Impact of low dose atorvastatin on development of new-onset diabetes mellitus in Asian population: Three-year clinical outcomes

- We investigate the impact of low dose atorvastatin (LDA) on diabetes mellitus.
- Propensity score matched analysis was performed to adjust potential confounders.
- The use of LDA tended to be a risk factor for new onset diabetes mellitus (NODM).
- The use of LDA reduced clinical events similar to the control group.
- Large randomized controlled trials will be needed to get the final conclusion.

BackgroundHigh dose atorvastatin is known to be associated with new onset diabetes mellitus (NODM) in patients with high risk for developing diabetes mellitus (DM). However, low dose atorvastatin is more commonly used as compared with high dose atorvastatin. The aim of this study is to investigate the impact of low dose atorvastatin (LDA, 10 mg or 20 mg) on the development of NODM up to three years in Asian patients.MethodsFrom January 2004 to September 2009, we investigated a total of 3566 patients who did not have DM. To adjust for potential confounders, a propensity score matching (PSM) analysis was performed using the logistic regression model. After PSM (C-statistics: 0.851), a total of 818 patients (LDA group, n = 409 patients and control group, n = 409 patients) were enrolled for analysis.ResultsBefore PSM, the cumulative incidence of NODM (5.8% vs. 2.1%, p < 0.001), myocardial infarction (0.5% vs. 0.1%, p-value = 0.007), and major adverse cardio-cerebral event (MACCE, 1.8% vs. 0.7%, p-value = 0.012) at three-years were higher in the LAD group. However, after PSM, there was a trend toward higher incidence of NODM (5.9% vs. 3.2%, p = 0.064) in the LDA group, but the incidence of MACCE (1.2% vs. 1.5%, p-value = 1.000) was similar between the two groups. In multivariable analysis, the LDA administration was tended to be an independent predictor of NODM (OR: 1.99, 95% CI: 1.00-3.98, p-value 0.050).ConclusionsIn this study, the use of LDA tended to be a risk factor for NODM in Asian patients and reduced clinical events similar to the control group. However, large-scale randomized controlled trials will be needed to get the final conclusion.