ABSTRACT
OBJECTIVE: 3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors (statins) are used to control blood cholesterol levels and reduce cardiovascular disease. It has been repeatedly reported that statins may cause new-onset diabetes mellitus (DM). However, limited evidence exists from direct head to head comparisons of statins on whether the risk of DM differs among statins. We investigated the risk of development of new-onset diabetes in subjects treated with different statins. METHODS: We retrospectively enrolled consecutive 3680 patients without DM or impaired fasting glucose who started receiving statin treatment for cholesterol control. We evaluated the incidence of new-onset diabetes according to the type of statin. RESULTS: The mean duration of follow-up was 62.6±15.3 months. The incidence of DM was significantly higher in the pitavastatin group (49 of 628; 7.8%) compared to that in the other statin groups [atorvastatin (68 of 1327; 5.1%), rosuvastatin (77 of 1191; 6.5%), simvastatin (11 of 326; 3.4%), and pravastatin (12 of 298; 5.8%); p=0.041]. The risk of diabetes was the highest in the pitavastatin group compared with that in the simvastatin group [hazard ratio (HR)=2.68, p=0.011]. Other statins showed no significant risk differences compared to that for simvastatin. Fasting blood glucose (FBG) level at baseline and body-mass index (BMI) were associated with the development of diabetes [FBG, HR=1.11, p<0.001; BMI, HR=1.02, p=0.005]. CONCLUSIONS: Among the five statins, pitavastatin showed the strongest effect on the development of new-onset diabetes.
