Reviews

PREVENTION OF CVD

Safety and efficacy of statin therapy

Bhavin B. Adhyaru1* and Terry A. Jacobson1,2*
Abstract | The 2013 ACC/AHA guidelines on blood cholesterol management were a major shift
in the delineation of the main patient groups that could benefit from statin therapy and emphasized
the use of higher-intensity statin therapies. In 2016, an expert consensus panel from the ACC
recommended the use of nonstatin therapies (ezetimibe and PCSK9 inhibitors) in addition to
maximally tolerated statin therapy in individuals whose LDL-cholesterol and non-HDL-cholesterol
levels remained above certain thresholds after statin treatment. Given the substantial benefits
of statin therapies in both primary and secondary prevention of cardiovascular disease, their
long-term safety has become a concern. The potential harmful effects of statin therapy on muscle
and liver have been known for some time, but new concerns have emerged regarding the risk of
new-onset diabetes mellitus, cognitive impairment and haemorrhagic stroke associated with the
use of statins and the risks of achieving very low levels of LDL cholesterol. The increased media
attention on the adverse events associated with statins has unfortunately led to statin therapy
discontinuation, nonadherence to therapy or concerns about initiating statin therapy. In this
Review , we explore the safety of statin therapy in light of the latest evidence and provide clinicians
with reassurance about the safety of statins. Overwhelming evidence suggests that the benefits
of statin therapy far outweigh any real or perceived risks.

Statin-associated muscle
symptoms
(SAMS). Muscle symptoms that
are the most prevalent adverse
effect reported with statin
therapy.
1
Division of General Internal
Medicine & Geriatrics,
Department of Medicine,
Emory University School of
Medicine, Atlanta, GA, USA.
2
Lipid Clinic and
Cardiovascular Risk
Reduction Program,
Department of Medicine,
Emory University School of
Medicine, Atlanta, GA, USA.
*e-mail: badhyar@emory.edu;
tjaco02@emory.edu
https://doi.org/10.1038/
s41569-018-0098-5
Strong evidence indicates that statins reduce the risk
of atherosclerotic cardiovascular disease (ASCVD) in
primary and secondary prevention1. A meta-analysis
from the Cholesterol Treatment Trialists (CTT) sug-
gested that each 1 mmol/l reduction in LDL-cholesterol
(LDL-C) levels in the plasma (approximately 39 mg/dl)
reduces the risk of ASCVD events by 22% after 5 years2,3.
However, data from surveys, registries and insurance
claims suggest that adverse effects of statins are com-
mon, which discourages patients from continuing statin
therapy at recommended doses4.
In this Review, we briefly discuss the patient popula-
tions that benefit from statin therapy, as well as the updated
recommendations on nonstatin therapy and the use of
LDL-C thresholds to reduce the risk of ASCVD. We dis-
cuss in detail the evidence base for several adverse effects
of statins, including  statin-associated muscle symptoms
(SAMS), and the risks of liver disease, diabetes melli-
tus, cognitive impairment and haemorrhagic stroke.
Given the high prevalence in clinical practice of mus-
cle symptoms associated with statin therapy, we further
explore statin intolerance and examine the risks associated
with achieving very low LDL-C levels.
Statin efficacy
In 2013, the ACC/AHA guidelines for the management
of blood cholesterol levels proposed a paradigm shift in
the management and treatment of lipid disorders on the
basis of a patient’s baseline risk of ASCVD and the inten-
sity of statin therapy required for risk reduction1. Statins
were recommended as first-line therapy for ASCVD risk
reduction on the basis of high-quality data, given the
consistent reductions in ASCVD events observed with
the use of statins in primary and secondary preven-
tion of ASCVD. Four major statin benefit groups were
identified: patients with clinical ASCVD (such as acute
coronary syndromes, history of myocardial infarction
(MI), stable or unstable angina, stroke, transient ischae-
mic attack (TIA) or peripheral artery disease); patients
without ASCVD or diabetes mellitus with primary eleva-
tion of LDL-C levels >190 mg/dl (>4.9 mmol/l); patients
with diabetes (type 1 or type 2) aged 40–75 years with
LDL-C levels of 70–189 mg/dl (1.8–4.9 mmol/l) and
without clinical ASCVD; and patients without diabe-
tes or clinical ASCVD and an estimated 10-year risk of
ASCVD risk ≥7.5%.
According to the 2013 ACC/AHA blood cholesterol
management guidelines, the recommended intensity of
statin therapy is based on both the risk of ASCVD and
the required percentage reduction in LDL-C levels to
reduce the risk of ASCVD1. The intensity of statin ther-
apy is determined by the predicted percentage reduction
in LDL-C levels: a reduction in LDL-C levels of >50%
is classified as high-intensity statin therapy (for exam-
ple, 40–80 mg of atorvastatin and 20–40 mg of rosuva­
statin daily), whereas reductions of 30–50% or <30% in

NATuRe RevIewS | CARDIOLOgy volume 15 | DECEMBER 2018 | 757

Reviews
Key points
• Moderate-intensity and high-intensity statin therapy has demonstrated benefits
in reducing the risk of atherosclerotic cardiovascular disease in both primary and
secondary prevention.
• The most commonly reported adverse effects of statins are statin-associated muscle
symptoms (SAMS), but whether these symptoms are caused by statin pharmacological
effects or by nocebo effects is controversial.
• The clinical tool SAMS-Clinical Index (SAMS-CI) might help clinicians to assess SAMS.
• Although hepatotoxicity has been a previous concern when using statin therapy,
data suggest that statins are safe and that levels of liver enzymes do not need to be
checked routinely.
• Other known adverse effects of statins include a small increase in the risk of diabetes
mellitus and a possible increase in haemorrhagic stroke in patients who have had a
previous stroke.
• Although concerns have been raised about reaching very low levels of LDL
cholesterol, data from trials on PCSK9 inhibitors suggest that low LDL-cholesterol
levels are safe.
LDL-C levels are classified as moderate-intensity and
low-intensity statin therapy, respectively. The calcula-
tion of the risk of ASCVD is based on a pooled cohort
equation derived from several patient populations in the
USA. Considerable controversy surrounded the 2013
ACC/AHA cholesterol guidelines with regard to the lack
of specific LDL-C goals, the use of nonstatin therapy and
the treatment of special patient populations. To address
these concerns, in 2016 and 2017, an expert consensus
panel from the ACC updated the recommendations on
the use of nonstatin therapies on the basis of the results
from the IMPROVE-IT5 and FOURIER6 studies and
further defined the importance of achieving LDL-C
thresholds or goals based on specific risk groups7–9.
The major statin benefit groups discussed above are
derived from multiple randomized, controlled trials
(RCTs) and meta-analyses of these RCTs. The landmark
2010 CTT meta-analyses examined individual-level
patient data from 26 RCTs on statin therapy that
included 170,000 individuals at various risk levels of
major vascular events, including major coronary events
(such as nonfatal MI and coronary heart disease (CHD)-
related death), stroke and revascularization procedures3.
One part of the meta-analysis included 129,526 patients
from 21 RCTs that compared statin versus control (pla-
cebo or usual care) in patients with a mean LDL-C level
of 3.7 mmol/l (143 mg/dl) over a median follow-up of
5 years. The other part of the meta-analysis compared
data on more-intensive versus less-intensive statin
therapy from five RCTs that included 39,612 patients.
The more-intensive statin regimen groups were pre-
scribed a daily dose of 80 mg of either atorvastatin
(PROVE-IT 10, TNT 11 and IDEAL 12) or simvastatin
(A to Z13 and SEARCH14). Overall, in the 26 trials, a
Statin intolerance reduction of 1 mmol/l (approximately 39 mg/dl) in
A clinical syndrome
LDL-C levels over 5 years reduced major vascular events
characterized by the inability
to tolerate multiple statins with by 22% (95% CI 20–24%, P < 0.0001), CHD death by 20%
either objectionable symptoms and all-cause mortality by 10% (Table 1). In the five trials
or abnormal laboratory of more-intensive versus less-intensive statin therapy,
determinations that are an additional reduction of 0.5 mmol/l in LDL-C levels
temporally related to statin
treatment and reversible upon
yielded a 15% further proportional reduction in the
statin discontinuation but rate of major vascular events, corresponding to a 28%
reproducible by rechallenge. risk reduction per 1 mmol/l reduction in LDL-C levels.
758 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio
In addition, analysis by type of event revealed that
coronary-related deaths or nonfatal MI were signifi-
cantly reduced by 13%, coronary revascularizations by
19% and ischaemic stroke by 16% in the more-intensive
statin group compared with the less-intensive statin
group. These results are very similar to the proportional
risk reduction in major vascular events per 1 mmol/l
reduction in LDL-C levels seen in the trials on statin ver-
sus control (22%), which was also observed in multiple
subgroups including secondary and primary prevention
cohorts and individuals with diabetes, hypertension, low
HDL-cholesterol (HDL-C) levels or advanced age. Data
from the 26 RCTs showed that patients aged >75 years
had less benefit in reduction of major vascular events
(rate ratio (RR) 0.84, 95% CI 0.73−0.97) with statin ther-
apy than younger patients, although the absolute num-
ber of patients in this subgroup was small3. However,
given that the absolute risks of major vascular events are
higher among old individuals, the absolute benefits of
statin therapy in these patients might be similar to those
in younger individuals.
The findings from the CTT meta-analysis suggest that
further lowering of LDL-C levels in high-risk patients
provides additional benefits and that the patient’s
risk of ASCVD should be assessed to decide whether
to initiate statin therapy. Given that the size of the risk
reduction is directly proportional to the absolute LDL-C
level reduction achieved, large LDL-C level reductions
of 2–3 mmol/l would be expected to yield risk reduc-
tions of 40–50%. Although LDL-C reductions induced
by statin therapy are independent of baseline LDL-C
levels, statins with greater LDL-C-lowering efficacy
such as the high-intensity statins (atorvastatin 40–80 mg
or rosuvastatin 20–40 mg) are often required for LDL
reductions of >50%. Of note, each additional doubling
of the dose of any statin will produce additional reduc-
tions in LDL-C levels of only 6%, which indicates that
the LDL-C-lowering efficacy of statins has limits, par-
ticularly with moderate-intensity statins (fluvastatin,
lovastatin, pitavastatin, pravastatin and simvastatin)4.
In a subsequent CTT meta-analysis published
in 2012, the benefits of statin therapy were evaluated in
patients at very low risk of vascular disease15. A pooled
analysis of a primary prevention patient cohort with
an ASCVD risk score of 5–10% showed a 31% relative
risk reduction in major vascular events over 5 years
in patients receiving statins compared with patients
receiving placebo and an 11% relative risk reduction
in all-cause death. Even patients with an ASCVD risk
score <5% had a 38% relative risk reduction in major
vascular events, but they showed no reductions in all-
cause mortality15. Of note, the role of statins in younger
patients (aged <40 years) and in elderly individuals
(aged >80 years) is unclear given the lack of high-quality
data. Some evidence from the CTT study indicates that
patients aged <75 years might benefit from statin therapy
even if they are in the lowest ASCVD risk group.
Statin therapy has been studied in particular patient
groups, and the outcomes have varied. For example, many
high-risk patients with heart failure (HF) were excluded
from previous RCTs. Two RCTs, the CORONA16 and
GISSI-HF 17 studies, were designed to examine the

