Articles

Effect of statin therapy on muscle symptoms: an individual

participant data meta-analysis of large-scale, randomised,
double-blind trials
Cholesterol Treatment Trialists’ Collaboration*

Summary
Lancet 2022; 400: 832–45 Background Statin therapy is effective for the prevention of atherosclerotic cardiovascular disease and is widely
Published Online prescribed, but there are persisting concerns that statin therapy might frequently cause muscle pain or weakness. We
August 29, 2022 aimed to address these through an individual participant data meta-analysis of all recorded adverse muscle events in
https://doi.org/10.1016/
large, long-term, randomised, double-blind trials of statin therapy.
S0140-6736(22)01545-8
See Comment page 791
Methods Randomised trials of statin therapy were eligible if they aimed to recruit at least 1000 participants with a
*Collaborators are listed at the
end of the report
scheduled treatment duration of at least 2 years, and involved a double-blind comparison of statin versus placebo or
Correspondence to:
of a more intensive versus a less intensive statin regimen. We analysed individual participant data from 19 double-
Cholesterol Treatment Trialists’ blind trials of statin versus placebo (n=123 940) and four double-blind trials of a more intensive versus a less intensive
Collaboration, Clinical Trial statin regimen (n=30 724). Standard inverse-variance-weighted meta-analyses of the effects on muscle outcomes were
Service Unit and Epidemiological conducted according to a prespecified protocol.
Studies Unit, Nuffield
Department of Population
Health, University of Oxford, Findings Among 19 placebo-controlled trials (mean age 63 years [SD 8], with 34 533 [27·9%] women, 59 610 [48·1%]
Oxford OX3 7LF, UK participants with previous vascular disease, and 22 925 [18·5%] participants with diabetes), during a weighted average
[email protected] median follow-up of 4·3 years, 16 835 (27·1%) allocated statin versus 16 446 (26·6%) allocated placebo reported
muscle pain or weakness (rate ratio [RR] 1·03; 95% CI 1·01–1·06). During year 1, statin therapy produced a 7%
relative increase in muscle pain or weakness (1·07; 1·04–1·10), corresponding to an absolute excess rate of 11 (6–16)
events per 1000 person-years, which indicates that only one in 15 ([1·07–1·00]/1·07) of these muscle-related reports
by participants allocated to statin therapy were actually due to the statin. After year 1, there was no significant excess
in first reports of muscle pain or weakness (0·99; 0·96–1·02). For all years combined, more intensive statin regimens
(ie, 40–80 mg atorvastatin or 20–40 mg rosuvastatin once per day) yielded a higher RR than less intensive or moderate-
intensity regimens (1·08 [1·04–1·13] vs 1·03 [1·00–1·05]) compared with placebo, and a small excess was present
(1·05 [0·99–1·12]) for more intensive regimens after year 1. There was no clear evidence that the RR differed for
different statins, or in different clinical circumstances. Statin therapy yielded a small, clinically insignificant increase
in median creatine kinase values of approximately 0·02 times the upper limit of normal.

Interpretation Statin therapy caused a small excess of mostly mild muscle pain. Most (>90%) of all reports of muscle
symptoms by participants allocated statin therapy were not due to the statin. The small risks of muscle symptoms are
much lower than the known cardiovascular benefits. There is a need to review the clinical management of muscle
symptoms in patients taking a statin.

Funding British Heart Foundation, Medical Research Council, Australian National Health and Medical Research
Council.

Copyright © 2022 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0
license.

Introduction LDL cholesterol reduction achieved, which corresponds

Atherosclerotic cardiovascular diseases, chiefly myo­ to the avoidance of approximately 50 major vascular
cardial infarction and ischaemic stroke, accounted for events in those with pre-existing vascular disease
approximately 18 million deaths worldwide in 2019,1 and (secondary prevention), and 25 major vascular events
low-density lipoprotein (LDL) cholesterol is a major when used for primary prevention, in every 1000 people
causal risk factor.2 Randomised controlled trials have administered this therapy for 5 years.2 Moreover, a more
shown that the long-term reduction of LDL cholesterol intensive statin regimen (ie, 40–80 mg atorvastatin once
concentrations with an 3-hydroxy 3-methylglutaryl-coen­ per day, or 20–40 mg rosuvastatin once per day), which
zyme A reductase inhibitor (ie, a statin) reduces the could reduce LDL cholesterol by 2 mmol/L, would
incidence of myocardial infarction and of ischaemic prevent twice as many major vascular events, and longer
stroke by approximately a quarter for every 1 mmol/L treatment yields greater benefits. For a given reduction

832 www.thelancet.com Vol 400 September 10, 2022

 Articles

Research in context
Evidence before this study examining the effects on particular symptoms, the timing of
We searched Medline and the Cochrane Central Register of any excess risk, and the variation in treatment effects in
Controlled Trials for randomised trials, meta-analyses, or review different types of patients. Statin regimens caused a small
articles published in any language between Jan 1, 1990, and relative increase (3%) in the number of first reports of muscle
June 1, 2021, that had specifically assessed the effects of statin symptoms, but the excess of reports due to statin therapy was
regimens on muscle symptoms. These previous randomised largely confined to the first year of treatment, during which
trials, and several meta-analyses of published data from such there was an absolute excess rate of 11 (95% CI 6–16) events
trials, indicated that statins cause a small excess risk of muscle per 1000 person-years. Statins had similar effects on a range of
pain, but these studies might be affected by biases due to reported muscle symptoms, including myalgia, muscle cramps
missing data. Observational studies, and meta-analyses or spasm, limb pain, and other musculoskeletal pain. There was
including such studies, have also estimated the prevalence of no evidence for variation in the relative effects of different
statin-associated muscle symptoms, but such studies are statins. The relative increase in the rate of muscle symptoms
subject to reporting biases and other biases, and cannot reliably was similar in a wide range of trial participants, and was
establish the causal contribution of statins to such symptoms. irrespective of the variation in the methods used to ascertain
There is a need, therefore, for a reliable assessment of the symptoms, suggesting that the observed relative effects of
effects of various statin regimens on muscle symptoms in statins on muscle symptoms are likely to be generalisable.
different clinical circumstances. Overall, for the regimens studied in these trials, the increase in
symptoms was greater for more intensive statin regimens, and
Added value of this study
a small excess risk was likely to persist for longer, than for
We were able to minimise biases by restricting our analyses to
less intensive or moderate-intensity regimens. Based on
large-scale, randomised, double-blind trials of statin therapy
previous analyses of these trials, this excess risk of muscle
versus placebo in which there was systematic and unbiased
symptoms is greatly outweighed by the known cardiovascular
event reporting. To overcome the potential bias arising from
benefits of statin therapy.
the selected publication of results, we obtained details of all
adverse events related to muscle recorded in each individual Implications of all the available evidence
trial participant, and coded them using the common These results from randomised double-blind trials suggest that
nosological standard (from the Medical Dictionary for when a patient reports muscle symptoms when taking a statin,
Regulatory Activities). The availability of individual participant the probability that it is actually caused by the statin is low
data permitted more detailed analyses of risk for each statin (<10%). Consequently, current recommendations on the
than have previously been possible, including analyses management of such muscle symptoms should be reviewed.

in LDL cholesterol, similar proportional reductions in to their limitations. In addition, most studies of statin
risk are seen across a wide range of patients (including intolerance, or of statin-associated muscle symptoms,
men and women, older and younger patients, and those report the proportions of patients who do not adhere to
with and without a previous history of cardiovascular the statin regimen because of symptoms they attribute to
disease).3–6 Consequently, statins are now used by a statin. However, such estimates have the potential to be
millions of people worldwide. misleading, because a proportion of such symptoms will
It is known that statins can, rarely, cause substantial not truly be due to the statin. Perhaps as a consequence,
muscle damage (ie, myopathy [approximately one extra there is widespread disinformation and confusion
case per 10 000 person-years]; or, in a more severe form, among patients about statin safety.13
rhabdomyolysis [approximately 2–3 cases per To provide more reliable information about the size,
100 000 person-years]) as indicated by muscle symptoms severity, and timing of any adverse effects caused by
accompanied by biochemical changes (eg, multi-fold various statin regimens, the Cholesterol Treatment
rises in creatine kinase).2 However, there are concerns Trialists’ (CTT) Collaboration sought new individual
regarding statin-related muscle adverse effects,7 although participant data on all recorded adverse events, together
reviews of the data from randomised trials (including with supporting information about the methods used to
N-of-1 trials) have shown that most such muscle record and classify such events in each trial.14 The current
symptoms are due to the so-called nocebo or drucebo8 analyses were restricted to trials in which the treatment
effect (that is, that they are not generally due to the was double blinded (to minimise reporting biases). We
statin).2,9 Moreover, on the basis of non-randomised selected such trials to avoid the limitations of
observational studies of routine health-care records, it observational studies, notably their susceptibility to
has been suggested that statin therapy is associated with biases and confounding.2 We also aimed to minimise
large excess risks of musculoskeletal disorders;10–12 and, other methodological limitations (eg, the selective
although such studies are susceptible to statistical bias reporting of adverse events) that might have biased
and confounding,2 they are often cited without reference previous meta-analyses of randomised trials using only

