined with superficial calcifications in 28% and
vance of MAC in arteries large and small.
24%, respectively; this finding potentially corre-
sponds to MACs and atherosclerotic intimal arterial PATHOPHYSIOLOGY
calcifications (IACs) (32).
The prevalence of MAC in the thoracic aorta is The current hypotheses of ACs, including both MAC
currently unknown, even though the distinction be- and IAC, span a wide range of molecular biology
tween MAC and atherosclerotic IAC is clinically pathways found in inflammation, apoptosis, disrup-
important in patients with porcelain aorta (33). In tions of CaP homeostasis, matrix vesicle (MV) extru-
patients with acute and chronic Stanford A or B sions, osteogenic transformations, extracellular
JACC VOL. 78, NO. 11, 2021 Lanzer et al. 1147
SEPTEMBER 14, 2021:1145–1165 Medial Arterial Calcification
matrix (ECM) degeneration, and genetic aberrations. experiments, magnesium has been shown to modu-
Although these processes overlap, the initial drivers late the development of phosphate-induced calcifi-
specific to MAC are unclear. Figure 1 provides an cation in a dose-dependent manner through the
overview of some of the proposed pathogenetic down-regulation of promoters and the up-regulation
principles to explain AC. of inhibitors of calcification (51,52).
CELLULAR AND MOLECULAR CONSIDERATIONS. It Certain genetic diseases with complex phenotypes,
is assumed that, with the proper building blocks and including MAC, originate from inactivating mutations
milieu conditions, CaP ions are prone to nucleate. of genes participating in the adenosine triphosphate
Thus, it has been proposed that to prevent ectopic (ATP) metabolic pathway and shed light on the initi-
nucleation within the ECM, certain proteins such as ating steps in MAC pathogenesis. ATP is released from
matrix g -carboxyglutamic acid (MGLA) protein may cells under numerous metabolic stresses and insults,
neutralize the crystallization process in a vitamin K– with ATP degradation products generally assisting
dependent manner (38). Although studies have cells and tissues to resist injury and to maintain ho-
shown that patients with T2DM have a higher serum meostasis (53). It has been assumed that mutations in
level of the inactive form of MGLA that correlated ectonucleotide pyrophosphatase/phosphodiesterase
with below-the-knee MAC (39), a more recent sys- 1 (ENPP1) and ATP-binding cassette subfamily C
tematic review found that “no single gamma- member 6 (ABCC6) drive MAC by the reduction in
carboxyglutamic protein species has demonstrated a extracellular pyrophosphate, a key endogenous in-
significant association with VC [vascular calcifica- hibitor of ectopic mineralization. Deficiency of ABCC6
tion]” (40). is considered pathogenetically linked to the devel-
In humans, inorganic phosphate serum levels are opment of disseminated calcifications in pseudox-
tightly regulated within the range of 2.5 to 4.5 mg/dL anthoma elasticum (54). ENPP1 breaks down ATP to
(0.81-1.45 mmol/L). In patients with CKD, profound adenosine monophosphate and pyrophosphate, and
impairment of phosphate homeostasis promotes the Ecto-5 0 -nucleotidase (CD73) converts extracellular
development of MAC (12,41). Extracellular levels of adenosine monophosphate into phosphate and
phosphate are sensed and taken up by the type III adenosine, which can signal through the 4 adenosine
sodium-dependent phosphate cotransporters PiT1 receptors. Lack of adenosine signaling promotes AC
and PiT2, as well as by the calcium-sensing receptor secondary to CD73 deficiency in affected patients
(CaSR). Mutations in PiT2 are found in idiopathic (20).
basal ganglia calcification, where calcification has Lack of CD73-mediated adenosine signaling causes
been hypothesized to be the result of excess accu- up-regulation of tissue nonspecific alkaline phos-
mulation of extracellular phosphate (22). More recent phatase (TNAP), a key component of ectopic miner-
studies have shown that these proteins not only alization (55). The lack of CD73-mediated adenosine
transport phosphate, but also appear to trigger signaling has led to increased levels of the transcrip-
intracellular signaling events that suppress osteo- tion factor FOXO1, which binds the TNAP promoter
genic programs (42,43) because deficiencies in PiT- and induces TNAP transcription (56). Importantly,
mediated signaling induced the up-regulation of this mechanism was also found in popliteal arteries
osteogenic genes and exacerbated MAC in a murine collected from non–CD73-deficient patients present-
CKD model (44,45). Nevertheless, the applicability of ing with MAC. This observation suggests that the
the experimental data to MAC evolution in humans mechanisms underlying CD73 deficiency—loss of
has been questioned (46-48), and the controversy adenosine signaling and up-regulation of FOXO1 ac-
concerning the recipe molding the ingredients into a tivity—may also have a role in a common MAC variety.
plausible molecular pathways hypothesis persists. Future studies should explore how CD73 and adeno-
The CaSR regulates CaP homeostasis and is also sine receptor protein levels change with age and
expressed within arteries with altered CaSR function whether their induced expression and activation may
impairing the transdifferentiation of vascular smooth protect against MAC.
muscle cells (VSMCs) to mineralizing cells (49). The Disruption of the ECM contributes to the develop-
CaSR is also expressed in monocytes, which have ment of MAC. This effect is appreciated in aneurysms,
been shown to prevent arterial calcification in vitro whereby calcification and disorganized and frag-
(50). Monocyte CaSR expression is decreased in CKD, mented aortic elastic lamina lead to dilatation and
and this is associated with an impaired ability of tortuosity (57). Patients with connective tissue dis-
monocytes to inhibit AC (50). orders, such as Marfan syndrome and several types of
Other minerals besides CaP ions may influence the Loeys-Dietz syndrome, also exhibit calcification in
development of MAC. In animal and laboratory tortuous vessels localized along damaged elastic
F I G U R E 1 Mechanisms Contributing to MAC Pathogenesis
1) Extracellular matrix (ECM) cues are detected by the discoidin domain receptor-1 (DDR1) promoting the activation of osteogenic genes. 2) DNA damage response
pathway leads to the accumulation of poly (adenosine monophosphate) ribose molecules. These molecules are secreted in matrix vesicles and act as niduses for
calcification. 3) Calcium or phosphate minerals can nucleate in the extracellular matrix, but this process is inhibited by g-carboxyglutamic acid. The type III sodium-
dependent phosphate cotransporter PiT1 or Pit2 (PiT1/2) removes inorganic phosphate (Pi) from the extracellular environment. 4) Tissue nonspecific alkaline phos-
phatase (TNAP) converts the endogenous inhibitor of mineral nucleation, pyrophosphate (PPi), into the mineral building block, inorganic phosphate. 5) Pyrophos-
phate and adenosine monophosphate (AMP) are produced from the activity of ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) in the extracellular
adenosine triphosphate (ATP) metabolic pathway. The substrate for adenosine triphosphate–binding cassette subfamily C member 6 (ABCC6) is thought to also be
adenosine triphosphate but currently is not known. 6) CD73-mediated A2b adenosine receptor (A2bAR) signaling functions to suppress media arterial calcification
(MAC) by cyclic adenosine monophosphate–mediated repression of the forkhead box O1 (FOXO1) gene. With a lack of this signaling, FOXO1 up-regulates alkaline
phosphatase (ALPL)/TNAP. MGP ¼ matrix g-carboxyglutamic acid protein; RUNX2 ¼ Runt-related transcription factor 2. Created using BioRender.
lamina, strikingly similar to the pathologic vascular switches in cell phenotypes (64). Stiffening of ECM
features of MAC observed in genetic diseases (58,59). transduced to VSMCs may trigger signaling events
These genetic aneurysmal diseases share increased through the collagen-binding discoidin domain re-
activity of the transforming growth factor-b signaling ceptor 1, which was found to promote osteogenic
pathway, and it has been proposed that the calcifi- differentiation and calcification (65). Exploration of
cation seen in these vessels is subsequent to the the significance of epigenetic imprinting stemming
aberrant remodeling of the vessel wall (60-63). from the developmental origin of individual vascular
Indeed, it is well known that cells are able to sense beds and their propensity to calcify the medial layer
the stiffness of their environment and that, provides an interesting example of prospective
conversely, the environment itself has may induce comprehensive disease modeling (66).
F I G U R E 2 Histology of MAC and Intimal (Atherosclerosis) Calcification
Medial arterial calcification (MAC) and the intimal atherosclerotic calcifications (Calc) are shown. The legends to each medial and intimal
calcification pattern are summarized in Table 1. The histologic sections shown in the left and right columns of the medial and intimal cal-
cifications were stained with (A to F and M to R) Movat pentachrome and (G to L and S to X) hematoxylin and eosin. The red boxes in the
Movat pentachrome sections indicate areas of magnifications shown in the hematoxylin sections and in stage IV (medial) and sheet (intimal)
Movat pentachrome sections. IEL ¼ internal elastic lamina (black arrows, black and white arrowheads). Modified and reproduced with
permission from https://doi.org/10.1016/j.jcmg.2018.08.039 and https://doi.org/10.1016/j.ejvs.2020.08.037.
Another mechanism driving MAC pathogenesis is such as senescence or cell death, but the product of
related to DNA damage response pathways. On DNA PARPs, poly[adenosine diphosphate-ribose] (PAR),
damage, the ataxia telangiectasia mutated protein itself can promote the MAC by 2 means: 1) activation
phosphorylates gH2AX, which marks the DNA lesion of osteogenic gene expression; and 2) amalgamation
and triggers DNA repair proteins (67). Among these and packaging of PAR in MVs that are delivered to the
response proteins are poly[adenosine diphosphate- ECM, where PAR and calcium and phosphate ions
ribose]polymerases (PARPs) which trigger events nucleate (65,68).
T A B L E 1 Stages of Progression: Comparison Between MAC and Intimal Calcifications
Medial Arterial Calcification Intimal Calcification
I Calcification of the internal elastic membrane with or Microcalcification (includes micro and punctate) is Micro and punctate
without extension into the media (A and G) identified by calcium particles ranging from >0.5 mm calcification
to <1 mm in diameter (M and S)
II Calcification coalescence and becomes confluent Small calcification is often accompanied by Fragment calcification
(varying in size from 1 to 3 mm), forming fragments inflammation, areas of microcalcification coalescence
of calcification (B and H) forming fragments of calcification that are >1 mm
but <3 mm in diameter (N and T)
III Calcification length > 3 mm and/or extending to involve Calcification of the intima >3 mm or >90o (O and U) Sheet calcification
>90o of the circumference (C and I) Calcification can extend and become circumferential
(P and V)
IV Calcifications of the media, spanning the entire
circumference (D and J)
Nodular Nodular calcification is rarely seen in medial wall, which Nodular calcification is composed of nodules of Nodular calcification
calcification is composed of nodules of calcification often calcification often accompanied by fibrin with a
accompanied by fibrin (E and K) fibrous cap (Q and W)
Bone formation Bone formation may be observed in fragmented and Bone formation can be observed within the regions of Bone formation
areas of sheet calcification (F and L) calcification (R and X)
Bone formation and rarely cartilaginous metaplasia may
be seen in late stages, most frequently in stages III
and IV, but rarely also in stage II
A to X correspond to Figures 2A to 2X. Modified and reproduced with permission from https://doi.org/10.1016/j.jcmg.2018.08.039; https://doi.org/10.1016/j.ejvs.2020.08.037.
The role of MVs in mineralization was first noticed significance (74). In a study of the UK Biobank par-
in cartilage (69), but it is now appreciated to ticipants whose arterial stiffness was measured, the
contribute to both MAC and IAC. In addition to arterial stiffness index was the subject of 2 GWAS
delivering packets of the mineral building blocks of analyses. These studies revealed 4 significant loci
HAP, MVs contain lipids, metabolites, microRNAs, (near TEX41, FOXO1, C1orf21, and MRVI1), with gene-
and proteins that promote the osteogenic switch of based analysis implying COL4A2 (75). Thus, although
cells (70). Although MVs play a role in atherosclerosis these studies provide preliminary evidence of a ge-
and MAC, the similarities and differences of their netic contribution to AC phenotypes, studies with
biogenesis, accumulation, and cargo have not been specific well-defined cohorts of patients with MAC
fully explored. are lacking.
In T2DM patients with sympathetic fiber damage,
HISTOPATHOLOGY
and in patients with sympathetic denervation
following sympathectomy, MAC is a common finding;
MAC affects predominantly muscular arteries. In
however, the mechanism causing calcification in such
large arteries, early stages are characterized by fine
patients remains to be clarified (71,72).
granulations of the CaP precipitates located within
In the setting of CKD, the potential role of micro-
the media in the proximity of the internal elastic
biome dysbiosis–derived uremic proteins in AC
membrane, later progressing in size and distribution.
pathogenesis has been proposed (73), although the
In advanced stages, large precipitates forming
proof of concept and confirmation in clinical studies
sheaths replacing large areas of the healthy media
of MAC are lacking.
and fragmentation of the internal elastic membrane
GENETIC CONSIDERATIONS. For rare mendelian are seen (7). In some cases, osteogenic transformation
disorders, the rare variants with large effect size and invasions of the intima can be seen (18). The
(mutations) directly cause the disease. In contrast, AC stages and differences between MAC and IAC have
associated with common clinical conditions such as been summarized in Figures 2A to 2X and in the cor-
CKD and DM appear to have a more complex etiology, responding Table 1.
with multiple low-frequency or common variants It is not known how far distally on the arterial tree
with small effect size interacting with nongenetic MAC can spread. Some studies suggest that in small
factors. For example, in patients with an ankle- arteries, numerous changes may occur, including se-
brachial index >1.3 suggestive of PAD, a meta- vere intimal hyperplasia and thrombotic occlusions
analysis of genome-wide association studies (7,31,76), along with CaP deposits (77). In amputated
(GWASs) from 21 population-based cohorts identified limbs of CLTI patients, a strong correlation was found
a single locus on chromosome 9 of genome-wide between MAC of the foot arteries and metatarsal artery
F I G U R E 3 MAC and Hemodynamics
Male patient with diabetes mellitus and gangrene of the right middle finger and x-ray angiography demonstrating advanced medial arterial
calcification (MAC) without evidence of stenotic lesions of the right upper extremity arteries. (A) High-fidelity blood pressure recordings at
the level of the mid brachial (Pa) and distal ulnar (Pd) arteries demonstrated a Pd/Pa ratio of 0.91 at baseline, (B) a Pd/Pa ratio of 0.80
following intravenous (140 mg/kg/min systemic infusion) and intra-arterial (150-mg) bolus administration of adenosine, and (C) a Pd/Pa ratio
of 0.76 following intra-arterial nitroglycerin (300 mg). The disproportional decrease in the distal diastolic pressures following vasodilatation
appears to suggest marked microvascular dysfunction. A hemodynamically significant lesion proximal to the target territory has been
excluded by x-ray angiography. FFR ¼ fractional flow reserve.
obstruction. The most common obstructive lesion was advanced MAC appears to cause an accentuated fall in
a combination of intimal thickening, advanced MAC, mean and diastolic arterial pressures. These data
and thrombosis that was striking “for the relatively appear to suggest a decrease in the driving force
small size of the metatarsal arteries” (78). resulting from the arterial wall rigidity and pathologic
peripheral vasodilatation (Figures 3A to 3C) (Lanzer
HEMODYNAMIC IMPAIRMENTS
et al, unpublished data, 2020). These early data seem
to corroborate the clinical observations in patients
MAC causes arterial wall stiffening, which is associ-
with CLTI secondary to MAC resulting in severe limb
ated with fundamental changes in central hemody-
ischemia (7).
namics: impaired windkessel function, increase in
left ventricular afterload, decrease in coronary artery INTERACTIONS WITH ATHEROSCLEROSIS
perfusion, and increase in blood flow pulsatility
(1,79,80). The limited studies of hemodynamics in The exact nature of interactions between MAC and
peripheral arteries of patients with MAC revealed atherosclerosis appears to be predominantly biome-
marked impairments suggestive of peripheral shunt- chanical. In healthy media, when atheroma develops,
ing (81). the luminal area is initially maintained by circum-
On the basis of our preliminary observations in ferential expansion (82). It is thought that this
patients with CLTI, following vasodilatation, compensatory stage lasts until the vessel’s cross-
F I G U R E 4 MAC in Ultrasound
Female patient with chronic kidney disease and medial arterial calcification (MAC) demonstrated by ultrasound. (A) B-mode ultrasound
longitudinal image of the left superficial femoral artery revealed on the near end arterial wall linear homogeneous echogenic band patho-
gnomonic for MAC; note the smooth endothelial interface of the superficial femoral artery; white arrowheads denote diffuse medial calci-
fications. (B) Color-coded Doppler image of A demonstrates normal arterial flow signal (blue); red corresponds to the accompanying vein;
white arrowheads as in A. (C) Color-coded Doppler and pulsed-wave Doppler images show a normal triphasic flow pattern indicating a
normal artery.
sectional area expands by approximately 40%. biomechanical factors (85), MAC occurs early in the
Beyond 40%, inward remodeling occurs, leading to a course of the disease along elastic lamellae. Thus,
decreasing lumen area. However, in the presence of whereas IACs of atherosclerosis occur in advanced
MAC, the compensatory stage may be prematurely disease and threaten patients with focal or multifocal
disrupted. MAC may encourage the inward phase of plaque evolution, rupture, and thromboembolic
remodeling to start earlier, thereby giving rise to an complications, MAC occurs early in the disease’s
acceleration of luminal stenosis related to the pro- course, at least in the peripheral arteries, and it is the
gression of atherosclerosis. How the enhanced inward main risk factor for progressive shutdown of circula-
remodeling occurs physiologically remains to be fully tion secondary to small artery disease.
characterized, but 1 hypothesis suggests a competi-
tion in stiffness between the intima and the media DIAGNOSTIC EVALUATIONS
(83). If the stiffness of the media is much greater than
that of the intima, the media will act as a stiff band MAC is most commonly seen in the arteries of the
surrounding a relatively compliant intima. With lower and upper extremities, but it has been reported
further growth, the intima is forced to encroach into in all major vascular territories (86), including the
the lumen, thus leading to a reduction in blood flow. temporal (87), facial (88), and mammary (89) arteries.
A stiffer media could also interfere with the vessel’s The diagnosis of MAC is frequently incidental in the
ability to stretch and recoil, disrupting the normal course of other diagnostic evaluations or a result of
development of pressure gradients. A long-term systematic screening in patients with PAD. In PAD,
disruption in hemodynamics may also promote an the ankle-brachial index has been the preferred first-
earlier onset of disease because low endothelial shear line method to screen for MAC; however the validity
stress is thought to be atherogenic (84). of this index to determine MAC has been questioned
Although IAC typically occurs in advanced (29). Recently, a simple MAC score was proposed on
atherosclerotic lesions often promoted by localizing the basis of a planar radiograph of the foot in 2
projections. Looking at 5 vascular sites and evalu-

F I G U R E 5 MAC in Native X-Ray and X-Ray Angiographic Images
ating the length of the “rail-tracking,” patients can be
divided into 3 MAC categories: absent, moderate, and
severe (90).
Multiple imaging modalities allowing differentia-
tion between the intimal and medial layers of the
arterial walls are suitable for MAC detection. In pe-
ripheral arteries, MAC and IAC can be clearly
differentiated by transcutaneous ultrasound imag-
ing. MAC is recognized on longitudinal views by
echogenic abluminal bands and typically smooth
endothelial interfaces (Figures 4A to 4C), whereas
IAC appears granular and spotty (91). On conven-
tional x-ray radiography, MAC radiopaque “rail-
tracking” type shadows are typical, in contrast to the
irregular and discrete “plaquelike” IAC (92). If con-
ventional x-ray radiography is combined with angi-
ography, smooth endothelial interfaces can be
appreciated (Figures 5A to 5D). Computed tomogra-
phy (CT) exquisitely depicts peripheral and coronary
artery calcifications as high- density signals, yet
because of its limited spatial resolution, MAC and
IAC cannot be differentiated on conventional CT
images. A distinction between these 2 types of
calcification on the basis of differences in patterns (A and B) Male patient with diabetes mellitus and medial arterial calcification (MAC)
demonstrated by x-ray imaging. (A) Native x-ray image of the right brachial artery
has been proposed on thin-slice CT images (93).
demonstrating the typical “railroad track” pattern of calcification parallel to the humerus
Compared with other imaging modalities, CT may (white arrowheads at the edge of calcifications). (B) X-ray angiographic image of the
provide whole body estimates of AC burden, as patient in A showing the smooth endothelial interface following contrast media filling;
shown in Figures 6A to 6F. heavy calcification are recognized as white stripes along the lumen (white arrowhead).
Coronary artery calcifications can be reliably visu- (C and D) Male patient with cardiac arrhythmias undergoing coronary angiography. (C)
High-resolution native x-ray image of the right coronary artery reveals a disseminated
alized by CT, and coronary calcium scores have been
pattern of calcifications (2 white arrowheads on the left) suggestive of medial arterial
used for detection and prognostication of coronary calcification seen in A; the image appears partly duplicated because of the rapid motion
artery disease for more than 2 decades (94). However, of the cardiac base (white arrowhead on the right). (D) X-ray coronary angiographic
to allow a distinction between MAC and IAC, either image of C showing the smooth endothelial interface demarcated by homogeneous
intravascular ultrasound or, even more optimally, contrast media filling (black arrow).
optical coherence tomography with axial resolution

