Introduction

Multiple large-scale randomized controlled trials (RCTs) and meta-analyses have shown that statin therapy is associated with consistent atherosclerotic cardiovascular disease (ASCVD) risk reduction in both secondary and primary prevention studies in women and men [1], those at low ASCVD risk [2], people with diabetes [3], severe hypercholesterolemia [4], and older adults up to 75 years of age [5]. In addition, a legacy risk reduction effect of statins that continues for at least 3 years after completion of a placebo-controlled RCT was demonstrated in intermediate-risk primary prevention subjects [6]. Despite these benefits, statin discontinuation is common in clinical practice and has been associated with adverse consequences, particularly in those with established ASCVD. Although there have been rare reports of neurocognitive disorders, hepatotoxicity, hemorrhagic stroke, and renal toxicity, a causal relationship has been confirmed only for statin-associated muscle symptoms (SAMS), temporary elevation in transaminases, and newly noted diabetes. This article will focus on statin intolerance attributed to muscle symptoms.

Diagnostic Criteria for SAMS

The lack of specific biomarkers makes the diagnosis of true SAMS difficult. Stakeholder organizations have suggested a variety of criteria to diagnose SAMS (Table 1). A point system has also been proposed to classify muscle symptoms as unlikely, possibly or probably due to statin therapy (Fig. 1) [12]. In clinical practice, SAMS are most often reported within weeks after initiation or increase in intensity of therapy, are more likely to occur in physically active individuals, affect proximal large muscle groups, and generally resolve within several weeks to as long as 8 weeks after discontinuation. When similar symptoms recur after re-challenge with an alternate statin, particularly in the starting dose, and resolve within weeks upon drug discontinuation, the diagnosis of statin-associated muscle symptoms is more likely [13].

Table 1 Comparison of guideline definitions of statin intolerance (for the purpose of this review, statin intolerance refers to adverse events attributed to muscle symptoms)
Full size table
Fig. 1
figure 1

An example of the Statin-Associated Muscle Symptom Clinical Index (SAMS-CI) for assessing the likelihood that patient’s muscle symptoms were caused or worsened by statin use. Adapted from Rosenson, R.S., et al. Cardiovasc Drugs Ther 31, 179–186 (2017). https://doi.org/10.1007/s10557-017-6723-4. Published on April 18th, 2017, with permission from Springer Nature [12]

Full size image

While some experts have suggested that the incidence of SAMS reported in meta-analyses of large RCTs largely represent misattribution [14], observational studies and a large web-based survey have reported muscle symptoms in 10 to 25% of studied subjects [15, 16]. More recently, a random effects meta-analysis including 176 studies (112 RCTs and 64 cohort studies) with 4,143,517 subjects reported that the incidence of statin intolerance was 9.1% (95% confidence interval (CI) 8.01–10%) [17••].

A major impediment to the determination of whether muscle complaints are truly statin-related is the nocebo affect, pre-existing patient belief that symptoms that occur while taking a medicine are due to that medicine [18]. The GAUSS-3 (Goal Achievement After Utilizing an Anti-PCSK9 Antibody in Statin Intolerant Subjects 3) trial was a 2-stage RCT, the purpose of which was to initially identify subjects with statin-induced muscle symptoms during a placebo-rechallenge procedure and then to compare the low-density lipoprotein cholesterol (LDL-C) lowering efficacy and tolerability of ezetimibe versus evolocumab at weeks 22 and 24 of treatment. During the randomization phase of this trial, 491 subjects with a history of intolerable muscle symptoms while taking each of two or more statins entered a double-blind phase in which 246 received placebo for 10 weeks and 245 atorvastatin, 20 mg daily, followed by cross-over to the alternate treatment group. During this phase, 26.5% of patients had muscle symptoms while taking placebo but not atorvastatin, and 42.6% had symptoms while taking atorvastatin but not placebo [19]. This study supported the observation that muscle symptoms, at least in some patients, are statin related.