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 833

 Articles

published data.14 The aim of the current meta-analysis biochemical evidence of muscle damage, and assessed
was to evaluate the effects of statin therapy on muscle the effect of the statin on the distribution of all creatine
effects of differing severity and to explore how any excess kinase values (reported as multiples of the trial-specific
risks varied over time, in different types of individual, upper limit of normal [ULN]). Although the protocol
and for different statin regimens. specified that myopathy would be defined as muscle pain
or weakness accompanied by creatine kinase more than
Methods ten times the ULN, it was redefined (before unmasking)
Search strategy and selection criteria as any event coded as myopathy or rhabdomyolysis owing
Methods and analyses were prespecified.14,15 Briefly, all to the frequent absence of creatine kinase data.
trials of statin therapy with more than 1000 participants The log-rank observed minus expected statistic (O–E)
and a scheduled mean follow-up of 2 years or more were and its variance were calculated for the first occurrence of
eligible if they involved a double-blind comparison of each outcome among participants randomly assigned into
statin versus placebo or of more intensive versus less each trial. The inverse-variance-weighted average of the
intensive statin regimens. Statin intensity (relative log of the rate ratio (log RR) across all trials was then
reduction in LDL cholesterol from baseline) for each calculated as S/V (with variance 1/V, and hence with a
statin regimen was defined according to the American 95% CI of S/V ±1·96/√V), where S is the sum of (O−E)
Heart Association and American College of Cardiology and V is the sum of variance over all trials.18,19 Prespecified
guidelines, with a low intensity classified as a less than subgroup analyses involved analyses within particular
30% LDL cholesterol reduction, moderate intensity as a participant characteristics (age, gender, race and ethnicity,
30 to less than 50% reduction, and high intensity as a history of vascular disease, history of diabetes, BMI, LDL
See Online for appendix 50% or more reduction (appendix p 6).16 The main results cholesterol, and estimated glomerular filtration rate),
were estimated separately for less intensive, moderate- analyses by year of treatment, and analyses of different
intensity, and more intensive statin regimens. Analyses statin regimens or intensities, or both. Tests for
involved only unconfounded trials (ie, those in which heterogeneity (or trend) across levels of each subgroup
there were no protocol-mandated differences between were performed. Further post-hoc subgroup analyses
randomly assigned groups other than those created by involved analyses of trials subdivided by whether they had
the random allocations). We requested individual an active run-in period, placebo run-in period, or no run-
participant data on all adverse events, including the in period before random assignment and by statin
timing of such events, the timing of and reasons for solubility (ie, hydrophilic vs lipophilic). For exploratory
stopping study treatment, non-trial statin use, use of analyses of the effects of statins on all events (ie, not just
other (non-trial) medications, comorbidities, and the first event), a negative binomial regression model
laboratory results (including creatine kinase for those (which provides additional flexibility compared with
regarding muscle; appendix p 2). Data on the exclusion Poisson regression) was fitted in each trial to estimate the
criteria of potential relevance to previous intolerance to a log RR and its SE. These estimates were then combined in
statin are shown in the appendix (p 3). Ethics approval a meta-analysis using the standard inverse-variance-
was granted by the UK National Health Service Health weighted method.
Research Authority (21/SC/0071). Only two trials allowed for a direct assessment of a more
intensive statin regimen versus placebo, but an indirect
Statistical analysis assessment of the effects of more intensive statin therapy
All variables for which the data were extracted were was made by combining the log event RR from the 15 trials
specified previously.15 Data were converted into a common of a moderate-intensity statin regimen versus placebo
format on the basis of the Clinical Data Interchange (SA/VA) with the log event RR from the two trials of more
Standards Consortium Study Data Tabulation Model.17 intensive versus moderate-intensity statin therapy (SB/VB).
Adverse events were mapped to a common dictionary (the Specifically, the log event RR for more intensive statin
For the Medical Dictionary for Medical Dictionary for Regulatory Activities version 20.0). therapy versus placebo was estimated indirectly by
Regulatory Activities see https:// The protocol prespecified that analyses of muscle SA/VA+SB/VB (which had the variance 1/VA+1/VB). The
www.meddra.org/
symptoms would consider reported muscle pain overall estimate of the effect of more intensive statin
(ie, myalgia) and weakness separately from myopathy. therapy versus placebo was then calculated as the inverse-
Before the unmasking of treatment allocation, variance-weighted average of the direct and indirect
non-myopathic muscle outcomes were categorised as: estimates.
myalgia, limb pain, other musculoskeletal pain, muscle To estimate the average absolute effect of statin therapy
cramp or spasm, any muscle pain (ie, the combination of on the underlying rate of specific outcomes in these trials,
the previous four outcomes), muscular fatigue or we applied the RR (or its lower and upper 95% CI limits)
weakness, and any muscle pain or weakness (ie, all the to the absolute rate in the appropriate control group. The
previous outcomes combined; appendix pp 4–5). We percentage of such outcomes reported by those who were
searched for creatine kinase values obtained within allocated statin therapy that could be attributed to such
2 weeks of the reported events to check for any therapy was calculated as 100 × (RR–1)/RR. Note that this

834 www.thelancet.com Vol 400 September 10, 2022

 Year of Number of Treatment Median Baseline LDL Baseline Number of Number of Number of Number of Baseline Baseline estimated
publication of patients comparison, mg follow-up, cholesterol, age, years women (%) White participants with a participants BMI, kg/m2 GFR, mL per min per
primary per day years mmol/L (mean participants history of vascular with a history (mean [SD]) 1·73 m2 (mean [SD])
results (mean [SD]) [SD]) (%) disease (%) of diabetes (%)
Statin versus placebo
4S20 1994 4444 Simvastatin 5·4 4·9 (0·7) 59 (7) 827 (19%) NA 4444 (100%) 202 (5%) 26·0 (3·3) NA
(20–40 mg) vs
placebo
WOSCOPS21 1995 6595 Pravastatin (40 mg) 4·8 5·0 (0·5) 55 (6) 0 NA 1066 (16%) 77 (1%) 26·0 (3·2) 77·8 (12·4)
vs placebo
CARE22 1996 4159 Pravastatin (40 mg) 4·9 3·6 (0·4) 59 (9) 576 (14%) 3851 (93%) 4159 (100%) 586 (14%) 27·6 (4·4) 67·2 (15·7)
vs placebo
AFCAPS/TexCAPS23 1998 6605 Lovastatin 5·0 3·9 (0·4) 58 (7) 997 (15%) 5860 (89%) 0 155 (2%) 26·9 (3·1) 65·4 (11·6)
(20–40 mg) vs
placebo
LIPID24 1998 9014 Pravastatin (40 mg) 5·9 3·9 (0·8) 61 (8) 1516 (17%) NA 9014 (100%) 782 (9%) 26·8 (3·8) 70·6 (16·3)
vs placebo
LIPS25 2002 1677 Fluvastatin (80 mg) 4·0 3·4 (0·8) 60 (10) 271 (16%) 1650 (98%) 1677 (100%) 202 (12%) 26·5 (3·3) 67·6 (15·5)
vs placebo

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022

HPS26 2002 20 536 Simvastatin (40 mg) 5·2 3·4 (0·8) 64 (8) 5082 (25%) 19 901 (97%) 17 386 (85%) 5963 (29%) 27·6 (4·4) 72·2 (16·5)
vs placebo
PROSPER27 2002 5804 Pravastatin (40 mg) 3·3 3·8 (0·8) 75 (3) 3000 (52%) NA 2565 (44%) 623 (11%) 26·8 (4·2) 56·7 (13·6)
vs placebo
ASCOT-LLA28 2003 10 240 Atorvastatin 3·3 3·4 (0·7) 63 (9) 1919 (19%) 9687 (95%) 1684 (16%) 2540 (25%) 28·6 (4·6) 68·4 (12·9)
(10 mg) vs placebo
ALERT29 2003 2102 Fluvastatin 5·5 4·1 (1·0) 50 (11) 715 (34%) 2039 (97%) 409 (19%) 396 (19%) 25·8 (4·5) 49·6 (17·0)
(40–80 mg) vs
placebo
CARDS30 2004 2838 Atorvastatin 4·2 2·9 (0·8) 61 (8) 909 (32%) 2676 (94%) 106 (4%) 2838 (100%) 28·8 (3·6) 64·2 (11·3)
(10 mg) vs placebo
4D31 2005 1255 Atorvastatin 2·7 3·3 (0·8) 66 (8) 578 (46%) 924 (74%) 1041 (83%) 1255 (100%) 27·6 (4·8) NA
(20 mg) vs placebo
ASPEN32 2006 2410 Atorvastatin 4·0 2·9 (0·7) 60 (8) 811 (34%) 2029 (84%) 747 (31%) 2410 (100%) 28·9 (3·8) 65·9 (12·8)
(10 mg) vs placebo
SPARCL33 2006 4731 Atorvastatin 4·9 3·5 (0·6) 63 (11) 1908 (40%) 4415 (93%) 4731 (100%) 794 (17%) 27·9 (5·2) 65·2 (13·8)
(80 mg) vs placebo
CORONA34 2007 4982 Rosuvastatin 2·7 3·6 (0·9) 72 (7) 1175 (24%) NA 4982 (100%) 1473 (30%) 26·4 (3·6) 55·4 (15·1)
(10 mg) vs placebo
GISSI-HF35 2008 4574 Rosuvastatin 3·9 3·1 (0·9) 68 (11) 1032 (23%) 4574 (100%) 4574 (100%) 1196 (26%) 27·1 (4·5) 66·3 (20·4)
(10 mg) vs placebo
JUPITER36 2008 16 714 Rosuvastatin 1·9 2·7 (0·5) 65 (8) 6374 (38%) NA 0 44 (<1%) 27·5 (3·6) 72·3 (14·8)
(20 mg) vs placebo
AURORA37 2009 2555 Rosuvastatin 3·9 2·6 (0·9) 64 (9) 969 (38%) NA 1025 (40%) 658 (26%) 24·8 (3·9) NA
(10 mg) vs placebo
HOPE-338 2016 12 705 Rosuvastatin 5·5 3·3 (0·9) 66 (6) 5874 (46%) 2546 (20%) 0 731 (6%) 27·1 (4·7) 79·6 (16·1)
(10 mg) vs placebo
Subtotal of the 19 NA 123 940 NA 4·3 3·5 (0·7) 63 (8) 34 533 (28%) 60 152 (81%)* 59 610 (48%) 22 925 (18%) 27·2 (4·1) 69·5 (15·0)
statin versus
placebo trials
(Table 1 continues on next page)
Articles

