RESUMEN
Background & objectives: The association between hyperglycaemia at admission, diabetes mellitus (DM) status and mortality in hospitalized SARS-CoV-2 infected patients is not clear. The purpose of this study was to determine the relationship between DM, at-admission hyperglycaemia and 28 day mortality in patients admitted with moderate-severe SARS-CoV-2 infection requiring intensive care. Methods: All consecutive moderate-to-severe patients with SARS-CoV-2 infection admitted to the intensive care units (ICUs) over six months were enrolled in this single-centre, retrospective study. The predicators for 28 day mortality were analysed from the independent variables including DM status and hyperglycaemia at-admission. Results: Four hundred and fifty two patients with SARS-CoV-2 were admitted to the ICU, with a mean age of 58.5±13.4 yr, 78.5 per cent being male, HbA1c of 7.2 per cent (6.3-8.8) and 63.7 per cent having DM. Overall, 28 day mortality was 48.9 per cent. In univariate analysis, mortality in diabetes patients was comparable with non-diabetes (47.9 vs. 50.6%, P=0.58), while it was significantly higher in hyperglycaemic group (60.4 vs. 35.8%, P<0.001). In multivariate Cox regression analysis, after adjusting for age, sex and comorbidities, hyperglycaemia at-admission was an independent risk factor of mortality [hazard ratio (HR) 1.45, 95% confidence interval (CI) (1.06-1.99), P<0.05]. Interpretation & conclusions: This study showed that the presence of hyperglycaemia at-admission in critically ill SARS-CoV-2 patients was an independent predictor of 28 day mortality. However, the findings may be susceptible to unmeasured confounding, and more research from prospective studies is required.