Journal of the American College of Cardiology

2013 by the American College of Cardiology Foundation

Published by Elsevier Inc.

Vol. 62, No. 21, 2013

ISSN 0735-1097/$36.00

CORRESPONDENCE

Research
Correspondence

Clinical Manifestations of Fibromuscular Dysplasia

Vary by Patient Sex
A Report of the United States Registry for Fibromuscular Dysplasia

To the Editor: Fibromuscular dysplasia (FMD) is an uncommon

arteriopathy which can result in stenosis, aneurysm, dissection, and/
or occlusion of arteries. It most commonly affects the renal, extracranial carotid, and vertebral arteries but can affect any artery.
Although FMD occurs primarily in middle-aged women, it can also
affect men and can occur across the life span. We recently reported
ndings of the rst 447 patients enrolled in a multicenter registry of
FMD involving 9 clinical centers in the United States (1). Although
these ndings included demographic information, presenting clinical symptoms, arterial bed involvement, and vascular events, there
may be important sex-related differences in the clinical manifestation
of FMD that have not been previously described. This present study
aimed to determine if there are differences in the clinical characteristics and arterial involvement of FMD between men and women.
The multicenter U.S. Registry for FMD was formed in
2008 with the Michigan Cardiovascular Outcomes Research and
Reporting Program serving as the coordinating center. Specics of
this registry are described in detail elsewhere (1). Since the publication of the rst registry report (1), one additional clinical center
has been added and an additional 168 patients have been enrolled
as of September 11, 2012.
Differences in clinical characteristics by sex were determined
using the Student t test and Fisher exact test with summary
statistics presented as mean
SD and percentages. All statistics
were analyzed by the coordinating center statistician (X.G.) using
SAS (SAS Institute, Inc., Cary, North Carolina).
There were 615 patients included in the analysis, including
52 men (8.5%). Mean age at diagnosis of FMD was 51.9 years
(men 52.1 years vs. women 51.9 years; p 0.94). Prevalence of
hypertension and headaches was high in the overall registry cohort
(73.1% and 63.2%, respectively), but there were no differences by
sex. Although women were signicantly more likely to have a family
history of hypertension or stroke, there were no differences in family
history of aneurysm, dissection, or sudden death (Table 1).
Hypertension and headache were the most common presenting
symptoms leading to the diagnosis of FMD overall and for both sexes,
but women were signicantly more likely to present with other
signs and symptoms of carotid and/or vertebral artery involvement
than men, with higher rates of pulsatile tinnitus (35.7% vs. 9.1%; p
0.0002), cervical bruit (26.8% vs. 4.5%; p 0.0004), and neck
pain (28.6% vs. 13.3%; p 0.034). Men compared with women
were signicantly more likely to present with signs and symptoms of
renal artery involvement including ank and/or abdominal pain
(43.8% vs. 14.3%; p < 0.0001), azotemia/renal insufciency (9.1% vs.
2.2%; p 0.026), and renal infarction (42.9% vs. 4.3%; p 0.0067)
(Table 1).
Although not all patients had every vascular bed evaluated with
imaging studies, differences in arterial bed distribution between men

