The association of diabetes and hyperglycemia with sepsis outcomes: a population-based cohort analysis.

Abstract

The independent association of diabetes and hyperglycemia on the outcomes of sepsis remains unclear. We conducted retrospective cohort analyses of outcomes among patients with community-onset sepsis admitted to Shamir Medical Center, Israel (08-12/2016). Statistical associations were queried by Cox and logistic regressions, controlled for by matched propensity score analyses. Among 1527 patients with community-onset sepsis, 469 (30.7%) were diabetic. Diabetic patients were significantly older, with advanced complexity of comorbidities, and were more often exposed to healthcare environments. Despite statistically significant univariable associations with in-hospital and 90-day mortality, the adjusted Hazard Ratios (aHR) were 1.21 95% CI 0.8-1.71, p = 0.29 and 1.13 95% CI 0.86-1.49, p = 0.37, respectively. However, hyperglycemia at admission (i.e., above 200mg/dl (was independently associated with: increased in-hospital mortality, aHR 1.48 95% CI 1.02-2.16, p = 0.037, 30-day mortality, aHR 1.8 95% CI 1.12-2.58, p = 0.001), and 90-day mortality, aHR 1.68 95% CI 1.24-2.27, p = 0.001. This association was more robust among diabetic patients than those without diabetes. In this study, diabetes was not associated with worse clinical outcomes in community-onset sepsis. However, high glucose levels at sepsis onset are independently associated with a worse prognosis, particularly among diabetic patients. Future trials should explore whether glycemic control could impact the outcomes and should be part of the management of sepsis, among the general adult septic population.