Editor's Note: Commentary based on Newman CB, Preiss D, Tobert JA, et al. Statin Safety and Associated Adverse Events: A Scientific Statement From the American Heart Association. Arterioscler Thromb Vasc Biol 2018.

Recently, the American Heart Association released a comprehensive scientific statement regarding the safety and tolerability of statin therapy. The review comes at an important time as societal guidelines continue to recommend the broader use of statin therapy. Statins remain among the most prescribed medications by US clinicians. Newman et al. performed a rigorous examination of the safety and tolerability of statins as a class, highlighting differences among the agents as appropriate. Utilizing data from randomized controlled trials, supplemented with observational data, this review covered both the general adult population as well as sub-groups potentially vulnerable to adverse events including the elderly, children, pregnant women and East Asians. It also discussed treatment of patients with chronic kidney and liver disease, HIV, and those undergoing organ transplantation.

This commentary is meant to clarify and highlight the salient clinical practice lessons we feel both primary and specialty providers should be aware of.

Muscle Related Symptoms

In the US about 10% of patients prescribed statin therapy stop their medication due to subjective complaints, most often due to muscle symptoms without a rise in creatine kinase. This is in contrast to randomized clinical trials in which the excess incidence of muscle symptoms without enzyme elevation due to statins is less than 1%. Although few drugs have adverse effects on skeletal muscle, all statins have been implicated as causing myopathy. These symptoms are typically bilateral and symmetrical and always confined to skeletal muscle.

The spectrum of adverse symptoms affecting muscle is defined across a range from statin associated muscle symptoms (SAMS) (defined as any muscle symptoms reported during use of statin therapy but not necessarily caused by statin), myalgias, myopathy (meaning unexplained muscle pains or weakness accompanied by CK elevations greater than 10 times the upper limit of normal) and very rarely rhabdomyolysis (severe form of myopathy with serum enzyme elevations typically greater than 40 times the upper limit of normal causing myoglobinuria and at times acute renal failure).

Large long-term randomized controlled trials show the excess risk of myopathy relative to placebo is typically up to 0.1%. This is with all currently marketed statins at up to their maximum recommended doses. Though individual trials report varied absolute difference in the rate of muscle related symptoms, up to 1.4% in some data, robust meta-analysis currently shows such variation to be non-significant. This small risk of myopathy is greatest within the first year of therapy and after either a dose increase or the addition of a known interacting drug. Risk factors for myopathy and rhabdomyolysis include hypothyroidism, preexisting muscle disease and renal impairment; other less robust associated factors include female gender, a pre-existing diagnosis of diabetes and East Asian descent.

Importantly, the authors address two common criticisms often raised regarding statin trials. The first is the use of an active run-in phase, intended to exclude those patients having a potential drug-related adverse event, and which theoretically leads to an underestimation of such events since susceptible patients may not get into the randomized trial. The second is the protocol driven or self-exclusion from these trials of individuals who report previous experiences of SAMS, a practice which may skew study populations away from those of common clinical practice.

Regarding the first point, of the 15 most robust placebo-controlled trials, only two used active run-in phases. Eleven of the trials utilized placebo run-in phases, which in theory should increase the observation of true adverse effects. The authors note that patients who do report prior muscle symptoms and are included in trials typically tolerate statins under double-blind placebo-controlled conditions. This finding offers reassurance that risks of adverse events is significantly lower than what is generally perceived by patients.

Muscle symptoms must be taken seriously by the clinician. Most drugs that carry rare but serious adverse risks often precipitate less-serious effects of the same nature in much more frequent fashion (anticoagulants and bleeding events are a good example of this); statins and muscle symptoms are no exception. While large and robust data in a broad array of patient types show that there is little difference in muscle symptoms between statins and placebo (at most 1%), patient expectations of harm may drive the onset of symptoms and resultant cessation of therapy. These symptoms may be severe despite the absence of a pharmacological or serological basis in the overwhelming majority of cases.

Clinicians should feel comfortable excluding myopathy based on physical exam and laboratory measures. In the absence of evident myopathy (CK <10 upper limit of normal), clinicians should also feel equally comfortable re-challenging with the same statin at a lower dose, a lower frequency of dosing such as every other day, or with a different statin near the previous level of intensity. Such practice can ideally restore therapy when most needed for those of our patients at greatest risk of an ASCVD event.