Table 1 | Meta-analysis of 26 trials on statin therapy in 170,000 patients3
Outcome
Coronary events
Nonfatal MI
CHD death
Any major coronary event
Revascularization procedures
CABG surgery
PTCA
Unspecified
Any coronary
revascularization
Stroke
Ischaemic stroke
Haemorrhagic stroke
Unknown stroke
Any stroke
Vascular events
Any major vascular event
CHD, coronary heart disease; MI, myocardial infarction; PTCA , percutaneous transluminal coronary
angioplasty ; RR , rate ratio. aRR per 1 mmol/l (39 mg/dl) reduction in LDL-cholesterol levels.
Creatine kinase
An enzyme that is found in
skeletal muscle, the heart, and
the brain, and increases in the
blood due to muscle injury,
inflammation or necrosis of
skeletal or heart muscle.
Myalgia
A condition involving muscle
aches, stiffness, soreness,
tenderness or cramps.
Rhabdomyolysis
A rare, life-threatening
condition, characterized by the
rapid destruction of skeletal
muscle typically with creatine
kinase levels >10 times the
upper limit of normal and
which often leads to acute
renal failure.
NATuRe RevIewS | CARDIOLOgy
Reviews
effect of statins was not significant in patients receiving
haemodialysis (HR 0.95, 95% CI 0.78–1.15). The AURORA
Number of events (% per year) RR (95% CI)a trial 19, which compared patients with end-stage
High-intensity Moderate-intensity renal disease receiving haemodialysis randomly assigned
statin statin to receive 10 mg rosuvastatin or placebo, showed no sig-
nificant benefit of statin therapy on the combined primary
3,485 (1.0) 4,593 (1.3) 0.73 (0.69–0.78) end point of death from cardiovascular causes, non­
fatal MI or nonfatal stroke (HR 0.96, 95% CI 0.84–1.11,
1,887 (0.5) 2,281 (0.6) 0.80 (0.74-0.87)
P = 0.59). The 4D trial20 to evaluate the effect of ator-
5,105 (1.4) 6,512 (1.9) 0.76 (0.73–0.78) vastatin (20 mg daily) versus placebo in 1,255 patients
with type 2 diabetes who were receiving haemodialysis
1,453 (0.4) 1,857 (0.5) 0.75 (0.69–0.82) also showed little evidence of benefit (RR 0.92, 95%
CI 0.77–1.10, P = 0.37). In conclusion, the role of statins
1,767 (0.5) 2,283 (0.7) 0.72 (0.65–0.80)
in patients receiving haemodialysis is unclear.
2,133 (0.6) 2,667 (0.8) 0.76 (0.70–0.82)
5,353 (1.5) 6,807 (2.0) 0.75 (0.72-0.78) Statin safety
Statin-associated muscle symptoms
Definition. SAMS are the most commonly reported
adverse effects of statins (present in 10–29% of patients
1,427 (0.4) 1,751 (0.5) 0.79 (0.72–0.87)
taking statin therapy according to observational
257 (0.1) 220 (0.1) 1.12 (0.88–1.43) studies)21,22. This observation is in sharp contrast with
618 (0.2) 709 (0.2) 0.88 (0.76–1.01) the results of RCTs, which often report minimal dif-
2,302 (0.6) 2,680 (0.8) 0.84 (0.79–0.89) ferences in the rates of muscle symptoms between
statin-treated and placebo-treated groups 4. These
adverse events are a very common reason for stopping
10,973 (3.2) 13,350 (4.0) 0.78 (0.76–0.80) statin therapy despite the well-known cardiovascular
benefits of the therapy22,23. Varying definitions and terms
are used to define the muscle-related symptoms of stat-
ins and often vary by guidelines24. Of note, SAMS often
effects of statins in patients with HF. The CORONA occur in the absence of an elevation in creatine kinase
study 16 included 5,011 patients (aged ≥60 years) plasma levels, and the symptoms can vary from myalgia
with systolic HF (New York Heart Association (NYHA) to rhabdomyolysis. The term SAMS has become the pre-
class II–IV) of ischaemic origin, with a mean baseline ferred term for muscle symptoms because it does not
LDL-C level of 137 mg/dl and mean ejection fraction necessarily imply causality.
(EF) of 27%, who were randomly assigned to receive Myalgias, cramps and weakness can be categorized
a daily dose of 10 mg rosuvastatin or placebo. No signi­ as SAMS. Symptoms are usually bilateral and affect large
ficant effects were observed with statin therapy for the muscle groups (thighs, calves, hip flexors or proximal
composite outcome of cardiovascular-related death, non- upper extremities). Symptoms often appear shortly
fatal MI or stroke (HR 0.92, 95% CI 0.83–1.02, P = 0.12). after starting statin therapy or after increasing the dose
The GISSI-HF study17 included a broader range of and generally resolve quickly after cessation of statin
patients in NYHA class II–IV, irrespective of EF or aetio­ therapy. If the symptoms persist for ≥2 months after
logy (n = 4,600), who were randomly assigned to receive drug cessation, other causes should be explored24.
rosuvastatin (10 mg daily) or placebo. The average age of Health-care providers should realize that although
the patients was 68 years and the study included younger the incidence of SAMS in patients taking statins ranges
patients than those in the CORONA study16 (mean age from 10% to 29%, the occurrence of myopathy with
73 years), and 10% of the patients had an EF >40%. The plasma creatine kinase elevation >10 times the upper
study revealed no significant difference in the mortality limit of normal (ULN) and rhabdomyolysis is very rare24.
or rate of hospitalization associated with cardiovascu- In patients taking statins, the incidence of myopathy
lar events between the rosuvastatin and placebo groups with plasma creatine kinase changes >10 times ULN
(HR 1.00, 95% CI 0.898–1.122, P = 0.943). In summary, is estimated to be approximately 1 in 1,000, whereas
these two studies did not show a clear benefit of statin the incidence of rhabdomyolysis is approximately
therapy in patients with HF with reduced EF. 1 in 10,000 (ref.24). Myopathy and rhabdomyolysis have
The role of statin therapy in patients receiving haemo­ been previously reviewed in the SAMS recommenda-
dialysis is also unclear. Three major RCTs, SHARP18, tions of the National Lipid Association (NLA) Safety
AURORA19 and 4D20, have evaluated statins in patients Task Force24, as well as in the recommendations of the
with chronic kidney disease. The SHARP trial18, which Canadian25 and European26 Working Groups.
included patients with chronic kidney disease (33%
of patients were receiving haemodialysis) randomly Evaluation. An initial step in the evaluation of SAMS is
assigned to receive 20 mg simvastatin plus 10 mg ezeti­ to assess plasma creatine kinase levels to rule out more
mibe daily or placebo, showed a 17% risk reduction in substantial muscle involvement, as seen in myopathy or
the end point of ASCVD events with statin therapy com- rhabdomyolysis. However, most SAMS are not accom-
pared with placebo (HR 0.83, 95% CI 0.74–0.94), but the panied by creatine kinase elevations, which makes the
volume 15 | DECEMBER 2018 | 759
Reviews
diagnosis more dependent on clinical characteristics,
including symptom location and distribution, temporary
onset upon initiation or rechallenge, and resolution of
symptoms upon dechallenge.
The next step in evaluating SAMS is statin discon-
tinuation and rechallenge with either a lower dose of
the same statin or the use of another statin. A washout
period of several weeks might be helpful before rechal-
lenge. Another strategy to consider after failing a third
statin at its lowest dose is to consider alternate-day dos-
ing of longer-acting statins, such as rosuvastatin or ator-
vastatin. A meta-analysis of 13 RCT studies showed that
alternate-day dosing can be as efficacious as daily dos-
ing in lowering LDL-C levels, but its effect on ASCVD
outcomes remains unclear27.
In 2014, the NLA proposed a clinical myalgia scor-
ing system, the Statin Myalgia Clinical Index (SMCI),
which might be useful for clinicians in determining the
probability that muscle symptoms are statin-related24.
The SMCI was based on several observational stud-
ies, including the PRIMO 21 and STOMP 28 studies.
The PRIMO trial21 included data from questionnaires
completed by 7,924 French patients treated with flu-
vastatin (80 mg daily), atorvastatin (40–80 mg daily),
pravastatin (40 mg daily) or simvastatin (40–80 mg
daily) for ≥3 months. Muscle symptoms were reported
by 10.5% of the patients. The STOMP study28 included
420 patients who were randomly assigned to receive
80 mg atorvastatin or placebo daily for 6 months.
The term myalgia was applied to patients who met all the
following criteria: new muscle symptoms that were not
associated with exercise, symptom duration >2 weeks,
symptom resolution within 2 weeks of statin cessation
and symptom recurrence within 4 weeks of drug rechal-
lenge. The results of the STOMP trial28 were used to cre-
ate a clinical profile of patients with true statin-related
myalgia. For example, patients with true statin-related
myalgia were more likely to have bilateral muscle symp-
toms in the calf and thigh muscles than patients in the
placebo group, who had more generalized or nonspecific
muscle symptoms. Muscle symptoms seemed to occur
faster among patients with true myalgia after statin
initiation than in the placebo group (typically 35 days
versus 61 days)28.
In 2017, the NLA SMCI was revised and renamed
the Statin-Associated Muscle Symptom Clinical Index
(SAMS-CI)29 (Box 1). This change was made to better
reflect the full breadth of SAMS in addition to myal-
gia and to prepare the instrument for clinical use.
In addition, the inter-rater reliability of the SAMS-CI
was assessed to increase its clinical utility29. The inves-
tigators conducted constructed in-depth interviews
with the original authors of the index, asked inde-
pendent physicians to develop inter-rater reliability
and sought feedback from external physicians on the
clinical use of the SAMS-CI in practice. This quali-
tative study revealed an inter-rater reliability of 0.77
(95% CI 0.66–0.85), which indicates a fairly high con-
cordance between raters. According to these findings,
the SAMS-CI might be an improved way to detect
SAMS and to optimize treatment for patients with
these symptoms29.
760 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio
In addition, the SAMS-CI was assessed retrospec-
tively in a validation study published in 2017 that used
data from a randomized, double-blind, crossover trial
in patients with SAMS who received simvastatin and
coenzyme Q10 or placebo30,31. Using a slightly revised
SAMS-CI scoring system, a low SAMS-CI score
(<4 instead of <6) had a negative predictive value of
91% in identifying patients who were unlikely to have
true myalgia 31. Although patients with confirmed
SAMS who were treated with simvastatin had higher
mean scores (6.7) than patients with muscle symptoms
who received placebo (2.2), the index did not differ-
entiate those individuals who had muscle symptoms
when receiving either simvastatin or placebo. These
results suggest that the SAMS-CI is a more efficient
tool for identifying patients who are unlikely to have
true SAMS than for identifying patients with true
SAMS. This tool might also be helpful for clinicians to
motivate patients to adhere to statin therapy even in the
presence of muscle symptoms. The SAMS-CI and scor-
ing system need to be further validated prospectively
in other studies30.
Causes. Several factors can increase the risk of SAMS,
including high doses and increased serum concentra-
tion of statins, the use of statin-interacting drugs that
inhibit statin catabolism, hypothyroidism, reduced
muscle mass and increased physical activity 21,31–33.
In addition, advanced age, female sex, physical frailty
and alcohol use can increase the risk of SAMS33. Statins
are metabolized by the cytochrome P450 pathway, which
transforms lipophilic compounds into hydrophobic
compounds in the liver for excretion34; the metabolism
of various statins is detailed in Table 2. Atorvastatin,
lovastatin and simvastatin are metabolized primarily
by cytochrome P450 3A4 (CYP3A4). Medications that
are also metabolized by CYP3A4, including azole anti-
fungals, macrolide or mycin antibiotics, protease inhib-
itors, Ca2+ channel blockers and warfarin, can increase
serum concentrations of statins34. Other statins, such
as fluvastatin, pitavastatin and rosuvastatin, are min­
imally metabolized by CYP2C9 and have fewer statin–
drug interactions. Pravastatin is not metabolized by the
cytochrome P450 system. Gemfibrozil can also interact
with statin metabolism and significantly increase the
risk of SAMS34,35. The statins with the longest half-lives
(atorvastatin and rosuvastatin) can reduce LDL-C
levels even when taken on a nondaily basis, making
these statins particularly attractive for individuals with
statin intolerance.
Given that the prevalence of SAMS in observational
studies is much higher (10–29%) than the prevalence
reported in RCTs (1–2%), some investigators have pos-
tulated that SAMS could be a result of a nocebo effect.
The nocebo effect is observed when subjective adverse
events are more likely to be attributed to a treatment
that is believed to cause harm. The investigators of
the ASCOT-LLA study36,37 examined various adverse
events of statins in the blinded and unblinded phases
of the trial. The first phase of the study was a blinded
phase that included 10,180 patients randomly assigned
to receive atorvastatin (10 mg daily) or placebo. After
 Reviews

