Myalgia

Myalgia is by far the most common symptom associated with statins. It is the symptom that has probably generated the most lay interest and is often cited as a major contraindication, or at least concern, to the use of statins. Yet its frequency is the issue that the existing literature has failed to satisfactorily answer; a voluminous literature has by and large concluded that the prevalence of myalgia is much the same in statin and placebo treated groups. However, with the exception of the STOMP trial, the literature has to be largely dismissed because of inadequate approaches to data collection.3 Myalgia is an extremely common complaint in general practice, and the nocebo effect in statin users may be relatively common.7 The STOMP study used stricter definitions than preceding studies, including the obvious temporal relationship between starting a statin (atorvastatin in this study) and the onset of myalgia, resolution on stopping the statin and recurrence of the same symptoms on rechallenge. Furthermore, those taking statins had a consistent pattern of myalgia, mainly involving the proximal lower limb muscles, whereas muscle symptoms in the placebo group, including myalgia, were more diverse. In STOMP, 19/203 (9%) of people treated with atorvastatin had myalgia. None of them had evidence of impaired strength or exercise function. The serum CK was modestly elevated in 40/203 patients on atorvastatin but it was not stated how many of the 19 patients with myalgia had a raised CK. No patient had a marked (>10 times upper limit of normal) increase in CK. Numerous other studies have shown that statins may increase CK, but it is difficult to interpret these data because of very poorly defined study methods. It is clear that not all patients with myalgia have a raised CK and that not all those with a raised CK have any muscular complaints.

Myalgia on physical exercise may be more prominent in those on statins, without certain evidence of impaired muscle function,9 but statins in professional athletes may limit performance.10

There has been much debate as to whether low concentrations of coenzyme Q (CoQ) or vitamin D increase the risk of myalgia and if their supplementation improves it. Statins inhibit the metabolic pathway for which CoQ is a distal component and statins lower tissue CoQ concentrations. However, the evidence to date provides little evidence of a significant association with muscle symptoms and does not support using CoQ supplements.11

Several studies have reported an association between myalgia and low serum vitamin D concentrations, with symptom improvement with supplementation to restore normal levels,12–15 whereas others found no association.16 17 We clearly need further studies but in the meantime it is appropriate to check serum vitamin D concentrations in people with myalgia and to give replacement therapy as required (box 2).

An occasional issue is the patient with a ‘very high’ serum CK, either with or without muscle symptoms, and the concern as to whether persistence may eventually lead to permanent muscle damage. There is little evidence to suggest that any such damage ensues but many physicians feel uncomfortable with a serum concentration above, say, 1000 iu/L (depending on definitions/age/sex/race, this may be 2–5 times the quoted upper limit of normal). The possibility of a statin-induced immune-mediated necrotising myopathy is discussed below. In practice, in such patients with marked CK elevation, it is probably reasonable to follow the same path as with a patient with troublesome myalgia—with or without a raised CK—discussed below.

Patients often tolerate mild myalgia and non-specific aches and pains. If not, the options include a trial of vitamin D supplementation if serum concentrations are low, withdrawal of the statin, dose reduction or trial of a different statin (box 2). If there is uncertainty as to whether it is a statin-induced problem, or the serum CK is very high, then I suggest stopping the statin and observing the response. For milder symptoms, one might in the first instance just try reducing the dose. If the myalgia or high CK persists, then alternative explanations need to be sought. A frequent practical problem is that serum CK is rarely measured before starting a statin. The elevated CK may therefore have preceded the introduction of the statin and reflect a pre-existing myopathy that may have been asymptomatic, or have been oligosymptomatic, and the significance of symptoms unrecognised. There are cases in the literature, and in my own practice, of McArdle’s disease and other genetic myopathies that have been identified only when patients reported muscle problems after starting a statin, but subsequent assessment revealed that they had in fact had longstanding but unrecognised classical symptoms.

If myalgia resolves on statin withdrawal, and assuming that the patient needs to be on a statin, then the options include either reintroducing the same statin at a lower dose, with monitoring of clinical response, serum CK and adequate reduction in cholesterol or trying an alternative statin. Despite some reports that a particular statin appears to cause fewer problems than others or theoretical arguments about differing lipid solubility between different statins, there is little evidence to favour any one agent, and the choice can be made on personal experience and, increasingly, availability of specific drugs from local formularies, determined in part by cost. If need be, several different drugs can be tried.