of approximately 100 and 10 m m, respectively, is
reports on patients undergoing long-term dialysis
needed (95). By using optical coherence tomography,
suggests that the benefit of non–calcium-based
the artery wall layers containing calcifications are
phosphate binders is not limited to phosphate
clearly visualized (Figures 7A to 7C). However, given
reduction, but rather it extends to reduction in the
their invasiveness, both techniques are applied only
calcium load that patients receive from calcium-
if clinically indicated in selected patients. Figures 8A
based binders and an effective control of para-
to 8F demonstrate the typical native and angio-
thyroid hormone (98). Indeed, parathyroid hormone
graphic x-rays and clinical findings of MAC in the limb
was shown to have direct procalcifying effects in an-
arteries.
imal experiments (99). Additionally, these binders
The typical laboratory test values and biomarkers
enhanced bone mineral density while slowing AC
for MAC are summarized in Supplemental Appendix B.
progression (100).
THERAPY Other experimental therapeutic approaches
included vitamin K supplementation, pyrophosphate,
PHARMACOLOGIC THERAPY. Slowing of MAC pro- tenapanor, and acetazolamide. The Valkyrie study
gression in patients with CKD has been demonstrated demonstrated a nominal but nonsignificant slower
with the use of phosphate binders and calcimimetic progression of calcifications in patients with atrial
agents (96,97). An extensive body of published fibrillation who were undergoing hemodialysis and
F I G U R E 6 MAC in CT
A 3-dimensional computed tomography (CT) reconstruction surface rendering in a male patient with medial arterial calcification (MAC). (A) Whole body scan reveals
calcifications (white spots) within the abdominal aorta and the pelvic and leg arteries. (B) Enlarged section of the internal carotid artery from the image in A appears
relatively free from calcifications. (C) Enlarged section of the superficial femoral artery with heavy diffuse calcification typical of medial arterial calcification. (D) Calcium
windowing highlights the calcifications of the vasculature. (E) Coronary computed tomography angiography shows extensive calcifications of the proximal left anterior
descending coronary artery and lesser calcifications of the proximal right and left circumflex coronary arteries. (F) Enlarged section of the proximal left anterior
descending artery shows heavy confluent, partly circumferential calcifications.
who were randomized to rivaroxaban with or without Pyrophosphate inhibits calcium and phosphate
vitamin K supplementation (101). deposition in bone and soft tissue. Bisphosphonates
Tenapanor, an inhibitor of the sodium-hydrogen are analogues of pyrophosphate, and animal studies
isoform 3 protein that regulates paracellular absorp- in the 1970s suggested the inhibitory activity of these
tion of sodium and phosphate in the intestine, agents on AC. Early forms of this drug (etidronate)
demonstrated several beneficial effects, such as have successfully halted the extensive calcification of
reduction in sodium and phosphorus absorption, newborns with generalized arterial calcification of
reduction in serum levels of phosphorus and fibro- infancy, thereby allowing their survival (104).
blast growth factor-23, and reduced heart mass in rats Currently, a clinical trial (Etidronate for Arterial Cal-
with chemically induced CKD. Additionally, it signif- cifications Due to Deficiency in CD73 [ACDC];
icantly decreased ectopic aortic calcification (102). In NCT01585402) is under way to test whether this
2 human trials of patients undergoing hemodialysis, therapy may also be useful in patients with CD73
tenapanor showed a clear dose-dependent ability to deficiency. Randomized controlled trials in patients
reduce serum phosphate levels, although the trials before beginning dialysis and in patients undergoing
did not measure the effect of this drug on AC (103). dialysis failed to reaffirm this hypothesis (105), and
F I G U R E 7 MAC in OCT
On optical coherence tomography (OCT), medial arterial calcification (MAC) appears as regions with homogeneous low signal intensity and sharply demarcated border
zones. (A to C) Cross-sections of the coronary arteries. (A) Medial arterial calcification spans more than one-half of the circumference (12 to 7 o’clock). (B) Extensive,
partly eccentric medial arterial calcification (1 to 3 o’clock). (C) Highly eccentric medial arterial calcification (10 to 12 o’clock). Note the smooth and regular intima in all 3
images. Courtesy and Copyright, Abbott, Inc.
the drug therapy appeared to promote osteomalacia does not cause regression of calcification (115).
(106). Etidronate, a bisphosphonate with a mecha- Although CaP overload represents a valid target in
nism of action similar to that of pyrophosphate, did patients with CKD, in disorders with preserved sys-
slow peripheral AC in a recent trial (107). temic CaP homeostasis, with the exception of pseu-
Aldosterone expands the procalcifying activity of doxanthoma elasticum, no therapy targets have been
hyperphosphatemia and facilitates osteoblastic established.
transformation of VSMCs in vitro (108), whereas spi- INTERVENTIONAL THERAPY. AC is associated with
ronolactone inhibited calcification in a Klotho- an increased risk of complications during surgical and
deficient mice model (109). Because patients with percutaneous coronary and peripheral revasculariza-
CKD often have hyperphosphatemia and concomitant tion procedures (14-16). Numerous surgical and
hyperaldosteronism, human trials with mineralocor- endovascular strategies were designed to improve
ticoid antagonists are currently under way. outcomes. For example, in endovascular procedures,
The anticalcifying activity of magnesium was noncompliant, scoring, and cutting balloons, excimer
demonstrated in 2 small human trials where magne- laser ablation, rotational and orbital atherectomy,
sium supplementations slowed the progression of and, more recently, intravascular lithotripsy have
coronary artery calcium in predialysis patients (110) been used to manage coronary and peripheral artery
and were associated with a decreased propensity to calcifications (116-118). Although these procedures do
AC development in patients undergoing hemodialysis not remove calcium, they fracture the superficial
(111). and/or deep calcifications, and by modifying the le-
The most interesting recent development in treat- sions they improve stent placement and provide ac-
ment of AC involves a direct inhibitor of HAP, the cess to distal target sites. Figures 9A to 9M review the
final step in all calcification processes. SNF472, histologic features of calcified lesions treated with
myoinositol hexaphosphate (112), was shown to slow endovascular procedures.
the progression of aortic valve and coronary artery
calcification significantly in patients with end-stage FUTURE PERSPECTIVES
CKD (113).
Whether any of the modest effects described here EXPERIMENTAL DATA AND PLAUSIBILITY
will translate into meaningful clinical benefits in CONSIDERATIONS. Currently, AC pathogenesis has
patients with MAC remains to be determined (114). been studied with a variety of in vitro and in vivo
Besides pharmacologic interventions, renal trans- experimental models using numerous experimental
plantation seems to slow the progression of MAC, but it protocols replicating some but not all of the AC
F I G U R E 8 MAC in Limb Arteries
(A to C) Male patient with the gangrene of the left great toe. (A) Native x-ray image of the left foot; heavy calcifications of the metatarsal and toe arteries, particularly
of the first metatarsal and great toe arteries, are shown (white arrowheads). (B) Angiographic x-ray image corresponding to the image in A; shown is the extensive
destruction of the small arteries; the paucity of the distal arterial supply is highlighted in the context of the great toe (oval broken line). (C) Photograph of the left foot
corresponding to the image in A; shown is extensive necrosis of the left great toe. (D to F) Male patient with the gangrene of the left middle finger. (D) Native x-ray
image of the left hand; heavy calcifications of the metacarpal and finger arteries are shown (white arrowheads). (E) Angiographic x-ray image corresponding to the
image in A; shown are multiple occlusions of the metacarpal and finger arteries and virtual absence of arterial blood flow to the fingers (exemplified by the middle
finger, oval broken line). (F) Photograph of the left hand corresponding to the image in A; shown is distal necrosis of the left middle finger. MAC ¼ medial arterial
calcification.
features encountered in human diseases (Supplemental Appendix C). To avoid these in-
(Supplemental Appendix A). Albeit providing impor- consistencies, experimental designs conforming to
tant insights to date, these studies have not uncov- the laws of thermodynamics governing the CaP pre-
ered novel and targetable molecular machinery that cipitation must be observed, and first principles
may be used to design pharmacologic prevention and thinking should be applied (Supplemental Appendix
therapy for AC. D). Identification of the triggers of CaP nucleation,
Given the extreme sensitivity of the CaP precipi- along with the potentially reversible stage of the
tation processes to the ionic strength of calcium and amorphous phosphate formation, and the irreversible
phosphorus ions modified by promoters and in- point of no return following HAP crystallization will
hibitors of mineralization, aberrations of in vivo be critical to define the targets of pharmacologic
conditions in experimental protocols bear the risk of therapy (27). Replication or direct observation of the
factitious results and errors in interpretations (27) in vivo conditions, while conserving the principles of
thermodynamics and focusing on the medial layer, genetic data, and conduct in silico and laboratory
will allow insights into the pathobiology of MAC, and studies to discover causal variants and identify
it holds the potential for developing causal candidate genes for MAC.
treatments. Third is the need for functional validation of GWAS
GENETICS. Genetic analysis provides an important discoveries. The International Knockout Mouse Con-
direction of research to unravel MAC molecular sortium, whose objective is to establish knockout
pathogenesis. Future genetic analyses of MAC will mouse lines for each of the protein-coding genes
depend on the availability of whole exome (123), may provide an opportunity to functionally
sequencing and the identification of appropriate validate candidate genes identified from GWASs and
patient cohorts, such as multigenerational families whole exome sequencing or whole genome
with affected members, to help identify monogenic sequencing studies. Systematic phenotyping for AC in
disorders (119,120). The genes identified in mono- these knockout mouse lines could yield novel genes
genic disorders may also be relevant for polygenic involved in VSMC function and the pathogenesis of
disease. However, robust characterization of genetic MAC. However, it is established that mice are not
variants contributing to polygenic forms of MAC re- good models for human MAC (41). The challenge lies
quires the establishment of large cohorts excluding in establishing noninvasive in vivo imaging modal-
secondary causes of AC, as outlined later in ities for the high-throughput detection and quantifi-
this section. cation of AC in mice. Fluorine-18–sodium fluoride
Genetic variants that promote the development of positron emission tomography has potential as a
MAC as a function of comorbidities such as DM and sensitive in vivo imaging technique (124). If greater
CKD are not easily identified by familial studies. resolution is required, ex vivo micro-CT analysis
Thus, discovery of genetic variants influencing MAC could also be considered to detect microcalcification
risk should follow several related paths. First, to (55).
discover low-frequency variants of high effect size, In addition to MAC manifesting as a comorbidity,
whole exome sequencing or whole genome primary systemic MAC characterized by the absence
sequencing should be undertaken using large and of predisposing factors has been reported (8). In the
diverse groups. Both of these techniques are limited absence of known risk factors for MAC, the possibility
by the ability to parse variants easily into benign or of inherited genetic causes is likely, thereby
pathogenic categories (121). providing an opportunity to search for additional
Second, GWASs could be undertaken to look for monogenic causes of MAC in these patients. The
common variants of small effect size that could availability of blood samples and vascular tissues
nevertheless yield important insights into the biology from patients with primary MAC would greatly
of AC. For example, a GWAS of >9,400 individuals simplify research protocols by providing valuable
identified an association with the histone deacetylase opportunities to identify the causes of MAC in such
9 (HDAC9) gene and atherosclerotic aortic calcifica- patients.
tion. This study showed HDAC9 to increase expres- Furthermore, the striking differences in the topo-
sion of RUNX family transcription factor 2 (RUNX2), graphic distribution of MAC among individual
which promotes the differentiation of VSMCs into vascular beds with strong predisposition for the ar-
cells with an osteogenic phenotype (122). Moreover, teries of the extremities appear to coincide with the
HDAC9 overexpression was shown to increase calci- differences in the embryologic origins of VSMCs (125).
fication in cultured VSMCs of mice, whereas defi- This finding potentially points toward an epigenetic
ciency of HDAC9 decreased calcification (122). Thus, predisposition worthy of being explored.
whereas HDAC9 was demonstrated as a potential OMICS. Although applications of omics approaches to
therapeutic target for atherosclerotic AC, comparable VC are still in their infancy, the research into MAC
data on MAC are not available. A major limitation of and IAC will benefit from a variety of studies using
the GWAS is the need for well-phenotyped cohorts different omics, including proteomics (126), metab-
with a sample size in the thousands. As mentioned olomics (127), and, most recently, single- cell tran-
earlier, discerning between IAC and MAC is not often scriptomics (128).
determined routinely. Furthermore, associations The in-depth characterization of calcifying blood
should be replicated in an independent cohort, again vessels or cells could yield important insights into
necessitating the collection of thousands more cases the evolving disease processes, mainly by providing
and control subjects. It is therefore clear that a large, unbiased tissue profiles. As depicted in Figure 10,
international, multidisciplinary team would be omics-based tissue profiles can have different
required to gather sufficient cases, analyze the resolutions.
F I G U R E 9 Histology of Calcified Plaques and Endovascular Interventions
Drug-Eluting Stent
A B C
Orbital Atherectomy
D E
Subintimal space
(A to M) Examples of histologic findings of intimal and medial calcifications (Ca) following endovascular interventions. (Top) Common femoral artery and superficial
femoral artery following drug-eluting stent placement and orbital atherectomy device in an ex vivo setting, respectively. (Bottom) Coronary artery treated by a
rotational atherectomy device, superficial femoral artery (SFA) treated with an intravascular lithotripsy device, and plaque debris obtained from directional atherectomy
device. (A) Histologic image of a stented common femoral artery with intimal nodular calcification. (B) A high-power image of the red-boxed area in A. Some stent
struts remain uncovered, others are partially covered, and a few are surrounded by fibrin. (C) A high-power image of the blue-boxed area in A. Stent struts are
surrounded by fibrin. (D) Histologic image of a superficial femoral artery chronic total occlusion (CTO) lesion treated with an orbital atherectomy device in an ex vivo
setting. (E) Note that calcified plate (Caþþ) is broken into multiple pieces and the blue arrow shows subintimal space. (F and G) Histologic images of coronary artery
sections stained with Movat pentachrome showing sheet calcium and an area of nodular calcification. The lesion was treated 12 hours earlier by a rotational athe-
rectomy device; note the sharp edge of the calcified plaque (red arrowheads) where the burr interacted with the plaque. Medial dissection is observed (blue ar-
rowheads). (H) A histologic image of SFA treated with an intravascular lithotripsy device. (I) A high-power image of the green-boxed area in H shows fracture of the
medial calcification resulting from lithotripsy. (J) A coregistered micro-CT image with the histologic image H. Red arrows show the fracture sites in MAC. (K) A gross
image of plaque tissues removed by directional atherectomy device. (L and M) A histologic image of the collected plaque with calcification (L) and thrombus (Th) (M).
A to D, H, I, L, and M are stained with hematoxylin and eosin. E to G are stained with Movat pentachrome. Caþþ ¼ calcium; CT ¼ computed tomography;
MAC ¼ medial arterial calcification. Modified and reproduced with permission from https://doi.org/10.1016/j.jcin.2019.10.060. (F and G) Modified and reproduced with
permission from https://doi.org/10.1016/0002-8703(9590384-4).
To date, a few important bulk-omics–based studies cardiovascular diseases, including heart failure (129)
on AC have been published. Long-noncoding RNAs and atherosclerosis (130). Using an in vitro model, key
(LncRNAs) are gene regulatory transcripts longer than regulatory LncRNAs were detected by transcriptomics
200 bases that have been associated with major in calcifying rat VSMCs (131), and the identified
Rotational Atherectomy
F G
Lithotripsy
H I J
Directional Atherectomy
K L M
candidates were assayed on the basis of earlier samples in patients with early and late stages of a
knowledge. In this experimental set-up, LncRNA broad spectrum of cardiovascular diseases. To iden-
Lrrc75a-as1 was confirmed to have a mechanistic tify potentially beneficial drugs, unbiased drug sig-
impact on in vitro calcification. A similar pipeline was natures (Connectivity MAP) (132) were applied to
applied by Schanstra et al (119), who analyzed the match the proteomic disease signature, assuming that
proteomic difference between human arterial tissue the drug could reverse the pathogenic phenotype. By
F I G U R E 1 0 MAC and Omics
Spatial Omics
Extracellular matrix degradation Promoter/inhibitor detection
Functional contextualization
Osteogenic differentiation
Single Cell Omics of promoters/inhibitors
Failed anticalcific effects Drug repurposing
Genetics Cellular interplay within

the vessel wall
Bulk Omics
Matrix vesicle accumulation
Functional cell states
Pi/Ca2+ homeostastis imbalance

Spatial context in calcification
Apoptosis and inflammation Calcifying Intima

lesion
Media
Adjacent tissue Adventitia
Overview of pathophysiology of medial arterial calcification (MAC) and the proposed omics approaches to study it. (Left) Different proposed pathophysiologic prin-
ciples suggested triggering and promoting vascular calcifications, including medial arterial calcification. (Middle) Calcified sites in the vasculature can be studied with
omics at different degrees of resolutions. Bulk omics (bottom) provides a gross tissue profile. Single- cell omics (middle) provides profiles of individual cells. Spatial
omics (top) yields profiles of regions represented as the 2-dimensional sections of the tissues. (Right) The proposed approaches have not been systematically applied
to study medial arterial calcification but hold promise to address numerous aspects of calcifications processes providing tangible insights and hypotheses for mo-
lecular pathogenesis of medial arterial calcification.
using this approach, cytosolic phospholipase A 2 in- fibroblasts, pericytes, and transitional cell types
hibitor was predicted to prevent AC, a prediction that termed myofibroblasts (136). Single-cell technologies
was subsequently confirmed experimentally in vivo. are likely to provide important insights into the state
Regarding the evaluation of complex biologic sys- of cellular heterogeneity of VSMCs and transitional
tems and the integration molecular cascades by cells in their functional evolution, along with their
network analysis, Song et al (133) assembled a calcifying potential.
network from protein-protein interactions in VSMCs. ACs exhibit a strong topographic aspect. They begin
Known proteins involved in AC were used as seed locally restricted to specific loci of the vasculature and
genes and mapped onto the resulting network. This are also present within specific layers of the vessel
was repeated for other endophenotypes such as wall. In MAC, calcification sites occur predominantly
fibrosis and inflammation to evaluate their relation- in the arteries of the extremities, and here, within the
ship with the AC endophenotype through the protein- media as the characteristic single target. Thus, spatial
protein interaction network. The network was further aspects of pathogenesis are expected to be highly
used for drug repurposing by assessing the network influential and should be addressed with proper sam-
distance between drug targets and the calcification pling strategies, as demonstrated for calcific aortic
module. Three candidates, including the mammalian valve disease (137) and by using spatially resolved
target of rapamycin (mTOR) inhibitor everolimus, omics. These novel technologies assess molecular
were experimentally validated to reduce VSMC profiles while retaining their spatial information.
calcification. Although novel technologies are under development,
Single-cell omics allows analysis across the spatial transcriptomics are the most advanced at pre-
complexity spectrum down to a single cell. This ap- sent (138,139). The blood vessels, small and well-
pears to be of particular value to research on AC structured organs, constitute excellent study sub-
because it has enabled insights into vessel wall jects for these technologies. Here, the functional
biology (128), especially into cell type and cell state analysis of calcifying sites could substantially improve
composition in atherosclerotic plaques (134,135). The our understanding of how and why calcification occurs
medial layer of arteries contains mainly VSMCs, at certain sites while other sites are spared.
C ENTR AL I LL U STRA T I O N Medial Arterial Calcification: A Systemic Vascular Disorder Devastating

