Abstract
OBJECTIVE To determine the independent association between diabetes and surgical site infection (SSI) across multiple surgical procedures. DESIGN Systematic review and meta-analysis. METHODS Studies indexed in PubMed published between December 1985 and through July 2015 were identified through the search terms "risk factors" or "glucose" and "surgical site infection." A total of 3,631 abstracts were identified through the initial search terms. Full texts were reviewed for 522 articles. Of these, 94 articles met the criteria for inclusion. Standardized data collection forms were used to extract study-specific estimates for diabetes, blood glucose levels, and body mass index (BMI). A random-effects meta-analysis was used to generate pooled estimates, and meta-regression was used to evaluate specific hypothesized sources of heterogeneity. RESULTS The primary outcome was SSI, as defined by the Centers for Disease Control and Prevention surveillance criteria. The overall effect size for the association between diabetes and SSI was odds ratio (OR)=1.53 (95% predictive interval [PI], 1.11-2.12; I2, 57.2%). SSI class, study design, or patient BMI did not significantly impact study results in a meta-regression model. The association was higher for cardiac surgery 2.03 (95% PI, 1.13-4.05) compared with surgeries of other types (P=.001). CONCLUSIONS These results support the consideration of diabetes as an independent risk factor for SSIs for multiple surgical procedure types. Continued efforts are needed to improve surgical outcomes for diabetic patients. Infect. Control Hosp. Epidemiol. 2015;37(1):88-99.