Polymorphisms in genes coding for transporter proteins have been associated with an increased risk of statin-induced myalgia and rhabdomyolysis. An SLCO1B1 gene polymorphism (521T>C) particularly associates with simvastatin-induced problems.18 However, the situation is more complex than that and sex and ethnicity are also relevant. Thus, the same SLCO1B1 polymorphism has recently been shown to be associated with increased myotoxicity in Chinese people receiving rosuvastatin.19 There is as yet no evidence to support the cost effectiveness of screening for this polymorphism or for those in other genes, before starting statins.

In contrast to statin-induced immune-mediated necrotising myopathy, discussed below, the presence of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) antibodies is not associated with myalgia.20

Rhabdomyolysis

While there is no universally accepted definition, rhabdomyolysis describes the widespread breakdown of muscle fibres, which releases CK and myoglobin into the circulation. The kidneys excrete myoglobin and this may discolour the urine dark brown/red—frequently referred to as cola-coloured or tea-coloured urine—and often misdiagnosed by visual inspection and dipstick testing as haemoglobinuria. Involvement may be restricted to one muscle or group, but is usually more widespread. The typical clinical presentation is with pain, weakness and sometimes muscle swelling evolving subacutely over hours to a few days (see Case History). With a typical clinical picture and gross elevation of the serum CK, the diagnosis is usually straightforward. There is not always visible urine discolouration, and there is no great value in estimating urinary myoglobin. As with the literature on statins and myalgia, that referring to statin-induced rhabdomyolysis is also of poor quality, with definitions that would not be accepted by muscle specialists (eg, raised serum CK without muscle symptoms). Despite these reservations, there is no doubt that any statin at sufficiently high dose may precipitate rhabdomyolysis. Some may be more of a risk than others, and cerivastatin was withdrawn from the market after reports of a cluster of cases, including several deaths in a short period of time. The exact mechanism of statin-induced rhabdomyolysis is unknown but probably relates to a consequence of downstream effects from blockage of the metabolic pathway inhibited by statins at the HMGCR enzyme. The most important factors triggering rhabdomyolysis are high dose and iatrogenically induced high serum concentrations precipitated by drugs that inhibit statin breakdown (box 3). Additional risk factors, knowledge of which is not particularly helpful clinically, include older age, hypothyroidism, renal impairment, other comorbidities and polypharmacy.

Statins, with the exception of pravastatin, are metabolised by the cytochrome P-450 enzyme system, as are numerous other therapeutic drugs. There is a huge potential for drug-to-drug interactions; in the present context, this means that there are many commonly used drugs that inhibit statin metabolism and may therefore precipitate rhabdomyolysis. Clinically, the major problems relate to concurrent use of various cardiovascular drugs, fibrates (often coprescribed with statins in those with severe hypercholesterolaemia and other vascular risk factors) and antimicrobials; box 4 lists some of the more common culprits, but for excellent comprehensive summaries see Wiggins et al and Hylton et al.21 22 Cardiovascular drugs are of course likely to be prescribed long term, whereas most antimicrobial agents are used only briefly; arguably in the latter situation, the statin could be temporarily withheld if there is no alternative antibiotic. Cardiovascular drugs such as amlodipine can be used safely if the dose of statin is restricted (eg, simvastatin restricted to maximum dose of 20 mg with amlodipine).21 Besides these two major drug groups, many other drugs may precipitate rhabdomyolysis through cytochrome p450 inhibition;23 these include ciclosporin, which is fairly often used in association with statins (eg, for immune suppression following kidney transplantation in a diabetic).

The management of statin-induced rhabdomyolysis primarily involves statin withdrawal (box 5). Myoglobinuria carries the risk of acute tubular necrosis and renal failure. Although forced alkaline diuresis may seem appropriate in theory, there is little evidence to support it and I simply recommend relatively high fluid input, to induce diuresis, for several days. Patients may also need pain relief. Particularly in the elderly, it is important to have early physiotherapy input to reduce risk of contractures and to maintain range of movement when very weak and then to aid mobilisation at the earliest opportunity.

Muscle biopsy is unnecessary if the clinical diagnosis is secure. Rarely, and arguably most often considered because of ignorance of the mechanism or failure to take a detailed drug history, muscle biopsy may help to exclude an inflammatory myopathy that would require use of corticosteroids. Dermatomyositis can have a similar subacute onset and could conceivably be confused with drug-induced rhabdomyolysis. Statin-induced immune-mediated necrotising myopathy would rarely present so acutely.

An episode of statin-induced rhabdomyolysis does not necessarily contraindicate the reintroduction of statins. If there is a recognised specific interaction (eg, simvastatin and erythromycin, see Case Report), then there is no reason to withhold the original statin. In other situations (eg, simvastatin and amlodipine), it may be appropriate to reduce the dose of amlodipine. Pravastatin, which is not cytochrome p450 metabolised, may be considered in combination with other drugs that are cytochrome p450 metabolised, but it is not free of drug-to-drug interactions and, for example, is contraindicated in combination with gemfibrozil, since both are metabolised via OAT P1.