Newly Diagnosed Diabetes

The trend of newly-diagnosed or new-onset diabetes is another oft-cited effect of statin therapy. A potential causal relationship between statin therapy and the risk of developing diabetes was first noted in a post-hoc analysis of the WOSCOPS (West of Scotland Coronary Prevention Study) trial, which revealed a borderline significant lower association of new diagnosis over 5 years (HR 0.7, placebo). The first prospective analysis occurred in the JUPITER (Crestor 20mg Versus Placebo in Prevention of Cardiovascular Events) trial, which had a prespecified outcome of newly diagnosed diabetes; it showed an incident increase of 0.6% (relative increase 24%) over 1.9 years by physician report with notably no change in fasting glucose levels over this time.

Subsequent large meta-analyses depending on study-level data have differed in their conclusions, showing proportional increases (OR) for the new diagnosis of diabetes to be around 10%. All of these analyses are limited by the varied definition and criterion for diabetes between trials, as well as the lack of diabetes as a pre-specified outcome in the overwhelming majority of studies.

The diabetogenic risk of statin therapy appears largely confined to patients with obesity, metabolic syndrome and prediabetes. One interpretation is that statin therapy accelerates the onset of diabetes amongst those most at risk for insulin resistance. What remains unclear is whether the duration of statin therapy affects this possible risk; whether the diabetogenic effect of statins would be reversible at all is an active area of research. In summary, statin therapy seems to modestly increase the risk of developing diabetes, albeit via mechanisms that are not well understood, with intensive dosing over 5 years.

The absolute increase in average hemoglobin A1c is small and of questionable clinical significance, particularly in light of clinical evidence that suggests A1c and blood sugar control alone are coarse measures for cardiovascular and macrovascular outcomes in diabetes. The increased risk of diabetes appears highest for a population similarly at risk for CVD (preexisting obesity, metabolic syndrome). Since statins are well established to substantially reduce cardiovascular events in those with and without diabetes, there is no reason to stop therapy. Clinical care should focus on increased efforts toward lifestyle modification, periodic diabetes screening and statin initiation and persistence.

Liver Function

Statins act in the liver to inhibit HMG-CoA reductase which temporarily depletes intracellular cholesterol and in turn induces production of LDL receptors. Statins have liver effects that range from mild transaminase elevation to very rare hepatotoxicity with severe liver injury. In about 1% of patients statins cause asymptomatic and dose-related elevations in transaminases greater than 3 times upper limit of normal, although this does not indicate either hepatocellular injury or liver synthetic dysfunction. Such increases nearly always demonstrate an ALT greater than AST, important in distinguishing liver from muscle related sources of the latter.

While no clear mechanism has been clarified as to why low-level transaminase elevation occurs in some and not others, to-date no clinical sequelae have been noted. Clinically significant statin hepatotoxicity is an extremely rare event, occurring in about 0.001% of patients. No clear pattern of prior transaminase elevation has been found in those patients, and it is no longer recommended to routinely monitor transaminase levels on statin therapy. Some experts suggest that the clinician should obtain baseline liver function studies prior to statin initiation, either for future comparison or to identify those individuals with pre-existing dysfunction who may be at theoretical risk of drug-related injury. It is not currently possible to predict which patients will develop hepatotoxicity, and providers must be alert to symptoms and signs of this rare complication.

Neurologic Effects

Some epidemiologic studies have found an inverse relationship between cholesterol levels and the risk of hemorrhagic stroke. The available aggregate data show no such increased risk in a primary prevention population. Fall risks may be increased in secondary stroke prevention, though the absolute risk appears quite small relative to the benefit of overall stroke and vascular event rate reduction. Reports of nonserious reversible forgetfulness and other forms of mild cognitive impairment often delineated in are also not uncommon in older adults. The available evidence indicates that statins do not increase the risk of such disorders. Lastly, an association between the development of peripheral neuropathy, a common clinical problem, and statin use is noted in observational studies. Currently, large epidemiologic studies are inconsistent and randomized controlled trial support no causal relationship.