Another way of studying whether muscle symptoms are actually due to the statin is to employ N of 1 trials, RCTs in which an individual patient serves as their own control during blinded administration of statin or placebo [20, 21]. Two such trials examined patient populations from England. In one study, 200 individuals, who were either considering statin discontinuation or stopped statin therapy during the previous year because of muscle symptoms, were treated with atorvastatin 20 mg daily or placebo over six 2-month randomly assigned treatment periods. The authors, using a muscle score symptoms methodology, reported no differences in the mean muscle symptom score between the statin and placebo periods (mean difference statin minus placebo − 0.11). By the end of the trial, 88% of subjects felt that the trial had been helpful in their decision-making about whether to restart a statin, and 66% reported readiness or intention to re-initiate statin therapy [22].

A second double blind N of 1 trial enrolled 60 subjects who had discontinued statins because of side effects reported within 2 weeks of initiating therapy. The purpose of the trial was to determine whether symptoms would be induced by atorvastatin 20 mg daily or placebo. Each patient received four bottles of atorvastatin 20 mg, placebo, or empty bottles, and they were asked to take the content of each bottle for one month and to use a smartphone app to report symptom intensity on a scale from 0 to 100. In this cohort, the mean symptom intensity was 8.0 during no-tablet months, 15.4 during placebo months compared to no-tablet months, and 16.3 with atorvastatin administration compared to no-tablet months. Six months after completion of the trial, 50% agreed to restart statin therapy. The authors observed that among those who had discontinued therapy because of side effects, 90% of reported symptoms attributed to statin therapy were also reported while taking the placebo [23]. The Understanding Statin Use in America and Gaps in Education (USAGE) survey assessed the practices of current and former statin users.in 10,138 individuals. Muscle-related side effects were reported by 60% of former users, which was the primary reason for discontinuation. Patients not satisfied with physician discussion were more likely to discontinue statins highlighting the importance of shared decision making [16]. However, selection bias among those agreeing to participate and lack of demographic and racial diversity limit the generalizability of these findings.

Factors Associated with Increased Risk for Statin-Associated Muscle Symptoms

Statin use is associated with variability in the risk of SAMS in different racial populations [24]. The SLCO1B1*15 allelic variant, which occurs primarily in Japanese individuals, results in reduced function of the organic anion transporter that regulates the hepatic uptake of simvastatin, leading to higher serum levels and a higher rate of myopathy in Japanese subjects [25, 26]. Similarly, the 421C > A polymorphism in the drug efflux transporter, ATP binding cassette G2 gene (ABCG2), results in a plasma concentration of rosuvastatin that is twice as high as in those without this variant [27]. Higher plasma levels of statins and their active metabolites increases the risk of myopathy or rhabdomyolysis [28•].

Among major guidelines, hypothyroidism, vigorous muscular exercise, vitamin D deficiency, preexisting muscle disease, renal impairment, drug-drug interactions, female sex, and East Asian ethnicity are recognized as risk factors for SAMS [7–9, 28•]. Polypharmacy is a common cause and should be questioned. Increased use of grapefruit juice in those taking statins metabolized by CYP P450 3A4, including lovastatin, simvastatin, and atorvastatin may have higher statin blood levels [29], but this effect should be balanced against the cardioprotective effects of higher intensity statin therapy. Alcohol consumption should be assessed. Although there is no evidence of an interaction between statins and alcohol, alcohol intake may affect liver function and predispose to hepatotoxicity.

The most recent and largest meta-analysis of studies of statin intolerance identified increased odds with older age (odds ratio (OR) 1.33, 95% CI: 1.26 to 1.42), female sex (OR 1.48, 95% CI: 1.39 to 1.54), Asian race (OR 1.25, 95% CI: 1.20 to 1.41), Black race (OR 1.29, 95% CI: 1.21 to 1.45), obesity (OR 1.31, 95% CI: 1.20 to 1.56), diabetes mellitus (OR 1.27, 95% CI:1.19 to 1.47), renal impairment (OR 1.25, 95% CI: 1.18 to 1.48), chronic liver failure (OR 1.24, 95% CI: 1.18 to 1.54), alcohol use (OR 1.22, 95% CI:1.10 to1.54), calcium channel blocker use (OR 1.31, 95% CI: 1.12 to 1.49), antiarrhythmic agent use (OR 1.36, 95% CI: 1.21 to 1.63), and increased statin dose (OR 1.38, 95% CI: 1.25 to 1.56) [17••]. These results should be interpreted with caution as details including the under-representation of Hispanic and Black races, the quantity of alcohol consumption, control of diabetes mellitus and severity of chronic liver or kidney disease were not reported. In addition, the contribution of the nocebo effect could not be determined.