and women were seen (using conventional angiography, computed

tomography angiography, magnetic resonance angiography, or
duplex ultrasound) (Table 1). Renal and extracranial carotid arteries
were the most commonly affected arterial beds in the overall cohort
(75.3% and 72.7%, respectively), but there were signicant differences in involvement by sex: men were more likely to have renal
involvement (89.7% vs. 74.1% in women; p 0.032), whereas
women were more likely to have extracranial carotid involvement
(74.9% vs. 44.1% in men; p 0.00043). Men and women had similar
average number of arterial beds involved (mean 1.8 beds; p 0.84).
Arterial dissection and aneurysm were common among patients
in the registry, with 21.7% of patients having at least one dissection
and 22.2% of patients with one or more arterial aneurysms. Men
had higher prevalence of both aneurysm (40.8% vs. 20.4% in
women; p 0.002) and dissection (39.6% vs. 20.0% in women;
p 0.0031) (Table 1). The most common sites of dissection for
both sexes were the carotid (15.7%), vertebral (4.2%), and renal
(4.1%) arteries. Renal artery dissections were signicantly more
common in men (18.5% vs. 2.7% in women; p < 0.0001).
We identied important sex-related differences in family history,
clinical presentation, and location of arterial involvement in FMD.
Men represent a minority of patients in the registry (8.5%), but men
with FMD had higher rates of arterial aneurysm and dissection
compared with women. Women with FMD were more likely to
present with signs and symptoms related to cervical artery (carotid
and/or vertebral) involvement, whereas men were more likely to
present with signs and symptoms related to renal artery involvement.
There have been sex-related differences reported in the clinical
manifestations and survival in other arteriopathies associated with
arterial aneurysm and dissection, such as Loeys-Dietz syndrome, an
autosomal dominant genetic disorder caused by mutations in the
transforming growth factor beta receptor 1 or 2 genes (2). Unique
genetic pathways for FMD have not yet been identied, however,
and the prevalence of mutations associated with other arteriopathies, including those in Col3A1, TGFBR1/2, and ACTA2, is
low in FMD (3,4).
As discussed in the rst registry publication (1), there are
several limitations to this study. A particularly pertinent limitation is the potential for surveillance bias because there was
no standardized algorithm for vascular imaging in all patients.
Signicant corresponding differences in clinical presentation
between sexes, however, make a compelling argument that the
imaging ndings are likely to be real. Practice guidelines for the
care of patients with FMD, including recommendations for
imaging to screen for disease in asymptomatic vascular territories, are eagerly anticipated.
In conclusion, there are important sex-related differences in the
family history, presentation, and arterial involvement of FMD.

JACC Vol. 62, No. 21, 2013

November 19/26, 2013:202630

Table 1

2027

Correspondence

History, Clinical Presentation, and Arterial Involvement of Patients With FMD

Age at diagnosis, yrs

All
(N 615)

Men
(n 52)

Women
(n 563)

51.9
13.5

52.1
16.6

51.9
13.1

p Value
0.94

Family history
Hypertension

404/532 (75.9)

27/47 (57.4)

377/485 (77.7)

0.0032

Stroke

227/490 (46.3)

10/39 (25.6)

217/451 (48.1)

0.0072

Aneurysm

105/493 (21.3)

9/42 (21.4)

96/451 (21.3)

Dissection

10/474 (2.1)

1/39 (2.6)

9/435 (2.1)

1.00
0.58

Sudden death

70/472 (14.8)

3/38 (7.9)

676/434 (15.4)

0.34

Presenting sign/symptom
Hypertension

375/563 (66.6)

36/49 (73.5)

339/514 (66.0)

0.34

Headache

305/537 (56.8)

22/47 (46.8)

283/490 (57.8)

0.17

Pulsatile tinnitus

168/503 (33.4)

4/44 (9.1)

164/459 (35.7)

0.0002

Cervical bruit

123/496 (24.8)

2/44 (4.5)

121/452 (26.8)

0.0004

Neck pain

124/492 (27.2)

6/45 (13.3)

128/447 (28.6)

0.034

50/523 (9.6)

2/46 (4.3)

48/477 (10.1)

0.29

Hemispheric TIA
Stroke

42/532 (7.9)

2/47 (4.3)

40/485 (8.3)

0.57

Flank/abdominal pain

85/494 (17.2)

21/48 (43.8)

64/446 (14.3)

<0.0001

Abdominal bruit

53/491 (10.8)

0/43 (0)

53/448 (11.8)

0.009

Renal insufciency*

14/507 (2.8)

4/44 (9.1)

10/463 (2.2)

0.026

Renal infarction
No. of arterial beds involved

7/100 (7.0)
1.8
1.0
1 (12)

3/7 (42.9)

4/93 (4.3)

1.8
1.2
1 (12)

1.8
0.9
1.5 (12)

0.0067
0.84

Arterial bed involved

Renal

382/507 (75.3)

35/39 (89.7)

347/468 (74.1)

0.032

Extracranial carotid

346/476 (72.7)

15/34 (44.1)

331/442 (74.9)

0.00043

Intracranial carotid
Vertebral
Mesenteric

48/281 (17.1)

8/22 (36.4)

40/259 (15.4)

0.033

110/329 (33.4)

6/27 (22.2)

104/302 (34.4)

0.29

63/292 (21.6)

11/32 (34.4)

52/260 (20.0)

0.071

Vascular complication
Any arterial dissection

123/567 (21.7)

19/48 (39.6)

104/519 (20.0)

0.0031

Any arterial aneurysmy

124/559 (22.2)

20/49 (40.8)

104/510 (20.4)

0.002

Values are mean
SD, n/N (%), or median (Q1 to Q3). The denominator of reported responses is reported as different than total cohort size of 615 patients to account for missing data. *Renal insufciency
was dened as a decrease in the glomerular ltration rate or increase in serum creatinine level and determined in accordance with each institutions individual criterion for diagnosis. yAneurysm was
checked yes in the registry if there was the documented presence of any aneurysm. The denition of aneurysm was determined by the site investigator.
FMD bromuscular dysplasia; TIA transient ischemic attack.