the primary end point of combined fatal CHD or non-
fatal MI was reached, the study was then extended to an
unblinded phase in which all patients were offered sta-
tin therapy. The investigators found that in the blinded
phase, the annual rate of muscle-related events was sim-
ilar in the placebo (2.00%) and statin (2.03%) groups. In
the unblinded phase, the annual rate of muscle-related
symptoms was 1.00% in the group not taking stat-
ins compared with 1.26% in the group receiving statins
(HR 1.41, 95% CI 1.10–1.79, P = 0.006). These results
suggest the possibility of a nocebo effect with respect
to increased muscle-related symptoms in patients who
agreed to take statins. However, the interpretation of the
results is controversial because the study was a post hoc
safety analysis, the design was observational and not an
RCT, and individuals with muscle symptoms taking stat-
ins during the blinded phase were less likely to agree to
take statins in the nonblinded phase38. Finally, the obser-
vation that the annual rate of muscle symptoms was sig-
nificantly lower in the unblinded phase (1.00–1.26%)
than in the blinded phase (2.00–2.03%) also casts doubt
on the possibility of a nocebo effect38.
The GAUSS-3 study39, which used a blinded rechal-
lenge approach, is the largest study to date to assess the
ability of patients with a history of SAMS to tolerate sta-
tin therapy. In the initial phase of the study, 491 patients
with SAMS received both 20 mg atorvastatin and pla-
cebo using a crossover design of 10-week duration per

Box 1 | The statin-associated muscle symptom clinical index29

Overview
The SAMS-Clinical Index (SAMS-CI) was designed to help
clinicians determine the likelihood that a patient’s muscle
symptoms (myalgia or myopathy) were caused by or
associated with statin use24.
Questionnaire domains
The SAMS-CI evaluates muscle symptoms using several
domains that encompass the statin regimen, including
location and pattern of muscle symptoms, timing of
muscle symptom onset in relation to starting
the statin regimen, timing of muscle symptom
improvement after withdrawal of the statin, and timing
of muscle symptom recurrence after rechallenge with
a different statin regimen. The score is calculated on
the basis of the patterns in the different domains
(see the figure).
Instructions
• Use for patients who have had muscle symptoms that
were new or increased after starting a statin regimen.
• A statin regimen includes any statin at any dose or
frequency.
• Muscle symptoms might include aches, cramps,
heaviness, discomfort, weakness, or stiffness.
• Interpret the score in light of other possible causes of the
muscle symptoms including recent physical exertion,
changes in exercise patterns, hypothyroidism, drug
interaction with statins, concurrent illness, and underlying
muscle disease.
Interpretation (score: likelihood of SAMS)
• 2–6: unlikely
• 7–8: possible
• 9–11: probable

If one regimen of statin involved If ≥2 regimens of statin involved

Questions regarding the regimen before
Questions regarding this regimen the most recent regimen
Location and pattern of muscle symptoms Score Location and pattern of muscle symptoms Score
Symmetric, hip flexors or thighs 3 Symmetric, hip flexors or thighs 3
Symmetric, calves 2 Symmetric, calves 2
Symmetric, proximal upper extremity 2 Symmetric, proximal upper extremity 2
Asymmetric, intermittent or not 1 Asymmetric, intermittent or not 1
specific to an area specific to an area

Timing of muscle symptom onset Timing of muscle symptom onset

<4 weeks 3 <4 weeks 3
4–12 weeks 2 4–12 weeks 2
>12 weeks 1 >12 weeks 1

Timing of symptom improvement after Timing of symptom improvement after

statin withdrawal statin withdrawal
<2 weeks 2 <2 weeks 2
2–4 weeks 1 2–4 weeks 1
No improvements after 4 weeks 0 No improvements after 4 weeks 0

Rechallenge with a statin regimen Questions regarding the most recent regimen
Timing of recurrence of similar muscle Timing of recurrence of similar muscle
symptoms after starting second regimen symptoms after starting second regimen
<4 weeks 3 <4 weeks 3
4–12 weeks 1 4–12 weeks 1
>12 weeks or symptoms did not reoccur 0 >12 weeks or symptoms did not reoccur 0

Total: Total:
All four scores above must be All four scores above must be
entered before totalling entered before totalling

Figure adapted from ref.29, Springer Nature Limited.

NATuRe RevIewS | CARDIOLOgy volume 15 | DECEMBER 2018 | 761

Reviews

Table 2 | Pharmacokinetic properties of statins

Statin Bioavailability, % Lipophilicity Protein Metabolism Metabolites t1/2, h Urinary Faecal
binding, % excretion, % excretion, %
Atorvastatin 12 Yes 80–90 CYP3A4; hydroxylation, Active 15–30 2 70
oxidation
Fluvastatin 19–29 Yes >99 CYP2C9 (CYP3A4/2C8) Inactive 0.5–2.3 NR 90
(minor)
Lovastatin 5 Yes >95 CYP3A4 Active 2.9 10 83
Pitavastatin >60 Yes 99 CYP2C9/2C8 (minor); Inactive 12 15 79
glucuronidation
Pravastatin 18 No 55 Non-CYP; sulfation, Inactive 1.3–2.8 20 71
hydroxylation, oxidation
Rosuvastatin 20 No 88 CYP2C9/2C19 (minor); Active 19 10 90
biliary excretion (minor)
Simvastatin 5 Yes 94–98 CYP3A4 Active 2–3 13 58
CYP, cytochrome P450; NR , not reported; t1/2, drug half-life. Adapted with permission from ref.82, Elsevier.

