Error in USPSTF Report on Statin Use
The 2016 review for the US Preventive Services Task Force on statins for prevention of cardiovascular disease in adults had errors in the analysis of statins vs placebo and cardiovascular mortality. For the JUPITER trial, we interpreted “MI, stroke or cardiovascular death” as reported in the main trial publication as “myocardial death, stroke death, or cardiovascular death,” when it meant “nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death.” Therefore, the analysis erroneously included nonfatal myo cardial infarction and stroke events (83/8901 vs 157/8901 in the rosuvastatin vs placebo groups, respectively) in the analysis of cardiovascular mortality (Figure 3, panel B in the evidence review). The US Food and Drug Administration (FDA) review of atorvastatin reported 29 vs 37 cardiovascular mor tality events in the rosuvastatin vs placebo groups, respectively, in JUPITER. However, a subsequent publication from 2 of the original JUPITER authors reported numbers of confirmed cardiovascular deaths of 35 vs 43 (not including 16 vs 25 cases of sudden death) in the rosuvastatin vs placebo groups . For the ASTRONOMER trial, data for cardiovascular deaths were transposed from another trial (2/103 vs 12/79 for statins vs placebo); the correct data are 2/134 vs 5/135.