The impact of diabetes mellitus medication on the incidence of endogenous endophthalmitis

PLoS One. 2020 Jan 10;15(1):e0227442. doi: 10.1371/journal.pone.0227442. eCollection 2020.

Abstract

Purpose: This study aimed to evaluate the relationship between diabetic mellitus (DM) treatment and the incidence rate of endogenous endophthalmitis (EE).

Design: This study used a matched cohort design. We utilized the Longitudinal Health Insurance Database to identify outpatients and inpatients who were diagnosed with DM and treated with medication from 2000 to 2010.

Methods: Several factors and different DM medications were also investigated. The influence of DM medication on the incidence of EE was examined by using Cox proportional hazards regression models, and the hazard ratios and 95% confidence intervals were determined.

Results: The cumulative incidence of EE was lower in DM patients treated with medication than in subjects in the control group (P = 0.002). The adjusted hazard ratio (AHR) was 0.47-fold lower in the treatment group than in the control group (P = 0.004). With respect to DM medication, single-agent therapy with insulin, metformin, gliclazide, glimepiride, or repaglinide and combination therapy with glimepiride/metformin or repaglinide/metformin were associated with decreased AHRs (0.257-0.544, all P<0.05).

Conclusions: Diabetic patients treated with medication had lower AHRs than those in the control group. Further stratification indicated that liver abscess, liver disease DM patients who were treated with medication had a lower risk of developing EE. Several specific DM medications may decrease the incidence of EE.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Case-Control Studies
  • Databases, Factual
  • Diabetes Complications / epidemiology
  • Diabetes Complications / etiology
  • Diabetes Mellitus / drug therapy*
  • Diabetes Mellitus / epidemiology
  • Endophthalmitis / epidemiology
  • Endophthalmitis / prevention & control*
  • Female
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Proportional Hazards Models
  • Retrospective Studies

Grants and funding

This study was supported by grants from the Tri-Service General Hospital Research Foundation (TSGHC107-004 and TSGH-C107-090) and the Ministry of National Defense Medical Affairs Bureau (MAB-108-049). The sponsors had no roles in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.