Although FMD less commonly affects men, the frequency of

arterial aneurysm and dissection is signicantly higher among men
with FMD. Although FMD can affect any arterial bed in both
sexes, FMD should be particularly suspected in women who
present with headache, neck pain, pulsatile tinnitus, or cervical
bruit and in men who present with renal dissection or infarct or
with arterial aneurysm.

*Esther S. H. Kim, MD, MPHy

Jeffrey W. Olin, DOz
James B. Froehlich, MD, MPHx
Xiaokui Gu, MAx
J. Michael Bacharach, MDk
Bruce H. Gray, DO{
Michael R. Jaff, DO#
Barry T. Katzen, MD**
Eva Kline-Rogers, MS, RN, NPx
Pamela D. Mace, RNyy
Alan H. Matsumoto, MDzz
Robert D. McBane, MDxx
Christopher J. White, MDkk
Heather L. Gornik, MD, MHSy

*Department of Cardiovascular Medicine

Cleveland Clinic
Desk J3-5
9500 Euclid Avenue
Cleveland, Ohio 44195
E-mail: kims@ccf.org
http://dx.doi.org/10.1016/j.jacc.2013.07.038

From the yDepartment of Cardiovascular Medicine, Cleveland

Clinic, Cleveland, Ohio; zMount Sinai Medical Center, New
York, New York; xUniversity of Michigan, Ann Arbor, Michigan;
kNorth Central Heart, Sioux Falls, South Dakota; {Greenville
Hospital System, Greenville, South Carolina; #Massachusetts
General Hospital, Boston, Massachusetts; **Miami Baptist
Cardiac/Vascular Institute, Miami, Florida; yyFibromuscular
Dysplasia Society of America, Rocky River, Ohio; zzUniversity of
Virginia Health System, Charlottesville, Virginia; xxMayo Clinic,
Rochester, Minnesota; and the kkOchsner Medical Center, New
Orleans, Louisiana.
Please note: This work was supported by the Fibromuscular Dysplasia Society of
America, a nonprot organization. Dr. Kim has served as a consultant for Philips
Ultrasound and has received grant support from General Electric. Dr. Froehlich has

2028

JACC Vol. 62, No. 21, 2013

November 19/26, 2013:202630

Correspondence

been a consultant for Pzer; Janssen Pharmaceuticals, Merck, and Boehringer Ingelheim. Drs. Olin and Gornik are unpaid members of the medical advisory board of the
FMD Society of America, a nonprot organization. Dr. Gray has a consultant
agreement without nancial benet with Abbott Vascular and Trivascular and has
received research support without nancial benet with SilkRoad, Medtronic, Abbott,
and WL Gore. Dr. Jaff is a board member of the Fibromuscular Dysplasia Society of
America and VIVA Physicians, a 501 c 3 not-for-prot education and research
organization. Dr. Katzen has served on advisory boards for Boston Scientic, Medtronic, and WL Gore. Ms. Mace is an employee of the FMD Society of America,
a nonprot organization. Dr. Matsumoto has received consultant fees/honoraria from
St. Jude and WL Gore; has received fees for participation in review activities for
Bolton Medical; is a board member of Boston Scientic; has served as an unpaid
consultant for Crux Medical; and has received grant support from Insightec. All other
authors have reported that they have no relationships relevant to the contents of this
paper to disclose.

