Statin treatment and muscle symptoms: series of randomised, placebo controlled n-of-1 trials
Emily Herrett, Elizabeth Williamson, Kieran Brack, Danielle Beaumont, Alexander Perkins, Andrew Thayne, Haleema Shakur-Still, Ian Roberts, Danielle Prowse, Ben Goldacre, Tjeerd van Staa, Thomas M MacDonald, Jane Armitage, Jon Wimborne, Paula Melrose, Jayshireen Singh, Lucy Brooks, Michael Moore, Maurice Hoffman, Liam Smeeth, StatinWISE Trial Group - BMJ, 2021
In a series of n-of-1 RCTs of atorvastatin 20 mg versus placebo among people who had previously reported statin-associated muscle symptoms, there was no overall difference in muscle-symptom scores between statin and placebo periods (mean difference -0.11 on a 0-10 scale, 95% CI -0.36 to 0.14, P=0.40). Withdrawals for intolerable muscle symptoms were similar (9% statin vs 7% placebo), and two thirds of those completing the trial reported restarting statins. The null result indicates that muscle symptoms many patients attribute to statins are largely not caused by the statin itself.
Key findings
- In a series of n-of-1 RCTs of atorvastatin 20 mg vs placebo among 200 UK primary-care patients with prior statin-associated muscle symptoms (151 in the primary analysis), there was no overall difference in muscle-symptom scores between statin and placebo periods (mean difference -0.11 on a 0-10 scale, 95% CI -0.36 to 0.14, P=0.40).
- Withdrawals for intolerable muscle symptoms were similar: 18 participants (9%) during a statin period and 13 (7%) during a placebo period.
- Two thirds of those completing the trial reported restarting long-term statin treatment.
- The findings indicate that muscle symptoms many patients attribute to statins are largely not caused by the statin itself (nocebo-consistent).
Effect measures
- Other: Mean difference (statin minus placebo) -0.11 on a 0-10 visual-analogue muscle-symptom scale95% CI -0.36 to 0.14; P=0.40
- Risk Difference: Withdrawal for intolerable muscle symptoms: 9% (18/200) statin vs 7% (13/200) placebo
Why this evidence tier (High)
- Risk of bias:
- Randomized, double-blind, placebo-controlled series of n-of-1 trials across 50 UK primary-care sites; within-patient crossover design with blinded symptom scoring.
- Precision:
- Larger than SAMSON (200 enrolled, 151 analysed) with a tight, null confidence interval around the mean difference (-0.11, 95% CI -0.36 to 0.14).
- Directness:
- Directly tests whether muscle symptoms differ on statin vs placebo, in exactly the patients who report statin-associated muscle symptoms.
- Consistency:
- Concordant with SAMSON (Wood/Francis 2020) and the broader statin nocebo literature.
- Funding / COI:
- Funded by the NIHR (HTA programme) with the London School of Hygiene and Tropical Medicine; all authors completed the ICMJE disclosure form and declared no relevant financial relationships. Independent (non-industry) funding.
High certainty that, on average, atorvastatin does not increase muscle symptoms versus placebo in people who report statin-associated muscle symptoms.
- Population:
- 200 adults in UK primary care (50 sites, December 2016 to April 2018) who had recently stopped or were considering stopping statins because of muscle symptoms; 151 provided scores for at least one statin and one placebo period and were included in the primary analysis.
- Conflicts of interest:
- All authors completed the ICMJE uniform disclosure form. Funded by the National Institute for Health Research (NIHR) HTA programme with support from the London School of Hygiene and Tropical Medicine. Authors declared no financial relationships with organisations that might have an interest in the submitted work in the previous three years, and no other relationships that could appear to have influenced it. Trial registration ISRCTN30952488 / EudraCT 2016-000141-31. No retraction or erratum was found.
- Funding:
- National Institute for Health Research (NIHR) HTA programme, with the London School of Hygiene and Tropical Medicine.
Limitations
- Enrolled patients who had recently stopped or were considering stopping statins for muscle symptoms, so results address that group rather than all statin users.
- Of 200 enrolled, 151 contributed to the primary analysis; some participants did not complete enough periods.
- Muscle-symptom scores are self-reported on a subjective 0-10 scale.
- The nocebo interpretation is an editorial characterization of the null result, though well supported.