We appreciate the comments by Cariou et al. (1) regarding our investigation (2) and their intermediate analysis of the CORONADO (CORONAvirus SARS-CoV-2 and Diabetes Outcomes) study (3), and we thank them for sharing important points. Currently, joint early efforts reveal that patients with diabetes and coronavirus disease 2019 (COVID-19) undergo more severe inflammation and worse prognosis, and no evidence suggests the need for stopping any glucose-lowering therapy (2,3), since high blood glucose during hospitalization has also been suggested as a risk factor for poor outcomes in patients with diabetes with COVID-19 (4).
We acknowledge Cariou et al. for sharing additional independent predictive factors associating with early death on day 7 after admission (D7) based on a larger cohort and stepwise-selected multivariable analyses (1). Nevertheless, the common independent risk factors in both studies, including older age and increased CRP, are worth specific attention.
Consistent with our findings (2), Cariou et al. also pointed out that patients with diabetes with worse outcomes were more frequently prescribed insulin, possibly due to their advanced stage of diabetes with more frequent comorbidities (1). Both studies identified no association between in-hospital death and insulin use (2,3). However, in our study, insulin users were prone to poor prognosis including progress to more severe illness. A recent study that included 952 COVID-19 patients with type 2 diabetes found that in-hospital mortality was significantly higher in patients with poorly controlled blood glucose (assessed by glycemic variability during hospitalization) than in those with well-controlled blood glucose. Importantly, a larger proportion of patients were prescribed insulin in the poorly controlled group than in the well-controlled group (54.6% vs. 14.2%, P < 0.001) (4); whether this difference was due to effects of insulin or to characteristics of the patients for whom insulin was prescribed remains unclear. We also noticed that the insulin users in our study showed higher admission blood glucose than those not taking insulin (2), suggesting more efforts are necessary for their blood glucose control, possibly due to unsatisfactory basic characteristics. We have underlined this result to indicate that more attention is needed for these patients concerning their worse clinical presentations (2). In addition, a consensus from the diabetes community suggested that insulin therapy should not be stopped, and monitoring of blood glucose is encouraged and careful adjustment is needed if appropriate (5).
Cariou et al. performed intermediate analysis on data by D7 after admission with composite primary outcome noted in 29.0% of participants, while we did not include patients who were still hospitalized during our observation time. In our study, the median length of hospital stay was 22.0 days (interquartile range 16.0–31.8) in 136 patients with diabetes and COVID-19, and their early death rate (5 [3.7%] deaths on D7) was lower than that in CORONADO (10.6%) (2,3). Curious about the association between characteristics at admission and longer-term COVID-19 prognoses, as well as clinical courses in association with glucose-lowering therapies during longer hospitalization, we are looking forward to further information from CORONADO.
Y.C. and D.Y. contributed equally.
Article Information
Funding. This work is supported by the National Key R&D Program of China (grants 2019YFA0802701 and 2018YFA0800700) and the National Natural Science Foundation of China (grants 91957114, 31971066, 31871381, and 31671195).
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
