After intensive insulin treatment, many obese African American patients with new-onset diabetic ketoacidosis (DKA) and severe hyperglycemia are able to achieve near-normoglycemia remission. The optimal treatment to prevent hyperglycemic relapses after remission is not known.

This prospective, 4-year, placebo-controlled study randomly assigned 48 African American subjects with DKA and severe hyperglycemia to metformin 1,000 mg daily (n = 17), sitagliptin 100 mg daily (n = 16), or placebo (n = 15) after normoglycemia remission. Hyperglycemic relapse was defined as fasting glucose >130 mg/dL (7.2 mmol/L) and HbA >7.0% (53 mmol/mol). Oral glucose tolerance tests were conducted at randomization and at 3 months and then every 6 months for a median of 331 days. Oral minimal model and incremental area under the curve for insulin (AUCi) were used to calculate insulin sensitivity (Si) and β-cell function, respectively. Disposition index (DI) was calculated as a product of Si and incremental AUCi.

Relapse-free survival was higher in sitagliptin and metformin (P = 0.015) compared with placebo, and mean time to relapse was significantly prolonged in the metformin and sitagliptin groups compared with the placebo group (480 vs. 305 days, P = 0.004). The probability of relapse was significantly lower for metformin (hazard ratio 0.28 [95% CI 0.10-0.81]) and sitagliptin (0.31 [0.10-0.98]) than for placebo. Subjects who remained in remission had a higher DI (P = 0.02) and incremental AUCi (P < 0.001) than those with hyperglycemia relapse without significant changes in Si.

This study shows that near-normoglycemia remission was similarly prolonged by treatment with sitagliptin and metformin. The prolongation of remission was due to improvement in β-cell function.