835
836
Articles

Year of Number of Treatment Median Baseline LDL Baseline Number of Number of Number of Number of Baseline Baseline estimated
publication of patients comparison, mg follow-up, cholesterol, age, years women (%) White participants with a participants BMI, kg/m2 GFR, mL per min per
primary per day years mmol/L (mean participants history of vascular with a history (mean [SD]) 1·73 m2 (mean [SD])
results (mean [SD]) [SD]) (%) disease (%) of diabetes (%)
(Continued from previous page)
More intensive versus less intensive statin regimen (double blind)
PROVE-IT39 2004 4162 Atorvastatin 2·1 2·6 (0·7)† 58 (11) 911 (22%) 3776 (91%) 4162 (100%) 762 (18%) 29·5 (5·7) 78·8 (18·7)
(80 mg) vs
pravastatin (40 mg)
A to Z40 2004 4497 Simvastatin (40 mg) 2·0 2·1 (0·7)† 60 (11) 1100 (24%) 3825 (85%) 4497 (100%) 1059 (24%) 27·6 (4·8) 68·4 (16·0)
then simvastatin
(80 mg) vs placebo
then simvastatin
(20 mg)
TNT41 2005 10 001 Atorvastatin 5·0 2·5 (0·5) 61 (9) 1902 (19%) 9410 (94%) 10001 (100%) 1501 (15%) 28·8 (6·1) 65·0 (12·4)
(80 mg) vs
atorvastatin (10 mg)
SEARCH42 2010 12 064 Simvastatin (80 mg) 7·0 2·5 (0·6) 64 (9) 2052 (17%) 11 854 (98%) 12 064 (100%) 1267 (11%) 28·1 (4·1) 77·2 (17·1)
vs simvastatin
(20 mg)
Subtotal of the four NA 30 724 NA 4·9 2·5 (0·6) 62 (9) 5965 (19%) 28 865 (94%) 30 724 (100%) 4589 (15%) 28·4 (5·1) 72·2 (15·6)
more intensive
versus less intensive
trials
Total of all trials NA 154 664 NA 4·4 3·1 (0·7) 63 (8) 40 498 (26%) 89 017 (85%)* 90 334 (58%) 27 514 (18%) 27·5 (4·3) 70·1 (15·1)
Estimated GFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation except for CORONA and JUPITER, where the estimated GFR was supplied by trialists (and the absence of data meant that the estimated GFR could not
be derived by the Cholesterol Treatment Trialists’ Collaboration); estimated GFR statistics were not presented for 4D and AURORA because these were trials done in participants who were on regular maintenance dialysis. AURORA, CORONA, and
JUPITER supplied age in categorical bands, and so the midpoint of each categorical band was used as a surrogate for baseline age in all analyses. A small number of participants in the AURORA (n=218), CORONA (n=27), and JUPITER (n=1088) trials
withdrew consent for use of their data post-trial, and hence data from these participants is excluded. The ASCOT-LLA trial excludes 65 patients for whom data were not available due to protocol violations. Note that some baseline characteristics might
differ from previous publications because of the receipt of updated data and a broader definition of baseline vascular disease being applied in these analyses. 4D=ie Deutsche Diabetes Dialyse Studie. 4S=Scandinavian Simvastatin Survival Study.
AFCAPS/TexCAPS=Air Force/Texas Coronary Atherosclerosis Prevention Study. ALERT=Assessment of Lescol in Renal Transplantation. ASCOT-LLA=Anglo-Scandinavian Cardiac Outcomes Trial-Lipid Lowering Arm. ASPEN=Atorvastatin Study for
Prevention of Coronary Heart Disease Endpoints in Non-Insulin-Dependent Diabetes Mellitus. A to Z=Aggrastat to Zocor. AURORA=A Study to Evaluate the Use of Rosuvastatin in Subjects on Regular Hemodialysis: An Assessment of Survival and
Cardiovascular Events. CARDS=Collaborative Atorvastatin Diabetes Study. CARE=Cholesterol And Recurrent Events. CORONA=Controlled Rosuvastatin Multinational Trial in Heart Failure. GFR=glomerular filtration rate. GISSI-HF=Gruppo Italiano per lo
Studio della Sopravvivenza nell’Insufficienza cardiaca. HOPE-3=Heart Outcomes Prevention Evaluation-3 trial. HPS=Heart Protection Study. JUPITER=Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin study
group. LDL=low density lipoprotein. LIPID=Long-term Intervention with Pravastatin in Ischaemic Disease. LIPS=Lescol Intervention Prevention Study. NA=not available. PROSPER=PROspective Study of Pravastatin in the Elderly at Risk. PROVE-
IT=Pravastatin or Atorvastatin Evaluation and Infection Therapy. SEARCH=Study of the Effectiveness of Additional Reductions in Cholesterol and Homocysteine. SPARCL=Stroke Prevention by Aggressive Reduction in Cholesterol Levels. TNT=Treating
to New Targets. WOSCOPS=West of Scotland Coronary Prevention Study. *Percentages were calculated after excluding the seven trials where information on race and ethnicity was not provided (the relevant denominators are therefore 73 832 for the
12 trials of statin vs placebo and 104 556 for all 16 trials with information on race and ethnicity). †These two trials did not have active run–in periods; the values shown are the estimated on-treatment LDL cholesterol concentrations in the standard
statin group.

Table 1: Characteristics of the trials included in the meta-analysis

www.thelancet.com Vol 400 September 10, 2022

 Articles

attributable proportion is unaffected by the absolute rate 59 610 [48%] participants with previous vascular disease,
of muscle adverse event reporting in the underlying and 22  925 [18%] participants with diabetes),
population. 16 835 (27·1%) participants assigned a statin versus
We used the Wilcoxon rank-sum test to compare 16 446 (26·6%) participants assigned placebo reported at
creatine kinase measurements recorded during follow- least one episode of muscle pain or weakness during a
up between randomly assigned groups. Overall RRs were median of 4·3 years. This corresponded to a 3% relative
reported with 95% CIs, but all other RRs were reported increase (RR 1·03; 95% CI 1·01–1·06; figure 1;
with 99% CIs to provide some allowance for multiple appendix p 9). The RRs were similar (heterogeneity
comparisons. Analyses of all categories of muscle p=0·43) for each term used to categorise muscle
symptoms were done by intention-to-treat analysis and symptoms (myalgia, 1·03 [0·99–1·08]; limb pain,
implemented using SAS version 9.4 and R version 4.1.3. 1·00 [0·92–1·09]; other musculoskeletal pain, 1·03 For R see www.R-project.org
[0·99–1·08]; muscle cramp or spasm, 1·09 [1·00–1·19];
Role of the funding source any muscle pain, 1·03 [1·01–1·06]; and muscle fatigue or
The funder of the study had no role in study design, data weakness, 1·10 [0·92–1·31]). Figures showing trial
collection, data analysis, data interpretation, or writing of specific findings for each category of muscle event can be
the report downloaded from the CTT website. For the CTT website see https://
Statin therapy produced a 7% relative increase in www.cttcollaboration.org

Results muscle pain or weakness during the first year (RR 1·07;
Individual participant data were available from 95% CI 1·04–1·10), but no significant increase thereafter
19 randomised double-blind trials of any statin regimen (0·99; 0·96–1·02; p value for heterogeneity for year 1 vs
versus placebo (123 940 patients) and four randomised all subsequent years was 0·0005; figure 2). An increased
double-blind trials of more intensive versus less intensive risk was already present within the first 3 months after
statin therapy (30 724 patients; table 1). Of the 19 double- random assignment to treatment (1·08, 99% CI
blind trials of any statin regimen versus placebo, one study23 1·02–1·15; figure 2). The aggregate rate of reporting of
compared a less intensive statin regimen versus placebo any muscle pain or weakness (ie, for statin and placebo
(6605 patients), 16 studies20–22,24–32,34,35,37,38 compared a groups combined) during the first year of treatment
moderate-intensity statin regimen versus placebo varied substantially by trial (appendix p 10), from 1·0%
(95 890 patients), and two studies33,36 compared a more to 60·5%, reflecting heterogeneity in how actively
intensive statin regimen versus placebo (21 445 patients). information was sought. Despite this, there was no
In the 19 double-blind trials of statin versus placebo evidence that the RR for the comparison of statin versus
(mean age 63 years [SD 8], with 34 533 [28%] women, placebo depended on the absolute reporting rate. By

Events (%) O−E Var(O−E) Rate ratio (95% CI or 99% CI)

Statin or more Placebo or less

intensive statin intensive statin
group group

(A) Statin vs placebo (n=62 028) (n=61 912)

Myalgia 7446 (12·0%) 7269 (11·7%) 120·1 3657·4 1·03 (0·99–1·08)
Limb pain 1850 (3·0%) 1836 (3·0%) 3·6 921·3 1·00 (0·92–1·09)
Other musculoskeletal pain 8245 (13·3%) 8037 (13·0%) 131·3 4066·1 1·03 (0·99–1·08)
Muscle cramp or spasm 1697 (2·7%) 1553 (2·5%) 71·2 812·4 1·09 (1·00–1·19)
Any muscle pain 16 656 (26·9%) 16 281 (26·3%) 274·8 8206·8 1·03 (1·01–1·06)
Muscle fatigue or weakness 445 (0·7%) 406 (0·7%) 19·4 212·7 1·10 (0·92–1·31)
Any muscle pain or weakness 16 835 (27·1%) 16 446 (26·6%) 283·1 8292·7 1·03 (1·01–1·06)

(B) More vs less intensive statin (n=15 390) (n=15 334)

Myalgia 3485 (22·6%) 3380 (22·0%) 70·2 1712·7 1·04 (0·98–1·11)
Limb pain 619 (4·0%) 603 (3·9%) 7·3 305·5 1·02 (0·88–1·19)
Other musculoskeletal pain 1721 (11·2%) 1628 (10·6%) 47·8 836·9 1·06 (0·97–1·16)
Muscle cramp or spasm 515 (3·3%) 495 (3·2%) 10·3 252·5 1·04 (0·89–1·22)
Any muscle pain 5490 (35·7%) 5274 (34·4%) 132·0 2686·6 1·05 (1·01–1·09)
Muscle fatigue or weakness 158 (1·0%) 148 (1·0%) 4·9 76·5 1·07 (0·79–1·43)
Any muscle pain or weakness 5558 (36·1%) 5342 (34·8%) 132·8 2720·5 1·05 (1·01–1·09)
99% CI 95% CI
0·8 1·0 1·2 1·4 1·6

Favours statin or more intense statin Favours placebo or less intense statin

Figure 1: Effect on muscle adverse events in trials of any statin regimen versus placebo (A) and more versus less intensive statin regimens (B)
Bold data are the totals or subtotals. O–E=observed minus expected. Var=variance.