Statin-induced necrotizing
autoimmune myopathy
(SINAM). A rare, immune-
mediated myopathy
characterized by proximal
muscle weakness, muscle
necrosis with markedly elevated
creatine kinase levels, and the
presence of autoantibodies to
hydroxyl-methyl-glutaryl-
coenzyme A reductase, in which
muscle symptoms persist
despite statin discontinuation.
treatment. During rechallenge, 42.6% of the patients
reported intolerable muscle symptoms with atorvasta-
tin but not with placebo, whereas 26.5% of the patients
reported muscle symptoms with placebo but not with
atorvastatin. Therefore, the majority of patients accurately
reported statin-related symptoms. However, the modest
difference in the incidence of muscle symptoms between
the two groups suggests that muscle symptoms are
common in many patients and are not always related to
statin use.
The mechanisms of SAMS are unclear, but several
hypotheses have been proposed26, including a possi-
ble alteration in mitochondrial function and cellular
energy utilization related to the depletion of coenzyme
Q10, which would lead to ATP depletion with aug-
mentation of oxidative stress by the increased prod­
uction of reactive oxygen species. Other potential
mechanisms include decreased isoprenoid biosyn-
thesis as a result of decreased mevalonate production
and induction of mitochondria-dependent apoptosis.
Individuals could also have a genetic susceptibility to
SAMS26. Most SAMS resolve after cessation of statin
therapy, with the exception of the very rare and idio-
syncratic statin-induced necrotizing autoimmune myopathy
(SINAM). SAMS might be explained by both a pharma-
cological effect of statins and a nocebo effect. Therefore,
health-care providers should work closely with patients
to continue statin therapy, even at a lower dose or inten-
sity or on a nondaily basis. In the hands of expert clini-
cal lipidologists, 70–80% of patients deemed to be statin
intolerant have been able to tolerate a certain dose of
statins on a daily or nondaily basis40. Useful recommen-
dations from health-care providers include switching to
a different statin, using a lower or nondaily dose and
being able to communicate more effectively about the
benefits and risks of statin therapy41.
Statin intolerance
Although statins are considered to be one of the most
effective strategies to reduce cardiovascular morbidity
and mortality, many patients are unable to tolerate sta-
tin therapy. Therefore, the management of patients with
statin intolerance remains a major clinical challenge.
The definition of statin intolerance varies between
guidelines but generally includes the inability to tolerate
at least two different statins, with one statin assessed at
its lowest effective dose24–26,42,43.
In 2014, the NLA defined statin intolerance as a clini-
cal syndrome characterized by the inability to tolerate at
least two statins, with one statin tested at the lowest start-
ing daily dose and another statin assessed at any daily
dose, that manifests itself by either objectionable symp-
toms (real or perceived) or abnormal laboratory results
temporally related to statin treatment, reversible upon
statin discontinuation and reproducible by statin rechal-
lenge43. The assessment was to be made after excluding
other known determinants of statin-associated symp-
toms or laboratory abnormalities, such as hypothyroid-
ism, known liver or renal disease, concurrent illnesses,
use of interacting drugs, substantial changes in physical
activity or exercise, or underlying muscle disease. The
hallmark of statin intolerance between the different
guidelines is generally the patient’s response to both sta-
tin dechallenge and rechallenge. Data suggest that statin
intolerance leads to increased health-care costs and to a
higher risk of nonfatal cardiovascular events44. The most
common cause of statin intolerance is SAMS.
Muscle complaints during statin treatment are
common, and convincing patients that their muscle
symptoms might be due to causes other than the statin
treatment is sometimes difficult. Some patients might
experience the nocebo effect, and clinicians should not
prematurely abandon statin therapy before investigating
other possible causes, including the nocebo effect.
If patients continue to have statin intolerance, non-
statin therapy could be considered, especially in light
of the 2016 ACC consensus guidelines8. Studies have
shown that patients with statin intolerance can tolerate
ezetimibe45. In addition, data from the IMPROVE-IT
trial5 showed that the combination of ezetimibe and
simvastatin did not increase the incidence of myopathy,
rhabdomyolysis or myalgia with creatine kinase eleva-
tions >5 times ULN compared with simvastatin alone5.
The GAUSS-3 (ref.39) and ODYSSEY ALTERNATIVE46
studies demonstrated that the PCSK9 inhibitors evolo-
cumab and alirocumab, respectively, can be used safely

762 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio

 Reviews

in patients with statin intolerance without increases in
adverse muscle-related events39,46,47.
Risk of diabetes mellitus
The design of early, large trials on statin therapy did not
include an assessment of newly diagnosed diabetes as a
clinical end point because this condition was not consid-
ered clinically relevant. In 2008, the landmark JUPITER
study48 provided evidence that rosuvastatin (20 mg daily)
significantly increased the relative risk of newly diag-
nosed diabetes by 25% compared with placebo. Previous
studies had not prospectively examined changes in phys­
icians’ diagnosis of diabetes or changes in glycaemic
control, which prompted a meta-analysis of large sta-
tin RCTs in 2010 (ref.49) (Table 3). The pooled analysis
showed a 9% increase in the incidence of new-onset
diabetes with statin therapy, with little heterogeneity
in incidence. Overall, the increase in absolute risk was
very small, with a number needed to harm (NNH) of
255 over 4 years49. A subsequent analysis of the JUPITER
study48 suggested that the cardiovascular and mortality
benefits exceeded the risk of diabetes even in patients at
high risk of developing diabetes50.
The SPARCL trial51, which was not included in the
2010 meta-analysis that evaluated new-onset diabetes,
showed a 44% relative increase in new-onset diabetes in
patients with prior stroke or TIA who were treated
with atorvastatin (80 mg daily) compared with patients
treated with placebo, which suggested dose-dependent
effects. Five studies, the TNT11, IDEAL12, PROVE-IT
TIMI10, A to Z13 and SEARCH14 studies, demonstrated
that more-intensive statin therapy was associated
with a significantly higher risk of new-onset diabe-
tes (12% increase in pooled risk) than less-intensive
statin therapy and with a 16% reduction in the inci-
dent risk of cardiovascular disease10–14,52. Again, the
benefit of statin therapy exceeded the risk of new-onset
diabetes (Table 4).
Additionally, statin therapy did not seem to have a
large effect on measures of glycaemic control, and the
small weight gain that could be associated with statin
therapy might not be clinically relevant. The benefits of
statins in patients with diabetes were similar to those in
other statin-treated subgroups of patients52.
The mechanism of new-onset diabetes associated
with statin use is not clear, but statins might interfere with
peripheral insulin signalling and pancreatic β-cell func-
tion. With regard to genetic factors, a 2015 study exam-
ined single nucleotide polymorphisms in the HMGCR
gene as proxies for 3-hydroxy-3-methylglutaryl-CoA
reductase (HMG-CoA reductase) inhibition by statins53.
The study showed that HMG-CoA inhibition was asso-
ciated not only with a reduction in plasma LDL-C levels
but also with a small increase in body weight, waist
circumference and plasma insulin and glucose concen-
trations. The data suggested that the increased risk of
diabetes was in part due to HMG-CoA inhibition. Over
the past decade, PCSK9 inhibitors have been developed
and have shown significant effects in reducing LDL-C
levels and the risk of cardiovascular disease. Although
the large-scale FOURIER trial54 of cardiovascular out-
comes in 27,564 individuals treated with evolocumab
did not identify an increase in new-onset diabetes after
a median duration of 2.2 years, several questions with
regard to new-onset diabetes associated with the use of
this drug class have been raised. A study published in
2017 to assess the association between PCSK9 genetic
variants and type 2 diabetes showed that, similarly
to the study on HMCGR, every 1 mmol/l reduction
in LDL-C levels was associated with a small increase in
the waist:hip ratio (0.006), body weight (1.02 kg), and
fasting blood glucose levels (0.009 mmol/l) and an

Table 3 | Incidence of diabetes mellitus in major trials on statin therapy49

Study (year) Statin therapy Patients Rate of new-onset diabetesa OR (95% CI) Refs
(n)
Statin Placebo or
group control group
ASCOT-LL A (2003) Atorvastatin 10 mg 7 ,773 11.9 10.5 1.14 (0.89–1.46) 37

HPS (2003) Simvastatin 40 mg 14,573 9.2 8.0 1.15 (0.98–1.35) 83

JUPITER (2008) Rosuvastatin 20 mg 17 ,802 16.0 12.8 1.26 (1.04–1.51) 48

WOSCOPS (2001) Pravastatin 40 mg 5,974 5.2 6.5 0.79 (0.58–1.10) 84

LIPID (2003) Pravastatin 40 mg 6,997 6.0 6.6 0.91 (0.71–1.71) 85

CORONA (2007) Rosuvastatin 20 mg 3,534 20.9 18.5 1.14 (0.84–1.55) 16

PROSPER (2002) Pravastatin 40 mg 5,023 20.5 15.8 1.32 (1.03–1.69) 72

MEGA (2006) Pravastatin 10–20 mg 6,086 10.8 10.1 1.07 (0.86–1.35) 86

AFCAPS/TEXCAPS Lovastatin 20–40 mg 6,211 4.5 4.6 0.98 (0.70–1.38) 87

(1998)
4 S (1994) Simvastatin 20–40 mg 4,242 17.3 16.8 1.03 (0.84–1.28) 88

ALLHAT (2002) Pravastatin 40 mg 6,087 16.4 14.4 1.15 (0.95–1.41) 89

GISSI-HF (2008) Rosuvastatin 10 mg 3,378 34.8 32.1 1.10 (0.89–1.35) 17

GISSI-PREV (2000) Pravastatin 20 mg 3,460 27.5 30.6 0.89 (0.67–1.20) 90

Meta-analysisb All statins 91,140 12.23 11.25 1.09 (1.02–1.17) 49

a
Events per 1,000 patient-years. Pooled odds ratio from a meta-analysis of the 13 trials.
b

NATuRe RevIewS | CARDIOLOgy volume 15 | DECEMBER 2018 | 763

Reviews

Table 4 | Diabetes mellitus and cardiovascular disease in trials on intensive versus moderate statin therapy52
Study (year) Intensive and moderate Incident diabetes mellitus Incident CVD Refs
doses
Intensive- Moderate- OR (95% CI) Intensive- Moderate- OR (95% CI)
therapy group therapy group therapy group therapy group
PROVE-IT Atorvastatin 80 mg; 101/1,707 99/1,688 1.01 315/1,707 355/1,688 0.85 10

TIMI (2004) pravastatin 40 mg (5.9%) (5.9%) (0.76–1.34) (18.4%) (21.0%) (0.72–1.01)
A to Z (2004) Simvastatin 40–80 mg; 65/1,768 47/1,736 1.37 212/1,768 234/1,736 0.87 13

placebo or simvastatin 20 mg (3.7%) (2.7%) (0.94–2.01) (12.0%) (13.5%) (0.72–1.07)

TNT (2005) Atorvastatin 80 mg; 418/3,798 358/3,797 1.19 647/3,798 830/3,797 0.73 11

atorvastatin 10 mg (11.0%) (9.4%) (1.02–1.38) (17.0%) (21.9%) (0.65–0.82)

IDEAL (2005) Atorvastatin 80 mg; 240/3,737 209/3,724 1.15 776/3,737 917/3,724 0.80 12

simvastatin 20–40 mg (6.4%) (5.6%) (0.95–1.40) (20.8%) (24.6%) (0.72–0.89)

SEARCH Simvastatin 80 mg; 625/5,398 587/5,399 1.07 1,184/5,398 1,214/5,399 0.97 14

(2010) simvastatin 20 mg (11.6%) (10.9%) (0.95–1.21) (21.9%) (22.5%) (0.88–1.06)

Meta-analysisa Various 1,449/16,408 1,300/16,344 1.12 3,134/16,408 3,550/16,344 0.84 52

(8.8%) (8.0%) (1.04–1.22) (19.1%) (21.7%) (0.75–0.94)

CVD, cardiovascular disease. aPooled odds ratio from a meta-analysis of the five trials.