Peripheral Circulation
Lanzer, P. et al. J Am Coll Cardiol. 2021;78(11):1145–1165.
In health, calcium phosphate homeostasis is maintained, and crystallization is prevented. In medial arterial calcification calcium phosphate homeostasis breaks down,
resulting in progressive mineralization and, in more advanced stages, bone formation within the medial layer. Certain pathogenetic principles, including smooth
muscle cell osteogenic differentiation, apoptosis, inflammation, and molecular defects of matrisome, have been reported to regulate the calcification process. Medial
arterial calcification impairs hemodynamics and often causes chronic limb-threatening ischemia. Omics approaches hold distinct promise to define medial arterial
calcification molecular pathogenesis and design treatments.
Although omics have the potential to reshape assumptions may not always reflect biologic
research on AC, important limitations should be reality. Nevertheless, omics holds a great promise
considered. The major limitations include repro- providing data-driven hypotheses requiring valida-
ducibility batch effects, cost, and coverage. Single- tion experiments and allowing causal conclusions,
cell and spatially resolved omics and the computa- and as such, omics needs to be systematically used
tional methods used for data analysis are still in to support biochemical, genetic, and experimental
early stages of development, and mechanistic approaches to understand AC (24) and specifically
insight from omics studies relying on biologic MAC.
The definition of the MAC molecular biology will received support from Informatics for Life funded by the Klaus
Tschira Foundation; has received funding from GSK and Sanofi; and
ultimately need to be addressed by interdisciplinary
expects consultant fees from Travere Therapeutics. Dr Millan has
teams with complementary expertise (24,140,141). received financial support from the Spanish Ministry of Science,
The Central Illustration summarizes the key points Innovation, and Universities (grant no: PGC2018_095795_B_I00). Dr
pertaining to the pathophysiology and clinical mani- Sato has received institutional research support from NIH-HL141425,
Leducq Foundation Grant, 480 Biomedical, 4C Medical, 4Tech,
festation of MAC.
Abbott, Accumedical, Alivas, Amgen, Biosensors, Boston Scientific,
Canon USA, Cardiac Implants, Celonova, Claret Medical, Concept
SUMMARY Medical, Cook, CSI, DuNing, Inc, Edwards LifeSciences, Emboline,
Endotronix, Envision Scientific, Lutonix/Bard, Gateway, Lifetech,
Limflo, MedAlliance, Medtronic, Mercator, Merill, Microport Medical,
MAC is a chronic systemic vascular disorder distinct
Microvention, Mitraalign, Mitra assist, NAMSA, Nanova, Neovasc,
from atherosclerosis that is frequently but not always NIPRO, Novogate, Occulotech, OrbusNeich Medical, Phenox, Profusa,
associated with DM, CKD, and aging. It is part of Protembis, Qool, ReCor Medical, Senseonics, Shockwave, Sinomed,
several complex phenotypes observed in numerous Spectranetics, Surmodics, Symic, Vesper, W.L. Gore, and Xeltis. Dr
Virmani has received institutional research support from NIH-
less common diseases. The hallmarks of MAC include
HL141425, Leducq Foundation Grant, 480 Biomedical, 4C Medical,
disseminated and progressive precipitation of CaP 4Tech, Abbott, Accumedical, Alivas, Amgen, Biosensors, Boston Sci-
within the medial layer, compromise of hemody- entific, Canon USA, Cardiac Implants, Celonova, Claret Medical,
namics, a prolonged and clinically silent course, and Concept Medical, Cook, CSI, DuNing, Inc, Edwards LifeSciences,
Emboline, Endotronix, Envision Scientific, Lutonix/Bard, Gateway,
CLTI. The pathogenesis of MAC is unknown. We re-
Lifetech, Limflo, MedAlliance, Medtronic, Mercator, Merill, Microport
view the current state of knowledge about MAC and Medical, Microvention, Mitraalign, Mitra assist, NAMSA, Nanova,
briefly outline directions for future research. Neovasc, NIPRO, Novogate, Occulotech, OrbusNeich Medical, Phe-
nox, Profusa, Protembis, Qool, ReCor Medical, Senseonics, Shock-
ACKNOWLEDGMENT Dr Lanzer acknowledges fruit-
wave, Sinomed, Spectranetics, Surmodics, Symic, Vesper, W.L. Gore,
ful discussions on coronary calcifications with Dr and Xeltis; is a consultant for Abbott Vascular, Boston Scientific,
Gary Mintz of Cardiovascular Research Foundation, Celonova, OrbusNeich Medical, Terumo Corporation, W.L. Gore,
Edwards Lifesciences, Cook Medical, CSI, ReCor Medical, SinoMedical
New York, during the late phases of drafting the
Sciences Technology, Surmodics, and Bard BD; and is a Scientific
manuscript. Advisory Board Member for Medtronic and Xeltis. Dr St. Hilaire has
received support from National Institutes of Health grants HL142932
and HL117917. All other authors have reported that they have no re-
lationships relevant to the contents of this paper to disclose.
Dr Furniss is supported by the National Institute for Health Research
(NIHR) Oxford Biomedical Research Centre. Dr Schuchardt has
received support from Charité 3R and the Bundesministerium für ADDRESS FOR CORRESPONDENCE: Dr Peter Lanzer,
Bildung und Forschung. Dr J.D. Lanzer has received support from Mitteldeutsches Herzzentrum–Standort Bitterfeld,
Informatics for Life funded by the Klaus Tschira Foundation. Dr
Health Care Center Bitterfeld-Wolfen gGmbH,
Thakker has received support from a Wellcome Trust Senior Investi-
gator Award, an NIHR Senior Investigator Award, and the NIHR Ox- Friedrich-Ludwig-Jahn-Strasse 2, D-06749 Bitterfeld-
ford Biomedical Research Centre Programme. Dr Saez-Rodriguez has Wolfen, Germany. E-mail: lanzer.peter@web.de.
REFERENCES
1. Reesink KD, Spronck B. Constitutive interpreta- 5. Pisani I, De Troia A, Allegri L, Corradi D, 10. Lehto S, Niskanen L, Suhonen M, Rönnemaa T,
tion of arterial stiffness in clinical studies: a Vaglio A. Malignant Mönckeberg medial calcific Laakso M. Medial artery calcification. A neglected
methodological review. Am J Physiol Heart Circ sclerosis. Intern Emerg Med. 2018;13:615–617. harbinger of cardiovascular complications in non-
Physiol. 2019;316:H693–H709. insulin-dependent diabetes mellitus. Arterioscler
6. Lanzer P, Boehm M, Sorribas V, et al. Medial
Thromb Vasc Biol. 1996;16:978–983.
2. Weber T, Chirinos JA. Pulsatile arterial haemo- vascular calcification revisited: review and
dynamics in heart failure. Eur Heart J. 2018;39: perspectives. Eur Heart J. 2014;35:1515–1525. 11. London GM, Guérin AP, Marchais SJ, Métivier F,
3847–3854. 7. Mustapha JA, Diaz-Sandoval LJ, Saab F. Infra- Pannier B, Adda H. Arterial media calcification in
popliteal calcification patterns in critical limb end-stage renal disease: impact on all-cause and
3. Chirinos JA, Segers P, Hughes T, Townsend R.
ischemia: diagnostic, pathologic and therapeutic cardiovascular mortality. Nephrol Dial Transplant.
Large-artery stiffness in health and disease: JACC
implications in the search for the endovascular 2003;18:1731–1740.
state-of-the-art review. J Am Coll Cardiol.
2019;74:1237–1263. holy grail. J Cardiovasc Surg (Torino). 2017;58: 12. Nelson AJ, Raggi P, Wolf M, Gold AM,
383–401. Chertow GM, Roe MT. Targeting vascular calcifi-
4. Mowafy KA, Soliman M, Hammoda AM,
8. Lanzer P. Mediakalzinose Mönckeberg. cation in chronic kidney disease. J Am Coll Cardiol
Soliman RM. Bilateral lower limb disabling
Z Kardiol. 1998;87:586–593. Basic Trans Science. 2020;5:398–412.
claudication in a young man: a case of Mön-
ckeberg’s arteriosclerosis. 2019. Vasc Endovasc 9. Niskanen L, Siitonen O, Suhonen M, 13. Pescatore LA, Gamarra LF, Lieberman M.
Rev. 2019;2(1):48–52. Accessed June 6, 2020. Uusitupa MI. Medial artery calcification predicts Multifaceted mechanisms of vascular calcification
https://www.verjournal.com/articles/Disabling- cardiovascular mortality in patients with NIDDM. in aging. Arterioscler Thromb Vasc Biol. 2019;39:
Claudication-Monckebergs-Arteriosclerosis Diabetes Care. 1994;17:1252–1256. 1307–1316.
14. Généreux P, Madhavan MV, Mintz GS, et al. not a valid proxy for peripheral medial arterial 44. Chavkin NW, Chia JJ, Crouthamel MH,
Ischemic outcomes after coronary intervention of calcification. Atherosclerosis. 2021;323:13–19. Giachelli CM. Phosphate uptake-independent
calcified vessels in acute coronary syndromes. signaling functions of the type III sodium-
30. Narula N, Dannenberg AJ, Olin JW, et al. Pa-
Pooled analysis from the HORIZONS-AMI dependent phosphate transporter, PiT-1, in
thology of peripheral artery disease in patients
(Harmonizing Outcomes with Revascularization vascular smooth muscle cells. Exp Cell Res.
with critical limb ischemia. J Am Coll Cardiol.
and Stents in Acute Myocardial Infarction) and 2015;333:39–48.
2018;72:2152–2163.
ACUITY (Acute Catheterization and Urgent Inter-
45. Yamada S, Leaf EM, Chia JJ, Cox TC, Speer MY,
vention Triage Strategy) TRIALS. J Am Coll Cardiol. 31. O’Neill WC, Han KH, Schneider TM,
Giachelli CM. PiT-2, a type III sodium-dependent
2014;63:1845–1854. Hennigar RA. Prevalence of nonatheromatous le-
phosphate transporter, protects against vascular
sions in peripheral artery disease. Arterioscler
15. Bourantas CV, Zhang Y-J, Garg S, et al. calcification in mice with chronic kidney disease
Thromb Vasc Biol. 2015;35:439–447.
Prognostic implications of severe coronary fed a high-phosphate diet. Kidney Int. 2018;94:
32. Mintz GS, Popma JJ, Pichard AD, et al. Patterns 716–727.
calcification in patients undergoing coronary
of calcification in coronary artery disease. A sta-
artery bypass surgery: an analysis of the SYNTAX 46. Hortells L, Sosa C, Guillén N, Lucea S, Millán A,
tistical analysis of intravascular ultrasound and
study. Catheter Cardiovasc Interv. 2015;85:199– Sorribas V. Identifying early pathogenic events
coronary angiography in 1155 lesions. Circulation.
206. during vascular calcification in uremic rats. Kidney
1995;91:1959–1965.
16. Misare BD, Pomposelli Jr FB, Gibbons GW, Int. 2017;92:1384–1394.
33. Abramovitz Y, Jilaihawi H, Chakravarty T,
Campbell DR, Freeman DV, LoGerfo FW. Infrapo- 47. Komaba H, Fukagawa M. Phosphate: a poison
Mack MJ, Makkar RR. Porcelain aorta; a compre-
pliteal bypasses to severely calcified, unclampable for humans? Kidney Int. 2016;90:753–763.
hensive review. Circulation. 2020;131:827–836.
outflow arteries: two-year results. J Vasc Surg.
48. Villa-Bellosta R, Millan A, Sorribas V. Role of
1996;24:6–15. 34. de Jong PA, Hellings WE, Takx RAP, Isgun I,
calcium-phosphate deposition in vascular smooth
van Herwaarden JA, Mali WPTM. Computed to-
17. Villa-Bellosta R, Egido J. Phosphate, pyro- muscle cell calcification. Am J Physiol Cell Physiol.
mography of aortic wall calcifications in aortic
phosphate, and vascular calcification: a question 2011;300:C210–C220.
dissection patients. PLoS One. 2014;9(7):e102036.
of balance. Eur Heart J. 2017;38:1801–1804.
https://doi.org/10.1371/journal.pone.0102036 49. Alam M, Kirton JP, Wilkinson FL, et al. Calci-
18. Deneke T, Langner K, Grewe PH, Harrer E, fication is associated with loss of functional
35. Everhart JE, Pettitt DJ, Knowler WC, Rose FA,
Müller KM. Ossification in atherosclerotic carotid calcium-sensing receptor in vascular smooth
Bennett PH. Medial arterial calcification and its
arteries. Z Kardiol. 2001;90(Suppl 3):106–315. muscle cells. Cardiovasc Res. 2009;81:260–268.
association with mortality and complications of
19. Boström K, Watson KE, Horn S, Wortham C, diabetes. Diabetologia. 1988;31:16–23. 50. Mary A, Objois T, Brazier M, et al. Decreased
Herman IM, Demer LL. Bone morphogenetic pro- monocyte calcium sensing receptor expression in
36. Smith DC, Bilmen GJ, Iqbal S, Robey S,
tein expression in human atherosclerotic lesions. patients with chronic kidney disease is associated
Pereira M. Medial artery calcification as an indi-
J Clin Invest. 1993;91:1800–1809. with impaired monocyte ability to reduce vascular
cator of diabetic peripheral vascular disease. Foot
calcification. Kidney Int. 2021;99:1382–1391.
20. St Hilaire C, Ziegler SG, Markello TC, et al. Ankle Int. 2008;29:185–190.
Nt5e mutations and arterial calcifications. N Engl J 51. Louvet L, Büchel J, Steppan S, Passlick-
37. Losurdo F, Ferraresi R, Ucci A, Zanetti A,
Med. 2011;364:432–442. Deetjen J, Massy ZA. Magnesium prevents
Clerici G, Zambon A. Association of infrapopliteal
phosphate-induced calcification in human aortic
21. Rutsch F, Ruf N, Vaingankar S, et al. Mutations medial arterial calcification with lower-limb am-
vascular smooth muscle cells. Nephrol Dial
in ENPP1 are associated with ’idiopathic’ infantile putations in high-risk patients: a systematic review
Transplant. 2013;28:869–878.
arterial calcification. Nat Genet. 2003;34:379–381. and meta-analysis. Vasc Med. 2021;26(2):164–173.
https://doi.org/10.1177/1358863X20979738 52. Xu J, Bai Y, Jin J, et al. Magnesium modulates
22. Wang C, Li Y, Shi L, et al. Mutations in slc20a2
the expression levels of calcification-associated
link familial idiopathic basal ganglia calcification 38. Schurgers LJ, Teunissen KJ, Knapen MH, et al.
factors to inhibit calcification in a time-
with phosphate homeostasis. Nat Genet. 2012;44: Novel conformation-specific antibodies against
dependent manner. Exp Ther Med. 2015;9:1028–
254–256. matrix gamma-carboxyglutamic acid (Gla) protein:
1034.
undercarboxylated matrix Gla protein as marker
23. Lebwohl M, Halperin M, Phelps RG. Occult
for vascular calcification. Arterioscler Thromb Vasc 53. Newby AC. Adenosine and the concept of
pseudoxanthoma elasticum in patients with pre-
Biol. 2005;25:1629–1633. ‘retaliatory metabolites’. Trends Biochem Sci.
mature cardiovascular disease. N Engl J Med.
1984;9:42–44.
1993;329:1237–1239. 39. Dalmeijer GW, van der Schouw YT,
Magdeleyns EJ, et al. Matrix Gla protein species 54. Li Q, van de Wetering K, Uitto J. Pseudox-
24. Rogers MA, Aikawa E. Cardiovascular calcifi-
and risk of cardiovascular events in type 2 diabetic anthoma elasticum as a paradigm of heritable
cation: artificial intelligence and big data accel-
patients. Diabetes Care. 2013;36:3766–3771. ectopic mineralization disorders: pathomechan-
erate mechanistic discovery. Nat Rev Cardiol.
isms and treatment development. Am J Pathol.
2019;16:261–274. 40. Barrett H, O’Keeffe M, Kavanagh E, Walsh M,
2019;189:216–225.
O’Connor EM. Is matrix Gla protein associated with
25. Schantl AE, Ivarsson ME, Leroux JC. Investi-
vascular calcification? A systematic review. Nutri- 55. Jin H, St Hilaire C, Huang Y, et al. Increased
gational pharmacological treatments for vascular
ents. 2018;10(4):415. https://doi.org/10.3390/ activity of TNAP compensates for reduced aden-
calcification. Adv Ther. 2019;2:1800094. https://
nu10040415 osine production and promotes ectopic calcifica-
doi.org/10.1002/adtp.201800094
tion in the genetic disease ACDC. Sci Signal.
41. Joolharzadeh P, St Hilaire C. CD73 (cluster of
26. Herrmann J, Babic M, Tolle M, van der Giet M, 2016;9(458):ra121.
differentiation 73) and the differences between
Schuchardt M. Research models for studying
mice and humans. Arterioscler Thromb Vasc Biol. 56. Moorhead 3rd WJ, Chu CC, Cuevas RA, et al.
vascular calcification. Int J Mol Sci. 2020;21(6):
2019;39:339–348. Dysregulation of FOXO1 (forkhead box O1 protein)
2204.
drives calcification in arterial calcification due to
42. Bon N, Couasnay G, Bourgine A, et al. Phos-
27. Millan A, Lanzer P, Sorribas V. The thermody- deficiency of CD73 and is present in peripheral
phate (Pi)-regulated heterodimerization of the
namics of medial vascular calcification. Front Cell artery disease. Arterioscler Thromb Vasc Biol.
high-affinity sodium-dependent Pi transporters
Dev Biol. 2021;9:633465. 2020;40:1680–1694.
PiT1/Slc20a1 and PiT2/Slc20a2 underlies extra-
28. Krüger K. Epidemiologie der peripheren cellular Pi sensing independently of Pi uptake. 57. Schlatmann TJ, Becker AE. Pathogenesis of
arteriellen Verschlusskrankheit in Deutschland; J Biol Chem. 2018;293:2102–2114. dissecting aneurysm of aorta. Comparative histo-
Was ist gesichert und was ist offen? Hamostaseo- pathologic study of significance of medial
43. Voelkl J, Lang F, Eckardt KU, et al. Signaling
logie. 2006;26:193–196. changes. Am J Cardiol. 1977;39:21–26.
pathways involved in vascular smooth muscle cell
29. Hoek AG, Sabine R Zwakenberg SR, Petra J M calcification during hyperphosphatemia. Cell Mol 58. Wanga S, Hibender S, Ridwan Y, et al. Aortic
Elders PJM. An elevated ankle-brachial index is Life Sci. 2019;76:2077–2091. microcalcification is associated with elastin
fragmentation in Marfan syndrome. J Pathol. 74. Murabito JM, White CC, Kavousi M, et al. As- 2021;28(2):194–207. https://doi.org/10.1177/
2017;243:294–306. sociation between chromosome 9p21 variants and 1526602820966309
the ankle-brachial index identified by a meta-
59. Markello TC, Pak LK, St Hilaire C, et al. 91. Liu KH, Chu WC, Kong AP, et al. US assessment
analysis of 21 genome-wide association studies.
Vascular pathology of medial arterial calcifications of medial arterial calcification: a sensitive marker
Circ Cardiovasc Genet. 2012;5:100–112.
in NT5E deficiency: implications for the role of of diabetes-related microvascular and macro-
adenosine in pseudoxanthoma elasticum. Mol 75. Zekavat SM, Aragam K, Emdin C, et al. Genetic vascular complications. Radiology. 2012;265:294–
Genet Metab. 2011;103:44–50. association of finger photoplethysmography- 302.
derived arterial stiffness index with blood pres- 92. Lidbom A. Arteriosclerosis and arterial
60. Canadas V, Vilacosta I, Bruna I, Fuster V.
sure and coronary artery disease. Arterioscler thrombosis in the lower limb: a roentgenological
Marfan syndrome. Part 1: pathophysiology and
Thromb Vasc Biol. 2019;39:1253–1261. study. Acta Radiol Suppl. 1950;80:1–80.
diagnosis. Nat Rev Cardiol. 2010;7:256–265.
76. Goldenberg S, Alex M, Joshi RM, 93. Konijin LCD, van Overhagen H, Takx RAP, de
61. Neptune ER, Frischmeyer PA, Arking DE, et al.
Blumenthal HAT. Nonatheromatous peripheral Jong PA, Veger HTC, Mali WPTM. CT calcification
Dysregulation of TGF-beta activation contributes
vascular disease of the lower extremity in diabetes patterns of peripheral arteries in patients without
to pathogenesis in Marfan syndrome. Nat Genet.
mellitus. Diabetes. 1959;8:261–273. known peripheral arterial disease. Eur J Radiol.
2003;33:407–411.
77. Johnson RC, Leopold JA, Loscalzo J. Vascular 2020;128:108973. https://doi.org/10.1016/j.ejrad.
62. Gallo EM, Loch DC, Habashi JP, et al. Angio- 2020.108973
calcification. Circ Res. 2006;99:1044–1059.
tensin II-dependent TGF-beta signaling contrib-
78. Ferrier TM. Comparative study of arterial dis- 94. Greenland P, Blaha MJ, Budoff MJ, Erbel R,
utes to Loeys-Dietz syndrome vascular
ease in amputated lower limbs from diabetics and Watson KE. Coronary calcium score and cardio-
pathogenesis. J Clin Invest. 2014;124:448–460.
non-diabetics (with special reference to feet ar- vascular risk: JACC state-of-the-art review. J Am
63. Callewaert BL, Willaert A, Kerstjens- Coll Cardiol. 2018;72:434–447.
teries). Med J Aust. 1967;1:5–11.
Frederikse WS, et al. Arterial tortuosity syndrome:
95. Wang X, Matsumura M, Mintz GS, et al. In vivo
clinical and molecular findings in 12 newly identi- 79. McEniery CM, McDonnell BJ, So A, et al. Aortic
calcium detection by comparing optical coherence
fied families. Hum Mutat. 2008;29:150–158. calcification is associated with aortic stiffness and
tomography, intravascular ultrasound, and angi-
isolated systolic hypertension in healthy in-
64. Jaalouk DE, Lammerding J. Mechano- ography. J Am Coll Cardiol Img. 2017;10:869–879.
dividuals. Hypertension. 2009;53:524–531.
transduction gone awry. Nat Rev Mol Cell Biol.
96. Chen NC, Hsu CY, Chen CL. The strategy to
2009;10:63–73. 80. Mitchell GF. Aortic stiffness, pressure and
prevent and regress the vascular calcification in
flow pulsatility, and target organ damage. Appl
65. Ngai D, Lino M, Rothenberg KE, Simmons CA, dialysis patients. BioMed Res Int. 2017;2017:
Physiol. 2018;125:1871–1880.
Fernandez-Gonzalez R, Bendeck MP. DDR1 (dis- 9035193. https://doi.org/10.1155/2017/903519
coidin domain receptor-1)-RhoA (Ras homolog 81. Edmonds ME, Roberts VC, Watkins PJ. Blood
97. Raggi P, Chertow GM, Torres PU, et al. The
family member A) axis senses matrix stiffness to flow in the diabetic neuropathic foot. Dia-
ADVANCE study: a randomized study to evaluate
promote vascular calcification. Arterioscler Thromb betologia. 1982;22:9–15.
the effects of cinacalcet plus low-dose vitamin D
Vasc Biol. 2020;40:1763–1776. 82. Glagov S, Weisenberg E, Zarins CK, on vascular calcification in patients on hemodial-
66. Hou YC, Lu CL, Yuan TH, Liao MT, Chao CT, Stankunavicius R, Kolettis GJ. Compensatory ysis. Nephrol Dial Transplant. 2011;26:1327–1339.
Lu KC. The epigenetic landscape of vascular enlargement of human atherosclerotic coronary
98. Jamal SA, Vandermeer B, Raggi P, et al. Effect
calcification: an integrative perspective. Int J Mol arteries. N Engl J Med. 1987;316:1371–1375.
of calcium-based versus non-calcium-based
Sci. 2020;21(3):980. 83. Fok PW, Lanzer P. Media sclerosis drives and phosphate binders on mortality in patients with
localizes atherosclerosis in peripheral arteries. chronic kidney disease: an updated systematic
67. Jackson SP, Bartek J. The DNA-damage
PLoS One. 2018;13(10):e0205599. https://doi. review and meta-analysis. Lancet.
response in human biology and disease. Nature.
org/10.1371/journal.pone.0205599 2013;382(9900):1268–1277.
2009;461(7267):1071–1078.
84. Chatzizisis YS, Coskun AU, Jonas M, 99. Neves KR, Graciolli FG, Reis LM, et al. Vascular
68. Wang C, Xu W, An J, et al. Poly(ADP-ribose)
Edelman ER, Feldman CL, Stone PH. Role of calcification: contribution of parathyroid hormone
polymerase 1 accelerates vascular calcification by
endothelial shear stress in the natural history of in renal failure. Kidney Int. 2007;71:1262–1270.
upregulating Runx2. Nat Commun. 2019;10(1):
1203. https://doi.org/10.1038/s41467-019- coronary atherosclerosis and vascular remodeling. 100. Raggi P, James G, Burke SK, et al. Decrease in
09174-1 J Am Coll Cardiol. 2007;49:2379–2393. thoracic vertebral bone attenuation with calcium-
85. Stary HC, Chandler AB, Dinsmore RE, et al. based phosphate binders in hemodialysis. J Bone
69. Anderson HC. Electron microscopic studies of
A definition of advanced types of atherosclerotic Miner Res. 2005;20:764–772.
induced cartilage development and calcification.
lesions and a histological classification of athero- 101. De Vriese AS, Caluwé R, Pyfferoen L, et al.
J Cell Biol. 1967;35:81–101.
sclerosis. Circulation. 1995;92:1355–1374. Multicenter randomized controlled trial of vitamin
70. Blaser MC, Aikawa E. Roles and regulation of K antagonist replacement by rivaroxaban with or
86. Lachman AS, Spray TL, Kerwin DM, Shugoll GI,
extracellular vesicles in cardiovascular mineral without vitamin K2 in hemodialysis patients with
Robert WC. Medial calcinosis of Mönckeberg. Am J
metabolism. Front Cardiovasc Med. 2018;5:187. atrial fibrillation: the Valkyrie Study. J Am Soc
Med. 1977;63:615–622.
https://doi.org/10.3389/fcvm.2018.00187 Nephrol. 2020;1:186–196.
87. Castillo BV, Torczynski E, Edward DP. Mön-
71. Edmonds ME, Morrison N, Laws JW, 102. Leadbetter MR, Kozuka L, Kohler K, et al.
ckeberg’s sclerosis in a temporal artery biopsy
Watkins PJ. Medial arterial calcification and dia- Gastrointestinal inhibition of sodium-hydrogen
specimen. Br J Ophthalmol. 1999;83:1091–1092.
betic neuropathy. Br Med J (Clin Res Ed). exchanger 3 reduces phosphorus absorption and
1982;284:928–930. 88. Lee SJ, Choe YS, Lee JC, Park BC, Lee WJ, protects against vascular calcification in CKD. J Am
Kim DW. Two cases of Mönckeberg’s medial scle- Soc Nephrol. 2015;26:1138–1149.
72. Bourron O, Aubert CE, Liabeuf S, et al. Below-
rosis on the face. Ann Dermatol. 2007;19:31–34.
knee arterial calcification in type 2 diabetes: as- 103. Block GA, Rosenbaum DP, Yan A,
sociation with receptor activator of nuclear factor 89. Kim HJ, Greenerg JS, Javitte MC. Breast Chertow GM. Efficacy and safety of tenapanor in
kB ligand, osteoprotegerin, and neuropathy. J Clin calcification due to Mönckeberg medial calcific patients with hyperphosphatemia receiving main-
Endocrinol Metab. 2014;99:4250–4258. sclerosis. Radiographics. 1999;19:1401–1403. tenance hemodialysis: a randomized phase 3 trial.
J Am Soc Nephrol. 2019;30:641–652.
73. Filipska I, Winiarska A, Knysak M, Stompór T. 90. Ferraresi R, Ucci A, Pizzuto A, et al. A novel
Contribution of gut microbiota-derived uremic scoring system for small artery disease and medial 104. Rutsch F, Böyer P, Nitschke Y, et al. Hypo-
toxins to the cardiovascular system mineralization. arterial calcification is strongly associated with phosphatemia, hyperphosphaturia, and
Toxins (Basel). 2021;13(4):274. https://doi.org/10. major adverse events in patients with chronic bisphosphonate treatment are associated with
3390/toxins13040274 limb-threatening ischemia. J Endovasc Ther. survival beyond infancy in generalized arterial
calcification of infancy. Circ Cardiovasc Genet. implications. J Am Coll Cardiol. 2014;63:1703– 131. Jeong G, Kwon DH, Shin S, et al. Long non-
2008;1:133–140. 1714. coding RNAs in vascular smooth muscle cells
regulate vascular calcification. Sci Rep. 2019;9:
105. O’Neill C, Lomashvili KA. Recent progress in 118. Dini CS, Tomberli B, Mattesini A, et al. Intra-
5848.
the treatment of vascular calcification. Kidney Int. vascular lithotripsy for calcific coronary and pe-
2010;78:1232–1239. ripheral artery stenoses. EuroIntervention. 132. Lamb J, Crawford ED, Peck D, et al. The
2019;15:714–721. Connectivity Map: using gene-expression signa-
106. Nitta K, Ogawa T. Vascular calcification in
tures to connect small molecules, genes, and
end-stage renal disease patients. Contrib Nephrol. 119. Schanstra JP, Luong TTD, Makridakis M, et al.
disease. Science. 2006;313:1929–1935.
2015;185:156–167. Systems biology identifies cytosolic PLA2 as a
target in vascular calcification treatment. JCI 133. Song JS, Wang RS, Leopold JA, Loscalzo J.
107. Kranenburg G, de Jong PA, Bartstra JW, et al.
Insight. 2019;4(10):e125638. https://doi.org/10. Network determinants of cardiovascular calcifica-
Etidronate for prevention of ectopic mineralization
1172/jci.insight.125638 tion and repositioned drug treatments. FASEB J.
in patients with pseudoxanthoma elasticum. J Am
2020;34(8):11087–11100. https://doi.org/10.
Coll Cardiol. 2018;71:1117–1126. 120. Stranneheim H, Wedell A. Exome and
1096/fj.202001062R
108. Jaffe IZ, Tintut Y, Newfell BG, Demer LL, genome sequencing: a revolution for the discovery
and diagnosis of monogenic disorders. J Intern 134. Fernandez DM, Rahman AH, Fernandez NF,
Mendelsohn ME. Mineralocorticoid receptor acti-
Med. 2016;279:3–15. et al. Single-cell immune landscape of human
vation promotes vascular cell calcification. Arte-
atherosclerotic plaques. Nat Med. 2019;25:1576–
rioscler Thromb Vasc Biol. 2007;27:799–805. 121. Taylor JC, Martin HC, Lise S, et al. Factors
1588.
109. Lang F, Ritz E, Voelkl J, Alesutan I. Vascular influencing success of clinical genome sequencing
across a broad spectrum of disorders. Nat Genet. 135. Depuydt MAC, Prange KHM, Slenders L, et al.
calcification–is aldosterone a culprit? Nephrol Dial
2015;47:717–726. Microanatomy of the human atherosclerotic pla-
Transplant. 2013;28:1080–1084.
que by single-cell transcriptomics. Circ Res.
110. Sakaguchi Y, Hamano T, Obi Y, et al. 122. Malhotra R, Mauer AC, Lino Cardenas CL,
2020;127(11):1437–1455.
A randomized trial of magnesium oxide and oral et al. HDAC9 is implicated in atherosclerotic aortic
carbon adsorbent for coronary artery calcification calcification and affects vascular smooth muscle 136. Liu M, Gomez D. Smooth muscle cell pheno-
in predialysis CKD. J Am Soc Nephrol. 2019;30: cell phenotype. Nat Genet. 2019;51:1580–1587. typic diversity. Arterioscler Thromb Vasc Biol.
1073–1085. 2019;39:1715–1723.
123. International Mouse Knockout Consortium;
111. Bressendorff I, Hansen D, Schou M, Pasch A, Collins FS, Rossant J, Wurst W. A mouse for all 137. Schlotter F, Halu A, Goto S, et al. Spatiotem-
Brandi L. The effect of increasing dialysate mag- reasons. Cell. 2007;128:9–13. poral multi-omics mapping generates a molecular
nesium on serum calcification propensity in sub- atlas of the aortic valve and reveals networks
124. Rucher G, Cameliere L, Fendri J, et al. Mo- driving disease. Circulation. 2018;138:377–393.
jects with end stage kidney disease: a randomized,
lecular imaging of endothelial activation and
controlled clinical trial. Clin J Am Soc Nephrol. 138. Zhang M, Sheffield T, Zhan X, et al. Spatial
mineralization in a mouse model of accelerated
2018;13:1373–1380. molecular profiling: platforms, applications and
atherosclerosis. EJNMMI Res. 2019;9:80. https://
112. Grases F, Costa-Bauza A. Key aspects of myo- analysis tools. Brief Bioinform. 2021;22(3):
doi.org/10.1186/s13550-019-0550-5
inositol hexaphosphate (phytate) and pathological bbaa145. https://doi.org/10.1093/bib/bbaa145
calcifications. Molecules. 2019;24:4434. https:// 125. Sinha S, Santoro MM. New models to study
139. Asp M, Bergenstråhle J, Lundeberg J.
doi.org/10.3390/molecules24244434109 vascular mural cell embryonic origin: implications
Spatially resolved transcriptomes-next generation
in vascular diseases. Cardiovasc Res. 2018;114:
113. Raggi P, Bellasi A, Bushinsky D, et al. Slowing tools for tissue exploration. Bioessays.
481–491.
progression of cardiovascular calcification with 2020;42(10):e1900221.
SNF472 in patients on hemodialysis: results of a 126. Herrington DM, Mao C, Parker SJ, et al. Pro-
140. Epple M, Lanzer P. How much inter-
randomized phase 2b study. Circulation. 2020;141: teomic architecture of human coronary and aortic
disciplinarity is required to understand vascular
728–739. atherosclerosis. Circulation. 2018;137:2741–2756.
calcifications? Formulation of four basic principles
114. Hedayati SS. A novel treatment for vascular 127. Chen X, Liu L, Palacios G, et al. Plasma of vascular calcification. Z Kardiol. 2001;90(Suppl
calcification in patients with dialysis-dependent metabolomics reveals biomarkers of the athero- 3):2–5.
chronic kidney disease: are we there yet? Circula- sclerosis. J Sep Sci. 2010;33:2776–2783. 141. Gourgas O, Marulanda J, Zhang P,
tion. 2020;141:740–742. 128. Williams JW, Winkels H, Durant CP, Zaitsev K, Murshed M, Cerruti M. Multidisciplinary approach
115. Alappan HR, Vasanth P, Manzoor S, Ghosheh Y, Ley K. Single cell RNA sequencing in to understand medial arterial calcification. Arte-
O’Neill WC. Vascular calcification slows but does atherosclerosis research. Circ Res. 2020;126:1112– rioscl Thromb Vasc Biol. 2018;38:363–372.
not regress after kidney transplantation. Kidney 1126.
Int Rep. 2020;5(12):2212–2217. https://doi.org/10. KEY WORDS atherosclerosis, chronic limb-
129. Gomes CPC, Schroen B, Kuster GM, et al. EU-
1016/j.ekir.2020.09.039 threatening ischemia, genetics, medial
CardioRNA COST Action (CA17129). Regulatory
116. Mintz GS. Intravascular imaging of coronary RNAs in heart failure. Circulation. 2020;141:313– arterial calcification, omics, peripheral artery
calcification and its clinical implications. J Am Coll 328. disease, vascular calcification
Cardiol Img. 2015;8:461–471.
130. Zhang Z, Salisbury D, Sallam T. Long non-
117. Madhavan MV, Tarigopula M, Mintz GS, coding RNAs in atherosclerosis: JACC review topic A PP END IX For supplemental appendices and
Maehara A, Stone GW, Généreux P. Coronary ar- of the week. J Am Coll Cardiol. 2018;72:2380– tables, please see the online version of this
tery calcification: pathogenesis and prognostic 2390. paper.
JACC STATE-OF-THE-ART REVIEW
Exercise Intolerance in Older Adults

With Heart Failure With Preserved
Ejection Fraction
JACC State-of-the-Art Review
Ambarish Pandey, MD, MSCS,a Sanjiv J. Shah, MD,b Javed Butler, MD, MPH, MBA,c Dean L. Kellogg JR, MD, PHD,d
Gregory D. Lewis, MD,e Daniel E. Forman, MD,f Robert J. Mentz, MD,g Barry A. Borlaug, MD,h Marc A. Simon, MD,f
Julio A. Chirinos, MD,i Roger A. Fielding, PHD,j Elena Volpi, MD,k Anthony J.A. Molina, PHD,l
Mark J. Haykowsky, PHD,m Flora Sam, MD,n Bret H. Goodpaster, PHD,o Alain G. Bertoni, MD, MPH,p
Jamie N. Justice, PHD,p James P. White, PHD,q Jingzhone Ding, PHD,p Scott L. Hummel, MD, MS,r
Nathan K. LeBrasseur, PHD,h George E. Taffet, MD,s Iraklis I. Pipinos, MD,t Dalane Kitzman, MDp
ABSTRACT
Exercise intolerance (EI) is the primary manifestation of chronic heart failure with preserved ejection fraction (HFpEF), the
most common form of heart failure among older individuals. The recent recognition that HFpEF is likely a systemic,
multiorgan disorder that shares characteristics with other common, difficult-to-treat, aging-related disorders suggests
that novel insights may be gained from combining knowledge and concepts from aging and cardiovascular disease dis-
ciplines. This state-of-the-art review is based on the outcomes of a National Institute of Aging–sponsored working group
meeting on aging and EI in HFpEF. We discuss aging-related and extracardiac contributors to EI in HFpEF and provide the
rationale for a transdisciplinary, “gero-centric” approach to advance our understanding of EI in HFpEF and identify
promising new therapeutic targets. We also provide a framework for prioritizing future research, including developing a
uniform, comprehensive approach to phenotypic characterization of HFpEF, elucidating key geroscience targets for
treatment, and conducting proof-of-concept trials to modify these targets. (J Am Coll Cardiol 2021;78:1166–1187)
© 2021 by the American College of Cardiology Foundation.
From the aUniversity of Texas Southwestern Medical Center, Dallas, Texas, USA; bNorthwestern University Feinberg School of
Medicine, Chicago, Illinois, USA; cUniversity of Mississippi Medical Center, Jackson, Mississippi, USA; dUniversity of Texas Health
Science Center and GRECC, South Texas Veterans Affairs Health System, San Antonio, Texas, USA; eMassachusetts General
Hospital, Boston, Massachusetts, USA; fUniversity of Pittsburgh and VA Pittsburgh Healthcare System, Pittsburgh, Pennsylvania,
USA; gDuke Clinical Research Center, Durham, North Carolina, USA; hMayo Clinic, Rochester, Minnesota, USA; iUniversity of
Listen to this manuscript’s Pennsylvania, Philadelphia, Pennsylvania, USA; jTufts University, Boston, Massachusetts, USA; kUniversity of Texas Medical
audio summary by Branch at Galveston, Galveston, Texas, USA; lUniversity of California, San Diego, California, USA; m
University of Alberta,
Editor-in-Chief Edmonton, Alberta, Canada; nBoston University School of Medicine, Boston, Massachusetts, USA; oAdvent Health Translational
Dr. Valentin Fuster on Research Institute, Orlando, Florida, USA; pWake Forest School of Medicine, Winston-Salem, North Carolina, USA; qDuke Uni-
JACC.org. versity, Durham, North Carolina, USA; rUniversity of Michigan and the VA Ann Arbor Health System, Ann Arbor, Michigan, USA;
s
Baylor College of Medicine, Houston, Texas, USA; and the tUniversity of Nebraska Medical Center, Omaha, Nebraska, USA.
Manuscript received September 15, 2020; revised manuscript received July 9, 2021, accepted July 13, 2021.