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 837

 Articles

Events (%) O−E Var(O−E) Rate ratio (95% CI or 99% CI)

Statin (n=62 028) Placebo (n=61 912)

Months 0 to ≤3 3931 (6·3%) 3655 (5·9%) 152·8 1879·5 1·08 (1·02–1·15)

Months >3 to ≤6 2410 (4·2%) 2252 (3·9%) 91·4 1161·6 1·08 (1·00–1·17)
Months >6 to ≤9 1454 (2·6%) 1344 (2·4%) 64·2 698·3 1·10 (0·99–1·21)
Months >9 to ≤12 1404 (2·6%) 1422 (2·7%) 1·4 705·3 1·00 (0·91–1·10)
Year 1 9199 (14·8%) 8673 (14·0%) 309·8 4444·7 1·07 (1·04–1·10)
Year 2 3812 (7·4%) 3865 (7·5%) 1·1 1916·9 1·00 (0·94–1·06)
Year 3 2169 (5·1%) 2272 (5·3%) −35·8 1109·1 0·97 (0·90–1·05)
Year 4 1066 (3·0%) 1056 (3·0%) 7·1 530·0 1·01 (0·91–1·13)
Year 5+ 589 (2·1%) 580 (2·1%) 0·9 292·0 1·00 (0·86–1·17)
All after year 1 7636 (14·8%) 7773 (15·0%) −26·7 3848·0 0·99 (0·96–1·02)
All years 16 835 (27·1%) 16 446 (26·6%) 283·1 8292·7 1·03 (1·01–1·06)

99% CI 95% CI
0·7 0·8 0·9 1·0 1·2 1·4

Statin better Statin worse

Figure 2: Effect on any muscle pain or weakness, by duration of treatment, in trials of any statin regimen versus placebo
Bold data are the totals or subtotals. White squares indicate months, black squares indicate years. The test for heterogeneity in the log rate ratio between the first year
and all subsequent years combined: χ2 =12·1, p=0·0005. For each risk period, percentages shown are of those alive and still at risk of a first report of muscle pain or
weakness at the start of the risk period. O–E=observed minus expected. Var=variance.

applying the RRs in the first and subsequent years to p=0·0019; figure 4) and during the first year alone when
the average rates over these periods among patients most of the excess risk was observed (so subgroup
allocated to the placebo group, allocation to a statin analysis should be most sensitive to variation in risk;
regimen resulted in an absolute excess rate of muscle heterogeneity p=0·012; appendix p 14). RRs did not
pain or weakness of 11 (95% CI 6 to 16) per 1000 person- differ significantly by other patient characteristics
years in the first year and 0 per 1000 person-years (figure 4; appendix p 14), or within groups of trials that
(–2 to 1) in subsequent years (figure 3). Approximately 1 exclusively recruited patients with major underlying
in 15 (calculated as [1·07–1·00]/1·07) reports of muscle health conditions (data not shown).
symptoms (<10% of reports) during the first year were Analyses of statin intensity used data from two
attributable to statin therapy. randomised trials of more intensive statin therapy
When trials were subdivided by statin and statin dose, versus placebo33,36 combined with data from four
there was no evidence that the summary RRs varied randomised trials of more intensive versus less
significantly among different statins (heterogeneity intensive statin therapy that included 30 724 participants
p=0·10; appendix p 11). Nor was there any clear dose (median duration 4·9 years, mean age 62 years; SD 9
relationship for any specific statin (p value for trend all years; all with known vascular disease; table 1). As
non-significant). When categorised by statin intensity compared with a moderate-intensity regimen, more
(appendix p 6), less intensive and moderate-intensity intensive statin regi­ mens resulted in a 5% relative
regimens yielded a 3% relative increase in the rate of first increase in the rate of reporting of a first episode of
reports of muscle pain or weakness (RR 1·03; 95% CI muscle pain or weakness (two trials RR 1·05; 95% CI
1·00–1·05) with a 6% increase in the first year (1·06; 0·99–1·11; table 2; appendix pp 7,15).39,41 There was a
1·03–1·10) but no increase thereafter (0·98; 0·95–1·02; similar 5% relative increase in the comparisons of
appendix p 12). There was no evidence for differences in different statin regimens, where both regimens fell
the RR between different statins in year 1 (less intensive within the moderate-intensity range (two trials 1·05;
or moderate-intensity statin therapy, heterogeneity 1·00–1·11; appendix p 15).40,42 There were broadly
p=0·50; more intensive statin therapy, heterogeneity similar increases in any muscle pain or weakness for
p=0·28). any type of more-intensive versus less-intensive statin
The RR in year 1 did not appear to differ according to therapy, both in the first and subsequent years
whether there was an active or placebo pre- (heterogeneity p=0·61; appendix p 16). There was no
randomisation run-in period or no run-in, nor did it evidence that the increase in muscle symptoms varied
differ between whether the statin was hydrophilic or significantly within any of the subgroups, including sex
lipophilic (appendix p 13). The RR for muscle pain or (heterogeneity p=0·61; appendix p 17). To obtain a
weakness was greater in women for less intensive and reliable estimate of the effects of more intensive regimens,
moderate-intensity statin regimens, both during the we combined the results of a direct comparison of more
whole follow-up period (RRs 1·09, 99% CI 1·03–1·16 in intensive statin regimens versus placebo in two trials33,36
women vs 1·00, 0·97–1·04 in men; heterogeneity with the results of an indirect comparison involving

838 www.thelancet.com Vol 400 September 10, 2022

 Articles

A Year 1 Rate ratio (95% CI) Absolute excess rate

per 1000 person-years
(95% CI)

83 Statin (n=62 028)

Myalgia Placebo (n=61 912) 1·06 (1·02 to 1·11) 5 (1 to 8)
78
11
Limb pain 1·08 (0·97 to 1·21) 1 (0 to 2)
10
59
Other musculoskeletal pain 1·08 (1·03 to 1·13) 4 (1 to 7)
55
12
Muscle cramp or spasm 1·14 (1·02 to 1·27) 1 (0 to 3)
10
164
Any muscle pain 1·07 (1·04 to 1·10) 11 (6 to 16)
153
3
Muscle fatigue or weakness 1·14 (0·92 to 1·40) 0 (0 to 1)
3
166
Any muscle pain or weakness 1·07 (1·04 to 1·10) 11 (6 to 16)
155

0 50 100 150 200 250

Absolute rate per 1000 person-years

B After first year Rate ratio (95% CI) Absolute excess rate
per 1000 person-years
(95% CI)

15
Myalgia 0·99 (0·93 to 1·04) 0 (−1 to 1)
15
6
Limb pain 0·96 (0·89 to 1·04) 0 (−1 to 0)
6
26
Other musculoskeletal pain 1·00 (0·96 to 1·04) 0 (−1 to 1)
26
5
Muscle cramp or spasm 1·06 (0·97 to 1·16) 0 (0 to 1)
5
49
Any muscle pain 0·99 (0·96 to 1·03) 0 (−2 to 1)
50
1
Muscle fatigue or weakness 1·07 (0·89 to 1·27) 0 (0 to 0)
1
50
Any muscle pain or weakness 0·99 (0·96 to 1·02) 0 (−2 to 1)
50
0 50 100 150 200 250
Absolute rate per 1000 person-years

Figure 3: Rate ratio and absolute rate difference for muscle adverse events by duration of treatment, in trials of any statin regimen versus placebo

two trials of more intensive versus moderate-intensity A creatine kinase concentration was available in less
regimens39,41 and 16 trials of moderate-intensity regimens than 6·2% of reports of muscle pain or weakness, but
versus placebo.20–22,24–32,34,35,37,38 The direct comparison of for 96·7% of those cases the concentrations were less
more intensive statin regimens versus placebo during a than 3 times the ULN. After excluding those with a
median follow-up of 2·6 years yielded, for the rate of recorded myopathy event, allocation to statin therapy
reporting a first episode of muscle pain or weakness, an resulted in a small (approximately 0·02 times the ULN)
RR of 1·09 (95% CI 1·03–1·16; table 2, appendix pp 7, 12). increase in median creatine kinase values during
In the indirect comparison, the estimated RR for more follow-up: 0·43 (IQR 0·30–0·62) times the ULN for a
intensive statin was 1·07 (1·01–1·14; table 2). Combining moderate-intensity regimen versus 0·41 (0·29–0·59)
these direct and indirect comparisons yielded an overall times the ULN for placebo (Wilcoxon test p<0·0001),
RR of muscle pain or weakness with more intensive statin and 0·49 (IQR 0·34–0·71) times the ULN for high
regimens of 1·08 (1·04–1·13), with RRs of 1·11 (1·05–1·17) intensity regimen versus 0·45 (IQR 0·31–0·65) times
in year 1 and 1·05 (0·99–1·12) after 1 year. ULN for placebo (Wilcoxon test p<0·0001; appendix p 18).
The RRs over the whole follow-up period for the effects Only one trial26 provided information on treatment
of moderate-intensity or more intensive statin regimens adherence in relation to muscle symptoms. In both
on all reports (ie, not just the first report) of muscle pain patients allocated simvastatin and patients allocated
or weakness were similar to the RRs for first events only placebo, the distribution of recorded adherence after the
(appendix p 7). The RRs were also similar when analyses year 1 visit was nearly identical among those who did
were restricted to year 1 or to the period after year 1 versus those who did not report muscle pain or weakness
(appendix p 7). within the first year (appendix p 8).