Should Statin Therapy Be Restarted in Patients with Statin-Attributed Muscle Symptoms?

Statin intolerance is associated with recurrent myocardial infarction and coronary events, particularly in patients with proven ASCVD [30]. A study of 105,329 Medicare beneficiaries who began a moderate- or high-intensity statin after hospitalization for a myocardial infarction between 2007 and 2013 were followed up for a median follow-up of 1.9–2.3 years. The multivariate-adjusted hazard ratios (HR) comparing those with statin intolerance versus those with high adherence were 1.50 (95% confidence interval [CI]: 1.30 to 1.73) for recurrent myocardial infarctions, 1.51 (95% CI: 1.34 to 1.70) for coronary heart disease events, and 0.96 (95% CI: 0.87 to 1.06) for all-cause mortality. Thus, statin intolerance was, in this population, associated with recurrent myocardial infarction and coronary events, but not all-cause mortality [30]. Because of the established benefit of statins in such patients, the re-initiation of therapy should be attempted whenever possible.

While sub-optimal statin adherence is associated with worsened ASCVD outcomes in secondary prevention cohorts, the data remain less clear for primary prevention [31]. Based on the primary prevention MESA (Multi-Ethnic Study of Atherosclerosis) cohort, 49% [32] of borderline- or intermediate-risk adults who are prescribed statins based upon their Pooled Cohort Equations estimated 10-year ASCVD risk have a coronary artery calcium (CAC) score of 0 [33]. Such individuals derive minimal ASCVD risk reduction from statin initiation [34, 35]. An algorithm that incorporates CAC scoring into clinical decision-making about statin re-initiation in statin intolerant patients has been proposed and provides a practical approach to the use of this technology in such patients (Fig. 2) [36].

Fig. 2
figure 2

The role of CAC scoring to aid clinical decision-making in those with statin-attributed muscle symptoms. This article was published in J Clin Lipidol, Vol 15, Orringer CE et al. Coronary artery calcium scoring in patients with statin associated muscle symptoms: prescribing statins for those most likely to benefit, P782-788, Copyright Elsevier (2021) [36]

Full size image

Statin therapy is generally well-tolerated by patients when rechallenged. In a large US retrospective cohort study, 11,124 of 107,835 subjects (10%) discontinued statin treatment because of adverse events considered to be statin related. Among these individuals, 6579 (59%) were rechallenged and statin treatment was able to be reinitiated in > 90% of cases. At 12 months, only 515 (7.8%) of the 6579 rechallenged patients were not taking a statin [37].

How Should We Manage SAMS?

The traditional approach to the management of SAMS includes an evaluation of risk factors that predispose to musculoskeletal symptoms, including recent onset muscular activities, hypothyroidism, underlying metabolic, musculoskeletal or rheumatologic disorders, severe vitamin D deficiency or the initiation of drugs that alter statin metabolism [13]. The European Atherosclerosis Society statement on statin-associated muscle symptoms recommended that when muscle symptoms are present, creatine kinase (CK) levels should be measured, and if greater than 4 times the upper limit of normal, the statin should be discontinued for 2–4 weeks before re-challenge [9]. Various complementary therapies to overcome SAMS such as CoQ10 and vitamin D have been proposed but none are supported by randomized controlled trial evidence.

Alternative strategies for LDL-C lowering in such patients include the initiation of a vegan diet, the use of lower or starting doses of the same or other statins, less-than-daily long-acting statins [38, 39], ezetimibe or bempedoic acid monotherapy [40], ezetimibe/bempedoic acid combination therapy [41], either of these drugs in combination with less-than-daily long-acting statins, or in high or very high risk patients, the use of proprotein convertase subtilisin/kexin 9 (PCSK9) monoclonal antibodies [19, 42]. Inclisiran, a small interfering RNA targeting hepatic PCSK9 synthesis is another non-statin approved both by the European Medicines Agency and the U.S. Food and Drug Administration for LDL-C reduction, [43, 44] but has not been studied specifically in this patient population. The use of nutraceuticals with LDL-C lowering properties for treatment of statin intolerant patients has been proposed as an additional strategy in a position statement of the International Lipid Expert Panel, although with few exceptions, subject enrollment in such studies was small [45].