JACC VOL. 78, NO. 11, 2021 Pandey et al. 1167
SEPTEMBER 14, 2021:1166–1187 Exercise Intolerance and HFpEF Among Older Adults
of-the-art review is unique in that it offers a ABBREVIATIONS

HIGHLIGHTS AND ACRONYMS
new, comprehensive, evidence-based “road
In older patients with HFpEF, EI is com- map” of the under-recognized but critical
CR = caloric restriction
mon and is associated with adverse contributions of extracardiac factors and the
CRP = C-reactive protein
outcomes. array of research opportunities to advance
CVD = cardiovascular diseases
the diagnosis and treatment of this pervasive,
Several age-related extracardiac mecha- EI = exercise intolerance
debilitating condition in older adults.
nisms contribute to the manifestation of HF = heart failure
EI in older patients with HFpEF. NEW PARADIGM FOR HFpEF AS A
HFpEF = heart failure with
MULTISYSTEM GERIATRIC SYNDROME
preserved ejection fraction
A multidisciplinary approach focused on
PAI = plasminogen activator-
age-related drivers of EI are needed to HFpEF—originally described by a geriatrician, inhibitor
improve outcomes among older patients Robert Luchi, 38 years ago—is the most com- SASP = senescence-associated
with HFpEF. mon form of HF in community-dwelling older secretory phenotype
H
persons (1,6). Almost all HF cases encoun- VO2peak = peak aerobic power
eart failure with preserved ejection fraction tered among nonagenarians are HFpEF. Ag-
(HFpEF) is the most common form of heart ing is one of the strongest risk factors for HFpEF, and
failure (HF); its prevalence is increasing, its increasing prevalence is largely driven by the ag-
its prognosis is worsening, and pharmaceutical trials ing population. The impact of aging and circulating
in HFpEF have been neutral on their primary out- factors in promoting HFpEF is supported by hetero-
comes (1,2). Thus, there is an urgent need for an chronic parabiosis studies, whereby animals of
enhanced understanding of HFpEF. The recognition different ages are joined surgically leading to com-
that HFpEF is likely systemic and multiorgan in nature mon circulation, demonstrating reversal of some
and that it shares many characteristics with other com- HFpEF features in older mice after prolonged expo-
mon, difficult-to-treat aging disorders supports exam- sure to the circulating growth differentiation factor-11
ining the condition from a gerontological viewpoint as in young mice (7).
well as from a cardiovascular perspective (2,3). How- HFpEF was initially believed to be due solely to
ever, few efforts to date have capitalized on these abnormal relaxation of the left ventricle (LV) and
new insights. This state-of-the-art review is based on decreased LV compliance (8). However, our under-
an eponymous workshop sponsored by the National standing of the pathophysiology of HFpEF has
Institute of Aging (NIA) held September 12-13, 2019, evolved over time toward the concept of a systemic,
to address this critical knowledge gap. The workshop multiorgan geriatric syndrome (2). It is likely initiated
brought together a diverse group of experts in aging by inflammation and other circulating factors that
and/or HFpEF, including geroscientists, geriatricians, originate from increased adiposity, particularly
cardiologists, exercise physiologists, skeletal muscle excess intra-abdominal adipose tissue, in the setting
biologists, and metabolic and adipocyte experts. of multimorbidity, aging, and physical inactivity (5).
These disparate groups of investigators rarely interact, These factors foment loss of capillarity, sarcopenia,
yet each holds complementary and potentially valu- mitochondrial dysfunction, and endothelial
able knowledge and insights relevant to HFpEF. The dysfunction resulting in multiorgan dysfunction,
goal was to maximize sharing of information and con- frailty, and cardiac and skeletal myopathy. Consistent
cepts that could enhance understanding of HFpEF with other geriatric syndromes, comorbidities, frailty,
pathophysiology and management to forge a frame- and decline in functional capacity are central to the
work for new transdisciplinary research discoveries. clinical manifestations of HFpEF and contribute to
The meeting focused on extracardiac mechanisms the high burden of mortality, hospitalizations, and
underlying development of exercise intolerance (EI) poor quality-of-life (Central Illustration).
in HFpEF. EI is the primary manifestation of chronic The implications of the concept of HFpEF as a sys-
stable HFpEF, is severe and debilitating, and is temic geriatric syndrome are profound and may help
associated with poor quality-of-life and clinical out- explain why pharmacological intervention trials to
comes (2). However, the pathophysiology and optimal date have been neutral. They also redirect attention
management of EI in HFpEF are poorly understood. and resources from a purely cardiocentric approach to
Although the role of cardiac contributors in the that of an integrated, whole-patient approach. More-
development of HFpEF and EI has been discussed over, the new paradigm of HFpEF suggests a shift in
extensively in prior reviews (3-5), there has been far treatment focus toward integrated patient-reported
less focus on extracardiac determinants. This state- outcomes, including quality-of-life and symptom
1168 Pandey et al. JACC VOL. 78, NO. 11, 2021
Exercise Intolerance and HFpEF Among Older Adults SEPTEMBER 14, 2021:1166–1187
C E NT R AL IL L U STR AT IO N Pathophysiology and Outcomes in Heart Failure With Preserved Ejection Fraction
Pandey, A. et al. J Am Coll Cardiol. 2021;78(11):1166–1187.
Risk factors such as multimorbidity, aging, obesity, physical inactivity, and systemic inflammation lead to loss of capillarity, mitochondrial, and endothelial dysfunction
and sarcopenia, and result in downstream multiorgan dysfunction, frailty, and cardiac and skeletal muscle myopathy. This constellation of pathophysiological ab-
normalities contributes to the clinical manifestation of heart failure with preserved ejection fraction (HFpEF). QOL ¼ quality of life.
burden, as well as functional assessments, ie, exercise perform require near-maximal effort for patients with
tolerance, physical performance, and cognition. They HFpEF (Figure 1) (4). VO 2peak can be measured
also point to opportunities for testing novel agents for reproducibly with standardized cardiopulmonary ex-
treatment, including those that are anti-inflammatory, ercise testing with modest participant burden and
are capillary-sparing and rejuvenating, and can cost. VO 2peak can be measured as an absolute value
improve mitochondrial function. Thus, a geroscience (mL/min) or relative to the body size (mL/kg/min),
approach, which focuses on biological mechanisms of and their use depends on the setting. It is customary
aging to improve the treatment of age-related chronic to index VO 2peak to body size given its intimate rela-
conditions, may foster needed advances in the un- tionship with a wide range of morphometric mea-
derstanding and treatment of HFpEF. sures, including weight, body surface area, body mass
index, and skeletal muscle mass. Extremes of body
EI: A KEY MANIFESTATION OF HFpEF size, obesity, and weight loss interventions may
disproportionately influence body size–indexed
Low exercise capacity, measured objectively as peak measures of VO 2peak. In these settings, changes in
oxygen uptake (VO 2peak ) during maximal exercise, is VO2peak can be verified and compared with measures
an independent predictor of HFpEF development of exercise capacity that are not indexed to weight,
among older adults (9). On average, VO 2peak is including exercise time to exhaustion, 6-minute walk
reduced by 35% in HFpEF compared with healthy age- distance, and maximal workload achieved (12,13).
and sex-matched control subjects (10). Among older Healthy aging is associated with declines in
patients with HFpEF, VO 2peak during upright exercise VO2peak, but in the absence of disease does not cause
is on average 3-4 mL/kg/min below the VO 2 threshold EI, ie, shortness of breath and fatigue with normal
of functional independence (11). Thus, many activ- day-to-day activity. Development of HFpEF is asso-
ities of daily living that are trivial for healthy adults to ciated with accelerated functional decline and
F I G U R E 1 VO 2peak Required for ADL Relative to Average VO 2peak in HFpEF
188%
25
20 129%
121%
VO2 (mL/kg/min)
Peak
15
78%
70%
10 ~Anaerobic (lactate) threshold
59%
5 23%
0
HFpEF Rest Dressing/ Sweeping/ Walking Functional Carrying Jogging
Making Bed Vacuuming 3 mph Independence Wood/ 5 mph
Groceries
Among patients with HFpEF, the threshold of functional independence is 21% above the average peak VO2 (blue bar) of patients with HFpEF. Reproduced
with permission from Nayor et al (4). ADL ¼ activities of daily living; HFpEF ¼ heart failure with preserved ejection fraction.
lowering of the threshold for dyspnea and fatigue so HFpEF and may require different approaches than
that they occur even with usual normal daily activity. those utilized for CVD in younger persons. The strong
EI is a key, patient-centered outcome that is critical to association of HFpEF with multimorbidity is pre-
understand HFpEF pathophysiology and optimal dictable, because these changes related to the hall-
treatment, independent of clinical events. EI is an marks of aging trigger multiple diseases, including
important outcome for observational, mechanistic, frailty, cognitive changes, and other systemic effects.
and interventional studies. In this section we have Thus, traditional single organ–focused care may not
discussed the key attributes of HFpEF as a geriatric be optimal for understanding HFpEF that occurs in
syndrome and their implications in the pathogenesis the context of change in aggregate aging, multiple
of EI and its management. comorbidities, and a decline in physiological
AGING, ACCELERATED FUNCTIONAL DECLINE, AND reserve (Figure 2).
EI IN HFpEF. The strong predisposition of older EI in HFpEF likely depends not only on cardiac
adults to the development of HFpEF suggests changes specific to HFpEF but also on a broad range
potentially specific age-related vulnerabilities to of factors related to aging that sets the stage for the
HFpEF. Recently, molecular hallmarks of aging and development of EI in older patients with HFpEF. Prior
shifts in subcellular biology have become recognized studies among healthy adults have showed that
as fundamental determinants of age-related changes VO 2peak declines significantly with age (16,17). The
that can lead to chronic age-related diseases, rate of decline in VO 2peak accelerates in older age from
including overt cardiovascular diseases (CVD) (14). 3%-6% per decade in the 20s to 30s to >20% per
Paneni et al (15) described how progressive changes in decade after the age of 70 years (16). Age-related de-
geroscience-related subcellular mechanisms can lead clines in strength and balance exacerbate decrements
to increased CVD vulnerability with aging that likely in VO 2peak caused by alterations in skeletal muscle,
influence development of HFpEF. This includes including atrophy, loss of strength, and impaired
changes in sirtuins, autophagy, and cellular senes- mitochondrial respiration (18).
cence, which lead not only to changes in cardiac cells, SEX DIFFERENCES IN EI IN HFpEF. There are impor-
but also to broader changes in fat and skeletal muscle, tant sex differences in the pathophysiology and
autonomic regulation, hematology, endothelial clinical manifestations of HFpEF. Women have a
function, and other wide-ranging effects that affect 2-fold higher lifetime risk of HFpEF vs heart failure
F I G U R E 2 The Multimorbidity Model of Exercise Intolerance in HFpEF
Aging
Sarcopenia Pulmonary
Disease
HTN
CAD
Type 2 DM Atrial
fibrillation
Obesity CKD
Physiologic Exercise
reserve tolerance
HFpEF
Accumulation of cardiac and noncardiac comorbidities coupled with aging contributes to systemic inflammation, endothelial dysfunction, global reduction in physi-
ological reserve, and exercise intolerance in patients with HFpEF. CAD ¼ coronary artery disease; CKD ¼ chronic kidney disease; DM ¼ diabetes mellitus; HFpEF ¼ heart
failure with preserved ejection fraction; HTN ¼ hypertension.
with reduced ejection fraction (HFrEF), and the and perimenopausal adiposity changes in women
prevalence of HFpEF is much higher than HFrEF (25). Specifically, women with (vs without) HFpEF
among older women (19). In the Cardiovascular have 34% higher visceral adipose tissue (VAT). Higher
Health Study, >80% of incident HF in older women amount of VAT has been associated with a more
was caused by HFpEF (1). Women with HFpEF have a exaggerated increase in pulmonary capillary pressure
higher burden of symptoms, worse quality-of-life, with exercise in women but not in men (26).
and lower VO 2peak (w20%-30% lower VO2peak ) ROLE OF EXTRACARDIAC COMORBIDITIES IN
compared with men (20); however, percent predicted MANIFESTATIONS OF EI IN HFpEF. HFpEF is associ-
VO 2peak appears similarly reduced among men (68%) ated with, and partly driven by, extracardiac comor-
and women (66%) with HFpEF (21). Women have bidities. On average, patients with HFpEF have $5
lower exercise cardiac reserve and reduced peripheral coexisting comorbidities, with a significant increase
oxygen extraction (22). The lower VO 2peak among in the comorbidity burden over time (27). In older
women may be related to sex-specific differences in patients with HFpEF, noncardiac comorbidities are
cardiac remodeling patterns and adiposity distribu- involved in adverse outcomes more frequently
tion. Women (vs men) have small LV cavity size, compared with HFrEF (27,28). Higher burden of
higher LV stiffness, higher LV filling pressure, and comorbidities contributes to lower VO2peak and EI in
lower stroke volume at rest (22). Furthermore, patients with HFpEF through extracardiac mecha-
women have greater decline in LV contractility and nisms, as detailed in Table 1. The importance of
greater concentric remodeling with aging (23). Animal extracardiac contributors to EI in HFpEF is high-
studies have also demonstrated sex differences in lighted, because they account for the following:
cardiac response to increased pressure overload 1) w50% of the reduced VO 2peak (29); 2) w85% of the
(aortic banding) with greater LV dilation, loss of improvement in VO 2peak with exercise training (30);
concentric remodeling, and loss of contractile reserve and 3) skeletal myocytes, in contrast to cardiac myo-
in male rats (vs female rats) (24). Remodeling differ- cytes, which are terminally differentiated and have a
ences may be related, in part, because of ACE gene rapid regenerative and repair capacity (2,31). These
polymorphisms, sex-specific differences in mRNAs, observations may have important implications given
accepted measures of frailty, the Fried Frailty

T A B L E 1 Extracardiac Contributors to Exercise Intolerance in
Older Patients With Heart Failure With Preserved
phenotype model and the Frailty Index, have ad-
Ejection Fraction vantages and disadvantages in the context of HF, as
Hallmarks of aging described in Table 2 (33). The Fried Frailty phenotype
Cell senescence measures decline in physiological reserve across 5
Systemic inflammation
Oxidative stress physical function domains: weight loss, weakness,
Accelerated exercise capacity decline poor endurance, slowness, and low physical activity
Reduced physiological reserve
level (33). The Frailty Index quantifies frailty as cu-
Skeletal muscle abnormalities
Increased fatty deposition in skeletal muscle mulative deficit in multiple organ system resulting in
Sarcopenia
decreased physiological reserve (35).
Reduced microvascular density (capillarity)
Impaired skeletal muscle perfusion The prevalence of frailty in patients with HFpEF is
Reduced conductive and diffusive oxygen transport
reported to be up to 75% (36,37). Frailty in older
Impaired mitochondrial function and bioenergetics
Excess adiposity
patients with HFpEF is associated with poor quality-
Up-regulation of inflammatory pathways of-life, greater functional impairment, and reduced
Increased metabolic and neurohormonal dysfunction
Insulin resistance
submaximal aerobic exercise capacity. Frailty pre-
Natriuretic peptide deficiency disposes to severe EI in HFpEF through several
Plasma volume expansion
Systemic endothelial dysfunction
shared mechanisms, including a higher burden of
Reduced microvascular density (capillarity) comorbidities, chronic inflammation, sarcopenia, and
Impaired mitochondrial function
a global decrease in functional reserve. Frailty is also
Pulmonary dysfunction
Altered ventilatory reserve associated with a higher risk of hospitalization and
Impaired gas diffusion mortality (37,38).
Ventilation perfusion mismatch
Increased dead space ventilation Considering the prevalence and prognostic impact
Restrictive and obstructive deficits of frailty in HFpEF, it should be considered when
Obstructive and central sleep apnea
Pulmonary vascular remodeling designing treatment paradigms for older patients
Pulmonary hypertension with HFpEF with the goal of reducing symptom
Right heart dysfunction
burden, improving functional status and exercise
Kidney dysfunction
Excess perinephric fat capacity, and ultimately, improving clinical out-
Reduced glomerular filtration rate
comes. Treatment approaches targeting frail patients
Excess sodium retention and plasma volume expansion
Up-regulation of renin-angiotensin-aldosterone pathway with HFpEF are being evaluated; one study showed
Liver dysfunction significant improvement with a novel physical reha-
Nonalcoholic fatty liver disease
Up-regulation of proinflammatory pathways
bilitation intervention (39).
Vascular dysfunction
Increased aortic stiffness PATHOPHYSIOLOGY OF EI IN HFpEF
Adverse pulsatile hemodynamics and late systolic load
Microvascular rarefaction
Impaired vasodilatory reserve Both cardiac and extracardiac abnormalities
contribute to EI in HFpEF by reducing physiological
reserve capacity for the key determinants of VO 2peak :
that optimal management of comorbidities and their exercise cardiac output and peripheral oxygen
consequences may represent an important strategy extraction. An integrative approach to quantifying
for older patients with HFpEF. Consistent with this relative contributions to impaired VO 2peak in HFpEF
notion, lifestyle interventions such as exercise and can be achieved by tracing oxygen pathway from
weight loss, which have pleiotropic favorable effects “mouth to mitochondria.” Deficiencies in alveolar
on noncardiac and cardiac comorbidities, have been ventilation, lung diffusion capacity, cardiac output,
shown to improve exercise capacity and quality-of- hemoglobin, skeletal muscle diffusion capacity for O 2,
life in HFpEF primarily through improvements in and mitochondrial respiration can contribute to
extracardiac factors such as systemic inflammation, impaired VO2peak associated with HFpEF (11). There is
adiposity, and peripheral oxygen utilization in skel- heterogeneity of O 2 pathway derangements in pa-
etal muscle (12,30,32). tients with HFpEF, the majority of whom have mul-
FRAILTY AND EI IN HFpEF. Frailty is a biological tiple abnormalities in the O 2 pathway (11). Given the
syndrome reflecting decreased physiological reserve serial nature of the steps in the O2 pathway, targeting
and vulnerability to stressors (33), and it represents a single step may not result in a significant overall
the key factor underlying the heterogeneity in func- enhancement of O 2 utilization (11). Deficiencies in
tional decline with aging (34). The 2 most widely specific steps of the O2 pathway can be assessed by
T A B L E 2 Measures of Physical Frailty
Frailty Measure Description Advantages Disadvantages
Fried Frailty phenotype Quantifies the decline in physiological Most commonly used tool to Poor discriminatory power
model reserve across 5 domains: weight loss, assess frailty in the published Overlap in clinical features of
weakness, poor endurance, slowness, and reports HF and frailty
low physical activity levels Time and resource-intensive
Frailty Index Based on a “multiple hit” model Continuous nature of the The number of deficits assessed
Quantifies frailty as an accumulation of estimate with a wide range of is not standardized and vary
deficits across several health-related distribution according to available data
domains Ability to use pre-existing Focuses more of the number of
Estimated as the ratio of the total deficits data from medical records for deficits and does not account
present to the number of deficits assessed estimating frailty for the severity of deficits
across signs, symptoms, comorbidity in- Ability to estimate cutoffs for
dex, laboratory values, and ADL specific clinical populations
ADL ¼ activities of daily living; HF ¼ heart failure.
invasive hemodynamic monitoring during exercise right ventricular systolic and diastolic function,
and via tests of peripheral skeletal muscle blood flow, which may exceed corresponding changes in the left
O 2 diffusion, and utilization. heart (44). Aging and associated cardiometabolic
Hemodynamic measurements during exercise dysregulation may also contribute to impaired right
permit assessment of the relative contributions of ventricular function (Figure 3) (45). Measures of pul-
cardiac vs peripheral impairments and confirm the monary function and gas exchange deteriorate with
presence HF and influence of extracardiac abnor- aging; this may be related in part to aging-related
malities. Anticipated changes in response to treat- kyphotic changes, declines in respiratory muscle
ment interventions may then be modeled in the function, and reduction in functional alveolar-
context of the compound O 2 pathway deficits. Multi- capillary units (46). Additionally, structural and dy-
ple benefits can be achieved through exercise namic compliance of the pulmonary vasculature
training, which thus serves as an example of an likely is reduced with aging (47). These abnormalities
intervention that demonstrably improves EI in HFpEF are amplified in patients with HFpEF (48,49).
by improving impairments in multiple steps in the O2 S y s t e m i c v a s c u l a r c o n t r i b u t i o n s t o H F p E F . The
pathway in multiple organ systems (40). In the systemic arterial tree plays a central role in modu-
following section, we have discussed in detail the lating LV afterload, ventricular-arterial coupling,
underlying contribution of different cardiac and systolic and diastolic function, and the distribution of
extracardiac abnormalities toward EI in older patients the cardiac output at rest and during exercise (50).
with HFpEF. Thus, it has an important mechanistic role at several
CARDIOVASCULAR CONTRIBUTORS TO EI. I m p a i r m e n t steps of the oxygen consumption cascade and in the
i n l e f t a t r i a l a n d L V r e s e r v e . Abnormalities of left chronic maladaptive remodeling of the heart and
heart structure, function, and reserve play a pivotal other target organs (Figure 3). Due to widely varying
role in the pathophysiology of EI in HFpEF. These characteristics, individual territories within the arte-
components have been described in detail elsewhere rial tree are discussed separately in the following
(41) and are summarized in the Supplemental text.
Appendix and integrated into Figure 3. Aorta. A healthy arterial system delivers relatively
Right heart function and the pulmonary vasculature in steady flow to the microvasculature, despite inter-
HFpEF. Pulmonary vascular remodeling, both arterial mittent LV ejection, largely caused by the cushioning
and venous, is common and severe in HFpEF and is (windkessel) function of the aorta. Aging results in
related to several factors, including chronically stiffening of the aorta, and this is accelerated by
elevated left-sided filling pressures, vasoconstriction, comorbidities, particularly hypertension, obesity,
decreased nitric oxide availability, and metabolic and and diabetes. Aortic stiffness is higher in patients
inflammatory dysregulation (42). Pulmonary hyper- with established HFpEF compared with healthy age-
tension is prevalent in up to 80% of patients with matched control subjects. Greater large-vessel
HFpEF and is associated with increased morbidity, arterial stiffness at rest is associated with lower
mortality, and EI (43). Vo2peak in older patients with HFpEF (51). Moreover,
Several age-related changes likely underlie the central arterial stiffness worsens during exercise
previously mentioned findings, including decreased and contributes to EI in HF (52,53). The changes in
F I G U R E 3 Contribution of Cardiovascular Maladaptations to Exercise Intolerance in HFpEF
Left atrial and Ventricular

Right heart abnormalities Abnormalities
Chronotropic and stroke volume reserve
Pulmonary vascular remodeling
LV contractility
Pulmonary hypertension
LV end-diastolic stiffness
RV function
LV/LA filling pressure
Right sided filling pressure LV/LA remodeling
Atrial fibrillation
Reduced Peak VO2
Skeletal muscle and microvascular

abnormalities
Vascular abnormalities
Vasodilatory reserve
Aortic stiffness
Capillarity
LV afterload
Oxygen extraction
Wave reflection
Skeletal muscle mitochondrial function
Muscle mass Forward flow
Muscle fat infiltration
Patients with HFpEF have impairment in stroke volume reserve and chronotropic reserve, and increased afterload, all of which contribute to reduced forward flow. This
“central limitation’ coupled with impairment peripheral vasodilation and reduced oxygen extraction contributes to lower peak exercise oxygen uptake. HFpEF ¼ heart
failure with preserved ejection fraction; LA ¼ left atrial; LV ¼ left ventricular; RV ¼ right ventricular; VO2 ¼ oxygen uptake.
large arteries favors adverse patterns of pulse Muscular arteries. LV contraction generates a pulse
pressure with consequent changes in pulsatile LV wave that travels forward in the arteries and is
afterload, which promotes LV remodeling, fibrosis, partially reflected at sites of impedance mismatch,
dysfunction, and failure (54). Large artery changes such as points of branching or change in wall diam-
also lead to excessive delivery of pulsatile energy to eter or stiffness. Numerous peripheral reflections are
the microvasculature of target organs as a function summed into a larger reflected wave, which travels
of arteriolar resistance (55). Therefore, in addition to back to the LV. In older adults, lower vascular
its impact on the heart, large artery stiffening may compliance results in a reduced wave transit time
be one of the factors contributing to clustering of (pulse wave velocity) from the LV to reflection sites
HFpEF with various comorbidities in older persons, and back to the proximal aorta. In some cases, wave
such as chronic kidney disease and dementia. reflections return to the LV while the aortic valve is
profile (55). It is possible that the reduced coronary

T A B L E 3 Measures of Muscle Performance Relevant to Older Adults
microvascular perfusion reserve in HFpEF is multi-
Muscle Performance Measure Description factorial and caused by abnormalities in coronary
Strength Maximum capacity to generate force or tension microvasculature, such as myocardial microvascular
Can be assessed using isometric handgrip dynamometry
or a more dynamic measure of the 1-repetition maximum rarefaction (61), endothelial dysfunction, and
Power Rate of physical work performance abnormal central aortic hemodynamics.
Has a component of velocity that may be have clinical
relevance for older adults CONTRIBUTION OF SKELETAL MUSCLE MYOPATHY
Peak power is more closely associated with declines in TO EI IN HFpEF. O v e r v i e w o f s k e l e t a l m u s c l e
physical function than muscle strength in older adults
Endurance Ability to maintain a specific force output
a g i n g . Sarcopenia, the age-associated loss of
Related to functional capacity in older adults skeletal muscle mass and function, results in
impaired mobility, reduced physical function, and
increased disability and mortality (62). The causes
still open and ejecting blood in mid-to-late systole, of sarcopenia are multifactorial but include poor
resulting in increased afterload. Thus, pulse wave nutritional status, physical inactivity, inflammation,
reflections can increase the late systolic LV and chronic comorbidities, ie, the hallmarks of
workload and adversely shift the LV loading aging. The recognition of sarcopenia as a relevant
sequence from early to late systole, which promotes clinical syndrome of advancing age has led to
LV remodeling and dysfunction (54,55). refinements in its identification, prevention, and
Patients with HFpEF have increased late systolic treatment (62-64).
load from wave reflections during exercise, which Strength, power, and endurance each comprise
contributes to exaggerated afterload, increased LV unique physiological domains of muscle performance
filling pressure, reduced exercise cardiac output, and and are differentially related to the declines in
reduced VO 2peak. Furthermore, vasodilator therapies physical functioning with aging (Table 3).
such as inorganic nitrites and beet-root juice, have The link between peripheral vascular function
been shown to improve arterial compliance and a n d s a r c o p e n i a . Aging is associated with reduced
VO 2peak in patients with HFpEF. Thus, wave re- sensitivity of skeletal muscle proteins to all key
flections may represent a mechanistic therapeutic anabolic stimuli: general nutrition, dietary amino
target for novel therapies such as inorganic nitrates/ acids, insulin, and resistance exercise (65). This phe-
nitrites (52,56). nomenon, called anabolic resistance, is promoted by
Microvasculature. In addition to its role in deter- abnormal rapamycin (mTOR) complex 1 signaling (a
mining mean arterial pressure and resistive load to key regulator of the anabolic response in skeletal
the LV, microvascular function has an important role muscle), and contributes to sarcopenia. Endothelial
in peripheral oxygen delivery and utilization. Vaso- dysfunction may also play a fundamental role in the
dilatory responses during exercise mediate the pe- anabolic resistance of aging, as shown in Figure 4.
ripheral redistribution of flow to working muscle, a Because nutritive flow helps regulate skeletal
key component of the normal hemodynamic response muscle protein anabolism, the decrease in endothe-
to exercise, and depend on local vasodilatory mech- lial function with aging contributes to anabolic
anisms within muscle as well as neurovascular sym- resistance and sarcopenia in older adults. HFpEF is
pathetic control mechanisms (57). Hypoxic associated with worse endothelial function, which is
vasodilation is an important local mechanism that correlated with symptoms of EI and reduced VO 2peak
ensures adequate local blood flow and oxygen de- (66), and could exacerbate aging-associated impaired
livery to muscle tissue and allows the local vascula- muscle anabolism.
ture to overcome humoral and reflex-mediated Impairment in skeletal muscle architecture and
vasoconstriction during exercise (57). Microvascular oxygen transport in H F p E F . HFpEF-mediated
dilatory reserve is abnormal in HFpEF, contributes to reduction in the peripheral O 2 extraction reserve is
EI, and may be improved by exercise training (29,58). also secondary to impaired skeletal muscle O2 diffu-
Patients with HFpEF also have abnormalities of the sion capacity (ie, transport of O2 from microcircula-
coronary microcirculation, impaired coronary flow tion to muscle mitochondria) (11). Patients with
reserve, and greater burden of myocardial injury with HFpEF with slower VO 2 kinetics have more severely
exercise caused by oxygen supply-demand mismatch impaired peripheral oxygen utilization and higher
(59,60). Large artery stiffness and wave reflections cardiac output during submaximal exercise compared
are key determinants of myocardial perfusion pres- with patients with HFpEF with faster VO 2 kinetics as
sure through their effects on the diastolic pressure well as healthy control subjects, suggesting that the
F I G U R E 4 Microvascular Function and Sarcopenia
MRI cross-section
of the thigh
ANABOLIC
AEROBIC EXERCISE
NORMAL Improved endothelial function
MICROCIRCULATION Enhanced insulin + mixed nutrient delivery
Normal blood flow Aged healthy control RESISTANCE EXERCISE
Normal insulin + AA delivery (healthy muscle)
increased capillarization
Improved flow-mediated vasodilation
Enhanced insulin + mixed nutrient delivery
Aging
Comorbidities
HFpEF CATABOLIC
DISEASED
MICROCIRCULATION
Impaired blood flow Aged HFpEF
Impaired vasodilation (sarcopenia)
Capillary rarefaction
Aging
Endothelial dysfunction
Comorbidities
Abnormal insulin + AA delivery
HFpEF
Subcutaneous Fat
Skeletal Muscle
Intramuscular Fat
Comparison of the thigh compartment fat and muscle composition in a control vs HFpEF individual on MRI. Patients with HFpEF have greater intramuscular fat and
reduced skeletal muscle mass. Impaired microcirculatory function leads to sarcopenia, which is magnified by the catabolic effects of inflammation, insulin resistance,
and increased myostatin. AA ¼ amino acids; FFA ¼ free fatty acids; GH ¼ growth hormone; HFpEF ¼ heart failure with preserved ejection fraction; IGF-1 ¼ insulin-like
growth factor-1; MRI ¼ magnetic resonance imaging; TNF ¼ tumor necrosis factor.
limitation for oxygen during submaximal exercise is have significantly increased thigh intermuscular fat
not related to impaired cardiac output (67). Thus, and intramuscular fat to skeletal muscle area ratio
evaluation of VO 2 kinetics with brief, low-intensity compared with healthy control subjects, both of
exercise could help phenotype older patients with which were inversely related to VO2peak (25).
HFpEF with peripheral limitations. Furthermore, skeletal muscle biopsy studies have
Because most of the O 2 intake during exercise is shown that older patients with HFpEF have a shift in
consumed in the active skeletal muscles, the decline fiber type distribution with a reduction in percent
in VO 2peak in older patients with HFpEF may be type I (oxidative) fibers and reduced capillary to fiber
partially caused by a loss in muscle mass, which is ratio (69), and both of these alterations were associ-
reduced in HFpEF (68). Haykowsky et al (68) ated with reduced VO 2peak (69). This is in contrast to
demonstrated a strong association between lean mass changes in skeletal muscle architecture noted with
(percent total as well as leg lean mass) and peak healthy aging, whereby the loss of muscle mass in
VO 2peak . However, in addition to the muscle mass, healthy old vs young adults has been attributed to
Haykowsky et al (68) also implicated impaired muscle smaller type II muscle fiber size without substantial
quality in the reduced VO 2peak among patients with reduction in muscle fiber numbers (70). Taken
HFpEF, such that VO 2peak indexed to total and leg together, abnormalities in skeletal muscle mass and
lean mass was significantly lower and the change in quality contribute to EI in older patients with HFpEF
VO 2peak with increasing percent leg lean mass was (Figure 4) (25,68).
markedly reduced in HFpEF compared with healthy Skeletal muscle mitochondrial bioenergetics
control subjects (68). Older patients with HFpEF also and H F p E F . Older patients with HFpEF have
T A B L E 4 Therapeutic Strategies That Have Shown Promise for Improving Exercise Intolerance in Patients With HFpEF Via Extracardiac Factors
Therapeutic Approach Known and Potential Therapeutic Effects in HFpEF
Supervised exercise Improvement in peak VO2