Steroid Hormone Function

As inhibitors of steroid biosynthesis, a statin's theoretical effect on steroid hormone production and function has been investigated. Statins have minimal if any clinically relevant effect on steroidogenesis. Levels and rhythms of ACTH, cortisol, LH and FSH seem unaffected.

Only two controlled studies have examined male gonadal function, with one showing only a significant reduction in bioavailable testosterone (10%) with 80mg of simvastatin over 12 weeks (free and total testosterone were unaffected). There is no clear association with muscle-mass, sexual function or well-being is evident, and no connection to erectile dysfunction. With a wide range of clinically normal plasma testosterone levels in men, the impact of this finding is unclear, particularly in the setting of absent clinical symptoms.

Cataract Formation

A few animal studies suggested the formation of subscapular lens opacities when statins are given in doses well above maximal human doses. Observational and epidemiologic studies have shown both increased and decreased effect of small size. However, randomized controlled trials show no consistent increase in cataracts with the clinical use of statins.

Renal Function

The adverse renal effects of statins span asymptomatic proteinuria to severe AKI associated with rhabdomyolysis. Rosuvastatin can cause dipstick-positive proteinuria and microscopic hematuria at the maximal dose of 40 mg/day, an effect that is generally transient and not associated with reduced renal function. Meta-analyses and prospective trials (JUPITER) have shown no increase in renal injury nor decline in renal function with long-term rosuvastatin use (albeit on 20mg).

All statins have been shown to cause AKI, albeit via the mechanism of diffuse skeletal muscle injury, rhabdomyolysis and subsequent myoglobinuria. Such events are more likely with higher doses of statins and amongst patients with known interacting medications. However, in individuals without rare rhabdomyolysis, statins do not cause acute renal injury or worsen proteinuria long term.

Tendon Injury

Rare case reports of spontaneous tendinitis and tendon rupture in users of statins have been reported since the early 1990s. Achilles tenosynovitis can occur in people with FH, and initial statin treatment in these individuals can be associated with an increased risk. The possible risk for Achilles tenosynovitis in those with FH might be due to a rapid reduction of plasma cholesterol rather than statin effect, and clinicians should be aware of this uncommon phenomenon. All studies of statins which note tendinitis and tendon rupture are observational only and show no consistent difference between users and nonusers. There is no good evidence to suggest that statin use increases the risk of either tendonitis or tendon rupture.

Oncologic Risk

There is no current link between statins and cancer in humans. Appreciating the time constraints of our largest and most robust randomized controlled trials (currently 5 to 7 years), there is no current evidence that statins cause an increase in cancer. The quantity and quality of cancer incident data available for statins likely exceeds any other class of available drug.

Drug-Drug Interactions

All statins undergo hepatic first-pass metabolism, which accounts for their low bioavailability (as low as 5%), though notably pitavastatin has the highest bioavailability of the class at 50%. Simvastatin and lovastatin the only two of the class administered as prodrugs rather than as orally active metabolites, increasing the extent of their hepatic metabolism. Drugs which undergo extensive hepatic first pass metabolism are often vulnerable to drug-drug interactions mediated by CYP450 isoenzymes, and the effect of statins in the body is almost universally affected by the concurrent metabolism of other pharmacologic agents with induce or inhibit CYP450; rarely do statins affect the metabolism of another drug. The one known exception to this is the effect of some statins on vitamin K antagonists; warfarin will often require a dose reduction when given alongside statins.

In most cases, interacting drugs increase plasma concentrations of statins and their active metabolites, increasing the risk for myopathy and rhabdomyolysis. As prodrugs undergoing more extensive hepatic metabolism, simvastatin and lovastatin have the greatest and most significant number of drug-drug interactions. For both of these agents, avoidance of macrolide antibiotics, antifungal azoles and the immunosuppressant cyclosporine are important in clinical practice. Clinicians should be aware too of necessary dose-reductions of non-dihydropyridine calcium channel blockers.

Age and Ethnic Groups

Large randomized controlled trials have included individuals over 65 years as well those in their 70s and 80s for treatment periods of up to 5 years. Though the risk of myopathy and rhabdomyolysis are about twice that of younger individuals, absolute risk remains low. Clinicians must appreciate that more frequent comorbid conditions and longer medication lists in this age group raise risks of adverse events, particularly from drug-drug interactions.