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 839

 Articles

Events (%) O−E Var(O−E) Rate ratio (95% CI or 99% CI)

Statin (n=51 288) Placebo (n=51 207)

Age, years (Trend: χ21=1·2, p=0·28)

≤65 7887 (26·9%) 7800 (26·7%) 45·5 3908·8 1·01 (0·97–1·05)
>65 to ≤75 4757 (27·0%) 4619 (26·2%) 106·3 2333·3 1·05 (0·99–1·10)
>75 1707 (39·7%) 1720 (39·6%) 30·8 851·5 1·04 (0·95–1·13)
Sex (Heterogeneity: χ21=9·6, p=0·0019)
Female 3901 (29·6%) 3665 (28·0%) 163·3 1883·0 1·09 (1·03–1·16)
Male 10 458 (27·4%) 10 482 (27·5%) 17·7 5215·4 1·00 (0·97–1·04)
Race or ethnicity (Heterogeneity: χ23=2·1, p=0·55)
Asian 321 (8·9%) 287 (8·0%) 17·5 151·1 1·12 (0·91–1·38)
Black 110 (19·2%) 113 (19·8%) 1·3 55·0 1·02 (0·72–1·45)
White 8549 (30·6%) 8442 (30·3%) 76·5 4225·8 1·02 (0·98–1·06)
Other 210 (9·0%) 196 (8·4%) 10·7 101·1 1·11 (0·86–1·44)
Missing 5169 (30·6%) 5109 (30·3%) 74·0 2564·8 1·03 (0·98–1·08)
History of vascular disease (Heterogeneity: χ21=2·9, p=0·089)
No 5745 (24·1%) 5593 (23·5%) 140·3 2828·8 1·05 (1·00–1·10)
Yes 8614 (31·3%) 8554 (31·2%) 35·8 4268·9 1·01 (0·97–1·05)
History of diabetes (Heterogeneity: χ21=0·3, p=0·57)
No 11 090 (27·6%) 10 988 (27·3%) 117·1 5498·4 1·02 (0·99–1·06)
Yes 3269 (29·4%) 3159 (28·8%) 60·2 1599·4 1·04 (0·97–1·11)
BMI, kg/m2 (Trend: χ21=3·1, p=0·078)
≤25 4402 (27·3%) 4500 (27·4%) −0·4 2216·1 1·00 (0·95–1·06)
>25 to ≤30 6538 (27·6%) 6322 (27·2%) 87·8 3201·6 1·03 (0·98–1·08)
>30 3387 (29·9%) 3301 (29·1%) 94·4 1662·3 1·06 (0·99–1·13)
LDL cholesterol mmol/L (Trend: χ21=0·1, p=0·81)
≤3·0 3524 (28·0%) 3441 (27·4%) 73·5 1731·7 1·04 (0·98–1·11)
>3·0 to ≤4·0 5489 (26·7%) 5479 (26·6%) 30·8 2730·5 1·01 (0·96–1·06)
>4·0 5025 (30·5%) 4904 (29·8%) 77·4 2473·6 1·03 (0·98–1·09)
Estimated GFR, mL per min per 1·73 m2 (Trend: χ21=1·4, p=0·24)
On dialysis 432 (22·7%) 450 (23·6%) −4·8 220·3 0·98 (0·82–1·16)
≤60 4280 (31·6%) 4252 (31·1%) 55·1 2124·4 1·03 (0·97–1·09)
>60 to ≤90 7532 (28·9%) 7374 (28·4%) 82·5 3710·2 1·02 (0·98–1·07)
>90 1504 (26·0%) 1423 (25·0%) 60·4 726·3 1·09 (0·99–1·20)
Total 14 359 (28·0%) 14 147 (27·6%) 178·6 7099·6 1·03 (1·00–1·05)
99% CI 95% CI
0·7 0·8 0·9 1·0 1·2 1·4

Statin better Statin worse

Figure 4: Effect of less intensive or moderate-intensity statin regimens on any muscle pain or weakness, by participant characteristics
Bold indicates the overall summary result. White squares indicate missing data. Tests of heterogeneity (or trend) listed after each prognostic characteristic are of the
log rate ratio for each of the subgroups of that characteristic, and are uncorrected for multiple comparisons. GFR=glomerular filtration rate. LDL=low-density
lipoprotein. O–E=observed minus expected. Var=variance.

Myopathy (based on Medical Dictionary for Regulatory myopathy with 80 mg atorvastatin once per day, but
Activities coding) was reported by 0·08% of those there was clear evidence of an excess with 80 mg
assigned any statin regimen versus 0·04% of those simvastatin once per day (which is no longer approved)
assigned placebo (RR 1·74; 95% CI 1·11–2·74, p=0·016; in the SEARCH trial42 compared with 20 mg simvastatin
appendix p 19), which corresponded to an absolute once per day (6·45; 99% CI 3·26–12·77; appendix p 20).
excess of 0·08 (0·01–0·18) per 1000 person-years. The
RR was 3·04 (1·43–6·47; p=0·0039) in year 1 and 1·28 Discussion
(0·72–2·25; p=0·40) after year 1; thus, in year 1, statin Previous randomised trials and meta-analyses of trials,43–47
therapy was the cause of approximately two-thirds as well as N-of-1 trials of statin therapy,48,49 have not been
([3·04–1·00]/3·04) of myopathy cases reported by able to resolve whether statin therapy might lead to a
patients allocated to a statin. The RRs for less intensive small increased risk of muscle pain. We aimed to evaluate
and moderate-intensity (1·73 [1·05–2·87]) compared the causal effects of statin therapy on muscle events of
with more intensive regimens (1·78 [0·63–5·08]) were differing types and severity, and to explore how any
similar (appendix p 19). In analyses of more intensive excess risk varied over time, in different types of
versus less intensive regimens, there were few reports of individual, and for different statin regimens. On the

840 www.thelancet.com Vol 400 September 10, 2022

 Articles

basis of individual participant data on all recorded

Events (% py) Rate ratio (95% CI)
muscle events in large-scale, long-term statin trials, our
study yielded many findings that enhanced Statin or more Placebo or less
intensive statin intensive statin
understanding of the nature and risks of muscle group group
symptoms caused by statin therapy.
Direct assessments, 0–1 year
Notably, we have reliably shown that the small excess
High-intensity statin vs placebo (2 trials) 1495 (15·4% py) 1351 (13·8% py) 1·11 (1·03–1·20)
risk of muscle symptoms due to statin therapy largely
Moderate-intensity statin vs placebo 7668 (18·0% py) 7282 (17·0% py) 1·07 (1·03–1·10)
occurred within the first year of treatment. Other studies12 (16 trials)
have shown that the risks of statin-associated muscle High-intensity vs moderate-intensity 1259 (20·4% py) 1218 (19·6% py) 1·04 (0·96–1·12)
symptoms or statin intolerance are highest in the period (2 trials)
after commencing therapy with a statin, but since a large Indirect assessments, 0–1 year
proportion of such events are not due to statin therapy, High-intensity statin vs placebo NA NA 1·10 (1·01–1·20)
such studies do not help in furthering understanding of Overall (direct and indirect) assessments, 0–1 year
the causal role of statins over time. There were similar High-intensity statin vs placebo NA NA 1·11 (1·05–1·17)
excesses of risk in different types of muscle symptoms, Direct assessments, after 1 year
including those categorised clinically as myalgia, muscle High-intensity statin vs placebo (2 trials) 981 (6·8% py) 948 (6·4% py) 1·06 (0·97–1·16)
cramps or spasm, limb pain, other musculoskeletal pain, Moderate-intensity statin vs placebo 6616 (5·2% py) 6802 (5·4% py) 0·98 (0·95–1·02)
or muscle fatigue or weakness, but we found that muscle (16 trials)
symptoms caused by statin therapy were no more severe High-intensity vs moderate-intensity 1308 (8·5% py) 1248 (8·0% py) 1·06 (0·98–1·14)
than the average severity of symptoms not caused by a (2 trials)
statin. The different statins of equivalent LDL-lowering Indirect assessments, after 1 year
ability included in our meta-analysis had similar effects High-intensity statin vs placebo NA NA 1·04 (0·95–1·13)
on muscle symptoms. The relative excess was similar in Overall (direct and indirect) assessments, after 1 year
different types of patients, and irrespective of how High-intensity statin vs placebo NA NA 1·05 (0·99–1·12)
muscle symptoms were defined, hence our results are Direct assessments, all years
likely to be widely generalisable. Although we did not High-intensity statin vs placebo (2 trials) 2476 (10·3% py) 2299 (9·4% py) 1·09 (1·03–1·16)
find any clear evidence of a dose-response relationship, Moderate-intensity statin vs placebo 14 284 (8·4% py) 14 084 (8·3% py) 1·02 (1·00–1·05)
more intensive regimens caused a greater increase in (16 trials)
muscle symptoms than moderate-intensity regimens in High-intensity vs moderate- 2567 (11·9% py) 2466 (11·3% py) 1·05 (0·99–1·11)
the first year of treatment (11% vs 6%), and there was intensity (2 trials)