Improvement in quality of life
Improvement in peripheral oxygen extraction at peak exercise
Multidomain physical function intervention Improved physical function
(REHAB-HF) (39) Improvement in frailty burden
Weight loss Improvement in VO2peak
Reduction in visceral adiposity and systemic inflammation burden
Reduction in myostasis
IL-1 receptor blockers (anti-inflammatory Reductions in CRP and NT-proBNP levels
agents) Modest improvement in the treadmill exercise time from baseline to follow-up
Senotherapeutic agents Reduction in senescent cell burden and the SASP in human tissues
In mouse model, alleviation of age- and senescence-related cardiovascular dysfunction, lung diseases,
metabolic syndromes, and frailty, and improvement in survival
Therapeutic effects in animal models or patients with HFpEF are not available.
Novel pharmacotherapies for diabetes Improved in adverse cardiac remodeling patterns in animal models with use of liraglutide and dapagliflozin
(SGLT-2 inhibitors/GLP-1 agonists) No significant effects on empagliflozin on exercise capacity or quality of life in patients with HFpEF in small
RCT (EMPERIAL-Preserved; NCT03448406)
Inorganic nitrates/nitrites Single dose of beet root juice (inorganic nitrate donor) has been shown to improve VO2peak and reduce systemic
vascular resistance and arterial wave reflections with exercise
1-week of daily beet root juice dosing has been shown to improve aerobic capacity by 24%
Inorganic sodium nitrite has been shown to improve VO2peak, skeletal muscle oxygen conductance, VO2
kinetics, alveolar capillary membrane O2 conductance, and O2 utilization during submaximal exercise
No effects of inhaled nitrite on VO2peak in HFpEF in the INDIE-HFpEF (NCT02742129) trial may be related to
challenges with the drug delivery and short-acting nature of the therapy
Results from other ongoing studies awaited
CRP ¼ C-reactive protein; EMPERIAL-Preserved ¼ EMPagliflozin Compared With Placebo on ExeRcise Ability and Heart Failure Symptoms, In Patients With Chronic HeArt FaiLure With
Preserved Ejection Fraction; HFpEF ¼ heart failure with preserved ejection fraction; INDIE-HFpEF ¼ Inorganic Nitrite Delivery to Improve Exercise Capacity in HFpEF; NT-proBNP ¼ N-terminal
pro-brain natriuretic peptide; RCT ¼ randomized controlled trial; REHAB-HF ¼ A Trial of Rehabilitation Therapy in Older Acute Heart Failure Patients; SASP ¼ senescence associated secretory
phenotype; VO2peak ¼ peak exercise oxygen uptake.
abnormal skeletal muscle oxygen utilization that is CONTRIBUTION OF MYOSTEATOSIS TO EI IN HFpEF.

related to reduced VO 2peak (11,29,71). By magnetic Common risk factors associated with HFpEF, such as
resonance spectroscopy, skeletal muscle oxidative obesity, aging, and physical inactivity, contribute to
metabolism is reduced in patients with HFpEF (72). deranged energy metabolism and metabolic inflexi-
Using hemodynamic monitoring during exercise, the bility (76)—a loss in the capacity to switch effectively
ability to extract O 2 in the skeletal muscles and between fatty acids and carbohydrates as fuel for
reduce venous O 2 content is significantly impaired in energy. These are associated with impaired skeletal
HFpEF and is a major contributor to reduced VO 2peak muscle energetics and excess adipose accumulation
(11,73). in muscle tissue, or myosteatosis. Older patients with
Mitochondrial content and oxidative capacity are HFpEF have significant myosteatosis and abnormal
significantly reduced in skeletal muscle biopsy sam- skeletal muscle energetics, including impaired mito-
ples from older patients with HFpEF vs age-matched chondrial function, reduced mitochondrial density,
control subjects (74). These mitochondrial alter- and impaired peripheral O2 extraction with exercise
ations are related to measures of aerobic exercise compared with healthy, age-matched control sub-
capacity, such as VO 2peak and 6-minute walk distance. jects, and these are significantly correlated with their
Improved mitochondrial function contributes to im- severe EI (74,77,78). Furthermore, improvement in
provements in VO 2peak associated with exercise VO2peak in obese patients with HFpEF in response to
training, further supporting that skeletal muscle dietary weight loss and exercise training appear to be
mitochondrial abnormalities contribute to patho- mediated, at least partly, by improved skeletal mus-
physiology of EI in patients with HFpEF (2). Exercise cle mitochondrial function in tandem with reduction
training can increase mitochondrial electron trans- in adipose depots in the skeletal muscle tissue (12).
port chain enzyme function 2-fold, accompanied by a CONTRIBUTION OF SYSTEMIC METABOLIC
60% increase in mitochondrial mass (75), and has ABNORMALITIES TO EI IN HFpEF. O b e s i t y , r e g i o n a l
been shown to increase electron transport chain ac- adiposity, and their implications in senescence
tivity, mitochondrial DNA content, and citrate syn- and H F p E F . Obesity, particularly central and
thase activity. visceral adiposity, have emerged as major risk factors
F I G U R E 5 Contribution of Hallmarks of Aging to HFpEF and Exercise Intolerance
Inflammation Obesity
COPD Frailty
CKD Mobility
HFpEF
&
Aging CVD Exercise Arthritis
Biology
intolerance
Diabetes
Genome
Sarcopenia
Nuclear
Epigenome
Envelope
Fibrosis
Osteoporosis
Neurodegeneration
Cytoskeleton Damage/
/Repair
Aging
Physiology
Signaling Proteostasis
Metabolism Organelles
The “hallmarks of aging” refer to closely inter-related cellular and molecular processes implicated in aging and conserved across species, which could serve
as new therapeutic targets for many age-related chronic diseases and geriatric syndromes, including HFpEF. Adapted with permission from Burch et al
(101). COPD ¼ chronic obstructive pulmonary disease; CVD ¼ cardiovascular diseases; other abbreviations as in Figure 2.
for HFpEF (9,25,79,80). Nearly one-half (w47%) of SERPINE1, is a leading candidate caused by its known
patients with HFpEF are obese, and patients with association with VAT, accelerated senescence, and
HFpEF with body mass index in the “normal” range risk of incident HFpEF but not HFrEF in population-
typically have significant visceral adiposity (81). based studies (86-88). In animal models of diabetes
Visceral obesity is associated with abnormal cardiac and aging, PAI-1 levels are elevated, and genetic
mechanics before the development of HFpEF (82), deficiency or inhibition of PAI-1 prevents diabetes
and the addition of hypertension to obesity acceler- and extends lifespan (86). These findings have been
ates the development of cardiac dysfunction and replicated in humans with genetic PAI-1 deficiency
elevated filling pressures (83). (87).
Age-associated changes in body composition Aging is also associated with changes in the cellu-
include redistribution of fat mass mainly to the larity and function of subcutaneous adipose tissue
visceral compartment. There is also a shift in lipid (89). Brown adipose tissue declines with aging with a
storage from the subcutaneous to the VAT depot (84), resultant predominance of white adipose tissue,
which is thought to be caused by a decline in pro- which is proinflammatory and exacerbates aging-
genitor cell function and an accumulation of senes- associated inflammation. In white adipose tissue,
cent adipose tissue cells in subcutaneous fat (85). adipocytes undergo hypertrophy, adipocyte necrosis,
There is also redistribution of adipose tissue to the adipose tissue inflammation, and a switch to proin-
abdominal compartment, which is presumed to be flammatory classical cytokines and adipokines. In
caused by the decline in testosterone in men and addition, there is recruitment of monocytes and other
estrogen in women following menopause. VAT is pro- immune cells to remove necrotic adipocytes and to
inflammatory, vasoactive, and dysmetabolic, with participate in tissue remodeling in an attempt to limit
many factors secreted by visceral adipocytes that may lipid storage, but this eventually results in ectopic
contribute to pathogenesis of HFpEF. Of these, plas- lipid accumulation (eg, liver and skeletal muscle) and
minogen activator-inhibitor (PAI)-1, encoded by insulin resistance. Importantly, immune cells have
been shown to play a pivotal role in the pathogenesis Contribution of diabetes and insulin resistance
of HFpEF (90). The resulting state of chronic, low- t o E I i n H F p E F . The risk of HF increases substan-
grade inflammation and systemic metabolic inflam- tially with diabetes, and up to 40% of patients in
mation (90,91) and augmented nitrosative stress have HFpEF studies have diabetes (96). Patients with
been implicated in obesity-associated HFpEF and HFpEF with (vs without) diabetes have a higher
associated EI (83). In a preclinical model of normal- burden of comorbidities, lower VO2peak and submax-
weight HFpEF, cardiac natriuretic peptide (NP) imal exercise capacity, elevated levels of biomarkers
signaling caused alterations in energy expenditure of inflammation, and higher risk of adverse clinical
and metabolism, and promoted brown adipose-like events of follow-up (96). The lower VO 2peak in pa-
features in white adipose tissue depots (92). The tients with HFpEF with diabetes may be related to
sympathetic nervous system and NP signaling in- higher LV filling pressures, chronotropic incompe-
creases metabolic activity in adipose tissue by acti- tence, and impaired parasympathetic and sympa-
vating lipolysis and modulating the brown-fat thetic response to exercise caused by underlying
thermogenic program. NP signaling is an important autonomic neuropathy (97). Patients with diabetes
regulator of overall NP activity in adipose tissue in also have reduced delivery and extraction of oxygen
HFpEF, and its relevance in obese HFpEF and in aging in the exercising skeletal muscles because of higher
requires further exploration. prevalence of anemia, reduced vasodilator reserve,
Obesity, regional adiposity, and EI in HFpEF. higher peripheral vascular resistance from endothe-
Obese patients with HFpEF have reduced relative lial dysfunction and vasoconstriction, and impaired
VO 2peak (mL/kg/min) compared with their nonobese mitochondrial function (96,98,99). Considering the
counterparts (93). This is related to both peripheral higher burden of EI, patients with HFpEF with dia-
and central limitations in exercise induced oxygen betes are an important target for exercise training
uptake. Obese individuals have exaggerated LV filling strategies, which can improve VO 2peak through pe-
pressure and pulmonary artery pressure response to ripheral adaptations in the O 2 pathway. Recent small
exercise, which contributes to reduced exercise car- RCTs have evaluated the role of novel pharmaco-
diac output. Obese individuals also have greater therapies such as sodium glucose cotransporter-2 in-
infiltration of fatty tissue in peripheral skeletal mus- hibitors in patients with HFpEF with variable effects
cles leading to impairment in oxygen extraction (25). (Table 4). Larger outcomes trials using these therapies
Recent studies have also demonstrated that obese in HFpEF are currently underway.
individuals have depletion of myocardial energetic
source, namely, the Creatine Kinase shuttle, which is TENETS OF AGING AND GEROSCIENCE:
the main transfer mechanism for maintaining ATP IMPLICATIONS FOR EI IN HFpEF
delivery in myocardial mitochondria (94). As a result,
the delivery of ATP cannot be maintained with exer- HALLMARKS OF AGING: ROLE IN HEALTH AND
cise stress leading to reduced VO 2peak. Furthermore, DISEASE. Two foundational reviews established the
intentional weight loss is associated with restoration field of geroscience by identifying a short list of bio-
of the ATP delivery with exercise in the myocardium logical processes implicated in aging and conserved
(94). across species that could serve as new therapeutic
Among regional adipose depots, increased VAT is a targets for many age-related chronic diseases and
major contributor to reduced VO 2peak in obese pa- geriatric syndromes, including HFpEF. These curated
tients with HFpEF (25) and may partially explain the lists of 7 to 9 biological processes, referred to as
female predominance of HFpEF. A recent study “hallmarks” or “pillars” of aging, provide a framework
demonstrated that excess VAT is associated with for understanding the role of conserved biological
more exaggerated increase in pulmonary capillary processes in health and disease (Figure 5) (14,100).
wedge pressure with exercise in women but not in These processes include macromolecular damage,
men. VAT-associated EI is modifiable, and loss of metabolism, proteostasis, inflammation, adaptation to
visceral fat with caloric restriction (CR) is strongly stress, epigenetics, cell senescence, stem cells/regen-
associated with improvement in VO 2peak among pa- eration, and pleiotropic processes. As discussed
tients with HFpEF (12). In tandem with VAT, peri- earlier, aging is a universal biological phenomenon
cardial fat also increases, which is associated with that results in a progressive decline in integrated
more adverse hemodynamics and lower VO2peak in function over time. These age-related declines ulti-
HFpEF (95). Taken together, obesity and excess mately lead to an increased probability of death (101).
regional adipose depot may be important targets for Approximately 5-10 overlapping and highly inter-
improvement in VO 2peak in patients with HFpEF. twined molecular and cellular processes are
evolutionally conserved “common denominators” In cardiac tissue, aging is associated with reduction
that are implicated in biological aging (14). These in autophagy and tissue proteostasis (107). Although
biological processes are active during “normal aging,” relatively little is known regarding the role of cardiac
and when ameliorated, the phenotype of aging and its proteostasis in HFpEF, mouse models of HFpEF have
sequelae are retarded. In contrast, when the activity of shown a decline in proteostasis and autophagy in
a hallmark is augmented, phenotypic aging processes skeletal muscle and liver tissue (108). Exercise, which
are accelerated. The hallmarks are interdependent increases muscle autophagy in many other condi-
such that the effects of one biological process activates tions, does not increase autophagy in a young rat
others and leads to increased dysregulation across model of HFpEF (109). Taken together, both aging
multiple processes at progressively accelerating rates and HFpEF independently reduce proteostasis in
over time. Interventions to ameliorate the aging tissues involved in the pathogenesis of HFpEF.
sequelae of each of the different hallmarks have been INFLAMMATION, “OMICS” OF IMMUNE CELLS, AND
demonstrated in mice, suggesting that similar EI IN HFpEF. Chronic inflammation, one of the hall-
amelioration is possible in humans (102). marks of aging, has been implicated as the key
The geroscience hypothesis posits that incident pathway unifying multifactorial and intertwined
aging-related chronic disease can be prevented, mechanisms leading to HFpEF (110). Prior studies
delayed, or attenuated by therapeutically targeting have shown high levels of circulating inflammatory
these biological processes (103). One example is biomarkers, including interleukin (IL)-6 and C-reac-
mTOR, which has emerged as a key regulator of aging tive protein (CRP), are present in established HFpEF,
in laboratory studies, with the genetic or pharmaco- and are associated with risk for development of
logical inhibition of mTOR shown to delay aging in HFpEF, and with severity and prognosis of HFpEF
animal models (102,104). Extant evidence suggests (111). Furthermore, dietary weight loss in obese
that mTOR antagonists may ameliorate nearly all HFpEF results in improved symptoms, VO 2peak,
identified biological aging processes to some extent, quality-of-life, and the improvement is associated
with consistent attenuation of multiple positive with reduced inflammation biomarkers (12). A causal
feedback interactions between hallmarks (102). role for inflammation in HF is further supported by
Another example is cellular senescence and the results from a recent randomized clinical trial (RCT)
resulting generation of a senescence-associated showing that anti-inflammatory therapy targeting
secretory phenotype (SASP), which is hypothesized IL1b may reduce risk for HF (112). Among patients
to be at the nexus of several age-related chronic dis- with HFpEF, a 12-week, phase-II, placebo-controlled
eases and geriatric syndromes (100). The SASP is a trial demonstrated significant reductions in CRP and
consequence of altered gene expression that con- N-terminal pro-brain natriuretic peptide levels with
tributes to sterile inflammation and adverse tissue IL-1 receptor blockers, and a modest improvement in
remodeling, and further accumulation of senescent the treadmill exercise time, a measure of aerobic ex-
cells in vivo. Experimental therapies targeting ercise performance, within the treatment arm (113).
cellular senescence, or senotherapeutics, include Current understanding of mechanistic links be-
senolytic drugs (selective removal of senescent cells) tween inflammation and HFpEF has mainly been
and senomorphic drugs (modulate or reverse SASP), derived from animal models (114). The inflammatory
which represent an emerging strategy for treatment response is a protective mechanism in which activated
of age-related disease(s), including HFpEF. immune cells phagocytose and remove pathogens.
CONTRIBUTION OF PROTEOSTASIS AND AUTOPHAGY TO However, during chronic inflammation, innate im-
EI IN HFpEF. Proteostasis is a homeostatic pathway mune cells such as monocytes can be reprogrammed
involving the generation and removal of cellular and promote exaggerated inflammatory responses af-
proteins and organelles for regeneration purposes ter infiltrating the myocardium (114,115). Therefore,
and for the provision of energy substrates (105). The “omics” profiling of human immune cells, particularly
cellular machinery involved in proteostasis includes of easily accessible circulating immune cells, holds
the proteasome and the autophagosome/lysosome. great promise in the exploration of the cell-specific
Functional proteostasis is required for tissue health, mechanisms leading to HFpEF and may facilitate the
and its impairment is associated with tissue development of innovative interventions through
dysfunction and disease. Aging brings a global modulation of these cells. Additionally, “omics” ap-
reduction in cellular proteostasis, especially the proaches enable generation of a systemic view of
autophagy process, in various tissues including skel- fundamental mechanisms of the aging process and
etal muscle, cardiac tissue, and liver (106). their interconnection. For example, aging-associated
T A B L E 5 Key Knowledge Gaps in Our Understanding of the Extracardiac Contributors to Exercise Intolerance in HFpEF
Organ System/Domain Specific Pathway/Target Key Knowledge Gaps for Future Research
Skeletal muscle Sarcopenia 1. Integrating “pathophysiology of sarcopenia development” in drug designing for HFpEF
2. Examine structural and functional changes in muscle in HFpEF trials
3. Role of protein supplementation and/or resistance training in attenuating development of
sarcopenia
Oxygen utilization 1. Impact of exercise and weight loss intervention on skeletal muscle mitochondrial
functional
2. Role of VO2 kinetics during the rest to low-level exercise transition in phenotyping
patients with HFpEF with a “peripheral” limitation to exercise
3. Evaluating the contribution of peripheral adaptations to the improvement in VO2peak with
exercise training and/or CR in older patients with HFpEF
Endothelial dysfunction 1. The molecular mechanisms linking endothelial dysfunction and age-related muscle
anabolic resistance
2. If treatment of endothelial dysfunction improves muscle bulk and function in older adults
3. Effects of endothelial function and nutrition on skeletal muscle anabolism and physical
function in older patients with HFpEF
Pulmonary and right Right heart and pulmonary 1. Delineate the interaction of other comorbidities with pulmonary hypertension in HFpEF
ventricular system function 2. Examine the relative contributions of pulmonary venous vs arterial remodeling to EI
3. Improve understanding of the right ventricular response to aging and cardiopulmonary
disease
Metabolic disorders Obesity 1. Role of browning of white adipose tissue in development and outcome of HFpEF
2. Mechanisms whereby increased remote, intercellular, and intracellular adiposity
contribute to EI in older patients with HFpEF
3. Greater representation of obese patients in HFpEF trials
Diabetes and insulin resistance 1. Mechanisms underlying the increased risk of HFpEF in diabetes patients
2. Peripheral and central impairments in exercise reserve among patients with HFpEF with
diabetes or insulin resistance
3. Peripheral and central adaptations to exercise training in patients with HFpEF and
diabetes
4. Effects of SGLT-2 inhibitors on exercise capacity in HFpEF in larger trials
Geroscientific mechanism Senescence 1. Burden of senescence cells in HFpEF
2. Identify adipose, skeletal, and cardiac markers of senescence in patients with HFpEF
3. Evaluate role of senolytic agent in managing patient with HFpEF
Proteostasis 1. Role of cardiac and skeletal muscle proteostasis in HFpEF development
3. Role of proteostatic function as a possible molecular marker of peripheral dysfunction in
HFpEF and as a potential mechanism underlying EI
Inflammation 1. Inflammatory pathways specifically in HFpEF
2. Delineate “omics” profiling of human immune cells, particularly of easily accessible
circulating immune cells, to better understand the role of aging hallmarks on HFpEF
3. Role of anti-inflammatory therapies and therapies targeting immune cells specific for
HFpEF in improving EI and other outcomes in patients with HFpEF
Frailty 1. Test interventions designed to reduce physical frailty and its adverse consequences in
older patients with HFpEF
2. Develop and evaluate strategies to overcome barriers to frailty assessment in clinical care
of patients with HFpEF
3. Develop novel screening tools to efficiently detect frailty
Lifestyle intervention Calorie restriction, exercise, and 1. Examine effects of dietary and exercise on clinical outcomes in patients with HFpEF
multidomain physical function 2. Evaluate efficacy of multidomain physical function intervention for improving functional
interventions status (balance, strength, mobility, and endurance) and clinical outcomes in HFpEF
3. Optimal dietary weight loss strategies in patients with HFpEF, particularly among pa-
tients with modestly elevated body mass index
4. Optimal exercise training intensity (high-intensity interval vs moderate intensity
continuous), modality (resistance vs aerobic vs combination), and duration for patients
with HFpEF
5. Development of strategies to minimize loss of skeletal muscle during CR in patients with
HFpEF
6. Development of strategies for enhancing long-term adherence to dietary and exercise
regimens
CR ¼ caloric restriction; EI ¼ exercise intolerance; HFpEF ¼ heart failure with preserved ejection fraction.
decline in several intercorrelated gene modules rele- APPROACHES TARGETING AGING-RELATED

vant to mitochondrial function has been reported in a CONTRIBUTIONS TO EI IN HFpEF
transcriptomic study of human peripheral monocytes
(116). The down-regulation of mitochondrial function– EXERCISE TRAINING AND MULTIDOMAIN PHYSICAL
related genes is a common feature of aging in various FUNCTION INTERVENTIONS IN HFpEF. Short-term
tissues across humans and other species (117) and may exercise training has been associated with significant
contribute to EI in HFpEF (14). improvement VO 2peak (w20%) and quality-of-life in
F I G U R E 6 Proposed High-Priority Extracardiac Areas for Future Research in HFpEF
Discovery Translation Implementation
Identify geroscience targets Comprehensive phenotypic • Evaluate multimorbidity