odds ratio of 1.12 for newly diagnosed diabetes55. Despite
the study limitations, these data corroborate previous
studies indicating that clinicians should screen patients
for diabetes when starting statin therapy and concurrently
initiate lifestyle modification strategies.
Liver considerations
Abnormal liver enzyme tests. Statins were previously
thought to have potential hepatotoxicity owing to their
metabolism in the liver and the interactions with the
cytochrome P450 pathway. In the past, elevated plasma
levels in the liver enzymes aspartate trans­aminase
(AST) or alanine transaminase (ALT) were a com-
mon reason for statin discontinuation. Hepatologists
use Hy’s law, in which three criteria must be satisfied
to suggest drug-induced liver injury: elevation in AST
or ALT ≥3 times ULN, increase in total bilirubin levels
>2 times ULN, and no other demonstrable cause such
as cholestasis, viral hepatitis, pre-existing or acute hepato­
biliary disease, or use of another drug that can cause the
observed injury56. Rates of fulminant hepatic failure
related to statin use are rare; consequently, liver enzymes
are not routinely evaluated except for a first-time check
before initiation of statin therapy or periodically if the
patients are symptomatic.
If a patient has abnormal levels of liver enzymes
(ALT or AST), the NLA has developed an algorithm
that can be used to guide safe statin use and other clin-
ical decision-making56. The algorithm is based on the
degree of ALT or AST elevation: if ALT or AST elevation
is above the ULN but ≤ 3 times the ULN, bilirubin levels
should be evaluated; if ALT or AST levels are normal and
creatine kinase level is normal, the patient likely has non-
alcoholic fatty liver disease (NAFLD), and statin therapy
can be reasonably continued; if the elevation in bilirubin
level is indirect and not new and the creatine kinase level
is normal, the patient could have underlying Gilbert syn-
drome or NAFLD, and the statin can be continued; a new
elevation in bilirubin levels requires further evaluation,
and the statin should be stopped until further work-up
is completed; if the elevation in ALT or AST level is
>3 times ULN, a repeated evaluation of ALT or AST levels
should be considered, and statin therapy should be held
or stopped until further evaluation is completed. Data
are also available regarding the use of statins in specific
liver diseases, including those below57.
Hepatitis C. Hepatitis C affects 2% of the population in
the USA and leads to chronic liver disease-related mor-
bidity and mortality. The hepatitis C virus (HCV) and
lipid metabolism are interrelated, given that viral repli-
cation involves the LDL receptor (LDLR) and enzymes
involved in cholesterol biosynthesis. Statins might inter-
rupt this process. Several studies have shown that statin
use is linked to improvements in viral response to HCV
treatments, with a minimal effect on liver function tests.
Certain statins (such as rosuvastatin) should be avoided
during hepatitis C treatment with ledipasvir and/or
sofosbuvir because of drug–drug interactions58.
NAFLD and NASH. The prevalence of NAFLD and
nonalcoholic steatohepatitis (NASH) has increased with
the rise in obesity prevalence. Approximately 9–37% of
people worldwide have NAFLD, and 3–5% have NASH,
with higher frequencies in patients with obesity 56.
Statin use in patients with NAFLD or NASH is asso-
ciated with a reduction in transaminase plasma levels,
with improvements in steatosis and the liver necro­
inflammatory grade but no change in the liver fibrosis
grade59,60. Some evidence suggests that patients with
NAFLD have a higher 10-year risk of cardiovascular
disease than patients without NAFLD61. In the GREACE
study62, a post hoc analysis showed that statin therapy is
safe and can improve the plasma levels of liver enzymes
and reduce cardiovascular events in patients with mild
abnormal liver tests attributed to NAFLD.
Liver transplantation. Many medications used after liver
transplantation, such as immunosuppressive therapy,
can predispose patients to dyslipidaemia and other con-
ditions that can increase the risk of cardiovascular dis-
ease. One observational study of 598 patients with liver
or liver–kidney transplant showed that the prevalence of
hyperlipidaemia increased from 12% before transplanta-
tion to 22% at 7 years after transplantation63. The high-
est risk factors were the use of steroids and sirolimus.

764 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio


Statins interact with immunosuppressive agents through
their CYP3A4-related metabolism; however, statins
remain the mainstay treatment for hyperlipidaemia.
A retrospective study with lipid-lowering agents
(statins and fibrates) showed that 90% of the patients
reached target lipid levels, with no significant increases
in the plasma levels of liver enzymes. All patients who
developed SAMS were also receiving cyclosporine, a
CYP3A4 inhibitor64. Generally, caution is advisable
when using statins that are metabolized via the CYP3A4
pathway (atorvastatin, lovastatin and simvastatin)
in patients receiving immunosuppressants that use the
same pathway (for example, cyclosporine, sirolimus
and tacrolimus)65.
Primary biliary cirrhosis and cholestatic liver disease.
Primary biliary cirrhosis is a rare autoimmune choles-
tatic liver disease that affects women more than men and
notably alters cholesterol metabolism57. Limited data
are available on the use of statins in this patient popu­
lation; however, evidence suggests that statins are safe
and effective for lipid lowering in patients with primary
biliary cirrhosis66.
Cognition
Cognition is a very broad term involving four domains:
executive function, memory, language and visuospatial
ability. Cognitive dysfunction can be an impairment
in any of these domains and represents a spectrum of
disease67. Mild cognitive impairment (MCI) is a state
between normal cognition and dementia68. Several
observational studies and case reports have shown a pos-
sible association between statins and cognitive impair-
ment67. However, the assessment of statin-induced
cognitive impairment poses challenges. The mini-mental
status exam (MMSE) is a common test used to assess
cognition but does not evaluate executive function. The
Montreal Cognitive Assessment (MOCA) is more com-
prehensive than the MMSE, but the evaluation can take
more time. Other brief tests, including the Mini-Cog and
Functional Activities scale, have excellent sensitivity
and specificity for distinguishing MCI and dementia
from healthy cognitive ageing69,70. The pathophysio­
logy of statin-induced cognitive dysfunction remains
unclear. However, some studies in animal models sug-
gest that the reduction of cholesterol synthesis below
a critical level in oligodendrocytes, which leads to the
inhibition of myelination in the central nervous system,
results in cognitive deficits67. Animal data also suggest
that brain levels of some statins depend on their lipo-
philicity (sim­vastatin > lovastatin > atorvastatin > pravas-
tatin > rosuvas­tatin) and whether statins are rapidly
eliminated from the brain67.
Observational studies have shown mixed results,
with studies suggesting that statins are associated with
either cognitive impairment or improvement in cogni-
tion67. However, many of these studies are biased owing
to their observational design and the inclusion of vari-
ous risk populations, including elderly individuals and
individuals with pre-existing cognitive dysfunction.
A  2015 meta-analysis of 23 RCTs did not indicate
that statin therapy was associated with cognitive
NATuRe RevIewS | CARDIOLOgy
Reviews
impairment71. Two small studies have reported poten-
tial adverse events of statins on cognition. One study
of 209 patients to evaluate the potential psychological
effects of lovastatin showed that the placebo group had
greater improvements in cognitive measures than the
statin group at 6 months, but the difference was small
and of questionable clinical relevance. Another study
of 308 patients showed a potential detrimental effect of
simvastatin on cognition; however, this study was limited
by the type of cognitive testing used in the study71.
Few large RCTs that prospectively study statins and
cognition are available. The PROSPER72 and HPS73 stud-
ies are the only two large RCTs that have prospectively
studied cognitive function with the use of well-validated
neuropsychological measurements. The  PROSPER
study72 involved 5,804 men and women aged 70–82 years
who were at risk of vascular disease. Participants were
randomly assigned to receive either pravastatin (40 mg
daily) or placebo for 3 years. The patients were assessed
over the course of the trial using four validated neuro­
psychological performance tests for executive function,
attention, processing speed and immediate and delayed
memory. These well-validated cognitive performance
tests showed no effect of pravastatin on cognitive func-
tion at any point during the trial74. In the HPS study73
involving 20,536 patients randomly allocated to simvas-
tatin (40 mg daily) or placebo for an average of 5 years,
the effects on memory were systematically assessed
using the modified Telephone Interview for Cognitive
Status questionnaire. The test was used only once at the
end of the study and revealed no difference in the percent-
age of patients classified as cognitively impaired between
the simvastatin (23.7%) and placebo (24.2%) groups.
In addition, assessment of cognition in 5,806 patients
aged 75–85 years showed no difference in cognitive
impairment between the statin and placebo groups73.
In summary, the totality of the evidence, including
several meta-analyses of multiple RCTs, does not indi-
cate an association between statin use and cognitive
dysfunction. Given that mild cognitive dysfunction is
common in elderly individuals, the NLA has provided
some guidelines to help clinicians assess the risk of cog-
nitive dysfunction during statin use. Generally, a baseline
cognitive exam does not need to be performed before
starting statin therapy. Of note, if patients complain of
cognitive symptoms, symptoms should not be dismissed
if patients are receiving statin therapy. A thorough clinical
history and physical exam should be performed to look
for contributors to cognitive impairment, and a MOCA
test or other screening test could be used to assess cogni-
tion in the four domains. If a statin is suspected to alter
cognition, the reasonable option is to stop the drug for
1–2 months before considering a rechallenge67.
Haemorrhagic stroke
The 2010 pooled CTT meta-analysis of 26 RCTs involv-
ing individual patient data showed that statins reduce the
overall risk of stroke by 16% (RR 0.84, 95% CI 0.79–0.89)
and the risk of ischaemic stroke by 21% (RR 0.79, 95%
CI 0.74–0.85) per 1 mmol/l reduction in LDL-C lev-
els3. However, the increase in the risk of haemorrhagic
stroke associated with statin use was not significant
volume 15 | DECEMBER 2018 | 765
Reviews
(RR 1.12, 95% CI 0.93–1.35, P = 0.20) (Table 1). The
issue of an increased risk of haemorrhagic stroke asso-
ciated with lower levels of blood cholesterol has been
a concern ever since epidemiological studies suggested a
possible correlation.
The SPARCL trial51 was the first RCT to examine
high-intensity statin (atorvastatin at a daily dose of
80 mg) versus placebo in 4,731 patients with a history
of stroke or TIA who had LDL-C levels of 100–190 mg/dl.
Although the risk of nonfatal and fatal stroke was
reduced by 16% (RR 0.84, 95% CI 0.71–0.99), a small
increase in the incidence of haemorrhagic stroke (n = 55
versus n = 33; HR 1.66, 95% CI 1.08–2.55) was observed
in the statin group compared with the placebo group.
Subsequent analysis revealed that the increased risk of
haemorrhagic stroke was higher in individuals with a
previous haemorrhagic stroke at study entry (HR 5.65,
95% CI 2.82–11.30), in men (HR 1.79, 95% CI 1.13–2.84)
and in increasing 10-year age increment groups (HR 1.42,
95% CI 1.16–1.74)75. When the results of SPARCL51 and
CORONA16 were added to the 2010 CTT meta-analysis
discussed above, the proportional risk of haemorrhagic
stroke per 1 mmol/l of LDL-C level reduction increased
by 21% (HR 1.21, 95% CI 1.05–1.41, P = 0.01)3. However,
the small additional risk of haemorrhagic stroke is clearly
outweighed by the large reductions seen in the risk of
ischaemic stroke and major cardiovascular events.
A meta-analysis of 23 RCTs and 19 observational
studies including 248,391 patients with a total of 14,784
intracerebral haemorrhages revealed no association
between statin therapy and increased risk of intracerebral
stroke76. Analysis of RCTs (RR 1.10, 95% CI 0.86–1.41),
cohort studies (RR 0.94, 95% CI 0.81–1.10) or case–control
studies (RR 0.60, 95% CI 0.41–0.88) also showed no incre­
ased risk of intracerebral stroke with statin use. Although
the  study was not an analysis of individual patient
data, the meta-analysis also showed no correlation between
the risk of intracerebral stroke and the magnitude of
LDL-C level reduction. Data from the IMPROVE-IT
trial77 on simvastatin and ezetimibe showed a small,
but not significant, increase in haemorrhagic stroke
events (59 versus 43; HR 1.38, 95% CI 0.93–2.04) with
statin and ezetimibe use but no association between
haemorrhagic stroke events and low levels of LDL-C
(<30 mg/dl). Similarly, the FOURIER trial6 showed a
nonsignificant increase in haemorrhagic stroke in the
statin group compared with the control group (29 versus
25) but no association between haemorrhagic stroke and
LDL-C levels <0.5 mmol/l (39 mg/dl). In summary, the
SPARCL trial51 was the only trial to suggest a significant
increase in haemorrhagic stroke with atorvastatin use,
whereas adjunctive statin therapy with ezetimibe and
evolocumab leading to very low levels of LDL-C did not.
Therefore, any small possible additional risk of haemor-
rhagic stroke is clearly outweighed by the large reductions
seen in overall stroke and major cardiovascular events.
Low LDL-C levels and adverse events
In a post hoc analysis of the JUPITER study78, in which
patients with an LDL-C level of <130 mg/dl (3.4 mmol/l)
were randomly assigned to high-intensity statin treat-
ment with rosuvastatin (20 mg daily), patients who
766 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio
achieved an LDL-C level of <50 mg/dl (1.3 mmol/l)
(n = 4,000) were compared with patients who achieved
LDL levels >50 mg/dl (1.3 mmol/l) (n = 4,154). The
rates of myalgia, muscle weakness, diabetes, neuro­
psychiatric conditions, cancer and most other major
adverse effects were not significantly different among
rosuvastatin-treated participants with or without LDL-C
levels <50 mg/dl. Other RCTs also confirmed that stat-
ins did not increase adverse events despite LDL-C level
reductions well below the existing goal or threshold of
70 mg/dl (1.8 mmol/l)3.
A post hoc analysis of the IMPROVE-IT trial77, in
which 18,144 patients with previous acute coronary
syndrome were randomly assigned to either simvastatin
(40 mg daily) or a combination of simvastatin (40 mg
daily) and ezetimibe (10 mg daily), was performed to
compare individuals with an achieved LDL-C level of
<30 mg/dl (0.8 mmol/l) versus those with an LDL-C level
of >30 mg/dl at 1 month. During the 6 years of the trial,
individuals with an LDL-C level <30 mg/dl had a simi-
lar safety profile to individuals who had achieved higher
LDL-C levels. The analysis indicated no differences in
safety end points including serious muscle events, ele-
vation in plasma aminotransferase levels, gall-bladder
adverse events, neurocognitive events, haemorrhagic
stroke or cataracts.
With the advent of several new drugs, including
the PCSK9 inhibitors alirocumab and evolocumab, the 
opportunity now exists to reduce LDL-C to even lower
levels by adding these drugs to high-intensity statin
therapy. These drugs inhibit PCSK9, a protein that binds
to the LDLR and assists in the internalization of the LDLR
for degradation79. The FOURIER trial6 on evolocumab
was the first trial of a PCSK9 inhibitor in combination
with statin therapy to show improved cardiovascular out-
comes compared with statin therapy alone. In this trial,
LDL-C levels were reduced from 92 to 30 mg/dl (2.4 to
0.8 mmol/l) with no accompanying safety issues after a
median trial duration of 2.2 years. A further safety analy­
sis of the FOURIER trial54 demonstrated no increased
safety effects with LDL-C levels <20 mg/dl (<0.5 mmol/l).
In addition, a 2017 meta-analysis of pooled RCTs on
alirocumab evaluated the safety of achieving LDL-C
levels as low as <25 mg/dl (<0.65 mmol/l) or <15 mg/dl
(<0.4 mmol/l) and revealed that very low LDL-C levels
were not associated with increased adverse events inclu­
ding diabetes, neurocognitive or muscle-related events80.
However, a small increase in cataracts in the LDL-C
<25 mg/dl group compared with the propensity-matching
group was observed. More generally, this analysis
and other short-term and long-term PCSK9 studies add
to the evidence base that achieving low LDL-C levels is
very well-tolerated and is not associated with adverse
events. The ODYSSEY Outcomes trial81 with alirocumab
will add long-term safety information on the use of
PCSK9 inhibitors.
Conclusions
The 2013 ACC/AHA guidelines, along with other inter-
national guidelines, establish the primacy of statins in
reducing the risk of ASCVD in both primary and sec-
ondary prevention1. Even with evidence of the benefits of
 Reviews