some evidence that a small excess with more intensive Indirect assessments, all years

regimens might persist for longer than 1 year. High-intensity statin vs placebo NA NA 1·07 (1·01–1·14)
Based on the evidence available from the Heart Overall (direct and indirect) assessments, all years
Protection Study, the small excess in risk appeared to be High-intensity statin vs placebo NA NA 1·08 (1·04–1·13)
because of events that did not usually lead to treatment The table excludes one trial of a low intensity statin versus placebo (AFCAPS/TexCAPS) and two trials that compared
discontinuation (consistent with evidence from a two moderate-intensity statin regimens (SEARCH and A to Z). High intensity statin versus placebo trials: JUPITER and
SPARCL. Moderate-intensity statin versus placebo trials: ALERT, ASCOT-LLA, ASPEN, AURORA, CARDS, CARE, CORONA,
previous review of tabular data from statin trials),2 and 4D, GISSI-HF, HOPE-3, HPS, LIPID, LIPS, PROSPER, WOSCOPS, and 4S. High intensity versus moderate-intensity, double
did not result in a clinically significant change in creatine blind trials: PROVE-IT and TNT. py=per year.
kinase. This finding indicates that most of the episodes
Table 2: Effects of moderate-intensity and more intensive statin regimens on any muscle pain or
of muscle pain or weakness caused by a statin were weakness by period of follow-up
clinically mild. This result also indicates that, since
participants with muscle symptoms generally continued
treatment, the absence of any significantly increased risk events at 1 year. It is not clear why this is so (other than
of repeat reports of muscle symptoms after year 1 is not the absence of statistical power) since it might be
explained by patients with symptoms stopping treatment expected that the RR should tend towards unity in
after having muscle symptoms during the first year. those trials with higher recorded rates because of a
We were able to explore the generalisability of our larger proportion of misclassified events (so-called
findings in several ways. Since most excess risk appears noise). However, when taken together with the general
in the period immediately after treatment commences, absence of variation in the relative effects of statin
we restricted our analyses to the first year to increase therapy on muscle pain or weakness in the subgroups
their sensitivity. There was substantial variation in the studied, one practical conclusion is that the overall RR
methods used to record muscle symptoms, so the first- might be broadly generalisable in different clinical
year rates among participants allocated to the placebo circumstances. Likewise, the absence of a sex-related
group in the 19 trials of statin versus placebo also difference in the trials comparing more versus less
varied substantially. Nevertheless, there was little intensive regimens does not support the finding of a
evidence that the RR for muscle pain or weakness higher RR in women than in men in the placebo-
varied according to the cumulative frequency of muscle controlled trials.

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 841

 Articles

Our analyses also help to address multiple concerns that patients. The data are derived from large-scale, double-
have been raised about previous analyses of the effects of blind, randomised trials, which guarantees the avoidance
statins on muscle symptoms.50 We found no evidence that of both moderate random errors and moderate biases.52
previous estimates based on published data were seriously By contrast, non-randomised observational studies, in
biased by the incomplete reporting of adverse events in which outcomes are compared between individuals who
trial reports. There was also no evidence that the use of an received the treatment of interest and those who did not
active or placebo run-in period led to the underestimation are (irrespective of their size) prone to moderate biases
of the risk of muscle symptoms attributable to statin (especially where participants are aware of which drugs
therapy. Finally, there was no evidence that risk estimates they are taking), which cannot be guaranteed to be
were biased by the exclusion of patients with removed through statistical adjustment (eg, propensity
comorbidities,50 since the relative risks were similar in the score matching).2,52
trials that had recruited exclusively from groups of patients However, there are some potential limitations to our
with underlying conditions, such as those with diabetes,30,32 analyses. Firstly, there was considerable heterogeneity in
with New York Heart Association grade 2–4 heart failure,34,35 the methods used for the ascertainment of muscle
or patients with chronic kidney disease undergoing symptoms, and definitions varied from trial to trial.
haemodialysis,31,37 with no significant heterogenity in rate Nonetheless, as noted above, the relative excesses
ratios between trials recruiting participants with these appeared to be broadly consistent among the different
conditions (data not shown). trials and clinical circumstances. Secondly, although we
In these trials, statin therapy caused approximately sought adverse event data from all randomised, double-
11 additional reports of any muscle pain or weakness per blind trials included in the CTT, some data were not
1000 patients during the first year, but little excess available. However, the missing data represented less
thereafter (although there was some evidence that a risk than 1% of all adverse events34,36,37 (mainly because of data
persisted beyond the first year for more intensive privacy concerns). Furthermore, data were not
regimens). This finding is in contrast with the consistently available on whether muscle events led to
cardiovascular benefits of statin therapy, which are the discontinuation of allocated treatment (and thus
observed in year 1, but which are then twice as high planned on-treatment analyses for rare events were not
during each subsequent year that treatment continues as possible), and there was no reliable information on some
previously reported (9% [99% CI 3–15%] in year 1 vs 24% relevant comorbid conditions (such as hypothyroidism)
[95% CI 21–26%] in years 1 to ≥5).2 Consequently, or concomitant medications that might affect the risk of
provided that statin therapy is taken for some years, the having symptoms. Thirdly, most reports of muscle pain
absolute reductions in major vascular events will greatly or weakness were not accompanied by a measurement of
exceed any small excess of muscle symptoms that occurs creatine kinase value, so we were unable to assess
soon after treatment initiation. For example, for every whether some symptoms were associated with large
1000 people in whom LDL cholesterol is lowered by increases in creatine kinase concentrations. However,
approximately 1 mmol/L for 5 years (ie, an effect easily among the cases for which there were data available on
achievable in almost all patients with a moderate- creatine kinase concentrations, more than 96% were less
intensity statin), statins might cause 11 (generally mild) than 3 times the ULN, and there was no evidence that
episodes of muscle pain or weakness, but prevent (after removing myopathy cases) extreme creatine kinase
50 major vascular events in those with pre-existing values were more common among participants allocated
vascular disease (secondary prevention), and 25 major to the statin groups. Finally, although all trials excluded
vascular events in those without pre-existing vascular anyone known to have had a previous serious adverse
disease (primary prevention).2 Moreover, a high intensity reaction to statin therapy, and many took steps to exclude
regimen that can reduce LDL cholesterol by 50% would those with a previous statin sensitivity or hypersensitivity,
on average produce at least a 2 mmol/L reduction in LDL most of them did not seek to identify or exclude
cholesterol in people with an LDL cholesterol of participants who might now be categorised as statin
4 mmol/L or more (eg, which was the measured intolerant (most of the trials had completed enrolment
concentration in approximately a third of people not before statins were in common use and well before they
taking a statin in UK Biobank),51 and would be expected were available generically).
to prevent twice as many major vascular events in each Currently, the management of patients reporting
setting, but without any material increase in the excess muscle symptoms while taking statin therapy is
rate of muscle pain or weakness as compared with a challenging, since the belief that statin therapy often
regimen reducing LDL cholesterol by 1 mmol/L. causes such symptoms is encouraged by drug labelling
Moreover, treatment for longer than 5 years will yield and other misleading sources of information (contributing
even higher cardiovascular benefits. to the so-called nocebo or drucebo effect, where negative
The main strength of our analyses is that they provide expectations can lead to perceived adverse effects).8,53 By
the first reliable estimates of the causal contribution of contrast, our results confirm that, in the majority of cases,
statins to muscle symptoms reported by a wide range of statin therapy is not likely to be the cause of muscle pain