The National Lipid Association defines severe myonecrosis as a greater than or equal to 50 fold increase in creatine kinase above baseline levels or normative upper limit, adjusted for age, race and sex [46]. Clinical rhabdomyolysis is defined as myonecrosis with myoglobinuria or acute renal failure (increase in serum creatine of at least 0.5 mg/dL over baseline values) [46]. Most cases of rhabdomyolysis are due to non-statin related etiologies (trauma, exogenous toxins, alcohol and illicit drugs) [47]. Serious muscle toxicity attributed to statin therapy is rare [28•]. Risk factors for statin-associated myotoxicity include older age, hypertension, diabetes mellitus and the concomitant use of a variety of drugs associated with myotoxicity [48].

Statin drug-drug interactions (DDIs) are most often mediated by the cytochrome P450 enzyme system (CYP450) and the permeability glycoprotein (p-gp). The main role of CYP450, which is mainly expressed in the liver, but also in the gastrointestinal tract, kidney and other sites, is to detoxify medications and facilitate their elimination from the body. P-gp expression, localized primarily in the gastrointestinal tract and in hepatic, renal, and brain tissue, plays an important role in drug disposition by promoting secretion of substrates into the urine and bile. The American Heart Association published a scientific statement providing recommendations on the management of clinically significant DDIs with statins and select agents used in patients with cardiovascular disease [49]. Commonly used drugs in this category include amiodarone, amlodipine, colchicine, conivaptan, cyclosporine/tacrolimus/everolimus/sirolimus, diltiazem, dronedarone, gemfibrozil, ranolazine, ticagrelor, and verapamil [49]. Clinician awareness of these DDI will help to mitigate the risk of serious myotoxicity in individuals receiving statin therapy.

Initial treatment of statin-induced rhabdomyolysis involves prompt cessation of statin therapy and hydration, generally resulting in rapid reversal of clinical symptoms and laboratory abnormalities [28•]. Given the low incidence of statin-associated rhabdomyolysis, data on statin re-initiation are limited. A study of 54 patients with statin-induced muscle injury (3 cases of rhabdomyolysis) over a 5-year follow-up, suggested that statin therapy can be safely and effectively restarted after full resolution of symptoms and normalization of creatine kinase levels [50]. The decision to restart statin therapy in persons with statin-induced rhabdomyolysis should be individualized based on ASCVD risk, concomitant medical conditions, required drug therapy, and the potential for use of non-statin agents. If statin therapy is to be reinitiated, the starting dose should be utilized.

Statin-induced necrotizing autoimmune myopathy was initially described in 2007 in a case series of 8 patients who had progressive weakness and high creatine kinase levels that persisted long after statin discontinuation [51]. This rare disorder, with an estimated incidence of 1 in 100,000 statin users [52], is characterized by proximal muscle weakness, markedly elevated creatine kinase levels persisting after drug discontinuation, the presence of antibodies to HMGCoA (3-hydroxy-3-methylglutaryl coenzyme A) reductase and the finding of myonecrosis with few inflammatory cells on muscle biopsy [53, 54]. Therapy for this disorder includes oral or intravenous corticosteroids [55], intravenous immunoglobulin [56] and in some cases azathioprine, methotrexate or rituximab. While statin therapy should not be re-started in such patients, the use of alternate non-statin lipid lowering therapies may be considered [55].

Conclusions

The management of patients who develop statin-attributed muscle symptoms remains a challenge in clinical practice. The first step is to determine the patient’s absolute risk and then to consider statin re-initiation. In those in whom the complaints appear to be a result of the nocebo effect, a caring clinician-patient discussion is warranted. A careful evaluation of factors that may increase the risk of statin-attributed muscle symptoms should be undertaken and when present, these risk factors should be addressed. For secondary prevention patients, every effort should be made to restart statin therapy or if not feasible, to consider non-statins to maximally lower LDL-C. For primary prevention, careful reassessment of the need for a statin should be done. In all cases, the benefit of an open and collaborative clinician-patient discussion will help to determine the most appropriate clinical management strategy.