Benefits Adverse effects variety of other hepatobiliary diseases including NASH

Risk of stroke Cognitive dysfunction and hepatitis C.
• ↓ 16% for total stroke • No evidence Other adverse effects of statins such as the increase
• ↓ 21% for ischaemic stroke Risk of haemorrhagic stroke in new-onset diabetes over a 5-year time period are
• Small increase in individuals
with prior stroke very small (HR 1.1 or HR 1.2 with high-intensity statin
therapy). The risks are generally higher in individuals
Risk of major coronary events Liver symptoms/diseases at risk of developing diabetes, such as patients with
• ↓ 27% for nonfatal MI • Clinically insignificant liver
• ↓ 20% for CHD death enzyme elevations
metabolic syndrome or prediabetes. The mechanism
Risk of revascularization • Incidence of liver failure: underlying the increased risk of diabetes is currently
procedures 1/100,000 unknown, and therefore, maintaining lifestyle coun-
• ↓ 25%
Incidence of new-onset
selling during statin therapy, with an emphasis on diet,
diabetes mellitus exercise and weight loss, is prudent.
• Moderate-intensity statin Although statins have been shown to reduce the
therapy: 0.1% per year
• High-intensity statin therapy: risk of overall and ischaemic stroke in most population
0.2% per year groups studied, the SPARCL trial51 is the only study to
demonstrate a significant increase in the risk of haem-
Incidence of muscle
symptoms/diseases orrhagic stroke. The totality of the evidence suggests
• SAMS: 10–29% in observational that statins are safe in most patients with previous
studies and 1–2% in RCTs stroke, with the caveat that the data are not as robust
• Myopathy: 1/1,000
• Rhabdomyolysis: 1/10,000 in patients with a history of intracerebral haemorrhage.
Using thresholds levels of LDL-C for statin therapy
Fig. 1 | Clinical benefits and potential adverse effects of statin therapy. The has been controversial since the publication of the 2013
cardiovascular benefits, including the reduction in the risk of major coronary events and ACC/AHA guidelines on cholesterol management1.The
revascularization and the reduction in the risk of stroke, associated with statin therapy recommendation not to target LDL-C levels to a given
far outweigh the potential risks. Given this clinical benefit, providers should work goal or threshold was based on the absence of RCT
diligently with patients to ensure that patients adhere to therapy in a shared-decision
evidence to do so. However, data from the 2010 CTT
model. CHD, coronary heart disease; MI, myocardial infarction; RCT, randomized
controlled trial; SAMS, statin-associated muscle symptoms.
meta-analysis suggest that reducing LDL-C levels by
1 mmol/l (approximately 39 mg/dl) over 5 years results in
a reduction of 22% in major cardiovascular events, 20%
statin therapy, adverse effects have been cited as a major in CHD death and 10% in all-cause mortality3. The size
reason that limits patients from taking and adhering of the risk reduction in major vascular events is directly
to statin therapy. The most common reported adverse proportional to the absolute LDL-C level reduction
events of statin therapy are muscle-related and are often achieved. The findings suggest even larger risk reduc-
referred to as SAMS. The SAMS-CI is a clinical tool that tions of 40–50% with greater LDL-C level reductions of
might be useful in helping to evaluate muscle symptoms 2–3 mmol/l (77–116 mg/dl). Therefore, the goal of ther-
in patients taking statin therapy. How much of SAMS is apy is to achieve the largest LDL-C reductions without
caused by a true statin pharmacological effect versus a increasing the risk of adverse events. Evidence from
nocebo effect, or by a combination of both, is controver- the IMPROVE-IT5 and FOURIER6 trials suggests that
sial. However, the symptoms are very real for the patient, adjunctive therapy with ezetimibe and PCSK9 inhibi-
regardless of the aetiology, which gives the health-care tors in addition to high-intensity statin therapy results
provider the opportunity to explain to the patient the in a further reduction in cardiovascular risk. This risk
true benefits and harms of statin therapy using shared reduction is seen with levels of LDL-C as low as 30 mg/dl
decision-making (Fig. 1). (0.8 mmol/l) without any apparent harm. Despite con-
Although hepatic toxicity was previously consi­ cerns that achieving very low LDL-C levels could cause
dered a very concerning adverse effect of statin ther- harm, the results of these trials suggest that achieving
apy given the high frequency of elevated levels of liver very low LDL-C levels is generally safe and yields addi-
enzymes in the plasma, evidence has shown that statins tional cardiovascular benefits. Given how beneficial
are very safe with respect to the liver and can be used and safe statins are in reducing ASCVD events, provid-
safely in a wide variety of liver diseases. Liver enzyme ers should work diligently with patients to ensure that
levels do not need to be checked routinely in patients patients adhere to therapy in a shared-decision model.
receiving statin therapy unless the patient becomes
symptomatic. Statins have been shown to be safe in a Published online 30 October 2018

