An important and very interesting study within this field indeed. The study has several strengths and was randomized (multicenter) with quite a lot of patients in both study arms, (n=161).

The manuscript is very clear and well written. Maybe a new study later on using glargine U300 versus Degludec U100 at hospital? But the impact of this study is a bit limited because of the study group.

"Moreover, we did not observe a differential effect of degludec compared to glargine on subgroup analyses by HbA1c categories (< 75 mmol/mol vs >75 mmol/mol) or admission glucose (<11 1 vs > 11 1 mmol/L) suggesting that both insulin formulations are comparable in a wide and heterogeneous group of hospitalized patients with T2D."

This is as a matter of fact a rather special group of patients with a really high mean HbA1c (84 mol/mol) and BMI of 34! Totally 40% of the patients only had insulin before the inclusion and very few patients with more modern diabetes therapy (incretins/SGLT-2).

You did perform a post hoc analysis but limited to patients with HbA1c <75 mol/mol...,

What about more common patients with T2D? I think it is important to discuss more about this important issue in this excellent manuscript?

Ok, thank you for this improvement of the manuscript!