To test the hypothesis that adverse effects of statins are only reported in excess in observational studies and not in blinded randomized trials.Design and Method:
We collated all reported AEs in hypertensive patients in the Lipid-Lowering arm of the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT-LLA) during the randomised, double-blind phase (when atorvastatin was compared with placebo) and the subsequent non-randomised un-blinded LLA-extension phase (when patients were offered open-label statin). Among them we blindly and independently adjudicated 4 prespecified AEs (outcomes) that have been claimed to be associated with statin use: muscle-related, erectile dysfunction (ED), sleep disturbance and cognitive impairment . We also performed a sensitivity analysis for all reported AEs.Results:
During 5.5 years of follow-up, 20.8% of 10,180 patients reported one (or more) of the 4 AEs. During the double- blind phase, those on statins (vs. placebo) reported a similar incidence of muscle- related AEs (HR 1.03[0.88 to1.21, p = 0.72]), a non significant reduction in ED(0.88 [0.75 to 1.04, p = 0.13]) and a significant reduction in sleep disturbance (0.60 [0.56 to 0.85, p < 0.001]). During the un-blinded phase, a significant increase in muscle related AEs was reported among statin users (vs. non-users) (1.41 [1.10 to 1.79, p = 0.006]). There were no significant differences in the reporting of ED or sleep disturbance among users and non-users. There were too few cases of cognitive impairment to warrant analysis. On sensitivity analyses, during the blinded phase, there was no significant difference in the proportion of patients reporting one or more of all AE's among those allocated to a statin (87.8%) or a placebo (88.8%) (p = 0.12).Conclusions:
These data show that under blinded conditions there is no true increase in AEs with the use of statin. However, with un- blinded observations, there was a significant increase in reporting of muscle-related AEs -an example of the nocebo effect.