for treatment: characterization: management approaches
• Senescent cellular pathways • Human biospecimens • Test promising geroscience
• Epigenetic pathways interventions
• Biomarkers
• Adipose biology • Improve adherence to
Proof of concept intervention
effective lifestyle interventions
• Skeletal muscle microvascular trials to modify the geroscience
and mitochondrial dysfunction targets • Disseminate successful
strategies into community
practice
Future research should focus on advancing the understanding of the pathophysiology of EI, development of novel approaches to pathophysiology-guided phenotypic
classification of HFpEF, identification of novel targets for preventing and mitigating EI, and testing such interventions in animal and human studies. EI ¼ exercise
intolerance; HFpEF ¼ heart failure with preserved ejection fraction.
patients with HFpEF (40). Exercise training can intervention by patients, health care systems,
improve VO 2peak by favorably affecting multiple and payers.
pathophysiological impairments in the oxygen
pathway as discussed earlier and summarized in CR, WEIGHT LOSS, AND NUTRITION. Dietary CR has
Table 4. Despite the well-known benefits of exercise strong potential as a strategy to prevent obese HFpEF
training, the optimal intensity and duration of exer- and reduce EI when HFpEF develops. Intentional
cise training for patients with HFpEF are uncertain. In weight loss by CR and bariatric surgery in obese pa-
a recent large multicenter exercise training trial, tients is associated with a lower risk of HFpEF (119).
high-intensity and moderate continuous exercise In a 20-week, controlled trial of CR with and without
training for 3 months were associated with compara- exercise in 100 patients with obese HFpEF, CR pro-
ble improvements in VO 2peak among patients with duced significant weight loss, more than exercise
HFpEF (32). Long-term adherence to exercise training alone, and both CR and exercise produced strong,
in older patients with HFpEF is generally suboptimal. additive improvements in VO 2peak (mL/kg/min). Sig-
Older patients with HFpEF, particularly those nificant improvements in exercise capacity were
recently hospitalized, are often frail and have phys- confirmed by improvement in measures that are not
ical function impairments in multiple domains indexed to body weight such as VO 2 reserve, exercise
including balance, mobility, and strength as well as time to exhaustion, workload, and leg power sup-
endurance and may benefit from a novel, early, porting the notion of true improvement in exercise
tailored, multidomain physical function intervention capacity with weight loss. CR was also associated with
more than patients with HFrEF, as reported in the greater improvements in quality-of-life than exercise.
recent REHAB-HF (Rehabilitation Therapy in Older In the CR group vs exercise-only and control patients,
Acute Heart Failure Patients) trial (39,118). However, VAT decreased by 15%-20% and high-sensitivity CRP
such interventions need to be evaluated in larger levels declined, and reductions in both were associ-
RCTs of older patients with HFpEF to evaluate their ated with improved VO2peak (12).
efficacy in improving not only physical function and CR studies in animal models have shown
quality-of-life, but also clinical outcomes (rehospi- numerous, robust favorable effects on aging-related
talization, death), which are important to ensure up- pathways, such that the benefits of CR in HFpEF
take and broad implementation of such an could extend beyond the pathways examined in the
previous text (120). In mice, these include mecha- of these 2 agents reduces senescent cell burden and
nisms that have been directly implicated in the the SASP in human tissues. In animal models, the
pathophysiology of HFpEF; CR from young adulthood combination of dasatinib and quercetin alleviates a
prevents diastolic dysfunction, endothelial dysfunc- range of age- and senescence-related disorders in
tion, and increased arterial stiffness, and even in mice, including cardiovascular dysfunction, fibrosis-
older mice, CR reduces oxidative stress, improves related lung diseases, lipotoxicity, hypercholester-
endothelial function, and increases nitric oxide olemia, metabolic syndromes, depression, cancer,
bioavailability (121,122). When compared with age- frailty, and substantially increases survival (127).
matched control subjects, humans voluntarily Future animal and human studies are needed to test
participating in moderate, long-term CR (while similar interventions that target biological hallmarks
maintaining nutritional quality) have favorable of aging for management of HFpEF.
alterations in aging-related pathways including
KNOWLEDGE GAPS AND RECOMMENDATIONS
PI3 K/AKT and AMPK/SIRT and up-regulation of
FOR FUTURE RESEARCH
antioxidant defense, DNA repair, proteostasis, and
autophagy-related genes (123).
Several knowledge gaps pertaining to our under-
In RCT of CR, 10%-30% CR in young to middle-aged,
standing of the contribution of extracardiac factors
nonobese participants produced decreased resting
toward EI in patients with HFpEF have been identified
metabolic rate, improved insulin sensitivity, reduced
and summarized across different extracardiac do-
VAT, decreased oxidative stress and associated DNA
mains in Table 5. Future research should focus on 4 key
damage, and increased skeletal muscle mitochondrial
priorities in older patients with HFpEF: 1) better un-
DNA content, confirming the feasibility, safety, and
derstand the pathophysiology of EI; 2) develop novel
antiaging potential of long-term CR (123-125).
approaches to phenotypic classification of HFpEF
In the CR trials in nonobese adults as well as in the
based on underlying pathophysiology; 3) identify
obese patients with HFpEF, CR was associated with
novel, promising targets for preventing and mitigating
decreased skeletal muscle mass despite careful
EI; and 4) design and test transdisciplinary in-
attention to adequate protein and micronutrient
terventions to prevent and mitigate EI (Figure 6).
intake (12,124,125). However, CR produced gains in
A key step to advance understanding of the path-
endurance and strength despite decreased muscle
ophysiology of EI in HFpEF would be the develop-
mass. Future studies are needed to evaluate the long-
ment of a uniform, comprehensive approach to
term safety of CR in obese patients with HFpEF.
phenotypic characterization of patients with HFpEF
Among other dietary interventions, recent trials sug-
recognizing that the phenotypes may differ across the
gest that focusing on dietary quality among hospi-
age distribution. This would entail baseline and
talized patients with HF can improve quality-of-life,
follow-up assessment of physical function, metabolic
prevent readmissions, and reduce mortality (126).
function, and cardiopulmonary exercise reserve
Future studies are needed to evaluate the efficacy of
testing using noninvasive and invasive approaches
such dietary interventions in patients with obesity
and uniform collection of biospecimen samples from
and HFpEF.
cardiac, adipose tissue, skeletal muscle, endothelial,
SENOTHERAPEUTICS: A NOVEL POTENTIAL TREATMENT and blood cells. These efforts could be organized as
PARADIGM FOR EI IN HFpEF. Senotherapeutics and inter-related studies, a registry, a precision biobank,
senolytic drugs hold promise as therapeutic targets to or as ancillary studies to a clinical trial. The resultant
improve function, geriatric syndromes, and age- data could be examined by computational modeling
associated chronic diseases including HFpEF. and machine learning approaches to elucidate the
Several pro-survival senescent cell antiapoptotic biological mechanisms and clinically meaningful
pathways have been identified from bioinformatics phenotypes in older individuals with HFpEF.
analyses of senescent vs nonsenescent human cells Investigations should prioritize the assessment of
and RNA interference studies. Dasatinib, a tyrosine the role of biological hallmarks of aging in the
kinase inhibitor used clinically to treat leukemia, and development of EI in HFpEF with the goal of identi-
quercetin, one of the first senolytic agents identified, fying key geroscience therapeutic targets such as se-
is a natural product that targets B-cell lymphoma, nescent cellular pathways, epigenetic pathways,
insulin and insulin-like growth factor-1, and hypoxia- adipose biology, skeletal muscle biology, and mito-
inducible factor-1 a senescent cell antiapoptotic chondrial dysfunction. Efforts should leverage ap-
pathway network components (127). The combination proaches recently developed for use in
contemporaneous NIA-sponsored efforts, including workshop “A Gero-centric Approach to Exercise

SOMMA (Study of Muscle, Mobility and Aging), Intolerance and Heart Failure with Preserved Ejection
MoTrPAC (Molecular Transducers of Physical Activity Fraction (HFpEF) in Older Adults: Elucidating and
Consortium), LIFE (Lifestyle Interventions and Inde- Targeting Extra-cardiac Contributors” with Dr
pendence for Elders), and the Geroscience network Kitzman.
(128-131). Aligning approaches with these efforts
would also provide non-HFpEF comparison/control
groups. Efforts should also leverage other NIA re-
The workshop “A Gero-centric Approach to Exercise Intolerance and
sources, including the Pepper Center Network, Heart Failure with Preserved Ejection Fraction (HFpEF) in Older
Nathan Shock Center, and relevant ongoing clinical Adults: Elucidating and Targeting Extra-cardiac Contributors” was
funded by the National Institute on Aging. The contents of this paper
studies.
are solely the responsibility of the authors and does not necessarily
Once promising targets are identified, proof-of- represent the official views of the National Institutes of Health (NIH).
concept intervention trials can be designed to test Dr Pandey is supported by the Texas Health Resources Clinical
whether they are modifiable, and if so, whether this Scholarship, the Gilead Sciences Research Scholar Program, the Na-
tional Institute of Aging GEMSSTAR Grant (1R03AG067960-01), and
improves EI and other key outcomes, including hall-
Applied Therapeutics; has served on the advisory board for Roche
marks of aging, in older patients with HFpEF. Positive Diagnostics; and has received nonfinancial support from Pfizer and
proof-of-concept studies should then be scaled to Merck. Dr Butler has served as a consultant for Abbott, Adrenomed,
community-based and larger trials to confirm proof of Arena Pharma, Array, Amgen, Applied Therapeutics, AstraZeneca,
Bayer, Boehringer Ingelheim, Cardior, CVRx, Eli Lilly, G3 Pharma,
concept and to implement and disseminate the
Imbria, Impulse Dynamics, Innolife, Janssen, LivaNova, Luitpold,
interventions. Medtronic, Merck, Novartis, Novo Nordisk, Relypsa, Roche, Sequana
Studies are critically needed to address long-term Medical, V-Wave Limited, and Vifor. Dr Kellogg Jr is supported by
adherence, particularly to behavioral and lifestyle NIH grants P30 AG044271 and K01AG059837. Dr Mentz has received
research support and honoraria from Abbott, American Regent,
interventions such as exercise, CR, and nutrition.
Amgen, AstraZeneca, Bayer, Boehringer Ingelheim/Eli Lilly, Boston
Because these interventions have been the most Scientific, Cytokinetics, Fast BioMedical, Gilead, Medtronic, Merck,
successful for improving EI in older patients with Novartis, Roche, Sanofi, and Vifor. Dr Forman is funded by NIH grants
HFpEF, studies designed to understand their mech- R01AG060499, R01AG058883, R01AG051376, and P30AG024827. Dr
Borlaug is funded by NIH grant RO1 HL128526. Dr Simon has received
anisms of improvement could yield important in-
grant support from Aadi; and has served as a consultant for Accel-
sights that could be translated to other interventions. eron, Actelion, and United Therapeutics. Dr Chirinos is supported by
Studies should also carefully account for and address NIH grants R01-HL 121510, R33-HL-146390, R01-AG058969, 1R01-
HL104106, P01-HL094307, R03-HL146874, R56-HL136730, and
multimorbidity, which is a key feature of HFpEF in
1R01HL153646-01; has served as a consultant for Bayer, Sanifit,
older patients and strongly contributes to adverse Fukuda-Denshi, Bristol Myers Squibb, Johnson & Johnson, Edwards
outcomes and assess the presence and impact of Lifesciences, Merck, and the Galway-Mayo Institute of Technology; is
depression and cognitive dysfunction. named as inventor in a University of Pennsylvania patent for the use
of inorganic nitrates/nitrites for the treatment of heart failure and
The previously mentioned research initiatives will
preserved ejection fraction; has received University of Pennsylvania
require a broad range of approaches and funding research grants from National Institutes of Health, Fukuda-Denshi,
mechanisms, including exploratory, pilot, develop- Bristol Myers Squibb, Microsoft, and Abbott; has received payments
mental, and program projects; clinical conferences for for editorial roles from the American Heart Association, the American
College of Cardiology, and Wiley; and has received research device
developing consensus; infrastructure-building
loans from Atcor Medical, Fukuda-Denshi, Uscom, NDD Medical
grants; and development of registries, precision bio- Technologies, Microsoft, and MicroVision Medical. Dr Sam is sup-
banks, and large multicenter clinical trials. ported by NIH grant R01HL145985. Dr Molina is supported by NIH
grant R21 AG051077. Dr Pipinos is supported by R01AG062198. Dr
CONCLUSIONS Lewis has received research funding from the National Institutes of
Health R01-HL 151841, R01-HL131029, American Heart Association
15GPSGC-24800006, and Amgen, Cytokinetics, Applied Therapeutics,
EI is the primary manifestation of chronic HFpEF, the
AstraZeneca, and Sonivie in relation to projects that are distinct from
most common form of HF in the older population. Its this work; has received honoraria outside of the current study from
pathophysiology is not well understood, and there Pfizer, Merck, Boehringer Ingelheim, Novartis, American Regent,
are few proven treatments. Progress can be optimized Cyclerion, Cytokinetics, and Amgen; and receives royalties from
UpToDate for scientific content authorship related to exercise phys-
and accelerated with transdisciplinary teams that
iology. Dr Bertoni is partially supported by NIH grant 1R01HL127028-
include experts in the fields of both aging and CVD, 01; and has consulted with Premier/Merck on a diabetes quality
recognize key principles of both domains, and utilize improvement project. Dr Justice has received support from NIH
a broad range of approaches. grants K01AG059837 and P30AG021332. Dr Hummel is supported by
NIH grants R01AG062582 and R01HL139813, and by VA grant
ACKNOWLEDGMENTS Drs Susan Zieman and Lyndon I01CX001636. Dr Kitzman is supported in part by NIH grants
Joseph developed, organized, and colead the R01AG18915, R01AG045551, P30AG021332, and U24AG059624, and by
the Kermit G. Phillips Endowed Chair in Cardiovascular Medicine; has

received honoraria outside the present study as a consultant for ADDRESS FOR CORRESPONDENCE: Dr Dalane W.
AbbVie, Bayer, Merck, Medtronic, Relypsa, Merck, Corvia Medical,
Kitzman, Sections on Cardiovascular Medicine and Geri-
Boehringer Ingelheim, Novo Nordisk, AstraZeneca, and Novartis; has
received grant funding outside the present study from Novartis,
atrics/Gerontology, Wake Forest School of Medicine,
Bayer, Novo Nordisk, and AstraZeneca; and has stock ownership in Medical Center Boulevard, Winston-Salem, North Carolina
Gilead Sciences. All other authors have reported that they have no 27157-1045, USA. E-mail: dkitzman@wakehealth.edu.
relationships relevant to the contents of this paper to disclose.
Twitter: @ambarish4786, @UTSWCards.
REFERENCES
1. Kitzman DW, Gardin JM, Gottdiener JS, et al. of life in obese older patients with heart failure 24. Weinberg EO, Thienelt CD, Katz SE, et al.
Importance of heart failure with preserved systolic with preserved ejection fraction: a randomized Gender differences in molecular remodeling in
function in patients > or ¼ 65 years of age. CHS clinical trial. JAMA. 2016;315:36–46. pressure overload hypertrophy. J Am Coll Cardiol.
Research Group. Cardiovascular Health Study. Am 1999;34:264–273.
13. Villareal DT, Chode S, Parimi N, et al. Weight
J Cardiol. 2001;87:413–419.
loss, exercise, or both and physical function in 25. Haykowsky MJ, Nicklas BJ, Brubaker PH, et al.
2. Upadhya B, Pisani B, Kitzman DW. Evolution of obese older adults. N Engl J Med. 2011;364:1218– Regional adipose distribution and its relationship
a geriatric syndrome: pathophysiology and treat- 1229. to exercise intolerance in older obese patients who
ment of heart failure with preserved ejection have heart failure with preserved ejection fraction.
14. Lopez-Otin C, Blasco MA, Partridge L,
fraction. J Am Geriatr Soc. 2017;65:2431–2440. J Am Coll Cardiol HF. 2018;6:640–649.
Serrano M, Kroemer G. The hallmarks of aging.
3. Shah SJ, Borlaug BA, Kitzman DW, et al. Cell. 2013;153:1194–1217. 26. Sorimachi H, Obokata M, Takahashi N, et al.
Research priorities for heart failure with preserved Pathophysiologic importance of visceral adipose
15. Paneni F, Diaz Canestro C, Libby P, Luscher TF,
ejection fraction: National Heart, Lung, and Blood tissue in women with heart failure and preserved
Camici GG. The aging cardiovascular system: un-
Institute Working Group summary. Circulation. ejection fraction. Eur Heart J. 2021;42(16):1595–
derstanding it at the cellular and clinical levels.
2020;141:1001–1026. 1605.
J Am Coll Cardiol. 2017;69:1952–1967.
4. Nayor M, Houstis NE, Namasivayam M, et al. 27. Pandey A, Vaduganathan M, Arora S, et al.
Impaired exercise tolerance in heart failure with 16. Fleg JL, Morrell CH, Bos AG, et al. Accelerated
Temporal trends in prevalence and prognostic
preserved ejection fraction: quantification of longitudinal decline of aerobic capacity in healthy
implications of comorbidities among patients with
multiorgan system reserve capacity. J Am Coll older adults. Circulation. 2005;112:674–682.
acute decompensated heart failure: the ARIC
Cardiol HF. 2020;8:605–617. 17. Pandey A, Kraus WE, Brubaker PH, Study Community Surveillance. Circulation.
5. Shah SJ, Kitzman DW, Borlaug BA, et al. Kitzman DW. Healthy aging and cardiovascular 2020;142:230–243.
Phenotype-specific treatment of heart failure with function: invasive hemodynamics during rest and
28. Ather S, Chan W, Bozkurt B, et al. Impact of
preserved ejection fraction. Circulation. 2016;134: exercise in 104 healthy volunteers. J Am Coll
noncardiac comorbidities on morbidity and mor-
73–90. Cardiol HF. 2020;8:111–121.
tality in a predominantly male population with
6. Luchi RJ, Snow E, Luchi JM, Nelson CL, 18. Gonzalez-Freire M, Scalzo P, D’Agostino J, heart failure and preserved versus reduced ejec-
Pircher FJ. Left ventricular function in hospitalized et al. Skeletal muscle ex vivo mitochondrial tion fraction. J Am Coll Cardiol. 2012;59:998–
geriatric patients. J Am Geriatr Soc. 1982;30:700– respiration parallels decline in vivo oxidative ca- 1005.
705. pacity, cardiorespiratory fitness, and muscle
29. Haykowsky MJ, Brubaker PH, John JM,
strength: The Baltimore Longitudinal Study of
7. Loffredo FS, Steinhauser ML, Jay SM, et al. Stewart KP, Morgan TM, Kitzman DW. De-
Aging. Aging Cell. 2018;17:e12725.
Growth differentiation factor 11 is a circulating terminants of exercise intolerance in elderly heart
factor that reverses age-related cardiac hypertro- 19. Pandey A, Omar W, Ayers C, et al. Sex and race failure patients with preserved ejection fraction.
phy. Cell. 2013;153:828–839. differences in lifetime risk of heart failure with J Am Coll Cardiol. 2011;58:265–274.
preserved ejection fraction and heart failure with
8. Zile MR, Baicu CF, Gaasch WH. Diastolic heart 30. Haykowsky MJ, Brubaker PH, Stewart KP,
reduced ejection fraction. Circulation. 2018;137:
failure—abnormalities in active relaxation and Morgan TM, Eggebeen J, Kitzman DW. Effect of
1814–1823.
passive stiffness of the left ventricle. N Engl J endurance training on the determinants of peak
Med. 2004;350:1953–1959. 20. Mauricio R, Patel KV, Agusala V, et al. exercise oxygen consumption in elderly patients
9. Pandey A, Patel KV, Bahnson JL, et al. Associ- Sex differences in cardiac function, biomarkers with stable compensated heart failure and pre-
ation of intensive lifestyle intervention, fitness, and exercise performance in heart failure with served ejection fraction. J Am Coll Cardiol.
and body mass index with risk of heart failure in preserved ejection fraction: findings from the 2012;60:120–128.
overweight or obese adults with type 2 diabetes RELAX trial. Eur J Heart Fail. 2019;21:1476–
31. Kitzman DW, Haykowsky MJ, Tomczak CR.
mellitus: an analysis from the Look AHEAD Trial. 1479.
Making the case for skeletal muscle myopathy and
Circulation. 2020;141:1295–1306. 21. Lau ES, Cunningham T, Hardin KM, et al. Sex its contribution to exercise intolerance in heart
10. Kitzman DW, Little WC, Brubaker PH, et al. differences in cardiometabolic traits and de- failure with preserved ejection fraction. Circ Heart
Pathophysiological characterization of isolated terminants of exercise capacity in heart failure Fail. 2017;10:e004281.
diastolic heart failure in comparison to systolic with preserved ejection fraction. JAMA Cardiol.
32. Mueller S, Winzer EB, Duvinage A, et al. Effect
heart failure. JAMA. 2002;288:2144–2150. 2020;5:30–37.
of high-intensity interval training, moderate
11. Houstis NE, Eisman AS, Pappagianopoulos PP, 22. Beale AL, Nanayakkara S, Segan L, et al. Sex continuous training, or guideline-based physical
et al. Exercise Intolerance in heart failure with differences in heart failure with preserved ejection activity advice on peak oxygen consumption in
preserved ejection fraction: diagnosing and fraction pathophysiology: a detailed invasive he- patients with heart failure with preserved ejection
ranking its causes using personalized O2 pathway modynamic and echocardiographic analysis. J Am fraction: a randomized clinical trial. JAMA.
analysis. Circulation. 2018;137:148–161. Coll Cardiol HF. 2019;7:239–249. 2021;325:542–551.
12. Kitzman DW, Brubaker P, Morgan T, et al. Ef- 23. Oneglia A, Nelson MD, Merz CNB. Sex differ- 33. Fried LP, Tangen CM, Walston J, et al. Frailty in
fect of caloric restriction or aerobic exercise ences in cardiovascular aging and heart failure. older adults: evidence for a phenotype. J Gerontol
training on peak oxygen consumption and quality Curr Heart Fail Rep. 2020;17:409–423. A Biol Sci Med Sci. 2001;56:M146–M156.
34. Mitnitski AB, Graham JE, Mogilner AJ, ejection fraction. J Am Coll Cardiol HF. 2016;4: 64. Cruz-Jentoft AJ, Baeyens JP, Bauer JM, et al.
Rockwood K. Frailty, fitness and late-life mortality 490–498. Sarcopenia: European consensus on definition and
in relation to chronological and biological age. diagnosis: report of the European Working Group
50. Kitzman DW, Haykowsky MJ. Vascular
BMC Geriatr. 2002;2:1. on Sarcopenia in Older People. Age Ageing.
dysfunction in heart failure with preserved ejec-
2010;39:412–423.
35. Rockwood K, Mitnitski A. Frailty in relation to tion fraction. J Card Fail. 2016;22:12–16.
the accumulation of deficits. J Gerontol A Biol Sci 65. Fry CS, Drummond MJ, Glynn EL, et al. Aging
51. Kitzman DW, Herrington DM, Brubaker PH,
Med Sci. 2007;62:722–727. impairs contraction-induced human skeletal mus-
Moore JB, Eggebeen J, Haykowsky MJ. Carotid
cle mTORC1 signaling and protein synthesis. Skelet
36. Warraich HJ, Kitzman DW, Whellan DJ, et al. arterial stiffness and its relationship to exercise
Muscle. 2011;1:11.
Physical function, frailty, cognition, depression, intolerance in older patients with heart failure and
and quality of life in hospitalized adults $60 years preserved ejection fraction. Hypertension. 66. Borlaug BA, Olson TP, Lam CS, et al. Global
with acute decompensated heart failure with 2013;61:112–119. cardiovascular reserve dysfunction in heart failure
preserved versus reduced ejection fraction. Circ with preserved ejection fraction. J Am Coll Cardiol.
52. Reddy YNV, Andersen MJ, Obokata M, et al.
Heart Fail. 2018;11:e005254. 2010;56:845–854.
Arterial stiffening with exercise in patients with
37. Sanders NA, Supiano MA, Lewis EF, et al. The heart failure and preserved ejection fraction. J Am 67. Hearon CM, Jr Sarma S, Dias KA, Hieda M,
frailty syndrome and outcomes in the TOPCAT Coll Cardiol. 2017;70:136–148. Levine BD. Impaired oxygen uptake kinetics in
trial. Eur J Heart Fail. 2018;20:1570–1577. heart failure with preserved ejection fraction.
53. Hundley WG, Kitzman DW, Morgan TM, et al.
38. Pandey A, Kitzman D, Whellan DJ, et al. Heart. 2019;105(20):1552–1558.
Cardiac cycle-dependent changes in aortic area
Frailty among older decompensated heart failure and distensibility are reduced in older patients 68. Haykowsky MJ, Brubaker PH, Morgan TM,
patients: prevalence, association with patient- with isolated diastolic heart failure and correlate Kritchevsky S, Eggebeen J, Kitzman DW. Impaired
centered outcomes, and efficient detection with exercise intolerance. J Am Coll Cardiol. aerobic capacity and physical functional perfor-
methods. J Am Coll Cardiol HF. 2019;7:1079– 2001;38:796–802. mance in older heart failure patients with pre-
1088. served ejection fraction: role of lean body mass.
54. Chirinos JA. Deep phenotyping of systemic
39. Kitzman DW, Whellan DJ, Duncan P, et al. J Gerontol A Biol Sci Med Sci. 2013;68:968–975.
arterial hemodynamics in HFpEF (part 2): clinical
Physical rehabilitation for older patients hospital- and therapeutic considerations. J Cardiovasc Transl 69. Kitzman DW, Nicklas B, Kraus WE, et al.
ized for heart failure. N Engl J Med. 2021;385(3): Res. 2017;10:261–274. Skeletal muscle abnormalities and exercise intol-
203–216. erance in older patients with heart failure and
55. Chirinos JA, Segers P, Hughes T, Townsend R.
40. Pandey A, Parashar A, Kumbhani D, et al. Ex- preserved ejection fraction. Am J Physiol Heart Circ
Large-artery stiffness in health and disease: JACC
ercise training in patients with heart failure and Physiol. 2014;306:H1364–H1370.
state-of-the-art review. J Am Coll Cardiol.
preserved ejection fraction: meta-analysis of ran- 2019;74:1237–1263. 70. Nilwik R, Snijders T, Leenders M, et al. The
domized control trials. Circ Heart Fail. 2015;8:33– decline in skeletal muscle mass with aging is
56. Zamani P, Rawat D, Shiva-Kumar P, et al. Ef-
40. mainly attributed to a reduction in type II muscle
fect of inorganic nitrate on exercise capacity in
41. Pfeffer MA, Shah AM, Borlaug BA. Heart failure fiber size. Exp Gerontol. 2013;48:492–498.
heart failure with preserved ejection fraction. Cir-
with preserved ejection fraction in perspective. culation. 2015;131:371–380. discussion 380. 71. Pandey A, Khera R, Park B, et al. Relative im-
Circ Res. 2019;124:1598–1617. pairments in hemodynamic exercise reserve pa-
57. Poole DC, Hirai DM, Copp SW, Musch TI.
42. Gorter TM, Obokata M, Reddy YNV, rameters in heart failure with preserved ejection
Muscle oxygen transport and utilization in heart
Melenovsky V, Borlaug BA. Exercise unmasks fraction: a study-level pooled analysis. J Am Coll
failure: implications for exercise (in)tolerance. Am
distinct pathophysiologic features in heart failure Cardiol HF. 2018;6:117–126.
J Physiol Heart Circ Physiol. 2012;302:H1050–
with preserved ejection fraction and pulmonary H1063. 72. Bhella PS, Prasad A, Heinicke K, et al.
vascular disease. Eur Heart J. 2018;39:2825–2835. Abnormal haemodynamic response to exercise in
58. Borlaug BA, Melenovsky V, Russell SD, et al.
43. Vanderpool RR, Saul M, Nouraie M, heart failure with preserved ejection fraction. Eur J
Impaired chronotropic and vasodilator reserves
Gladwin MT, Simon MA. Association between he- Heart Fail. 2011;13:1296–1304.
limit exercise capacity in patients with heart failure
modynamic markers of pulmonary hypertension and a preserved ejection fraction. Circulation. 73. Dhakal BP, Malhotra R, Murphy RM, et al.
and outcomes in heart failure with preserved 2006;114:2138–2147. Mechanisms of exercise intolerance in heart failure
ejection fraction. JAMA Cardiol. 2018;3:298–306. with preserved ejection fraction: the role of
59. Obokata M, Reddy YNV, Melenovsky V, et al.
44. Obokata M, Reddy YNV, Melenovsky V, abnormal peripheral oxygen extraction. Circ Heart
Myocardial injury and cardiac reserve in patients
Pislaru S, Borlaug BA. Deterioration in right ven- Fail. 2015;8:286–294.
with heart failure and preserved ejection fraction.
tricular structure and function over time in pa- J Am Coll Cardiol. 2018;72:29–40. 74. Molina AJ, Bharadwaj MS, Van HC, et al.
tients with heart failure and preserved ejection Skeletal muscle mitochondrial content, oxidative
fraction. Eur Heart J. 2019;40:689–697. 60. Ahmad A, Corban MT, Toya T, et al. Coronary
capacity, and Mfn2 expression are reduced in older
microvascular dysfunction is associated with ex-
45. Harhay MO, Kizer JR, Criqui MH, et al. Adipo- patients with heart failure and preserved ejection
ertional haemodynamic abnormalities in patients
kines and the right ventricle: the MESA-RV Study. fraction and are related to exercise intolerance.
with heart failure with preserved ejection fraction.
PLoS One. 2015;10:e0136818. J Am Coll Cardiol HF. 2016;4:636–645.
Eur J Heart Fail. 2021;23(5):765–772.
46. Lowery EM, Brubaker AL, Kuhlmann E, 75. Holloszy JO. Biochemical adaptations in mus-
61. Mohammed SF, Hussain S, Mirzoyev SA,
Kovacs EJ. The aging lung. Clin Interv Aging. cle. Effects of exercise on mitochondrial oxygen
Edwards WD, Maleszewski JJ, Redfield MM. Cor-
2013;8:1489–1496. uptake and respiratory enzyme activity in skeletal
onary microvascular rarefaction and myocardial
muscle. J Biol Chem. 1967;242:2278–2282.
47. Kovacs G, Olschewski A, Berghold A, fibrosis in heart failure with preserved ejection
Olschewski H. Pulmonary vascular resistances fraction. Circulation. 2015;131:550–559. 76. Goodpaster BH, Sparks LM. Metabolic flexi-
during exercise in normal subjects: a systematic bility in health and disease. Cell Metab. 2017;25:
62. Studenski SA, Peters KW, Alley DE, et al. The
review. Eur Respir J. 2012;39:319–328. 1027–1036.
FNIH Sarcopenia Project: rationale, study descrip-
48. Obokata M, Olson TP, Reddy YNV, tion, conference recommendations, and final es- 77. Haykowsky MJ, Tomczak CR, Scott JM,
Melenovsky V, Kane GC, Borlaug BA. Haemody- timates. J Gerontol A Biol Sci Med Sci. 2014;69: Paterson DI, Kitzman DW. Determinants of exer-
namics, dyspnoea, and pulmonary reserve in heart 547–558. cise intolerance in patients with heart failure and
failure with preserved ejection fraction. Eur Heart reduced or preserved ejection fraction. J Appl
63. Reginster JY, Cooper C, Rizzoli R, et al. Rec-
J. 2018;39:2810–2821. Physiol (1985). 2015;119:739–744.
ommendations for the conduct of clinical trials for
49. Olson TP, Johnson BD, Borlaug BA. Impaired drugs to treat or prevent sarcopenia. Aging Clin 78. Haykowsky MJ, Kouba EJ, Brubaker PH,
pulmonary diffusion in heart failure with preserved Exp Res. 2016;28:47–58. Nicklas BJ, Eggebeen J, Kitzman DW. Skeletal
muscle composition and its relation to exercise 94. Rayner JJ, Peterzan MA, Watson WD, et al. endothelial inflammation. J Am Coll Cardiol.
intolerance in older patients with heart failure and Myocardial energetics in obesity. Circulation. 2013;62:263–271.
preserved ejection fraction. Am J Cardiol. 2020;141:1152–1163.
111. Kalogeropoulos A, Georgiopoulou V,
2014;113:1211–1216.
95. Koepp KE, Obokata M, Reddy YNV, Olson TP, Psaty BM, et al. Inflammatory markers and inci-
79. Rao VN, Zhao D, Allison MA, et al. Adiposity Borlaug BA. Hemodynamic and functional impact dent heart failure risk in older adults: the Health
and incident heart failure and its subtypes: MESA of epicardial adipose tissue in heart failure with ABC (Health, Aging, and Body Composition) study.
(Multi-Ethnic Study of Atherosclerosis). J Am Coll preserved ejection fraction. J Am Coll Cardiol HF. J Am Coll Cardiol. 2010;55:2129–2137.
Cardiol HF. 2018;6:999–1007. 2020;8:657–666. 112. Everett BM, Cornel JH, Lainscak M, et al. Anti-
80. Patel KV, Bahnson JL, Gaussoin SA, et al. As- 96. Lindman BR, Dávila-Román VG, Mann DL, inflammatory therapy with canakinumab for the
sociation of baseline and longitudinal changes in et al. Cardiovascular phenotype in HFpEF patients prevention of hospitalization for heart failure.
body composition measures with risk of heart with or without diabetes: a RELAX trial ancillary Circulation. 2019;139:1289–1299.
failure and myocardial infarction in type 2 dia- study. J Am Coll Cardiol. 2014;64:541–549. 113. Van Tassell BW, Trankle CR, Canada JM, et al.
betes: findings from the Look AHEAD Trial. Circu- IL-1 blockade in patients with heart failure with
97. Vinik AI, Ziegler D. Diabetic cardiovascular
lation. 2020;142:2420–2430. preserved ejection fraction. Circ Heart Fail.
autonomic neuropathy. Circulation. 2007;115:387–
397. 2018;11:e005036.
81. Caughey MC, Vaduganathan M, Arora S, et al.
Racial differences and temporal obesity trends in 114. Waddingham MT, Paulus WJ. Microvascular
98. Wu YW, Hsu CL, Wang SS, et al. Impaired ex-
heart failure with preserved ejection fraction. J Am paradigm in heart failure with preserved ejection
ercise capacity in diabetic patients after coronary
Geriatr Soc. 2021;69:1309–1318. fraction: a quest for proof of concept. Circ Heart
bypass surgery: effects of diastolic and endothelial
Fail. 2017;10(6):e004179.
82. Kondamudi N, Thangada N, Patel KV, et al. function. Cardiology. 2008;110:191–198.
Regional adiposity, cardiorespiratory fitness, and 115. Netea MG, Joosten LA, Latz E, et al. Trained
99. Bauer TA, Reusch JE, Levi M, Regensteiner JG.
left ventricular strain: an analysis from the Dallas immunity: a program of innate immune memory in
Skeletal muscle deoxygenation after the onset of
Heart Study. J Cardiovasc Magn Reson. 2021;23: health and disease. Science. 2016;352:aaf1098.
moderate exercise suggests slowed microvascular
78. 116. Reynolds LM, Ding J, Taylor JR, et al. Tran-
blood flow kinetics in type 2 diabetes. Diabetes
83. Schiattarella GG, Altamirano F, Tong D, et al. Care. 2007;30:2880–2885. scriptomic profiles of aging in purified human im-
Nitrosative stress drives heart failure with pre- mune cells. BMC Genomics. 2015;16:333.
100. Kennedy BK, Berger SL, Brunet A, et al.
served ejection fraction. Nature. 2019;568:351– 117. de Magalhaes JP, Curado J, Church GM. Meta-
Geroscience: linking aging to chronic disease. Cell.
356. analysis of age-related gene expression profiles
2014;159:709–713.
84. Schwartz RS, Shuman WP, Bradbury VL, et al. identifies common signatures of aging. Bioinfor-
101. Burch JB, Augustine AD, Frieden LA, et al. matics. 2009;25:875–881.
Body fat distribution in healthy young and older
Advances in geroscience: impact on healthspan
men. J Gerontol. 1990;45:M181–M185. 118. Mentz RJ, Whellan DJ, Reeves GR, et al.
and chronic disease. J Gerontol A Biol Sci Med Sci.
2014;69(Suppl 1):S1–S3. Rehabilitation intervention in older patients with
85. Tchkonia T, Morbeck DE, Von Zglinicki T, et al.
acute heart failure with preserved versus reduced
Fat tissue, aging, and cellular senescence. Aging
102. Johnson SC, Rabinovitch PS, Kaeberlein M. ejection fraction. J Am Coll Cardiol HF. Published
Cell. 2010;9:667–684.
mTOR is a key modulator of ageing and age- online June 29, 2021.
86. Eren M, Boe AE, Klyachko EA, Vaughan DE. related disease. Nature. 2013;493:338–345.
119. Persson CE, Bjorck L, Lagergren J, Lappas G,
Role of plasminogen activator inhibitor-1 in
103. Franceschi C, Garagnani P, Morsiani C, et al. Giang KW, Rosengren A. Risk of heart failure in
senescence and aging. Semin Thromb Hemost.
The continuum of aging and age-related diseases: obese patients with and without bariatric surgery
2014;40:645–651.
common mechanisms but different rates. Front in Sweden-a registry-based study. J Card Fail.
87. Vaughan DE, Rai R, Khan SS, Eren M, Med (Lausanne). 2018;5:61. 2017;23:530–537.
Ghosh AK. Plasminogen activator inhibitor-1 is a
104. Harrison DE, Strong R, Sharp ZD, et al. 120. Fontana L, Partridge L, Longo VD. Extending
marker and a mediator of senescence. Arterioscler
Rapamycin fed late in life extends lifespan in healthy life span—from yeast to humans. Science.
Thromb Vasc Biol. 2017;37:1446–1452.
genetically heterogeneous mice. Nature. 2010;328:321–326.
88. de Boer RA, Nayor M, deFilippi CR, et al. As- 2009;460:392–395.
121. Donato AJ, Walker AE, Magerko KA, et al.
sociation of cardiovascular biomarkers with inci- Life-long caloric restriction reduces oxidative
105. Vilchez D, Simic MS, Dillin A. Proteostasis and
dent heart failure with preserved and reduced stress and preserves nitric oxide bioavailability and
aging of stem cells. Trends Cell Biol. 2014;24:161–
ejection fraction. JAMA Cardiol. 2018;3:215–224. function in arteries of old mice. Aging Cell.
170.
89. Krotkiewski M, Bjorntorp P, Sjostrom L, 2013;12:772–783.
106. Klaips CL, Jayaraj GG, Hartl FU. Pathways of
Smith U. Impact of obesity on metabolism in men 122. Rippe C, Lesniewski L, Connell M, LaRocca T,
cellular proteostasis in aging and disease. J Cell
and women. Importance of regional adipose tissue Donato A, Seals D. Short-term calorie restriction
Biol. 2018;217:51–63.
distribution. J Clin Invest. 1983;72:1150–1162. reverses vascular endothelial dysfunction in old
107. Miyamoto S. Autophagy and cardiac aging. mice by increasing nitric oxide and reducing
90. Hulsmans M, Sager HB, Roh JD, et al. Cardiac
Cell Death Differ. 2019;26:653–664. oxidative stress. Aging Cell. 2010;9:304–312.
macrophages promote diastolic dysfunction. J Exp
Med. 2018;215:423–440. 108. Seiler M, Bowen TS, Rolim N, et al. Skeletal 123. Most J, Tosti V, Redman LM, Fontana L. Cal-
muscle alterations are exacerbated in heart failure orie restriction in humans: an update. Ageing Res
91. Oh A, Okazaki R, Sam F, Valero-Munoz M.
with reduced compared with preserved ejection Rev. 2017;39:36–45.
Heart failure with preserved ejection fraction and
fraction: mediated by circulating cytokines? Circ
adipose tissue: a story of two tales. Front Car- 124. Heilbronn LK, de Jonge L, Frisard MI, et al.
Heart Fail. 2016;9:e003027.
diovasc Med. 2019;6:110. Effect of 6-month calorie restriction on biomarkers
109. Bowen TS, Herz C, Rolim NPL, et al. Effects of longevity, metabolic adaptation, and oxidative
92. Valero-Munoz M, Li S, Wilson RM, et al. Heart
of endurance training on detrimental structural, stress in overweight individuals: a randomized
failure with preserved ejection fraction induces
cellular, and functional alterations in skeletal controlled trial. JAMA. 2006;295:1539–1548.
beiging in adipose tissue. Circ Heart Fail. 2016;9:
muscles of heart failure with preserved ejection
e002724. 125. Racette SB, Weiss EP, Villareal DT, et al, for
fraction. J Card Fail. 2018;24:603–613.
the Washington University School of Medicine
93. Carbone S, Canada JM, Buckley LF, et al.
110. Paulus WJ, Tschope C. A novel paradigm for CALERIE Group. One year of caloric restriction in
Obesity contributes to exercise intolerance in
heart failure with preserved ejection fraction: humans: feasibility and effects on body composi-
heart failure with preserved ejection fraction. J Am
comorbidities drive myocardial dysfunction and tion and abdominal adipose tissue. J Gerontol A
Coll Cardiol. 2016;68:2487–2488. remodeling through coronary microvascular Biol Sci Med Sci. 2006;61:943–950.
126. Hummel SL, Karmally W, Gilespie BG, et al. 129. MoTrPAC. Molecular Transducers of Physical LIFE study randomized clinical trial. JAMA.
Home-delivered meals postdischarge from heart Activity Consortium. Accessed June 8, 2020. 2014;311:2387–2396.
failure hospitalization: the GOURMET-HF Pilot https://www.motrpac.org/
Study. Circ Heart Fail. 2018;11:e004886.
130. Pahor M, Guralnik JM, Ambrosius WT, et al.
127. Zhu Y, Tchkonia T, Pirtskhalava T, et al. The KEY WORDS aging, exercise intolerance,
Effect of structured physical activity on prevention
Achilles’ heel of senescent cells: from tran- heart failure with preserved ejection fraction,
of major mobility disability in older adults: the
scriptome to senolytic drugs. Aging Cell. 2015;14: senescence, skeletal muscle
LIFE study randomized clinical trial. JAMA.
644–658.
2014;311:2387–2396.
128. University of Pittsburgh Center for Aging and
131. Pahor M, Guralnik JM, Ambrosius WT, et al. A PP END IX For details on impairment in left
Population Health. SOMMA Study of Muscle,
Effect of structured physical activity on prevention heart function reserve, please see the online
Mobility and Aging. Accessed June 8, 2020.
of major mobility disability in older adults: the version of this paper.
https://www.caph.pitt.edu/somma/
FELLOWS-IN-TRAINING & EARLY CAREER SECTION
How Cardiovascular Disease Fellows