REFERENCES

1. Olin JW, Froehlich J, Gu X, et al. The United States Registry for

Fibromuscular Dysplasia: results in the rst 447 patients. Circulation
2012;125:318290.
2. Tran-Fadulu V, Pannu H, Kim DH, et al. Analysis of multigenerational
families with thoracic aortic aneurysms and dissections due to TGFBR1
or TGFBR2 mutations. J Med Genet 2009;46:60713.
3. Marks SD, Gullett AM, Brennan E, et al. Renal FMD may not confer
a familial hypertensive risk nor is it caused by ACTA2 mutations.
Pediatr Nephrol 2011;26:185761.
4. Poloskey SL, Kim E, Sanghani R, et al. Low yield of genetic testing for
known vascular connective tissue disorders in patients with bromuscular
dysplasia. Vasc Med 2012;17:3718.

Letters to the Editor

Rapid Rate Nonsustained

Ventricular Tachycardia Found
on Implantable CardioverterDebrillator Interrogation
Relationship of Rapid Rate Nonsustained
Ventricular Tachycardia to Outcomes
in the SCD-HeFT Study
The paper by Chen et al. (1) on the relevance of rapid rate nonsustained ventricular tachycardia (RR-NSVT) noted on routine
debrillator follow-up is intriguing. Figure 3 in the article demonstrated an inappropriate shock for an episode of RR-NSVT that
self-terminated but for which the device was committed to
delivering therapy. A question for the authors is what role did such
shocks play in their nding of increased mortality with these
episodes? They note that nearly 5% of patients received inappropriate therapy. Did the mortality results change signicantly if
patients were stratied by shock appropriateness? This is particularly relevant in light of the dramatic ndings of the MADIT-RIT
(Multicenter Automatic Debrillator Implantation Trial-Reduce
Inappropriate Therapy) trial (2) and could provide further
evidence that a minimalist approach to debrillator therapy is best
for patient outcomes.

*Richard Z. Andraws, MD
*Gagnon Cardiovascular Institute at Atlantic Health and
Central New Jersey Cardiology
1511 Park Avenue
Suite 2
South Plaineld, New Jersey 07080
E-mail: randraws@yahoo.com
http://dx.doi.org/10.1016/j.jacc.2013.06.055
REFERENCES

1. Chen J, Johnson G, Hellkamp AS, et al. Rapid-rate nonsustained

ventricular tachycardia found on implantable cardioverter-debrillator
interrogation: relationship to outcomes in the SCD-HeFT (Sudden
Cardiac Death in Heart Failure Trial). J Am Coll Cardiol 2013;61:21618.
2. Moss AJ, Schuger C, Beck CA, et al., for the MADIT-RIT Trial
Investigators. Reduction in inappropriate therapy and mortality through
ICD programming. N Engl J Med 2012;367:227583.

Reply

Rapid Rate Nonsustained Ventricular

Tachycardia Found on ICD Interrogation:
Relationship of RR-NSVT to Outcomes
in the SCD-HeFT Trial
We appreciate the interest in our paper and the opportunity to reply
to Dr. Andraws most thoughtful comments (1).
It is very difcult to isolate the different facets of the rapid-rate
nonsustained ventricular tachycardia (RR-NSVT) and inappropriate shock-mortality relationships. In fact, it is likely that these
relationships are not entirely independent of one another. Patients
could have a widely varying number of RR-NSVT episodes as well
as multiple inappropriate (and appropriate) shocks, and these
events could have happened in any possible order over a range of
time intervals. Earlier events may have had an impact on later
events. A model that attempts to capture all of the variable
relationships would be impossible to interpret.
A recent study by Powell et al. (2), from the Boson Scientic
sponsored ALTITUDE database demonstrated that inappropriate
implantable cardioverter-debrillator (ICD) shocks from sinus
tachycardia, sustained ventricular tachycardia, noise artifact, or
oversensing did not increase mortality, whereas inappropriate ICD
shocks from NSVT did increase mortality (as well as shocks for
ventricular tachycardia/ventricular brillation/atrial brillation).
This suggests that it may be the substrate and arrhythmias, not the
shocks themselves, that are the primary contributors to the
increased mortality that has been observed in such patients.
In a previous analysis from the SCD-HeFT (Sudden Cardiac
Death in Heart Failure Trial), appropriate and inappropriate shocks
were both found to be independently associated with an increase in
mortality (3). In that analysis, NSVT was included in the category
of inappropriate shocks (example illustrated in Figure 3 of our
paper), where the RR-NSVT terminated, but the ICD delivered
shock therapy due to the conrmation algorithm used in that
generation of Medtronic ICD (model 7223). These data, taken
together, strongly support longer detection times in an effort to
minimize unneeded ICD therapy.