Immune-mediated necrotising myopathy

Although the least common statin-related myopathy, this is of considerable clinical importance and research interest. Whereas the disorders discussed earlier resolve on statin withdrawal, this condition may progress and requires specific treatment (immunotherapy).

Most immune-mediated myopathies are characterised pathologically by major inflammatory infiltrates in muscle, designated myositis, including dermatomyositis, polymyositis and antisynthetase syndromes. Clinically, they are characterised by progressive proximal weakness, and many are associated with myositis-specific or myositis-associated antibodies in the serum. The immune-mediated necrotising myopathies are a recently delineated group of disorders with a similar clinical presentation to myositis (ie, progressive weakness) but muscle biopsy shows few or no inflammatory infiltrates despite frequent necrotic and regenerating muscle fibres (figure 1). Critically, like myositis, they respond to immunotherapies. There are three main forms of immune-mediated necrotising myopathies; those associate with anti-signal recognition particle antibodies, those with anti-HMGCR antibodies and seronegative cases.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

H&E section showing widespread necrotic and regenerating fibres, but no significant inflammatory infiltrates. Photograph courtesy of Dr Monika Hofer.

The link between statin use and the development of progressive weakness, due to an immune-mediated necrotising myopathy associated with anti-HMGCR antibodies, was initially shown in 2010,24 but there have since been numerous confirmatory reports.25–27 Such patients may have been taking statins for months or occasionally years. They then develop slowly progressive proximal, mainly lower limb, weakness with a very high serum CK (typically >5000 iu/L). Some have myalgia, but the process may be painless. The symptoms usually progress despite stopping the statin. In two cases reported in this edition of Practical Neurology,28 the myopathy resolved on statin withdrawal without additional treatment; however, in an interesting variation on the theme, the myopathy recurred in one patient when statins were introduced 4 years later did not resolve on statin withdrawal the second time and so required immunosuppressant treatment. The muscle biopsy shows features of an immune-mediated necrotising myopathy. Intensive treatment with corticosteroids, with or without second-line agents such as methotrexate, may be effective, but there is increasing evidence that many patients in addition require intravenous immunoglobulin to induce remission, and in some patients, this has been used as monotherapy.29 There is also evidence that younger patients may have more severe disease and be more resistant to therapy than older patients, which may influence therapeutic approaches.30

Anti-HMGCR associated immune-mediated necrotising myopathy is strongly associated with HLA DRB 1*11:01.31

There is no evidence that HMGCR antibodies are directly pathogenic. Increasingly, they have been found in association with immune-mediated necrotising myopathies in patients who have never been exposed to statins, including children. Furthermore, there is mounting evidence of an increased incidence of cancer in people with HMGCR related immune-mediated necrotising myopathy, with or without a history of statin exposure.32 This needs further study, but on current evidence, it seems reasonable practice to consider the possibility of underlying malignancy in adults presenting with immune-mediated necrotising myopathy with HMGCR antibodies, as one should do with patients presenting with dermatomyositis. Thus, patients need clinical assessment, imaging (CT scan of chest/abdomen/pelvis or positron-emission tomography scanning), mammography, serum tumour markers; clinicians should maintain suspicion for 1–2 years after first presentation. The risk is probably greater in older patients.

The diagnosis of anti-HMGCR immune-mediated necrotising myopathy may not be considered initially because of the lack of an immediate temporal relationship with the starting of statins. When it is, identifying the characteristic features (box 6) should lead to rapid diagnosis. The statin should be stopped. If the weakness is mild, it is appropriate to wait a few weeks to see if there is spontaneous improvement, although that is unlikely.

Persisting myalgia after statin withdrawal

A sometimes challenging problem is the patient who develops myalgia when put on statins and the symptoms persist despite drug withdrawal. At present, the only recognised pathological mechanism to explain a persisting myopathy after statin withdrawal is the syndrome of immune-mediated necrotising myopathy, described above. Those patients may or may not have myalgia, but invariably have progressive weakness, elevated (usually marked) serum CK and HMGCR antibodies. The more common problem is the patient with persisting myalgia without weakness. Box 7 shows an approach to their assessment.

It is of course important to clarify that the myalgia did indeed start after starting the statin; noting that statin-induced myalgia typically develops very shortly after starting the treatment or increasing the dose. Myalgia developing months or years after starting treatment is much less likely to be related. As noted above, myalgia due to a pre-existing but unrecognised disorder may apparently become symptomatic only when the patient and doctor are looking out for statin-related problems.