1. Stone, N. J. et al. 2013 ACC/AHA guideline on the
treatment of blood cholesterol to reduce atherosclerotic
cardiovascular risk in adults: a report of the American
College of Cardiology/American Heart Association Task
Force on Practice Guidelines. J. Am. Coll. Cardiol. 63,
2889–2934 (2014).
2. Baigent, C. et al. Efficacy and safety of cholesterol-
lowering treatment: prospective meta-analysis of data
from 90,056 participants in 14 randomised trials of
statins. Lancet 366, 1267–1278 (2005).
3. Cholesterol Treatment Trialists Collaboration. Efficacy
and safety of more intensive lowering of LDL cholesterol:
a meta-analysis of data from 170,000 participants
in 26 randomised trials. Lancet 376, 1670–1681
(2010).
4. Collins, R. et al. Interpretation of the evidence for
the efficacy and safety of statin therapy. Lancet 388,
2532–2561 (2016).
5. Cannon, C. P. et al. Ezetimibe added to statin therapy
after acute coronary syndromes. N. Engl. J. Med. 372,
2387–2397 (2015).
6. Sabatine, M. S. et al. Evolocumab and clinical
outcomes in patients with cardiovascular disease.
N. Engl. J. Med. 376, 1713–1722 (2017).
7. Adhyaru, B. B. & Jacobson, T. A. Role of non-
statins, LDL-C thresholds, and special population
considerations: a look at the updated 2016
ACC Consensus Committee Recommendations.
Curr. Atheroscler. Rep. 19, 29 (2017).
8. Writing Committee. et al. 2016 ACC expert consensus
decision pathway on the role of non-statin therapies
for LDL-cholesterol lowering in the management of
atherosclerotic cardiovascular disease risk: a report
of the American College of Cardiology Task Force
on Clinical Expert Consensus documents. J. Am.
Coll. Cardiol. 68, 92–125 (2016).

NATuRe RevIewS | CARDIOLOgy volume 15 | DECEMBER 2018 | 767

Reviews
9. Lloyd-Jones, D. M. et al. 2017 focused update of the
2016 ACC expert consensus decision pathway on the
role of non-statin therapies for LDL-cholesterol lowering
in the management of atherosclerotic cardiovascular
disease risk: a report of the American College of
Cardiology Task Force on Expert Consensus Decision
Pathways. J. Am. Coll. Cardiol. 70, 1785–1822
(2017).
10. Cannon, C. P. et al. Intensive versus moderate lipid
lowering with statins after acute coronary syndromes.
N. Engl. J. Med. 350, 1495–1504 (2004).
11. LaRosa, J. C. et al. Intensive lipid lowering with
atorvastatin in patients with stable coronary disease.
N. Engl. J. Med. 352, 1425–1435 (2005).
12. Pedersen, T. R. et al. High-dose atorvastatin versus
usual-dose simvastatin for secondary prevention after
myocardial infarction: the IDEAL study: a randomized
controlled trial. JAMA 294, 2437–2445 (2005).
13. de Lemos, J. A. et al. Early intensive versus a delayed
conservative simvastatin strategy in patients with
acute coronary syndromes: phase Z of the A to Z trial.
JAMA 292, 1307–1316 (2004).
14. Study of the Effectiveness of Additional Reductions in
Cholesterol and Homocysteine (SEARCH) Collaborative
Group. Intensive lowering of LDL cholesterol with 80
mg versus 20 mg simvastatin daily in 12,064
survivors of myocardial infarction: a double-blind
randomised trial. Lancet 376, 1658–1669 (2010).
15. Cholesterol Treatment Trialists Collaboration. The
effects of lowering LDL cholesterol with statin therapy
in people at low risk of vascular disease: meta-analysis
of individual data from 27 randomised trials. Lancet
380, 581–590 (2012).
16. Kjekshus, J. et al. Rosuvastatin in older patients with
systolic heart failure. N. Engl. J. Med. 357, 2248–2261
(2007).
17. Tavazzi, L. et al. Effect of rosuvastatin in patients
with chronic heart failure (the GISSI-HF trial):
a randomised, double-blind, placebo-controlled trial.
Lancet 372, 1231–1239 (2008).
18. Baigent, C. et al. The effects of lowering LDL cholesterol
with simvastatin plus ezetimibe in patients with chronic
kidney disease (Study of Heart and Renal Protection):
a randomised placebo-controlled trial. Lancet 377,
2181–2192 (2011).
19. Fellstrom, B. C. et al. Rosuvastatin and cardiovascular
events in patients undergoing hemodialysis. N. Engl.
J. Med. 360, 1395–1407 (2009).
20. Wanner, C. et al. Atorvastatin in patients with type 2
diabetes mellitus undergoing hemodialysis. N. Engl.
J. Med. 353, 238–248 (2005).
21. Bruckert, E., Hayem, G., Dejager, S., Yau, C.
& Begaud, B. Mild to moderate muscular symptoms
with high-dosage statin therapy in hyperlipidemic
patients — the PRIMO study. Cardiovasc. Drugs Ther.
19, 403–414 (2005).
22. Cohen, J. D., Brinton, E. A., Ito, M. K. & Jacobson, T. A.
Understanding statin use in America and gaps in patient
education (USAGE): an internet-based survey of 10,138
current and former statin users. J. Clin. Lipidol. 6,
208–215 (2012).
23. Colantonio, L. D. et al. Adherence to high-intensity
statins following a myocardial infarction hospitalization
among medicare beneficiaries. JAMA Cardiol. 2,
890–895 (2017).
24. Rosenson, R. S. et al. An assessment by the Statin
Muscle Safety Task Force: 2014 update. J. Clin. Lipidol.
8, S58–71 (2014).
25. Mancini, G. B. et al. Diagnosis, prevention,
and management of statin adverse effects
and intolerance: Canadian Consensus Working
Group update. Can. J. Cardiol. 32, S35–S65
(2016).
26. Stroes, E. S. et al. Statin-associated muscle symptoms:
impact on statin therapy-European Atherosclerosis
Society Consensus Panel Statement on Assessment,
Aetiology and Management. Eur. Heart J. 36,
1012–1022 (2015).
27. Awad, K. et al. Efficacy and Safety of alternate-day
versus daily dosing of statins: a systematic review and
meta-analysis. Cardiovasc. Drugs Ther. 31, 419–431
(2017).
28. Parker, B. A. et al. Effect of statins on skeletal muscle
function. Circulation 127, 96–103 (2013).
29. Rosenson, R. S. et al. The statin-associated muscle
symptom clinical index (SAMS-CI): revision for clinical
use, content validation, and inter-rater reliability.
Cardiovasc. Drugs Ther. 31, 179–186 (2017).
30. Taylor, B. A. et al. Application of the statin-associated
muscle symptoms-clinical index to a randomized
trial on statin myopathy. J. Am. Coll. Cardiol. 70,
1680–1681 (2017).
768 | DECEMBER 2018 | volume 15 www.nature.com/nrcardio
31. Schech, S. et al. Risk factors for statin-associated
rhabdomyolysis. Pharmacoepidemiol. Drug Saf. 16,
352–358 (2007).
32. Banach, M. et al. Statin intolerance — an attempt at a
unified definition. Position paper from an International
Lipid Expert Panel. Expert Opin. Drug Saf. 14,
935–955 (2015).
33. Abd, T. T. & Jacobson, T. A. Statin-induced myopathy:
a review and update. Expert Opin. Drug Saf. 10,
373–387 (2011).
34. Guengerich, F. P. Cytochrome P450 and chemical
toxicology. Chem. Res. Toxicol. 21, 70–83 (2008).
35. Bellosta, S. & Corsini, A. Statin drug interactions and
related adverse reactions. Expert Opin. Drug Saf. 11,
933–946 (2012).
36. Gupta, A. et al. Adverse events associated with
unblinded, but not with blinded, statin therapy in
the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid-
Lowering Arm (ASCOT-LLA): a randomised double-
blind placebo-controlled trial and its non-randomised
non-blind extension phase. Lancet 389, 2473–2481
(2017).
37. Sever, P. S. et al. Prevention of coronary and stroke
events with atorvastatin in hypertensive patients who
have average or lower-than-average cholesterol
concentrations, in the Anglo-Scandinavian Cardiac
Outcomes Trial-Lipid Lowering Arm (ASCOT-LLA):
a multicentre randomised controlled trial. Lancet 361,
1149–1158 (2003).
38. Adhyaru, B. B. & Jacobson, T. A. Unblinded ASCOT
study results do not rule out that muscle symptoms
are an adverse effect of statins. Evid. Based Med. 22,
210 (2017).
39. Nissen, S. E. et al. Efficacy and tolerability of
evolocumab versus ezetimibe in patients with
muscle-related statin intolerance: the GAUSS-3
randomized clinical trial. JAMA 315, 1580–1590
(2016).
40. Mampuya, W. M. et al. Treatment strategies in
patients with statin intolerance: the Cleveland Clinic
experience. Am. Heart J. 166, 597–603 (2013).
41. Jacobson, T. A., Khan, A., Maki, K. C., Brinton, E. A.
& Cohen, J. D. Provider recommendations for
patient-reported muscle symptoms on statin therapy:
insights from the understanding statin use in America
and gaps in patient education survey. J. Clin. Lipidol.
12, 78–88 (2018).
42. Banach, M. et al. Statin intolerance — an attempt at a
unified definition. Position paper from an International
Lipid Expert Panel. Arch. Med. Sci. 11, 1–23 (2015).
43. Guyton, J. R., Bays, H. E., Grundy, S. M.
& Jacobson, T. A., The National Lipid Association
Statin Intolerance Panel. An assessment by the Statin
Intolerance Panel: 2014 update. J. Clin. Lipidol. 8,
S72–S81 (2014).
44. Graham, J. H. et al. Clinical and economic consequences
of statin intolerance in the United States: results from
an integrated health system. J. Clin. Lipidol. 11, 70–79
(2017).
45. Athyros, V. G. et al. Effectiveness of ezetimibe alone or
in combination with twice a week Atorvastatin (10 mg)
for statin intolerant high-risk patients. Am. J. Cardiol.
101, 483–485 (2008).
46. Moriarty, P. M. et al. Efficacy and safety of
alirocumab, a monoclonal antibody to PCSK9, in
statin-intolerant patients: design and rationale of
ODYSSEY ALTERNATIVE, a randomized phase 3 trial.
J. Clin. Lipidol. 8, 554–561 (2014).
47. Orringer, C. E. et al. Update on the use of PCSK9
inhibitors in adults: recommendations from an Expert
Panel of the National Lipid Association. J. Clin. Lipidol.
11, 880–890 (2017).
48. Ridker, P. M. et al. Rosuvastatin to prevent vascular
events in men and women with elevated C-reactive
protein. N. Engl. J. Med. 359, 2195–2207 (2008).
49. Sattar, N. et al. Statins and risk of incident diabetes:
a collaborative meta-analysis of randomised statin
trials. Lancet 375, 735–742 (2010).
50. Ridker, P. M., Pradhan, A., MacFadyen, J. G., Libby, P.
& Glynn, R. J. Cardiovascular benefits and diabetes
risks of statin therapy in primary prevention:
an analysis from the JUPITER trial. Lancet 380,
565–571 (2012).
51. Amarenco, P. et al. High-dose atorvastatin after stroke
or transient ischemic attack. N. Engl. J. Med. 355,
549–559 (2006).
52. Preiss, D. et al. Risk of incident diabetes with
intensive-dose compared with moderate-dose statin
therapy: a meta-analysis. JAMA 305, 2556–2564
(2011).
53. Swerdlow, D. I. et al. HMG-coenzyme A reductase
inhibition, type 2 diabetes, and bodyweight: evidence
from genetic analysis and randomised trials. Lancet
385, 351–361 (2015).
54. Giugliano, R. P. et al. Clinical efficacy and safety of
achieving very low LDL-cholesterol concentrations with
the PCSK9 inhibitor evolocumab: a prespecified
secondary analysis of the FOURIER trial. Lancet 390,
1962–1971 (2017).
55. Schmidt, A. F. et al. PCSK9 genetic variants and risk of
type 2 diabetes: a mendelian randomisation study.
Lancet Diabetes Endocrinol. 5, 97–105 (2017).
56. Bays, H., Cohen, D. E., Chalasani, N. & Harrison, S. A.,
The National Lipid Association’s Statin Safety Task
Force. An assessment by the Statin Liver Safety
Task Force: 2014 update. J. Clin. Lipidol. 8, S47–S57
(2014).
57. Herrick, C., Bahrainy, S. & Gill, E. A. Statins and the
liver. Cardiol. Clin. 33, 257–265 (2015).
58. Henderson, L. M., Patel, S., Giordano, T. P., Green, L.
& El-Serag, H. B. Statin therapy and serum
transaminases among a cohort of HCV-infected
veterans. Dig. Dis. Sci. 55, 190–195 (2010).
59. Gomez-Dominguez, E., Gisbert, J. P.,
Moreno-Monteagudo, J. A., Garcia-Buey, L.
& Moreno-Otero, R. A pilot study of atorvastatin
treatment in dyslipemid, non-alcoholic fatty liver
patients. Aliment. Pharmacol. Ther. 23, 1643–1647
(2006).
60. Hyogo, H. et al. Efficacy of atorvastatin for the
treatment of nonalcoholic steatohepatitis with
dyslipidemia. Metabolism 57, 1711–1718
(2008).
61. Motamed, N. et al. Non-alcoholic fatty liver disease
(NAFLD) and 10-year risk of cardiovascular
diseases. Clin. Res. Hepatol. Gastroenterol. 41,
31–38 (2017).
62. Athyros, V. G. et al. Safety and efficacy of long-term
statin treatment for cardiovascular events in patients
with coronary heart disease and abnormal liver tests
in the Greek Atorvastatin and Coronary Heart Disease
Evaluation (GREACE) Study: a post-hoc analysis.
Lancet 376, 1916–1922 (2010).
63. Parekh, J., Corley, D. A. & Feng, S. Diabetes,
hypertension and hyperlipidemia: prevalence over
time and impact on long-term survival after liver
transplantation. Am. J. Transplant. 12, 2181–2187
(2012).
64. Martin, J. E. et al. Incidence of adverse events
with HMG-CoA reductase inhibitors in liver
transplant patients. Clin. Transplant. 22, 113–119
(2008).
65. Wiggins, B. S. et al. Recommendations for
management of clinically significant drug-drug
interactions with statins and select agents used in
patients with cardiovascular disease: a scientific
statement from the American Heart Association.
Circulation 134, e468–e495 (2016).
66. Cash, W. J. et al. Randomized controlled trial
assessing the effect of simvastatin in primary biliary
cirrhosis. Liver Int. 33, 1166–1174 (2013).
67. Rojas-Fernandez, C. H. et al. An assessment by
the Statin Cognitive Safety Task Force: 2014 update.
J. Clin. Lipidol. 8, S5–S16 (2014).
68. Petersen, R. C. Clinical practice. Mild cognitive
impairment. N. Engl. J. Med. 364, 2227–2234
(2011).
69. Nasreddine, Z. S. et al. The Montreal Cognitive
Assessment, MoCA: a brief screening tool for
mild cognitive impairment. J. Am. Geriatr. Soc. 53,
695–699 (2005).
70. Robert, P. et al. Review of Alzheimer’s disease
scales: is there a need for a new multi-domain
scale for therapy evaluation in medical practice?
Alzheimers Res. Ther. 2, 24 (2010).
71. Ott, B. R. et al. Do statins impair cognition? A
systematic review and meta-analysis of randomized
controlled trials. J. Gen. Intern. Med. 30, 348–358
(2015).
72. Shepherd, J. et al. Pravastatin in elderly individuals
at risk of vascular disease (PROSPER): a randomised
controlled trial. Lancet 360, 1623–1630 (2002).
73. Heart Protection Study Collaborative Group. MRC/BHF
Heart Protection Study of antioxidant vitamin
supplementation in 20,536 high-risk individuals:
a randomised placebo-controlled trial. Lancet 360,
23–33 (2002).
74. Trompet, S. et al. Pravastatin and cognitive function in
the elderly. Results of the PROSPER study. J. Neurol.
257, 85–90 (2010).
75. Goldstein, L. B. et al. Hemorrhagic stroke in the
stroke prevention by aggressive reduction in
cholesterol levels study. Neurology 70, 2364–2370
(2008).