842 www.thelancet.com Vol 400 September 10, 2022

 Articles

in a person taking statin therapy. This finding is ALLIANCE (Aggressive Lipid-Lowering Initiation Abates New Cardiac
particularly true if the treatment has been well tolerated Events) Michael Koren; ASCOT (Anglo-Scandinavian Cardiac Outcomes
Trial) Björn Dahlöf, Ajay Gupta, Neil Poulter, Peter Sever, Hans Wedel;
for a year or more before developing symptoms; but, even ASPEN (Atorvastatin Study for the Prevention of Coronary Heart Disease
during the first year of a moderate-intensity statin Endpoints in Non-Insulin Dependent Diabetes Mellitus) Robert H Knopp
regimen, it is likely to be the cause in only approximately (deceased); AURORA (A study to evaluate the Use of Rosuvastatin in
one in 15 patients who report muscle symptoms, rising to subjects On Regular haemodialysis: an Assessment of survival and
cardiovascular events) Stuart Cobbe, Bengt Fellström, Hallvard Holdaas
approximately one in 10 in those who are taking a more (deceased), Alan Jardine, Roland Schmieder, Faiez Zannad; CARDS
intensive regimen. In other words, the statin is not the (Collaborative Atorvastatin Diabetes Study) D John Betteridge (deceased),
cause of muscle symptoms in more than 90% of Helen M Colhoun, Paul N Durrington, John Fuller (deceased), Graham
individuals who report such symptoms. A Hitman, Andrew Neil; CARE (Cholesterol And Recurrent Events Study)
Eugene Braunwald, Barry Davis, C Morton Hawkins, Lemuel Moyé,
In conclusion, this individual participant data meta- Marc Pfeffer, Frank Sacks; CORONA (Controlled Rosuvastatin
analysis of randomised trials has found that statin Multinational Trial in Heart Failure) John Kjekshus, Hans Wedel,
therapy causes a small proportional increase in reports John Wikstrand; 4D (Die Deutsche Diabetes Dialyse Studie)
of muscle pain, largely during the first year after Christoph Wanner, Vera Krane; GISSI (Gruppo Italiano per lo Studio della
Sopravvivenza nell’Infarto miocardico) Heart Failure and Prevention trials
treatment commences. These findings have shown that Maria Grazia Franzosi, Roberto Latini, Donata Lucci, Aldo Maggioni;
symptoms might differ from those observed in patients Roberto Marchioli, Enrico B Nicolis, Luigi Tavazzi, Gianni Tognoni;
with myopathy (patients might report myalgia, cramps, HOPE-3 Jackie Bosch, Eva Lonn, Salim Yusuf; HPS (Heart Protection
limb pain, fatigue or weakness, or some other Study) Jane Armitage, Louise Bowman, Rory Collins, Anthony Keech,
Martin Landray, Sarah Parish, Richard Peto, Peter Sleight (deceased);
musculoskeletal pain) with no good evidence that the IDEAL (Incremental Decrease in Endpoints through Aggressive Lipid-
proportional risk increases vary between different types lowering) John JP Kastelein, Terje R Pedersen; JUPITER (Justification for
of patients or statins of equivalent LDL-lowering the Use of Statins in Prevention: an Intervention Trial Evaluating
abilities, but some evidence that the proportional risk Rosuvastatin) Robert Glynn, Antonio Gotto Jr, John JP Kastelein,
Wolfgang Koenig, Jean Macfadyen, Paul M Ridker; LIPID (Long-term
increase is higher for more intensive statin regimens Intervention with Pravastatin in Ischaemic Disease) Anthony Keech,
than for moderate-intensity statin regimens. However, Stephen MacMahon, Ian Marschner, John Shaw (deceased), John Simes,
for all patients for whom statin therapy might be Andrew Tonkin, Harvey White; LIPS (Lescol Intervention Prevention
Study) Patrick W Serruys; Post-CABG (Post-Coronary Artery Bypass Graft
considered, the quantitative evidence from previous
Study) Genell Knatterud (deceased); PROSPER (Prospective Study of
analyses of large trials within the CTT Collaboration Pravastatin in the Elderly at Risk) Naveed Sataar, J Wouter Jukema,
clearly indicate that the risk is greatly outweighed by the Stella Trompet, Ian Ford; PROVE-IT (Pravastatin or Atorvastatin
cardiovascular benefits of statins. Our findings suggest Evaluation and Infection Therapy) Eugene Braunwald,
Christopher P Cannon, Sabina Murphy; SEARCH (Study of Effectiveness
that there is a need to review recommended strategies
of Additional Reductions in Cholesterol and Homocysteine) Rory Collins,
for managing such symptoms, and to revise the Jane Armitage, Louise Bowman, Sarah Parish, Richard Peto,
information in the drug label for statins. In particular, Peter Sleight (deceased); SPARCL (Stroke Prevention by Aggressive
for patients who report mild muscle symptoms when Reduction in Cholesterol Levels) Pierre Amarenco, K Michael Welch;
4S Scandinavian Simvastatin Survival Study) John Kjekshus,
taking a statin, our findings suggest that it is most likely
Terje R Pedersen, Lars Wilhelmsen; TNT (Treating to New Targets)
that the symptoms are not due to the statin, and statin Phlip Barter, Antonio Gotto JR, John La Rosa; John JP Kastelein,
therapy should continue until other potential causes James Shepherd (deceased); WOSCOPS (West of Scotland Coronary
have been explored. Prevention Study) Stuart Cobbe, Ian Ford, Sharon Kean,
Peter Macfarlane, Chris Packard, Michele Roberston, Naveed Sattar,
Cholesterol Treatment Trialists’ Collaborators James Shepherd (deceased), Robin Young.
Writing committee: Christina Reith*, Colin Baigent*, Lisa Blackwell, Other CTT Members: Hiroyuki Arashi, Robert Byington, Robert Clarke,
Jonathan Emberson, Enti Spata, Kelly Davies, Heather Halls, Marcus Flather, Shinya Goto, Uri Goldbourt, Jemma Hopewell,
Lisa Holland, Kate Wilson, Jane Armitage, Charlie Harper, David Preiss, Kees Hovingh, Patricia Kearney, George Kitas, Connie Newman,
Alistair Roddick, Anthony Keech, John Simes, Rory Collins Marc S Sabatine, Greg Schwartz, Liam Smeeth, Jonathan Tobert,
(*These authors contributed equally). John Varigos, Junichi Yamaguchi.
CTT secretariat: Jane Armitage, Colin Baigent, Elizabeth Barnes,
Lisa Blackwell, Rory Collins, Kelly Davies, Jonathan Emberson, Contributors
Jordan Fulcher, Heather Halls, Charlie Harper, William G Herrington, The authors accept full responsibility for the content of this paper and
Lisa Holland, Anthony Keech, Adrienne Kirby, Borislava Mihaylova, were responsible for the decision to submit the manuscript. CR, CB,
Rachel O’Connell, David Preiss, Christina Reith, John Simes, Enti Spata, LB, JE, ES, KD, HH, LH, KW, JA, DP, AK, JS, and RC conceived of and
Kate Wilson. designed the experiments. CR, CB, LB, JE, ES, KD, HH, LH, KW, CH,
Independent oversight committee: Emily Banks, Michael Blastland, and AR accessed and verified the data. LB, JE, and ES analysed the data.
Stephen Evans, Robert Temple, Peter Weissberg (chair), Janet Wittes. CR, CB, JE, DP, and RC wrote the first draft. All authors critically
Collaborating trials: A to Z trial (phase Z) Michael Blazing, Eugene revised the manuscript for important intellectual content.
Braunwald, James de Lemos, Sabina Murphy; Terje R Pedersen, All collaborators had an opportunity to provide input into the study
Marc Pfeffer, Harvey White, Stephen Wiviott; AFCAPS/TEXCAPS protocols, contribute to the interpretation of the results, and to critically
(AirForce/Texas Coronary Atherosclerosis Prevention Study) revise the manuscript for important intellectual content.
Michael Clearfield, John R Downs, Antonio Gotto Jr, Stephen Weis; Declaration of interests
ALERT (Assessment of Lescol in Renal Transplantation) Bengt Fellström, JA and DP report receiving a grant to their research institution from
Hallvard Holdaas (deceased), Alan Jardine, Terje R Pedersen; ALLHAT Novartis for the ORION 4 trial of inclisiran. AK reports receiving grants
(Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack for his institution from Abbott, Amgen, and Mylan; consulting fees from
Trial) David Gordon, Barry Davis; Curt Furberg, Richard Grimm, AstraZeneca; honoraria from Sanofi and Pfizer; and is a Data Safety
Sara Pressel, Jeffrey Probstfield, Mahboob Rahman, Lara Simpson; Monitoring Board member of the Kowa PROMINENT trial. JS reports