Statins are less commonly used for children and adolescents, most often for those with heterozygous FH. Chronic diseases, which may increase the risk of atherosclerosis, may also prompt consideration of long-term statin therapy in younger individuals, and include CKD, Kawasaki disease with associated coronary aneurysms, juvenile rheumatoid arthritis and lupus. Both the American Heart Association and American Academy of Pediatrics have recommended and approved statin therapy for children at high risk of lipid abnormalities as early as age 8.

Currently, there is no evidence that statins increase musculoskeletal, hepatic, renal or drug-interacting effects in this age group greater than their adult counterparts. Importantly, there seems to be no effects on either growth velocity or sexual maturation in children. Statins begun in early life though are likely to be continued for long durations and continued collection of long-term safety data is a top priority.

For woman of childbearing age, current practice mandates a high threshold to initiate statin therapy. With the paucity of current data, statins remain a category X (contraindicated) drug in pregnancy by the FDA and should be stopped 3 months before attempting to conceive. Available evidence from those women exposed to statins in early pregnancy shows no increased risk of fetal abnormalities, and women with exposure can be reassured that careful and standard pre-natal screening for congenital abnormalities can determine the safety and/or desirability of carrying a pregnancy to term. Statins remain contraindicated for breastfeeding mothers as well.

Lastly, East Asians have long being considered as a separate and unique population regarding tolerance and dosing of statin therapy. Since introduction of statins three decades ago, East Asian populations have generally been prescribed lower doses due to beliefs that they are either more sensitive to these medications or have increased therapeutic response relative to western populations; the current prescribing information for rosuvastatin and simvastatin suggest lower doses in these populations. Pharmacokinetic studies do suggest greater plasma concentrations of some statins and their active metabolites in this population. Clinicians should remain aware of this possible greater sensitivity of East Asians to statins in general.

Specific Disease Considerations

Statins are shown with robust data to be safe in patients with CKD stages 2 to 4, as well as for patients on dialysis. Notably there is no evidence for cardiovascular benefit or CVD reduction in patients on dialysis, likely due to the poor prognosis and competing risks in this group.

There may be a small increased absolute risk for intracranial hemorrhage when statins are used for secondary stroke prevention in those with a history of intracranial hemorrhage. In the SPARCL (Stroke Prevention by Aggressive Reduction in Cholesterol Levels) trial, the overall benefit of atorvastatin in reducing recurrent stroke was somewhat offset by risk of brain hemorrhage. However, the benefits of reducing overall stroke and other vascular events generally outweighs the possible small absolute risk of hemorrhage in secondary prevention populations.

Statins do not cause progression of liver dysfunction in those with nonalcoholic fatty liver disease or chronic viral hepatitis C. There is no need to avoid statin therapy in patients with stable chronic liver disease or normal modestly elevated transaminases levels up to three times the upper limit of normal. Currently no reliable data exists showing statins are safe to use in advanced or decompensated liver disease.

For patients who have received solid organ or bone marrow transplantation, the immunosuppressant cyclosporine is well known to interact with all statins. Pravastatin, fluvastatin and rosuvastatin are currently recommended by the FDA for safe use in patients treated with cyclosporine, each at reduced doses; other transplant immunosuppressive agents have no evident interactions with statins.

Lastly, with the development of robust and tolerable antiretroviral regimens for HIV, men and women living with this disease are now living longer and suffering the complications of cardiovascular disease due at least in part to the high prevalence of dyslipidemia. Statins are often subject to the effects on metabolism of the antiretroviral drugs, and many of these agents will increase plasma concentrations of statins. Patients receiving protease inhibitors or a pharmacokinetically boosted antiretroviral regimen should avoid simvastatin and lovastatin, the two prodrug formulations of statins. Pitavastatin, atorvastatin, rosuvastatin and pravastatin can be considered safe, though dosing changes may be needed based on the antiretroviral regimen. The clinician should work to choose an appropriate formulation and dose of statin, rather than altering the patient's antiretroviral regimen.

Concluding Views and Evidence

Newman et al.'s Scientific Statement is a thorough and trustworthy comprehensive review. The evidence relied upon is current and the questions of tolerability and safety are timely in an era of widespread vascular disease in need of good medical intervention. Their work appropriately characterizes the degrees of statin related adverse events as often exaggerated and provides clinicians with a founded assurance that the benefits of this generally well tolerated class far outweigh their infrequent risks. Specialist and primary care providers alike should look to this work in support of our combined efforts to use directed therapy toward cardiovascular prevention.