The specific characteristics, and in particular the distribution of the myalgia, can be helpful. As noted in the STOMP study, statin-induced myalgia mainly affects the lower limbs, both proximally and in the calves, whereas that occurring in people taking placebo tends to be more generalised throughout the body.3 The myalgia associated with vitamin D deficiency can also be generalised, so blood tests should include vitamin D as well as CK.

Probably the most common challenge is a patient with rather diffuse myalgia, present at rest but worsened by activity, who has no demonstrable weakness and a normal serum CK. In this situation, electromyography and muscle biopsy are very unlikely to provide an answer, but if undertaken—sometimes at the patient’s insistence—may confuse matters further because they show ‘minor changes of uncertain significance’. Clinicians should certainly consider electromyography and biopsy if the serum CK is clearly elevated, but even so, if myalgia is the sole feature, these tests are rarely productive. Nevertheless, many neuromuscular specialists have anecdotes about conditions such as myotonic dystrophy type 2, which can present with a ‘fibromyalgia-like’ syndrome in middle age,33 although more recent studies have challenged the association.34

Further research will possibly reveal a mechanism whereby statins may precipitate myalgia with persistence after withdrawal but as with myalgia in the general population, many cases remain unexplained.

Does a pre-existing myopathy increase the risk of developing a statin-related myopathy?

The simple, and clinical relevant answer, is ‘no’. Vladutiu and colleagues suggested an association between carrier status for McArdle’s disease or carnitine palmitoyltransferase deficiency, or homozygosity for myoadenylate deaminase deficiency, with an increased risk of statin-induced muscle symptoms, including pain, fatigue, weakness and rhabdomyolysis.35 However, numerous methodological concerns (especially patient self-reporting and most patients having no formal clinical assessments) greatly limit their conclusions, and no subsequent studies have confirmed such associations.

However, it would be naïve to ignore two truths. First, that many people with a neuromuscular disorder are understandably introspective about their condition and associated symptoms. Second, that there is a huge lay-press literature ‘warning’ of the dangers of statins (see comments about nocebo effect above). For these reasons, clinicians should take a precautionary approach to prescribing statins in these patients. This is summarised at the end of the following section considering the reverse phenomenon—the possibility that statins exacerbate the pre-existing neuromuscular disorder.

Can statins exacerbate a pre-existing myopathy?

The answer is a very guarded ‘yes’. However, there is probably no situation where somebody with a pre-existing neuromuscular disorder should withhold statins if there is a clear clinical need for them. As discussed immediately above, it is essential to engage fully with the patient in discussing issues and to follow the precautionary approach outlined below.

There is little literature concerning statins exacerbating a pre-existing myopathy; considering the many millions of doses prescribed, this in itself suggests that there is not a major problem. Statins impair CoQ synthesis and there has been much speculation, but no certain proof, that this may be one of the mechanisms by which statins cause myopathy. In many mitochondrial cytopathies, there is respiratory chain function impairment, and statin-induced lowering of CoQ might in theory exacerbate the underlying disorder. Two rather limited case reports of patients with the mtDNA A3243G mutation suggested that statins might have precipitated features of MELAS (mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes) syndrome in one36 and elevated serum CK in the other.37 I am aware of verbal anecdotal reports of statins apparently exacerbating fatigue and weakness in other mitochondrial disorders, but once again the nocebo effect needs to be considered, and there is a complete lack of literature to suggest that this is a major issue.

There is a small literature suggesting that statins may precipitate or exacerbate myasthenia gravis; however, myasthenia symptoms are notoriously variable, with patients frequently making associations between exacerbation of symptoms and possible environmental triggers. The most recent review identified fewer than 20 cases in the world literature, noting also that as many as 60 000 people in the USA alone may have myasthenia—with millions of the population receiving statins38 —any risk therefore seems to be very small.

Precautionary prescribing

Statins can be used in those with a pre-existing neuromuscular disorder, with only a low risk of either statin-related myopathy or exacerbation of the underlying disorder. Although many of the anxieties relating to statins may be unfounded, they are extensively discussed in the lay literature, and it is imperative to have an appropriate discussion with the patient and record this in their notes. Box 8 outlines a reasonable precautionary approach, and this should ensure that patients are not denied the potential benefits of statins and that any risks are minimal. It is always worth checking thyroid function if not done recently, since hypothyroidism can cause hypercholesterolaemia and a raised serum CK. It also increases the risk of statin-induced problems including myalgia and rhabdomyolysis. On current evidence, there is little if any difference in risk of myotoxicity between the different statins, so use whatever is local policy, but start at a low dose, increase the dose as needed depending on the serum cholesterol and continue monitoring clinically and biochemically.