Importance: Patients with diabetes are known to be at increased risk for infections including severe coronavirus disease 2019 (COVID-19) but the relationship between COVID-19 severity and specific pre-infection glucose levels is not known.

Objective: To assess the differential effects of pre-infection glucose levels on the risk for severe COVID-19 amongst patients with and without diabetes.

Design: Population based historical cohort study.

Setting: National state-mandated HMO.

Patients: All adult patients with a positive SARS-COV2 test between March-October 2020.

Exposure: Recent fasting blood glucose (FBG) and glycated hemoglobin (HBA1C), age, gender, body mass index (BMI) and diagnoses of diabetes, hypertension, ischemic heart disease.

Outcome: Risk for severe COVID-19, defined as resulting in ≥10 hospitalization days, ICU admission or death.

Results: 37,121 patients with a positive SARS-COV2 test were identified; 707 defined as severe (1.9%). Unadjusted risk factors for severe disease were age (OR = 1.1 for every year increase; 95% CI 1.09-1.11, p < 0.001), male gender (OR = 1.34, 95% CI 1.06-1.68, p = 0.012); BMI (OR = 1.02 for 1 kg/m2 increase, 95% CI 1.00-1.04, p = 0.025). Controlling for these factors, we found an association between pre-infection FBG and the risk of severe COVID-19, with a differential effect in patients with and without a diagnosis of diabetes. For patients without diabetes, elevated FBG in the pre-diabetes range (106-125 mg/dl) was associated with severe COVID-19 (OR 1.55 95% CI 1.04-2.26 p = 0.027). For patients with a diagnosis of diabetes, we found a J-shaped association between pre-infection glucose control and the risk for severe COVID-19 where the lowest risk for was for patients with FBG 106-125 mg/dl; the risk increased with higher pre-infection glucose levels but strikingly also for patients with a low pre-infection FBG (<100mg/dl) or HbA1C (<5.7%).

Conclusions and relevance: Elevated pre-infection blood glucose is a risk factor for severe COVID-19 even in non-diabetics. For patients with a diagnosis of diabetes both high as well as low pre-infection glucose levels are risk factors for severe COVID-19. Further research is required to assess whether these associations are causal, but we believe these findings can already have clinical implications for COVID-19 risk assessment and stratification.

The authors have declared that no competing interests exist.

👁 Fig 1

Fig 1

Left (A) : Percentage of…

Fig 1

Left (A) : Percentage of severe COVID-19 (death/ICU/≥10 hospitalization days) given prior fasting…

👁 Fig 2

Fig 2. Frequency of severe COVID-19 (death/ICU/…

Fig 2. Frequency of severe COVID-19 (death/ICU/ hospitalization ≥10 days) by pre-infection fasting blood glucose,…