Table 1: Estimated number of patients who experience benefit or harm if statintherapy is used in 10,000 individuals over 5 years and achieves a 77 mg/dL reduction in LDL-C levels.

Benefit	Estimated Number of Patients
Prevented first major vascular event (primary prevention)	500
Prevented recurrent major vascular event (secondary prevention)	1000
Harm
New diagnosis of diabetes mellitus	100
Statin associated muscle symptoms (without significant CK elevation)	<100
Myopathy (with CK elevation >10x ULN)	5
Rhabdomyolysis	1
Autoimmune myopathy	<1
Hemorrhagic stroke #	10
Severe liver disease	<1

References

1. Collins R, Reith C, Emberson J, et al. Interpretation of the evidence for the efficacy and safety of statin therapy. Lancet 2016;388:2532-61.
2. Koskinas KC, Siontis GCM, Piccolo R, et al. Effect of statins and non-statin LDL-lowering medications on cardiovascular outcomes in secondary prevention: a meta-analysis of randomized trials. Eur Heart J 2018;39:1172-80.
3. Mercado C, DeSimone AK, Odom E, Gillespie C, Ayala C, Loustalot F. Prevalence of cholesterol treatment eligibility and medication use among adults - United States, 2005-2012. MMWR Morb Mortal Wkly Rep 2015;64:1305-11.
4. Newman CB, Preiss D, Tobert JA, et al. Statin safety and associated adverse events: a scientific statement from the American Heart Association. Arterioscler Thromb Vasc Biol 2018. [Epub ahead of print].

Clinical Topics: Anticoagulation Management, Cardiovascular Care Team, Congenital Heart Disease and Pediatric Cardiology, Diabetes and Cardiometabolic Disease, Dyslipidemia, Prevention, Stable Ischemic Heart Disease, Vascular Medicine, Congenital Heart Disease, CHD and Pediatrics and Arrhythmias, CHD and Pediatrics and Prevention, CHD and Pediatrics and Quality Improvement, Lipid Metabolism, Nonstatins, Novel Agents, Statins, Diet, Chronic Angina

Keywords: Acute Kidney Injury, American Heart Association, Atherosclerosis, Adrenocorticotropic Hormone, Anti-Bacterial Agents, Biological Availability, Anticoagulants, Arthritis, Juvenile, Bone Marrow Transplantation, Blood Glucose, Calcium Channel Blockers, Carcinoma, Hepatocellular, Breast Feeding, Cataract, Cholesterol, Chronic Disease, Coronary Aneurysm, Coronary Aneurysm, Creatine Kinase, Cyclosporine, Diabetes Mellitus, Drug Interactions, Dyslipidemias, Epidemiologic Studies, Erectile Dysfunction, Ethnic Groups, Fatty Acids, Monounsaturated, Fasting, Glucose, Hematuria, Hepatitis C, HIV Infections, Hydrocortisone, Hydroxymethylglutaryl CoA Reductases, Hypothyroidism, Immunosuppressive Agents, Indoles, Insulin Resistance, Intracranial Hemorrhages, Isoenzymes, Life Style, Liver Diseases, Liver Neoplasms, Lovastatin, Macrolides, Metabolic Syndrome, Mucocutaneous Lymph Node Syndrome, Muscular Diseases, Muscle, Skeletal, Myalgia, Myoglobinuria, Neoplasms, Obesity, Pediatrics, Peripheral Nervous System Diseases, Pravastatin, Prediabetic State, Pregnancy, Prevalence, Primary Health Care, Primary Prevention, Prodrugs, Prognosis, Prospective Studies, Protease Inhibitors, Proteinuria, Quinolines, Receptors, LDL, Renal Dialysis, Renal Insufficiency, Chronic, Rhabdomyolysis, Risk Factors, Secondary Prevention, Sexual Maturation, Simvastatin, Stroke, Tendon Injuries, Tendons, Tenosynovitis, Testosterone, Transaminases, Vitamin K, Warfarin

< Back to Listings