76. Hackam, D. G. et al. Statins and intracerebral
hemorrhage: collaborative systematic review
and meta-analysis. Circulation 124, 2233–2242
(2011).
77. Giugliano, R. P. et al. Long-term safety and efficacy
of achieving very low levels of low-density lipoprotein
cholesterol: a prespecified analysis of the IMPROVE-IT
trial. JAMA Cardiol. 2, 547–555 (2017).
78. Hsia, J., MacFadyen, J. G., Monyak, J. & Ridker, P. M.
Cardiovascular event reduction and adverse events
among subjects attaining low-density lipoprotein
cholesterol <50 mg/dl with rosuvastatin. The JUPITER
trial (justification for the use of statins in prevention:
an intervention trial evaluating rosuvastatin). J. Am.
Coll. Cardiol. 57, 1666–1675 (2011).
79. Hess, C. N., Low Wang, C. C. & Hiatt, W. R. PCSK9
inhibitors: mechanisms of action, metabolic effects,
and clinical outcomes. Annu. Rev. Med. 69, 133–145
(2018).
80. Robinson, J. G. et al. Safety of very low low-density
lipoprotein cholesterol levels with alirocumab: pooled
data from randomized trials. J. Am. Coll. Cardiol. 69,
471–482 (2017).
81. Schwartz, G. G. et al. Effect of alirocumab,
a monoclonal antibody to PCSK9, on long-term
cardiovascular outcomes following acute coronary
syndromes: rationale and design of the ODYSSEY
outcomes trial. Am. Heart J. 168, 682–689
(2014).
82. Rosenson, R. S. et al. Optimizing cholesterol
treatment in patients with muscle complaints.
J. Am. Coll. Cardiol. 70, 1290–1301 (2017).
NATuRe RevIewS | CARDIOLOgy
83. Collins, R. et al. MRC/BHF Heart Protection Study of
cholesterol-lowering with simvastatin in 5963 people
with diabetes: a randomised placebo-controlled trial.
Lancet 361, 2005–2016 (2003).
84. Freeman, D. J. et al. Pravastatin and the development
of diabetes mellitus: evidence for a protective treatment
effect in the West of Scotland Coronary Prevention
Study. Circulation 103, 357–362 (2001).
85. Keech, A. et al. Secondary prevention of
cardiovascular events with long-term pravastatin in
patients with diabetes or impaired fasting glucose:
results from the LIPID trial. Diabetes Care 26,
2713–2721 (2003).
86. Nakamura, H. et al. Primary prevention of
cardiovascular disease with pravastatin in Japan
(MEGA Study): a prospective randomised controlled
trial. Lancet 368, 1155–1163 (2006).
87. Downs, J. R. et al. Primary prevention of acute
coronary events with lovastatin in men and women
with average cholesterol levels: results of AFCAPS/
TexCAPS. Air Force/Texas Coronary Atherosclerosis
Prevention Study. JAMA 279, 1615–1622 (1998).
88. Scandinavian Simvastatin Survival Study Group.
Randomised trial of cholesterol lowering in 4444
patients with coronary heart disease: the
Scandinavian Simvastatin Survival Study (4S).
Lancet 344, 1383–1389 (1994).
89. The ALLHAT Officers and Coordinators for the ALLHAT
Collaborative Research Group. Major outcomes in
moderately hypercholesterolemic, hypertensive
patients randomized to pravastatin versus usual care:
The antihypertensive and lipid-lowering treatment to
Reviews
prevent heart attack trial (ALLHAT-LLT). JAMA 288,
2998–3007 (2002).
90. GISSI Prevenzione Investigators (Gruppo Italiano
per lo Studio della Sopravvivenza nell’Infarto
Miocardico). Results of the low-dose (20 mg)
pravastatin GISSI Prevenzione trial in 4271 patients
with recent myocardial infarction: do stopped trials
contribute to overall knowledge? Ital. Heart J. 1,
810–820 (2000).
Author contributions
Both authors researched data for the article, contributed to
the discussion of content, and wrote, reviewed and edited the
manuscript before submission.
Competing interests
T.A.J. is a consultant for Amgen and Regeneron/Sanofi. B.B.A.
declares no competing interests.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
Review criteria
We conducted a literature search using Medline, PubMed and
Embase and limited the search to the years 2000–2017.
Search terms included “statin adverse effects”, “statin safety”,
“statin side effects” and “statin intolerance”. In addition,
search terms related to specific target organs and adverse
events were also used, including “statin AND muscle, diabe-
tes, glucose, cognitive function and stroke”. Publications were
fully reviewed if the abstract suggested that the study was
relevant and used only human data.
volume 15 | DECEMBER 2018 | 769