Stress hyperglycemia ratio, rather than admission blood glucose, predicts in-hospital mortality and adverse outcomes in moderate-to severe COVID-19 patients, irrespective of pre-existing glycemic status

Aim: To compare admission-blood-glucose (ABG) or stress-hyperglycemia-ratio (SHR) performs better in predicting mortality and worse outcomes in COVID-19 patients with (DM) and without known Type 2 Diabetes Mellitus (UDM). Methods: ABG and SHR were tested for 451 patients with moderate-severe COVID-19 infection [DM = 216,47.9%; pre-diabetes = 48,10.6%, UDM = 187,41.4%],who were followed-up to look for in-hospital -mortality (primary outcome) and secondary outcomes (ICU stay or mechanical ventilation, hospital-acquired -sepsis and multiple organ dysfunction syndrome [MODS]). Those with and without SHR >= 1.14 were compared; logistic regression was done to identify predictors of outcomes, with subgroup analysis based on pre-existing DM status and COVID-19 severity. Results: Those who died (n = 131) or developed >= 1 secondary outcomes (n = 218) had higher prevalence of SHR >= 1.14, ABG >= 180 mg/dl and higher median SHR (pall < 0.01). Those with SHR >= 1.14 had higher mortality (53.7%), higher incidence of >= 1 secondary outcomes (71.3%) irrespective of pre-existing diabetes status. SHR >= 1.14, but not ABG >= 180 was an independent predictor of mortality in the whole group (OR: 7.81,4.07-14.98), as also the DM (OR:10.51,4.34-25.45) and UDM (5.40 (1.57-18.55) subgroups. SHR >= 1.14 [OR: 4.41 (2.49-7.84)] but not ABG >= 180 could independently predict secondary outcomes AUROC of SHR in predicting mortality was significantly higher than ABG in all subgroups. Conclusion: SHR better predicts mortality and adverse outcomes than ABG in patients with COVID-19, irrespective of pre-existing chronic glycemic status.

Automated summary

This study compared the effectiveness of admission-blood-glucose (ABG) and stress-hyperglycemia-ratio (SHR) in predicting mortality and adverse outcomes in COVID-19 patients. The results showed that SHR is a better predictor than ABG, regardless of the patients' pre-existing diabetic status.