Can Promote Diversity and
Inclusion in Cardiology
Doing Our Part
Joyce N. Njoroge, MD,a Quentin R. Youmans, MD, MSC,b Sarah Chuzi, MD, MSCb
I n 2020, the medical community reached a greater

awareness of how bias, social determinants of
health, and racism harm our patients and health-
care systems (1). Cardiology is central to this discus-
power and privilege to create an inclusive environ-
ment for junior trainees, and advocate for change.
In this paper, we explore 4 illustrative scenarios
commonly encountered by cardiology FITs that
sion, not only because minority patients experience highlight opportunities to change culture and support
staggering disparities in cardiovascular mortality (2), diverse trainees through allyship and advocacy
but because the cardiology workforce suffers from a (Figure 1). These concrete, actionable strategies
significant lack of diversity (3). The data show that highlight the ways in which both faculty leadership
physicians who are under-represented in medicine and FITs can actively embrace diversity and
(UIM), including racial, ethnic, sexual, and gender inclusion.
minority groups, enhance the quality of patient care
and are more likely to work in underserved areas OPPORTUNITY 1: LEAD DIVERSE TEAMS OF
(4); thus, the diversification of our cardiology work- TRAINEES WITH INTENTION
force is an urgent and critical need. Cultivating an in-
clusive environment within the field is an important It is Rob’s first day as the FIT on the cardiology consult
step toward achieving this goal. service. He leads a team of internal medicine residents
Although efforts to enhance diversity and inclu- and medical students. Joan (medical student) is the
sion in cardiology have historically focused on the only Black team member. The morning is busy, and
roles of professional societies and academic leader- Rob notices that Joan is quiet in the team room and
ship, we believe cardiology fellows-in-training during rounds. Although the other medical students
(FITs) have a significant opportunity to contribute eagerly answer the attending’s questions, Joan waits
to this space. Cardiology FITs are omnipresent until she is called upon to speak. Although she answers
throughout most academic medical centers. We lead all questions correctly, she appears nervous, and
teams of junior trainees, not only exposing them to defers opportunities to see new patients to the other
the richness of clinical cardiology, but also repre- students.
senting the culture of the field. We, therefore, have Diverse trainees, particularly those who are UIM,
a unique opportunity to set an example, use our face profound barriers to success, including micro-
aggressions and differing performance expectations
(5). These barriers contribute to imposter syndrome
and social isolation, and impact behavior and clinical
From the aDivision of Cardiology, Department of Medicine, University of
California at San Francisco School of Medicine, San Francisco, California,
performance. In particular, the intersection of female
USA; and the bDivision of Cardiology, Department of Medicine, Feinberg sex and minority status may be associated with more
School of Medicine, Northwestern University, Chicago, Illinois, USA. negative experiences for trainees (6). Being cognizant
The authors attest they are in compliance with human studies
of these inherent challenges is only the first step
committees and animal welfare regulations of the author’s institution
and Food and Drug Administration guidelines, including patient consent toward being an effective leader and ally. FITs
where appropriate. For more information, visit the Author Center. must then commit to creating an inclusive team

JACC VOL. 78, NO. 11, 2021 Njoroge et al. 1189
SEPTEMBER 14, 2021:1188–1192 Fellows-in-Training & Early Career Section
F I G U R E 1 Challenges and Strategies
UIM trainees Lead diverse teams with intention

face barriers • Review expectations
to success on • Facilitate goal-setting
teams • Minimize rivalry among trainees
Respond to bias and mistreatment

• Step in
Mistreatment • Address the behavior
Challenges and during clinical • Emphasize the institution's values
Strategies for care is • Set expectations and boundaries
Promoting Diversity common • Report and document
& Inclusion for
Cardiology FITs
UIM trainees Mentor and sponsor aggressively

lack • Connect diverse trainees to mentors
adequate • Involve trainees in research projects
mentors and • Provide positive feedback to medical
sponsors school and residency leadership
• Serve as a resource for trainees
The cardiology
workforce lacks
diversity Advocate for change to leadership
• Be outspoken and persistent about
desire to change program culture
Challenges and strategies for promoting diversity and inclusion for the cardiology fellow-in-training (FITs). UIM ¼ under-represented
in medicine.
environment that fosters the unique perspectives (eg, medical student) is engaged first on a topic (eg,
each trainee brings to the field. what are the classic electrocardiographic signs of
We propose a few strategies FITs can use to be pulmonary embolism?). If that trainee is unsure,
inclusive leaders. First, set the tone for the team. engage the next most junior trainee, rather than
Review expectations for the rotation, including that asking another medical student. This provides op-
all trainees, staff, and patients are treated with portunities for trainees to teach each other in a
respect. Acknowledge one’s own fallibility and invite nonthreatening environment and minimizes trainee
learners to provide constructive feedback (eg, “This anxiety.
month, we will all make mistakes, myself included.
Let’s commit to giving each other regular, construc- OPPORTUNITY 2: RESPOND TO
tive feedback.”) Second, facilitate goal setting. Meet MICROAGGRESSIONS AND
with trainees individually at the start of a rotation to MACROAGGRESSIONS AGAINST
assess their goals for the block; then, follow-up on TEAM MEMBERS
their progress and show an interest in their growth.
Third, consider the use of humor in the team setting. Ann, the cardiology FIT on call, is asked to evaluate a
When used in a tactical, appropriate way, humor can patient (Mr M) in the emergency department. She
put team members at ease, reducing anxiety and brings an internal medicine resident, who is openly
stress, and enhancing the learning environment (7). transgender, to see the patient. After seeing the resi-
Finally, minimize rivalry among trainees. Consider a dent, Mr M immediately says he would like to be cared
“ladder” approach whereby the most junior trainee for by someone who “isn’t a queer.”
1190 Njoroge et al. JACC VOL. 78, NO. 11, 2021
Fellows-in-Training & Early Career Section SEPTEMBER 14, 2021:1188–1192
bias, discrimination, and lack of support (10). Cardi-

T A B L E 1 The SAFER Model for Responding to Mistreatment
ology FITs can fill this gap and serve as a resource for
Strategy Example Response trainees by connecting UIM trainees to research or
Step in when you observe “Help me understand your request. It clinical mentors with shared interests within the di-
inappropriate behavior sounds like you are asking to be
treated by another clinician.” vision, involving trainees in research projects, pro-
Address the behavior “All of our clinicians and staff here are actively reaching out to residency or medical school
qualified to care for you. This is not
program leadership about a trainees’ strengths and
an appropriate request.”
Focus on the institution’s “At our institution, we value our ability performance on cardiology rotations, and serving as a
values to provide excellent care to all resource, especially during application seasons.
patients. We also value treating all
patients and staff with respect and This scenario represents a missed opportunity to
dignity.” mentor, sponsor, and recruit a UIM trainee who may
Explain the institution’s “We do not change doctors/nurses/etc not have had equal access to research opportunities
expectations and set because of their race/ethnicity/
boundaries sexual orientation.” previously. FITs should be proactive and inclusive in
Report the incident and Incident is reported to supervisor their research endeavors, recognizing that mentoring,
document the event and/or risk management and is
documented appropriately.
sponsoring, and elevating junior trainees leaves a
lasting impact on the field.
OPPORTUNITY 4: ADVOCATE FOR CHANGE

Expressions of bias and mistreatment toward
TO PROGRAM LEADERSHIP
medical trainees are common and contribute to
burnout and moral distress (8). FITs are at the front-
Eric is a first-year FIT at his dream program. After
lines of clinical care and may witness many of these
Match day, he was disappointed to learn that his
harmful interactions. The impact of allyship from the
program did not match any UIM or female trainees.
cardiology FIT should not be underestimated, and it
He is passionate about gender and racial diversity in
is imperative that FITs develop the courage and skills
cardiology, but he is not sure what role he can play to
to address these encounters.
change practices at his institution.
Communication strategies and frameworks for
Cardiology FITs are often vocal about program lo-
addressing mistreatment against healthcare pro-
gistics (eg, call schedule, rotation responsibilities) but
fessionals have been developed. One framework is
may feel less sure about their role in advocating for a
SAFER (9), in which witnesses step in, address the
more diverse and inclusive environment. We argue
inappropriate behavior, emphasize the values and
that this is very much within our purview and
boundaries of the institution, and report the incident
encourage cardiology FITs to be outspoken and
(Table 1). Regardless of the framework used, FITs
persistent about their desires to change institutional
should take the opportunity to debrief with the
and program culture. Consider starting a fellow-led
trainee and any other witnesses to provide emotional
committee with the goal of promoting and support-
support and discuss follow-up. By taking a thought-
ing an inclusive climate within your division. Recruit
ful, constructive response to mistreatment, FITs can
program leadership, faculty, and fellows and hold
advocate for a more fully inclusive culture in
regular meetings to discuss the importance of pro-
cardiology.
gram diversity, develop initiatives to enhance
OPPORTUNITY 3: MENTOR AND SPONSOR recruitment of UIM and female applicants, and orga-
AGGRESSIVELY AND BROADLY nize educational initiatives for the division. This
group can serve as a hub for continued dialogue about
Jane is a cardiology FIT looking for a resident to assist the importance of diversity in cardiology and a way to
her with a research project. There are numerous bring like-minded, passionate individuals together
diverse residents who are interested in cardiology; within a division.
however, a co-fellow tells Jane that she must recruit 1 In conclusion, a diverse cardiology workforce
of 2 residents (both White) who have successfully benefits patients, communities, and clinicians, and
completed multiple research projects and are “super- the cardiology FIT is uniquely positioned to play a
stars.” Wanting to complete her project expeditiously, role in enhancing the diversity and inclusivity of our
Jane asks one of these residents to assist given their field. FITs must make it known that diversity matters.
“track record.” Through awareness, allyship, and advocacy, we must
Mentorship and sponsorship are invaluable in car- apply the same passion and creativity that we bring to
diology. However, UIM trainees face significant bar- research and patient care to innovative and fervent
riers to establishing mentors and sponsors due to endeavors to enhance the diversity of our workforce.
ACKNOWLEDGMENTS The authors thank Dr Amber

Johnson and Dr Katherine Berlacher for their valuable ADDRESS FOR CORRESPONDENCE: Dr Sarah Chuzi,
contributions to this piece. Division of Cardiology, Department of Medicine,

Northwestern University, Feinberg School of Medi-
cine, 676 N St Clair Street, Suite 600, Chicago, Illinois
The authors have reported that they have no relationships relevant to 60611, USA. E-mail: sarah-chuzi@northwestern.edu.
the contents of this paper to disclose. Twitter: @chuzisarah.
REFERENCES
1. Godlee F. Racism: the other pandemic. BMJ. United States. J Am Heart Assoc. 2021;10: in medical education. Med Educ. 2013;47:
2020;369:m2303. e018893. 40–48.
2. Virani Salim S, Alonso A, Aparicio Hugo J, et al. 5. Ackerman-Barger K, Valderama-Wallace C, 8. Fnais N, Soobiah C, Chen MH, et al. Harassment
Heart disease and stroke statistics—2021 update: a Latimore D, Drake C. Stereotype threat suscepti- and discrimination in medical training: a system-
report from the American Heart Association. Cir- bility among minority health professions students. atic review and meta-analysis. Acad Med. 2014;89:
culation. 2021;143:e254–e743. J Best Practices Health Professions Diversity. 817–827.
2016;9:1232–1246.
3. Santhosh L, Babik JM. Trends in racial and 9. Warsame R, Hayes S. Mayo Clinic’s 5-step policy
ethnic diversity in internal medicine subspecialty 6. Mocanu V, Kuper TM, Marini W, et al. Inter- for responding to bias incidents. AMA J Ethics.
fellowships from 2006 to 2018. JAMA Netw Open. sectionality of gender and visible minority status 2019;21:E521–E529.
2020;3:e1920482. among general surgery residents in Canada. JAMA
10. Orom H, Semalulu T, Underwood 3rd W. The
Surg. 2020;155:e202828.
4. Capers Q, Johnson A, Berlacher K, Douglas PS. social and learning environments experienced by
The urgent and ongoing need for diversity, inclu- 7. Lingard L. Language matters: towards underrepresented minority medical students: a
sion, and equity in the cardiology workforce in the an understanding of silence and humour narrative review. Acad Med. 2013;88:1765–1777.
RESPONSE: Promoting Equity, Diversity,

and Inclusion in Cardiology
Robert O. Roswell, MD
Department of Cardiology, Lenox Hill Hospital, Northwell Health, The Zucker School of Medicine at Hofstra/Northwell,
New York, New York, USA
E-mail: rroswell@northwell.edu
Twitter: @DrRobRoswell
Health disparities, under-representation in cardiol- field of cardiology by addressing some common

ogy, bias, and lack of inclusion in medicine are all downstream manifestations of structural inequities in
sequelae of structural inequities. Structural in- the workplace. The strength of their work lies in the
equities are systemic and unjust conditions, policies, communication of actionable steps to enhance EDI in
and cultural norms that create barriers to opportu- cardiology while also explaining EDI concepts. The
nities and health, yet they are ultimately rectifiable scenario-based “opportunities” are sensible in-
(1). For instance, structural inequities were associated terventions for FITs and faculty to advance EDI in
with increased out-of-hospital sudden death during relatable clinical contexts.
the coronavirus disease 2019 pandemic, which has “Leading teams with intention” by building psy-
disproportionately infected and killed Black, His- chological safety (Opportunity 1) is critically impor-
panic, and Indigenous people (2). The syndemic ef- tant to an effective, productive, safe, and inclusive
fect of this novel viral outbreak, overlaid on the long- learning and healthcare environment (3). As Njoroge
standing inequities of the United States and its et al. point out, performing these small steps can
healthcare system, has recently reanimated societal vastly change the culture, effectiveness, and perfor-
efforts to advance equity, diversity, and inclusion mance of a medical team. Conversely, team culture
(EDI). and performance can be deeply compromised by
Njoroge et al. have written an insightful article on microaggressions and macroaggressions. Macro-
how FITs and faculty can help promote EDI in the aggressions, where the discriminatory message and
its target are made clear, are more egregious than critical to securing a fellowship and being promoted
microaggressions, where the target may be less clear in cardiology and demonstrate how structural biases
and the message more subtly derisive. Yet, micro- can counteract well-intentioned EDI initiatives in
aggressions can actually cause more psychological cardiology.
harm because confusion about the intent of the To truly advance EDI in cardiology—and more
message can produce intrusive cognition, which re- broadly in medicine and society—our efforts must
fers to interruptive thoughts trying to decipher the transcend the confines of our fellowship programs
meaning of the statement or action (4). Allyship is and health institutions. As a medical community, we
crucial to interrupting both microaggressions and must build deep pipelines that support the education
macroaggressions; effective training will help FITs and recruitment of under-represented groups to
and faculty more readily recognize and interrupt medicine by engaging with their communities and
microaggressions when they arise (Opportunity 2). advocating for equity in the education system. We
EDI are interdependent. Without explicit inclu- must understand empirically and work to change the
sion, implicit exclusion will continue to tacitly stymie inequitable structures of our society and health sys-
equity and diversity. Explicit inclusion entails being tem, which fail far too many. Our challenges are great,
intentional about extending mentorship and spon- but so, too, are the rewards of a more equitable,
sorship opportunities to under-represented learners diverse, inclusive society. As the authors state, we all
and faculty (Opportunity 3). These opportunities are have to do our part.
REFERENCES
1. Itchhaporia D. Paving the way for health equity COVID-19 pandemic. Circ Arrhythm Electrophysiol. Qual Patient Saf. Published online September 20,
in cardiology: why does it matter? J Am Coll Car- 2021;14:e009646. 2020. https://doi.org/10.1016/j.jcjq.2020.09.
diol. 2021;77:2613–2616. 005
3. Jung OS, Kundu P, Edmondson AC, et al.
2. Mountantonakis SE, Epstein LM, Coleman K, Resilience vs. vulnerability: psychological safety 4. Sue DW, Capodilupo CM, Torino GC, et al. Racial
et al. The association of structural inequities and and reporting of near misses with varying prox- microaggressions in everyday life: implications for
race with out-of-hospital sudden death during the imity to harm in radiation oncology. Jt Comm J clinical practice. Am Psychol. 2007;62:271–286.
ACKNOWLEDGMENTS The authors thank Dr Amber

