Retrospective Study on the Effects of Glucose Abnormality on COVID-19 Outcomes in Japan

Introduction: To investigate the effects of glucose abnormality on outcomes of hospitalized coronavirus disease 2019 (COVID-19) patients in Japan.

Methods: This study retrospectively analyzed 393 COVID-19 patients admitted at Juntendo University Hospital. Patients were divided into subgroups according to history of diabetes and blood glucose (BG) levels and subsequently compared in terms of in-hospital death, invasive ventilation, or a composite of both.

Results: Patients with glucose abnormality demonstrated more risk factors for serious COVID-19, such as high body mass index, dyslipidemia, and hypertension, and higher biomarkers for inflammation compared to those with normal BG levels. Oxygen inhalation and steroid use were more frequent among patients with than without glucose abnormality. Invasive ventilation was more frequent in patients with diabetes (9.5% vs. 3.2%, p = 0.033) and BG ≥ 140 mg/dl (11.0% vs. 3.1%, p = 0.009) compared with those without diabetes and BG < 140 mg/dl, respectively. Logistic regression analysis showed that BG ≥ 140 mg/dl was a risk factor for invasive ventilation [odds ratio (OR) 2.87, 95% CI 1.04-7.68, p = 0.037] or the composite outcome (OR 3.03, 95% CI 1.21-7.38, p = 0.015) even after adjusting for by age and gender. Kaplan-Meier analysis showed that glucose abnormality was significantly associated with invasive ventilation and that BG ≥ 140 mg/dl was a risk factor for invasive ventilation [hazard ratio (HR) 2.68, 95% CI 1.05-6.82, p = 0.039] and the composite of death and invasive ventilation (HR 2.77, 95% CI 1.21-6.37, p = 0.016) regardless of age and gender.

Conclusions: Glucose abnormality, particularly BG ≥ 140 mg/dl, was associated with serious outcomes among Japanese COVID-19 patients, suggesting the need to consider high BG as a major risk factor for poor clinical course also in Japan.