Endocrine Abstracts

Introduction

Diabetes mellitus (DM) is one of the major comorbidities in patients with Coronavirus disease (COVID-19) leading to poorer outcomes. Previous evidence showed that poorly-controlled hyperglycemia increases the severity and mortality of COVID-19. Nevertheless, there is limited data on the role of in-hospital glucose control on the outcomes of elderly patients with type 2 diabetes mellitus (T2DM).

Objective

To access the impact of blood glucose control, assessed by the derived time in range (dTIR) and glycemic variability, on the mortality of elderly patients with T2DM and COVID-19.

Methods

We selected consecutive patients with laboratory confirmed COVID-19 who had been hospitalized in a general ward of our hospital between 25 March and 25 May 2020. From a total of 97 patients, we identified and included 38 patients with DM, with a median age of 80 years (IQR, 76–87). To assess glycemic control, all capillary blood glucose levels were extracted for each diabetic patient in the first seven days (four glucose tests per day). Individual derived time in range (percentage of time with plasma glucose between 70–180 mg/dl) was derived as the proportion of values within range (dTIR). dTAR (derived time above range) was derived as the proportion of values above range.

Results

The dTIR for all diabetic patients was 49%, and the dTAR was 52%. TIR > 70% was 36.8% for all diabetic patients. Nonsurvivors were more likely to have a lower TIR (38% vs . 73%, P = 0.020) and a higher TAR (62% vs . 27%, P = 0.020). Survivors were more likely to have TIR > 70% (50% vs . 14.3%, P = 0.030). There were no differences between groups regarding data estimates of glycemic variability: coefficient of variation (CV) (23.26 [17.46–35.76] vs . 29.15 [19.63–37.84], P = 0.526); high blood glucose index (HBGI) (13.78 [7.09–21.89] vs . 9.73 [3.35–17.03], P = 0.151); measure of stability of glycemia in comparison with an arbitrary assigned ’ideal’ glucose value, ’R, ’ set to 100 mg/dl (M-100 index) (297.41[194.23–386.75] vs . 216.94 [147.00–321.54], P = 0.123) or measure of quality of glycemic control calculated as 0.001 × [mean + Standard of deviation (SD)] (J-index) (73.20 [45.82–104.39] – 56.99 [31.05–88.34], P = 0.221).

Conclusion

A poorer glycemic control, assessed by lower dTIR during hospitalization, was associated with in-hospital mortality. Clinicians should maximize TIR even in elderly patients, using a basal-bolus or continuous insulin infusion whenever needed, with appropriate surveillance.