Abstract 16898: Statin, A Double-Edge Sword in Pre-Diabetes: A Propensity Score-Matched Outcome Study

Abstract

Background The Food and Drug Administration updated a safety warning concerning increased risk of diabetes by statins. Whether pre-diabetes is more susceptible to diabetogenic effects of statins and the adverse outcomes after statins prevail therapeutic advantages were investigated in the current analysis. Methods and Results We used Taiwan National Health Insurance research database to identify pre-diabetic beneficiaries between 2001 and 2010. Dispensing and medical claims were linked to ascertain the endpoints of diabetes mellitus and death. A 1:1 propensity score-matched model based on demographics and comorbidities was constructed for statin users and controls. Of 3,314 pairs of subjects, there were 1,669 incident diabetes and 884 deaths during a median follow-up of 4.7 years. The hazard ratio (HR) of diabetes after statins was 1.21 (95% confidence intervals (CIs): 1.10-1.33) compared to controls and the risk for all-cause death was 43% less (95% CIs: 0.49-0.65) in the statin group. The risk of incident diabetes was increased by 2% for per 10% increment in statin dosage (95% CIs: 1.01-1.03) and by 7% for per year increment in exposure. The diabetogenic effect was more prominent in subjects receiving persistent up-stream statin therapy (HRs and 95% CIs: 1.11 (0.77-1.58) for the statin use after pre-diabetes, 1.20 (1.07-1.35) for the non-persistent use, and 1.51 (1.26-1.80) for the persistent use) (Fig. A) whereas mortality rates were similar (HRs and 95% CIs: 0.49 (0.26-0.92) for the statin use after pre-diabetes, 0.57 (0.49-66) for the non-persistent use, and 0.58 (0.43-0.77) for the persistent use (Fig. B). Conclusions Diabetogenic effects of statins were observed among subjects receiving up-stream therapy. The statin use after pre-diabets was not associated with clinically significant diabetes. The outcomes favored statin therapy in the prediabetic subjects.