377 - Diabetes as a prognostic factor for mortality in Coronavirus Disease 19 (COVID-19): a systematic review and meta-analysis comprising 18,506 patients

Background: Diabetes Mellitus is one of the leading causes of morbidity and mortality in the world. Infectious diseases are more common and associated with worse outcomes among diabetics. Diabetes is considered a predictor of morbidity in patients with COVID-19.

Methods: Medline, Embase, Google Scholar, and medRxiv were systematically reviewed up to May 10^th, 2020 for observational studies on diabetic adult populations hospitalized for COVID-19 and that assessed possible correlation between diabetes and mortality.
A meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Heterogeneity among trials for each outcome was assessed with the I-squared test. Values < 25% indicated low, 25 to 70% moderate, and > 70% high heterogeneity. Egger test and funnel plots were used to assess for publication bias.

Results: Fourteen observational studies (12 retrospective and 2 prospective) met the prespecified criteria for inclusion in the analysis, including 18,506 patients (43% women): 3,713 diabetics (DM group) and 14,793 non-diabetics (no-DM group). The mean or median age was above 60 years in 12 studies. DM group had a higher risk of death compared to the no-DM group, heterogeneity was significant (OR: 1.65; 95% CI: 1.35–1.96; I² 77.4%). Sensitivity analysis for US studies only also revealed a higher chance of death among the DM group (OR: 1.34; 95% CI: 1.04–1.85; I² 73.7%).

Conclusion: In conclusion, death was 65% more likely among diabetic inpatients compared to non-diabetics. Further studies are needed to assess whether this association is independent or not, and to investigate to role of glucose control prior or during the disease.