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 843

 Articles

receiving grants for his institution from Abbvie, Astra Zeneca, Bayer, 13 Nissen SE. Statin denial: an internet-driven cult with deadly
Bristol Myers Squibb, Pfizer, and Roche. RC reports a patent for a statin- consequences. Ann Intern Med 2017; 167: 281–82.
related myopathy genetic test licensed to University of Oxford from 14 Cholesterol Treatment Trialists’ Collaboration. Protocol for analyses
Boston Heart Diagnostics (RC has waived any personal reward). of adverse event data from randomized controlled trials of statin
All other authors declare no competing interests. therapy. Am Heart J 2016; 176: 63–69.
15 Cholesterol Treatment Trialists’ Collaboration. Harmonisation of
Data sharing large-scale, heterogeneous individual participant adverse event
Individual patient data from each contributing trial have been provided data from randomised trials of statin therapy. Clin Trials 2022;
to the Cholesterol Treatment Trialists’ Collaboration on the published online July 9. https://doi.org/10.1177/17407745221105509.
understanding that they would be used only for the purpose of the 16 Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/AACVPR/
Cholesterol Treatment Trialists’ meta-analyses and would not be released AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA
to others. Requests for such data should be made directly to the data Guideline on the Management of Blood Cholesterol: a report of
custodians of each trial. The Cholesterol Treatment Trialists’ data policy the American College of Cardiology/American Heart Association
can be found on the website: https://www.cttcollaboration.org/. Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 2019;
73: e285–350.
Acknowledgments 17 Clinical Data Interchange Standards Consortium. Study Data
The work of the Cholesterol Treatment Trialists’ Collaboration was Tabulation Model. https://www.cdisc.org/standards/foundational/
funded by the British Heart Foundation (PG/18/16/33570 and sdtm (accessed March 29, 2022).
CH/1996001/9454), UK Medical Research Council (MC_UU_00017/4), 18 Peto R, Pike MC, Armitage P, et al. Design and analysis of
and Australian National Health and Medical Research Council randomized clinical trials requiring prolonged observation of
(GNT1037786 and GNT1150467). Although individual trials that were each patient. I. Introduction and design. Br J Cancer 1976;
contributing data to the analyses were funded by the pharmaceutical 34: 585–612.
industry, as well as by charities and government organisations, 19 Peto R, Pike MC, Armitage P, et al. Design and analysis of
the Cholesterol Treatment Trialists’ Collaboration has not received grant randomized clinical trials requiring prolonged observation of each
funding from industry. Further details are available at: www. patient. II. analysis and examples. Br J Cancer 1977; 35: 1–39.
cttcollaboration.org. We thank all participants who originally took part in 20 Scandinavian Simvastatin Survival Study Group. Randomised trial
the included trials, as well as the investigators who conducted those of cholesterol lowering in 4444 patients with coronary heart disease:
trials, without whom this research would not be possible. We also thank the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;
Samantha Briggs and Clare Mathews for their valuable administrative 344: 1383–89.
assistance in preparing this report. 21 Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart
disease with pravastatin in men with hypercholesterolemia.
References N Engl J Med 1995; 333: 1301–07.
1 Roth GA, Mensah GA, Johnson CO, et al. Global burden of 22 Sacks FM, Pfeffer MA, Moye LA, et al. The effect of pravastatin on
cardiovascular diseases and risk factors, 1990-2019: update from the coronary events after myocardial infarction in patients with average
GBD 2019 study. J Am Coll Cardiol 2020; 76: 2982–3021. cholesterol levels. Cholesterol and Recurrent Events Trial
2 Collins R, Reith C, Emberson J, et al. Interpretation of the evidence investigators. N Engl J Med 1996; 335: 1001–09.
for the efficacy and safety of statin therapy. Lancet 2016; 388: 2532–61. 23 Downs JR, Clearfield M, Weis S, et al. Primary prevention of acute
3 Cholesterol Treatment Trialists’ Collaboration. Efficacy and safety of coronary events with lovastatin in men and women with average
more intensive lowering of LDL cholesterol: a meta-analysis of data cholesterol levels: results of AFCAPS/TexCAPS. JAMA 1998;
from 170,000 participants in 26 randomised trials. Lancet 2010; 279: 1615–22.
376: 1670–81. 24 The Long-Term Intervention with Pravastatin in Ischaemic Disease
4 Cholesterol Treatment Trialists’ Collaboration. The effects of (LIPID) Study Group. Prevention of cardiovascular events and death
lowering LDL cholesterol with statin therapy in people at low risk of with pravastatin in patients with coronary heart disease and a broad
vascular disease: meta-analysis of individual data from range of initial cholesterol levels. The Long-Term Intervention with
27 randomised trials. Lancet 2012; 380: 581–90. Pravastatin in Ischaemic Disease (LIPID) Study Group. N Engl J Med
5 Cholesterol Treatment Trialists’ Collaboration. Efficacy and safety of 1998; 339: 1349–57.
LDL-lowering therapy among men and women: meta-analysis of 25 Serruys PW, de Feyter P, Macaya C, et al. Fluvastatin for prevention
individual data from 174,000 participants in 27 randomised trials. of cardiac events following successful first percutaneous coronary
Lancet 2015; 385: 1397–405. intervention: a randomized controlled trial. JAMA 2002;
6 Cholesterol Treatment Trialists’ Collaboration. Efficacy and safety of 287: 3215–22.
statin therapy in older people: a meta-analysis of individual 26 Heart Protection Study Collaborative Group. MRC/BHF Heart
participant data from 28 randomised controlled trials. Lancet 2019; Protection Study of cholesterol lowering with simvastatin in
393: 407–15. 20,536 high-risk individuals: a randomised placebo-controlled trial.
7 Byrne P, Jones M, Malhotra A. Rapid response: re: statin treatment Lancet 2002; 360: 7–22.
and muscle symptoms: series of randomised, placebo controlled 27 Shepherd J, Blauw GJ, Murphy MB, et al. Pravastatin in elderly
n-of-1 trials. BMJ 2021; published online Feb 24. https://doi. individuals at risk of vascular disease (PROSPER): a randomised
org/10.1136/bmj.n135/rr-1. controlled trial. Lancet 2002; 360: 1623–30.
8 Penson PE, Bruckert E, Marais D, et al. Step-by-step diagnosis and 28 Sever PS, Dahlöf B, Poulter NR, et al. Prevention of coronary and
management of the nocebo/drucebo effect in statin-associated muscle stroke events with atorvastatin in hypertensive patients who have
symptoms patients: a position paper from the International Lipid average or lower-than-average cholesterol concentrations, in the
Expert Panel (ILEP). J Cachexia Sarcopenia Muscle 2022; 13: 1596–622. Anglo-Scandinavian Cardiac Outcomes Trial--Lipid Lowering Arm
9 Newman CB, Preiss D, Tobert JA, et al. Statin safety and associated (ASCOT-LLA): a multicentre randomised controlled trial. Lancet
adverse events: a scientific statement from the American Heart 2003; 361: 1149–58.
Association. Arterioscler Thromb Vasc Biol 2019; 39: e38–81. 29 Holdaas H, Fellström B, Jardine AG, et al. Effect of fluvastatin on
10 Bruckert E, Hayem G, Dejager S, Yau C, Bégaud B. Mild to cardiac outcomes in renal transplant recipients: a multicentre,
moderate muscular symptoms with high-dosage statin therapy in randomised, placebo-controlled trial. Lancet 2003; 361: 2024–31.
hyperlipidemic patients--the PRIMO study. Cardiovasc Drugs Ther 30 Colhoun HM, Betteridge DJ, Durrington PN, et al. Primary
2005; 19: 403–14. prevention of cardiovascular disease with atorvastatin in type 2
11 Buettner C, Rippberger MJ, Smith JK, Leveille SG, Davis RB, diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS):
Mittleman MA. Statin use and musculoskeletal pain among adults multicentre randomised placebo-controlled trial. Lancet 2004;
with and without arthritis. Am J Med 2012; 125: 176–82. 364: 685–96.
12 Bytyçi I, Penson PE, Mikhailidis DP, et al. Prevalence of statin 31 Wanner C, Krane V, März W, et al. Atorvastatin in patients with
intolerance: a meta-analysis. Eur Heart J 2022; published online type 2 diabetes mellitus undergoing hemodialysis. N Engl J Med
Feb 16. https://doi.org/10.1093/eurheartj/ehac015. 2005; 353: 238–48.

844 www.thelancet.com Vol 400 September 10, 2022

 Articles

32 Knopp RH, d’Emden M, Smilde JG, Pocock SJ. Efficacy and safety 44 Ganga HV, Slim HB, Thompson PD. A systematic review of statin-
of atorvastatin in the prevention of cardiovascular end points in induced muscle problems in clinical trials. Am Heart J 2014;
subjects with type 2 diabetes: the Atorvastatin Study for Prevention 168: 6–15.
of Coronary Heart Disease Endpoints in non-insulin-dependent 45 Gupta A, Thompson D, Whitehouse A, et al. Adverse events
diabetes mellitus (ASPEN). Diabetes Care 2006; 29: 1478–85. associated with unblinded, but not with blinded, statin therapy in
33 Amarenco P, Bogousslavsky J, Callahan A 3rd, et al. High-dose the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid-Lowering
atorvastatin after stroke or transient ischemic attack. N Engl J Med Arm (ASCOT-LLA): a randomised double-blind placebo-controlled
2006; 355: 549–59. trial and its non-randomised non-blind extension phase. Lancet
34 Kjekshus J, Apetrei E, Barrios V, et al. Rosuvastatin in older patients 2017; 389: 2473–81.
with systolic heart failure. N Engl J Med 2007; 357: 2248–61. 46 Moriarty PM, Thompson PD, Cannon CP, et al. Efficacy and safety
35 Tavazzi L, Maggioni AP, Marchioli R, et al. Effect of rosuvastatin in of alirocumab vs ezetimibe in statin-intolerant patients, with a
patients with chronic heart failure (the GISSI-HF trial): a randomised, statin rechallenge arm: the ODYSSEY ALTERNATIVE randomized
double-blind, placebo-controlled trial. Lancet 2008; 372: 1231–39. trial. J Clin Lipidol 2015; 9: 758–69.
36 Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatin to prevent 47 Nissen SE, Stroes E, Dent-Acosta RE, et al. Efficacy and tolerability
vascular events in men and women with elevated C-reactive protein. of evolocumab vs ezetimibe in patients with muscle-related statin
N Engl J Med 2008; 359: 2195–207. intolerance: the GAUSS-3 randomized clinical trial. JAMA 2016;
37 Fellström BC, Jardine AG, Schmieder RE, et al. Rosuvastatin and 315: 1580–90.
cardiovascular events in patients undergoing hemodialysis. 48 Herrett E, Williamson E, Brack K, et al. Statin treatment and
N Engl J Med 2009; 360: 1395–407. muscle symptoms: series of randomised, placebo controlled
38 Yusuf S, Bosch J, Dagenais G, et al. Cholesterol lowering in n-of-1 trials. BMJ 2021; 372: n135.
intermediate-risk persons without cardiovascular disease. 49 Wood FA, Howard JP, Finegold JA, et al. N-of-1 trial of a statin,
N Engl J Med 2016; 374: 21. placebo, or no treatment to assess side effects. N Engl J Med 2020;
39 Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus 383: 2182–84.
moderate lipid lowering with statins after acute coronary 50 Abramson JD, Rosenberg HG, Jewell N, Wright JM. Should people
syndromes. N Engl J Med 2004; 350: 1495–504. at low risk of cardiovascular disease take a statin? BMJ 2013;
40 de Lemos JA, Blazing MA, Wiviott SD, et al. Early intensive 347: f6123.
vs a delayed conservative simvastatin strategy in patients with acute 51 Welsh C, Celis-Morales CA, Brown R, et al. Comparison of
coronary syndromes: phase Z of the A to Z trial. JAMA 2004; conventional lipoprotein tests and apolipoproteins in the prediction
292: 1307–16. of cardiovascular disease. Circulation 2019; 140: 542–52.
41 LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering 52 Collins R, MacMahon S. Reliable assessment of the effects of
with atorvastatin in patients with stable coronary disease. treatment on mortality and major morbidity, I: clinical trials. Lancet
N Engl J Med 2005; 352: 1425–35. 2001; 357: 373–80.
42 Armitage J, Bowman L, Wallendszus K, et al. Intensive lowering of 53 Tobert JA, Newman CB. The nocebo effect in the context of statin
LDL cholesterol with 80 mg versus 20 mg simvastatin daily in intolerance. J Clin Lipidol 2016; 10: 739–47.
12,064 survivors of myocardial infarction: a double-blind
randomised trial. Lancet 2010; 376: 1658–69.
43 Davis JW, Weller SC. Intensity of statin therapy and muscle
symptoms: a network meta-analysis of 153 000 patients. BMJ Open
2021; 11: e043714.

www.thelancet.com www.thelancet.com Vol 400 September 10, 2022 845