Hyperglycaemia in predicting severe COVID-19 at Tabanan general hospital, Bali

Background: Hyperglycaemia has been shown to be associated with disease progression and poor prognosis in Corona virus disease 2019 (COVID-19) patients. This study aims to find the effect of hyperglycaemia on disease severity and investigate whether high blood glucose levels on admission can predict severity of COVID-19 infection Methods: in this cross-sectional study, a total of 286 COVID-19 patients in Tabanan general hospital, Bali were retrospectively analysed. Data were obtained from medical records from January 1 to December 31, 2021. Hyperglycaemia was defined as random blood glucose (RBG) >140 mg/dl. The severity of COVID-19 was determined according to the 4th edition of the Indonesian COVID-19 management guidelines. Clinical and biochemical characteristics of COVID-19 patients with or without diabetes were compared. Receiver operating characteristic (ROC) analysis was used to identify optimal admission plasma glucose levels to predict COVID-19 severity. Results: 47.2% of subjects had hyperglycaemia at admission, 67.5% experienced severe COVID-19, of which 68.4% died. Admission RBG values were positively correlated with leukocyte and NLR values. In ROC analysis, admission RBG >145 mg/dl can predict severe COVID-19 with sensitivity of 56% and specificity of 76% (AUC 0.663, p<0.01). Conclusions: Hyperglycaemia is an independent predictor of severe COVID-19 and impose a significantly higher mortality rate compared to normoglycemic patients regardless of diabetic status. Early measurement of plasma glucose levels upon admission can help identify patients who are likely to experience a worse clinical course.