Johnson and Dr Katherine Berlacher for their valuable ADDRESS FOR CORRESPONDENCE: Dr Sarah Chuzi,
contributions to this piece. Division of Cardiology, Department of Medicine,

Northwestern University, Feinberg School of Medi-
cine, 676 N St Clair Street, Suite 600, Chicago, Illinois
The authors have reported that they have no relationships relevant to 60611, USA. E-mail: sarah-chuzi@northwestern.edu.
the contents of this paper to disclose. Twitter: @chuzisarah.
REFERENCES
1. Godlee F. Racism: the other pandemic. BMJ. United States. J Am Heart Assoc. 2021;10: in medical education. Med Educ. 2013;47:
2020;369:m2303. e018893. 40–48.
2. Virani Salim S, Alonso A, Aparicio Hugo J, et al. 5. Ackerman-Barger K, Valderama-Wallace C, 8. Fnais N, Soobiah C, Chen MH, et al. Harassment
Heart disease and stroke statistics—2021 update: a Latimore D, Drake C. Stereotype threat suscepti- and discrimination in medical training: a system-
report from the American Heart Association. Cir- bility among minority health professions students. atic review and meta-analysis. Acad Med. 2014;89:
culation. 2021;143:e254–e743. J Best Practices Health Professions Diversity. 817–827.
2016;9:1232–1246.
3. Santhosh L, Babik JM. Trends in racial and 9. Warsame R, Hayes S. Mayo Clinic’s 5-step policy
ethnic diversity in internal medicine subspecialty 6. Mocanu V, Kuper TM, Marini W, et al. Inter- for responding to bias incidents. AMA J Ethics.
fellowships from 2006 to 2018. JAMA Netw Open. sectionality of gender and visible minority status 2019;21:E521–E529.
2020;3:e1920482. among general surgery residents in Canada. JAMA
10. Orom H, Semalulu T, Underwood 3rd W. The
Surg. 2020;155:e202828.
4. Capers Q, Johnson A, Berlacher K, Douglas PS. social and learning environments experienced by
The urgent and ongoing need for diversity, inclu- 7. Lingard L. Language matters: towards underrepresented minority medical students: a
sion, and equity in the cardiology workforce in the an understanding of silence and humour narrative review. Acad Med. 2013;88:1765–1777.
RESPONSE: Promoting Equity, Diversity,

and Inclusion in Cardiology
Robert O. Roswell, MD
Department of Cardiology, Lenox Hill Hospital, Northwell Health, The Zucker School of Medicine at Hofstra/Northwell,
New York, New York, USA
E-mail: rroswell@northwell.edu
Twitter: @DrRobRoswell
Health disparities, under-representation in cardiol- field of cardiology by addressing some common

ogy, bias, and lack of inclusion in medicine are all downstream manifestations of structural inequities in
sequelae of structural inequities. Structural in- the workplace. The strength of their work lies in the
equities are systemic and unjust conditions, policies, communication of actionable steps to enhance EDI in
and cultural norms that create barriers to opportu- cardiology while also explaining EDI concepts. The
nities and health, yet they are ultimately rectifiable scenario-based “opportunities” are sensible in-
(1). For instance, structural inequities were associated terventions for FITs and faculty to advance EDI in
with increased out-of-hospital sudden death during relatable clinical contexts.
the coronavirus disease 2019 pandemic, which has “Leading teams with intention” by building psy-
disproportionately infected and killed Black, His- chological safety (Opportunity 1) is critically impor-
panic, and Indigenous people (2). The syndemic ef- tant to an effective, productive, safe, and inclusive
fect of this novel viral outbreak, overlaid on the long- learning and healthcare environment (3). As Njoroge
standing inequities of the United States and its et al. point out, performing these small steps can
healthcare system, has recently reanimated societal vastly change the culture, effectiveness, and perfor-
efforts to advance equity, diversity, and inclusion mance of a medical team. Conversely, team culture
(EDI). and performance can be deeply compromised by
Njoroge et al. have written an insightful article on microaggressions and macroaggressions. Macro-
how FITs and faculty can help promote EDI in the aggressions, where the discriminatory message and
its target are made clear, are more egregious than critical to securing a fellowship and being promoted
microaggressions, where the target may be less clear in cardiology and demonstrate how structural biases
and the message more subtly derisive. Yet, micro- can counteract well-intentioned EDI initiatives in
aggressions can actually cause more psychological cardiology.
harm because confusion about the intent of the To truly advance EDI in cardiology—and more
message can produce intrusive cognition, which re- broadly in medicine and society—our efforts must
fers to interruptive thoughts trying to decipher the transcend the confines of our fellowship programs
meaning of the statement or action (4). Allyship is and health institutions. As a medical community, we
crucial to interrupting both microaggressions and must build deep pipelines that support the education
macroaggressions; effective training will help FITs and recruitment of under-represented groups to
and faculty more readily recognize and interrupt medicine by engaging with their communities and
microaggressions when they arise (Opportunity 2). advocating for equity in the education system. We
EDI are interdependent. Without explicit inclu- must understand empirically and work to change the
sion, implicit exclusion will continue to tacitly stymie inequitable structures of our society and health sys-
equity and diversity. Explicit inclusion entails being tem, which fail far too many. Our challenges are great,
intentional about extending mentorship and spon- but so, too, are the rewards of a more equitable,
sorship opportunities to under-represented learners diverse, inclusive society. As the authors state, we all
and faculty (Opportunity 3). These opportunities are have to do our part.
REFERENCES
1. Itchhaporia D. Paving the way for health equity COVID-19 pandemic. Circ Arrhythm Electrophysiol. Qual Patient Saf. Published online September 20,
in cardiology: why does it matter? J Am Coll Car- 2021;14:e009646. 2020. https://doi.org/10.1016/j.jcjq.2020.09.
diol. 2021;77:2613–2616. 005
3. Jung OS, Kundu P, Edmondson AC, et al.
2. Mountantonakis SE, Epstein LM, Coleman K, Resilience vs. vulnerability: psychological safety 4. Sue DW, Capodilupo CM, Torino GC, et al. Racial
et al. The association of structural inequities and and reporting of near misses with varying prox- microaggressions in everyday life: implications for
race with out-of-hospital sudden death during the imity to harm in radiation oncology. Jt Comm J clinical practice. Am Psychol. 2007;62:271–286.
ISSN 0735-1097/$36.00
Letters
square testing at an alpha level of 0.05 and beta of

A Transatlantic 0.2). All patients who reported a history of CVD were
Comparison of allocated a 10-year CVD risk of 20% in continuous
analyses or to the $10% category in categorical ana-
Patient-Reported lyses. Because the survey was anonymous, the rele-
Access to and Use of vant ethics committees waived the requirement for
informed consent.
Aspirin in Contemporary Between November 2019 and September 2020, we
Preventive Cardiology received 300 complete survey responses (29% of
those invited); 150 at each site (Figure 1). Almost one-
half of the European cohort (49%) reported a personal
Aspirin is effective in the secondary prevention of history of CVD compared with 32.7% of the U.S.
cardiovascular disease (CVD). By contrast, aspirin’s cohort (P ¼ 0.01). However, the median 10-year CVD
role in primary prevention is controversial, with the risk in the primary prevention subgroup was similar
most recent data demonstrating a reduction in at both sites (United States 19% vs Europe 16%; P ¼
nonfatal CVD that is largely counterbalanced by 0.06).
excess risk of major bleeding (1). Current guidelines A greater percentage of the U.S. primary preven-
for primary prevention aspirin differ on both sides of tion subgroup was taking aspirin compared with
the Atlantic (2,3). Accordingly, patients may have Europe (46% vs 26%; P ¼ 0.01). Almost all participants
heard mixed messages about aspirin use in primary who reported aspirin use were taking low-dose
prevention, and some may even now be uncertain aspirin (#81 mg daily); however, 96% of the Euro-
about its use in the secondary prevention of CVD. We pean group reported obtaining aspirin via prescrip-
aimed to survey contemporary patients with a history tion, whereas 84% of the U.S. group obtained aspirin
of or risk factors for CVD to evaluate their use and as an over-the-counter (OTC) medication. Many par-
understanding of aspirin. We enrolled participants in ticipants in both the United States and Europe were
the United States and Europe so that comparisons unable to quantify the relative magnitude of aspirin’s
could be made in responses by location. benefits (61% reported uncertainty) and risks (69%
We developed an anonymous survey instrument to reported uncertainty). Just 62% of all participants
collect demographic and clinical information on par- correctly reported that aspirin was routinely indi-
ticipants; estimate their 10-year risk of future CVD; cated for secondary prevention, whereas 47% incor-
record their use of and access to aspirin; and examine rectly answered that aspirin was routinely indicated
their understanding of the risks, benefits, and infor primary prevention, meeting the prespecified
dications for aspirin. Taking a convenience sample level of statistical significance for the primary outcome
approach, consecutive patients attending hospital- (P ¼ 0.048). Notably, there was also poor agreement
based cardiology or internal medicine outpatient between participants’ self-reported and calculated
clinics at a tertiary care academic center in both the cardiovascular risk using QRISK3 (Cohen’s Kappa 0.1
United States (Johns Hopkins Hospital) and Europe [with values <0.4 considered low]). Major limitations
(National University of Ireland, Galway, Ireland) were of the study are that results from these tertiary centers
invited to participate. Estimated 10-year CVD risk was were self-reported and may also not be generalizable to
calculated using QRISK3, which is validated and all of the United States or Europe.
allowed calculation using patient-reported data alone This survey demonstrated that a significant pro-
(4). We prespecified that 300 patients would be portion of patients continue to take aspirin for pri-
needed to determine a >15% difference in the primary mary prevention, particularly in the United States
outcome of accurate patient understanding regarding where aspirin is obtained OTC far more often than in
guideline recommendations for routine aspirin use in Europe. In both the United States and Europe, the
primary vs secondary CVD prevention (assuming chi- following was true of many patients: 1) they did not
1194 Letters JACC VOL. 78, NO. 11, 2021
SEPTEMBER 14, 2021:1193–1195
F I G U R E 1 A Transatlantic Survey on Aspirin Use in CVD Prevention
A Participant Demographic and Clinical Data
300 total participants

Mean ± SD age 65 ± 10 years
40% female
41% with history of CVD
NATIONAL UNIVERSITY OF
JOHNS HOPKINS HOSPITAL IRELAND, GALWAY
150 participants (response 150 participants (response
Key demographic and clinical
rate 25%), Mean age rate 32%), Mean age
characteristics that differ between
68 years, 63 years, 97% White
U.S. and European cohorts
39% Black participants, participants, 49% with
33% with history of CVD history of CVD
Mean risk of entire sample = 21% Mean risk of entire sample = 19%
Similar calculated 10-year CVD
risk in U.S. and European cohort
Median risk of primary Median risk of primary
prevention sample = 19% prevention sample = 16%
B Participant Aspirin Use, Understanding of Aspirin, and Understanding of their Cardiovascular Risk
Primary prevention: 46% using ASA Primary prevention: 26% using ASA
Aspirin use, dose, source
Secondary prevention: 90% using ASA Secondary prevention: 81% using ASA
and understanding
Most using 81 mg daily (95%) Most using 75 mg daily (96%)
of its role in CVD
84% obtained over the counter 98% obtained via prescription
Poor understanding of the role of aspirin in the prevention of CVD at both locations
47% incorrectly reported that aspirin was routinely indicated for primary prevention
62% correctly reported that aspirin was routinely indicated for secondary prevention
AGREEMENT BETWEEN PERCEIVED AND CALCULATED 10-YEAR FUTURE CVD RISK

Actual Risk Actual Risk Cohen’s
AMONG ALL PARTICIPANTS*
<1 in 10 (<10%) ≥1 in 10(≥10%) Kappa
Overall (both study sites) 0.097
Perceived Risk <1 in 10 (<10%) 16 65
Perceived Risk ≥1 in 10 (≥10%) 5 53
Johns Hopkins Hospital 0.12
National University of Ireland, Galway 0.068
(A) Participants at both sites had similar age and 10-year CVD risk. (B) Most participants used low-dose aspirin. However, more primary prevention adults in America
were taking aspirin than in Europe. Aspirin was predominantly obtained over-the-counter in the United States but by prescription in Europe. At both sites, there was
poor agreement between perceived CVD risk and actual (calculated) CVD risk using QRISK3. *Of all 300 participants, 161 selected "don't know/not sure" when asked to
estimate their CVD risk and were not included in the analysis of Cohen’s Kappa. ASA ¼ acetylsalicylic acid; CVD ¼ cardiovascular disease.
JACC VOL. 78, NO. 11, 2021 Letters 1195
SEPTEMBER 14, 2021:1193–1195
know the difference between aspirin guideline Moyola Lane

recommendations for primary and secondary CVD Galway, Ireland
prevention; 2) they misunderstood the relative risks E-mail: johnwilliam.mcevoy@nuigalway.ie
and benefits of aspirin; and 3) they lacked insight into Twitter: @johnwmcevoy
their own CVD risk. https://doi.org/10.1016/j.jacc.2021.07.015
These findings challenge the widespread avail- Ó 2021 by the American College of Cardiology Foundation. Published by Elsevier.
ability of OTC aspirin for CVD prevention in America,
The authors have reported that they have no relationships relevant to the contents
particularly when compared with Europe, where of this paper to disclose. The authors thank Dominque Ashen, PhD, for her contri-
bution to enrollment.
either a prescription or a risk-benefit discussion with
The authors attest they are in compliance with human studies committees and
a pharmacist is necessary to gain access to low-dose animal welfare regulations of the authors’ institutions and Food and Drug
Administration guidelines, including patient consent where appropriate. For more
aspirin from behind the pharmacy counter.
information, visit the Author Center.
Alan P. Jacobsen, MB, BCh, BAO REFERENCES

Zi Lun Lim, MB, BCh, BAO
1. Raber I, McCarthy CP, Vaduganathan M, et al. The rise and fall of aspirin in
Blair Chang, BA the primary prevention of cardiovascular disease. Lancet. 2019;393(10186):
Kaleb D. Lambeth, MD 2155–2167. https://doi.org/10.1016/s0140-6736(19)30541-0
Thomas M. Das, MD 2. Arnett DK, Blumenthal RS, Albert MA, et al. 2019 ACC/AHA guideline on the
Colin Gorry, MB, BCh, BAO primary prevention of cardiovascular disease: a report of the American College
of Cardiology/American Heart Association Task Force on Clinical Practice
Michael McCague, MSc
Guidelines. J Am Coll Cardiol. 2019;74(10):e177–e232. https://doi.org/10.
William Wijns, MD, PhD 1016/j.jacc.2019.03.010
Patrick W.J.C. Serruys, MD, PhD 3. Piepoli MF, Hoes AW, Agewall S, et al. 2016 European guidelines on car-
Roger S. Blumenthal, MD diovascular disease prevention in clinical practice. Eur Heart J. 2016;37(29):
Seth S. Martin, MD, MHS 2315–2381. https://doi.org/10.1093/eurheartj/ehw106
*John W. McEvoy, MB, BCh, BAO, MHS 4. Hippisley-Cox J, Coupland C, Brindle P. Development and validation of
QRISK3 risk prediction algorithms to estimate future risk of cardiovascular
*National Institute for Prevention and Cardiovascular Health disease: prospective cohort study. BMJ. 2017;357:j2099. https://doi.org/10.
National University of Ireland Galway 1136/bmj.j2099
ISSN 0735-1097/$36.00
Letters
Calcium Score to add the aortic valve calcium score to appropriately

select study patients.
Specify Assessment of
*Gudrun Lamm, MD
Low-Flow Aortic Johann Auer, MD
Stenosis Severity *Department of Internal Medicine 3
Karl Landsteiner University of Health Sciences
University Hospital St. Pölten
We read with great interest the paper by Jean et al Dunant Platz 1
(1) on moderate aortic stenosis (AS) in patients with St. Pölten A-3100, Austria
heart failure and reduced ejection fraction. Some E-mail: gudrun.lamm@stpoelten.lknoe.at
major limitations of this nonrandomized trial have https://doi.org/10.1016/j.jacc.2021.06.047
already been mentioned by the authors. We wish to Ó 2021 by the American College of Cardiology Foundation. Published by Elsevier.
add the challenging issue to correctly separate
The authors have reported that they have no relationships relevant to the con-
moderate AS from true severe low-flow low-gradient tents of this paper to disclose.
AS. It is well known that calculating the aortic valve The authors attest they are in compliance with human studies committees and
animal welfare regulations of the authors’ institutions and Food and Drug
area by the continuity equation is prone to mea- Administration guidelines, including patient consent where appropriate. For
surement errors and invalid results. Although the more information, visit the Author Center.
authors performed stress echocardiography in all

patients, they may have missed some patients with REFERENCES
severe AS (in particular some cases with borderline 1. Jean G, Van Mieghem NM, Gegenava T, et al. Moderate aortic stenosis in
aortic valve area or in patients with limited con- patients with heart failure and reduced ejection fraction. J Am Coll Cardiol.
2021;77(22):2796–2803.
tractile reserve) (2). Therefore, it might have been
2. Lamm G. New strategies in the management of valvular heart disease: a
useful to calculate the aortic valve calcium score on
critical appraisal on the top 10 messages of the 2020 ACC/AHA guidelines for
cardiac computed tomography to add another piece the management of patients with valvular heart disease. Wien Klin
of information and to reduce the risk of misclassi- Wochenschr; 2021 May 31. https://doi.org/10.1007/s00508-021-01879-y
fication of heavily calcified true severe aortic valves 3. Otto CM, Nishimura RA, Bonow RO, et al. 2020 ACC/AHA guideline for
the management of patients with valvular heart disease: a report of the
(aortic valve calcium score >1,300 in women and
American College of Cardiology/American Heart Association Joint Com-
>2,000 in men) as moderate AS (3). Hence, up- mittee on Clinical Practice Guidelines. J Am Coll Cardiol. 2021;77(4):e25–
coming randomized trials on moderate AS should e197.
ISSN 0735-1097/$36.00
REPLY: Calcium Score to Specify Guillaume Jean, MD

Assessment of Low-Flow Aortic Philippe Pibarot, DVM, PhD
Stenosis Severity *Marie-Annick Clavel, DVM, PhD
*Institut Universitaire de Cardiologie et de Pneumologie
We thank Drs Lamm and Auer for their interest in our Université Laval
work (1). We agree that correct classification of aortic 2725 Chemin Ste-Foy #A-2047
stenosis (AS) severity is challenging in patients with Québec G1V4G5, Canada
low ejection fraction (2). However, in our study, we E-mail: marie-annick.clavel@criucpq.ulaval.ca
took particular care to ascertain that the stenosis was https://doi.org/10.1016/j.jacc.2021.07.011
moderate by having concordant moderate grading of
AS at rest or under dobutamine stress echocardiog- Ó 2021 by the American College of Cardiology Foundation. Published by Elsevier.
raphy (DSE). Indeed, to be included in the study, The authors have reported that they have no relationships relevant to the con-
patients had to fulfill the criteria of a mean gradient tents of this paper to disclose.
The authors attest they are in compliance with human studies committees and
(MG) <40 mm Hg with an aortic valve area (AVA) animal welfare regulations of the authors’ institutions and Food and Drug
>1 cm 2. Among our 262 patients, 223 had a moderate Administration guidelines, including patient consent where appropriate. For
more information, visit the Author Center.
AS at rest (ie, rest AVA >1.0 cm 2 and rest
MG <40 mm Hg) and only 39 had a pseudo-severe AS
at rest (ie, rest AVA <1.0 cm 2) with confirmation of REFERENCES
moderate AS using DSE (ie, stress AVA >1.0 cm 2 with 1. Jean G, Van Mieghem NM, Gegenava T, et al. Moderate aortic stenosis in
stress MG <40 mm Hg). Patients with pseudo-severe patients with heart failure and reduced ejection fraction. J Am Coll Cardiol.
2021;77:2796–2803.
AS (ie, discordant grading at rest) in whom DSE was
2. Clavel MA, Burwash IG, Pibarot P. Cardiac imaging for assessing low-
not performed or was inconclusive (lack of contractile
gradient severe aortic stenosis. J Am Coll Cardiol Img. 2017;10:185–202.
reserve, for example) were not included in the study.
3. Clavel MA, Messika-Zeitoun D, Pibarot P, et al. The complex nature of
Aortic valve calcium (AVC) scoring is, as we discordant severe calcified aortic valve disease grading: new insights from
demonstrated (3,4), a robust and flow-independent combined Doppler-echocardiographic and computed tomographic study. J Am
Coll Cardiol. 2013;62:2329–2338.
marker of AS severity. AVC is thus useful to confirm
AS severity in patients with discordance among 4. Clavel MA, Pibarot P, Messika-Zeitoun D, et al. Impact of aortic valve
calcification, as measured by MDCT, on survival in patients with aortic ste-
echocardiographic markers of AS severity (2), and nosis: results of an international registry study. J Am Coll Cardiol. 2014;64:
especially in patients with low ejection fraction and 1202–1213.
inconclusive DSE. Unfortunately, because AVC 5. Otto CM, Nishimura RA, Bonow RO, et al. 2020 ACC/AHA guideline for the
scoring was only introduced in the guidelines recently management of patients with valvular heart disease: executive summary: a
report of the American College of Cardiology/American Heart Association
(5), it was not available in these patients but will joint committee on clinical practice guidelines. J Am Coll Cardiol. 2021;77:450–
undoubtably be included in future trials. 500.

JACC Volume 78, Issue 11 September

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

JACC Volume 78, Issue 11 September

Uploaded by

Copyright:

Available Formats

SEPTEMBER 14, 2021

SEE ADDITIONAL CONTENT ONLINE

SEE ADDITIONAL CONTENT ONLINE

Technology-Assisted Self-Selection of Candidates for Nonprescription Statin Therapy 1114

SEE ADDITIONAL CONTENT ONLINE

Innate Lymphoid Cells Promote Recovery of Ventricular Function After 1127

SEE ADDITIONAL CONTENT ONLINE

JACC STATE-OF-THE- Medial Arterial Calciﬁcation: JACC State-of-the-Art Review 1145

SEE ADDITIONAL CONTENT ONLINE

SEE ADDITIONAL CONTENT ONLINE

LETTERS A Transatlantic Comparison of Patient-Reported Access to and 1193

ONLINE FEATURE Calcium Score to Specify Assessment of Low-Flow Aortic e71

ONLINE FEATURE Reply

THESE ARTICLES ONLY APPEAR IN THE ONLINE VERSION OF THE ISSUE.

ª 2021 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

High-Sensitivity C-Reactive Protein

Listen to this manuscript’s

ISSN 0735-1097/$36.00 https://doi.org/10.1016/j.jacc.2021.07.016

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

enhanced immunonephelometric assay (N High

distributed variables were reported as mean SD.

ﬁltration rate. Sensitivity analysis was performed by

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

F I G U R E 1 Cardiovascular Risk Across Clinical Strata of Lp(a)

A hsCRP <2 mg/L

<50 mg/dL 268/1,960 (13.7) 1.00 (reference) NA

Continued on the next page

B hsCRP <2 mg/L

Cumulative Incidence of CVD (%)

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

F I G U R E 2 Cardiovascular Risk Classiﬁed by Lipoprotein(a) and hsCRP

A Risk Groups Events/Total (%) HR (95% CI) P Value

DISCUSSION A recent post hoc analysis suggested that hsCRP

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

T A B L E 3 HR for Cardiovascular Events by Race

African American Non-African American

F I G U R E 3 Cumulative Incidence of Cardiovascular Events in African Americans and Non-African Americans

Cumulative Incidence of CVD (%)

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

T A B L E 4 Hazard Ratio for Cardiovascular Events by Sex

C E NT R AL IL L U STR AT IO N Cumulative Incidence of Cardiovascular Events in Women and Men

Cumulative Incidence of CVD (%)

Log-Rank P < 0.001

hsCRP Modiﬁes Lp(a)-Associated ASCVD Risk SEPTEMBER 14, 2021:1083–1094

ª 2021 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

Lifetime Risk Estimation in

Xavier Rossello, MD, PHD

D espite signiﬁcant advances in the diagnosis

ISSN 0735-1097/$36.00 https://doi.org/10.1016/j.jacc.2021.07.035

Lifetime Risk Estimation in ASCVD SEPTEMBER 14, 2021:1095–1096

ª 2021 THE AUTHORS. PUBLISHED BY ELSEVIER ON BEHALF OF THE AMERICAN

COLLEGE OF CARDIOLOGY FOUNDATION. THIS IS AN OPEN ACCESS ARTICLE UNDER

THE CC BY LICENSE (http://creativecommons.org/licenses/by/4.0/).

Phenotypic Expression and Outcomes in

ISSN 0735-1097 https://doi.org/10.1016/j.jacc.2021.07.017

Impact of HCM-Associated Variants on the Adult Population SEPTEMBER 14, 2021:1097–1110

WES = whole exome

F I G U R E 1 Cardiac Image Analysis in the UK Biobank

A Cardiac Image Analysis

3-Dimensional cardiac segmentation

Cardiac strain analysis

B Distribution of Maximum Left Ventricular Wall Thickness (n = 21,272)

Impact of HCM-Associated Variants on the Adult Population SEPTEMBER 14, 2021:1097–1110

F I G U R E 2 Flowchart of UKBB Participants