Medical Student Brown University Providence, Rhode Island, United States
Objective : To evaluate the impact of SH on health outcomes in COVID-19 hospitalizations without a prior diagnosis of diabetes.
Methods: Data from 2068 individuals with no diabetes admitted in Rhode Island with COVID-19 infection during the first-wave (March 1-June 30, 2020) and second-wave (July 1, 2020-February 28, 2021) was analyzed. Multivariate logistic regression was conducted to compare SH (defined as ≥ 2 instances of BG above 180 mg/dL) status for length of stay (LOS), ICU admission, mechanical ventilation (MV) and in-hospital mortality, and was adjusted for age, race/ethnicity, gender, insurance, and wave. Model investigating risk factors associated with poor outcomes (ICU admission, MV or in-hospital mortality) in hospitalized individuals without Diabetes was additionally adjusted to include BMI, Chronic kidney disease, hypertension (HTN), pulmonary disease, cardiovascular diseases, and SH.
Results: 8.2% (170) of the study cohort had stress hyperglycemia. Gender (52.8% male) and race/ethnicity (Caucasian 61.9%, Hispanic 23.5%, Black 8.6% and other race 6.0%) distribution was similar among individuals with/without stress hyperglycemia. However, individuals with stress hyperglycemia were older (66.7±16.0 vs 62.8±19.8 years, p-value=0.0141) and more likely to require LOS ≥1-week (OR 6.3[4.3-9.1]), ICU admission (9.7[6.9-13.8]) and MV (18.9[12.8-28.0]). Stress hyperglycemia was also associated with higher in-hospital mortality (11.2[7.7-16.5]). The risk of poor outcomes was significantly higher in individuals with stress hyperglycemia compared to individuals without stress hyperglycemia (ICU admission 47.7 vs 9.0%, MV 45.3 vs 4.6%, Mortality 41.8 vs. 5.9%, Composite 62.9% vs 13.0%). Stress hyperglycemia (10.8[7.49-15.71]), male gender (1.76[1.32-2.34]) and admission during the first-wave (1.56[1.14-2.15]) were identified as independent risk factors for poor outcomes in individuals without diabetes.
Discussion/Conclusion: Our study showed that hospitalized individuals with COVID-19 infection, without known diabetes, who experienced SH, had worse outcomes (i.e., MV, admission to the ICU, longer LOS and death) when compared to individuals without SH. Male gender and admission during the first-wave were identified as independent risk factors for poor outcomes.
