Effect of statins on fasting glucose in non-diabetic individuals: nationwide population-based health examination in Korea

Abstract

Background: Increasing evidence suggest that statin therapy has a diabetogenic effect. Individual types of statin may have a different effect on glucose metabolism. Using the repeated nationwide population-based health screening data in Korea, we investigated the longitudinal changes in fasting glucose level of non-diabetic individuals by use of statins.

Methods: From the National Health Screening Cohort, we included 379,865 non-diabetic individuals who had ≥ 2 health screening examinations with fasting blood glucose level measured in 2002-2013. Using the prescription records of statins in the database, we calculated the proportion of days covered (PDC) and average number of defined daily doses per day (anDDD) by statins. We constructed multivariate linear mixed models to evaluate the effects of statins on the changes in fasting glucose (Δglu).

Results: High PDC by statins had a significant positive effect on Δglu (coefficient for PDC 0.093 mmol/L, standard error 0.007, p < 0.001). anDDD by statins was also positively associated with Δglu (coefficient for anDDD 0.119 mmol/L, standard error 0.009, p < 0.001). Unlike statins, the PDC by fibrate and ezetimibe were not significantly associated with Δglu. There was no significant interaction effect on Δglu between time interval and statin. Considering individual types of statins, use of atorvastatin, rosuvastatin, pitavastatin, and simvastatin were significantly associated with increase of Δglu. Pravastatin, lovastatin, and fluvastatin were also positively associated with Δglu, but were not statistically significant.

Conclusions: More adherent and intensive use of statins was significantly associated with an increase in fasting glucose of non-diabetic individuals. In subgroup analysis of individual statins, use of atorvastatin, rosuvastatin, pitavastatin and simvastatin had significant association with increase in fasting glucose. Pravastatin, lovastatin, and fluvastatin had non-significant trend toward an increased fasting glucose. Our findings suggest the medication class effect of statins inducing hyperglycemia.

Keywords: Diabetes mellitus; Glucose; Health examination; Statin.

Fig. 1

Flow chart of patients according to inclusion and exclusion criteria. NHIS-HEALS, the National Health Insurance Service-National Health Screening Cohort in Korea. *Those who met the exclusion criteria in any time of study period were excluded

Fig. 2

Effects of statin therapy on the change in fasting glucose over time. Plots illustrate the estimated change in fasting glucose (line) and 95% confidence interval (shadow) based on the multivariate linear mixed models adjusted for the variables listed in Table 3. a According to adherent use of statin therapy (PDC). b According to intensity of statin therapy (anDDD). PDC, proportion of days covered by statins for the time period between baseline and serial measurements of fasting glucose; anDDD, average number of defined daily doses per day for the time period between baseline and serial measurements of fasting glucose

Fig. 3

Effects of individual statins on the change in fasting glucose. Data are coefficients (β) and 95% CIs of individual statins on the change in fasting glucose derived by the multivariate linear-mixed models adjusted for the variables listed in Table 3. PDC, proportion of days covered by statins for the period between baseline and serial measurements of fasting glucose; anDDD, average number of defined daily doses per day; CI confidence interval