Introduction Subcutaneous administration of insulin is the preferred method for achieving glucose control in non-critically ill patients with diabetes. Glucose-based titration protocols were widely applied in clinical practice. However, most of these algorithms are experience-based and there is considerable variability and complexity. This study aimed to compare the effectiveness and safety of a weight-based insulin titration algorithm versus glucose-based algorithm in hospitalized patients with type 2 diabetes mellitus (T2DM).

Research design and methods This randomized clinical trial was carried out at four centers in the South, Central and North China. Inpatients with T2DM were randomly assigned (1:1) to receive weight-based and glucose-based insulin titration algorithms. The primary outcome was the length of time for reaching blood glucose (BG) targets (fasting BG (FBG) and 2-hour postprandial BG (2hBG) after three meals). The secondary outcome included insulin dose for achieving glycemic control and the incidence of hypoglycemia during hospitalization.

Results Between January 2016 and June 2019, 780 patients were screened, and 575 completed the trial (283 in the weight-based group and 292 in the glucose-based group). The lengths of time for reaching BG targets at four time points were comparable between two groups. FBG reached targets within 3 days and 2hBG after three meals within 4 days. There is no significant difference in insulin doses between two groups at the end of the study. The total daily dosage was about 1 unit/kg/day, and the ratio of basal-to-bolus was about 2:3 in both groups. The incidence of hypoglycemia was similar in both groups, and severe hypoglycemia was not detected in either of the groups.

Conclusions Weight-based insulin titration algorithm is equally effective and safe in hospitalized patients with T2DM compared with glucose-based algorithm.

Trial registration number NCT03220919.