Lower risk of hypoglycemia with sitagliptin compared to glipizide when either is added to metformin therapy: a pre-specified analysis adjusting for the most recently measured HbA(1c) value.

Background: In a previously-published study, adding sitagliptin or glipizide to ongoing metformin therapy provided similar HbA(1c) improvement (both groups, -0.7%) after 52 weeks in patients with type 2 diabetes (T2DM). Significantly fewer patients experienced symptomatic hypoglycemia with sitagliptin (5% of 588 patients) compared to glipizide (32% of 584 patients). Glycemic efficacy and patient characteristics may influence hypoglycemic events. The present analysis evaluated the risk of hypoglycemia with sitagliptin or glipizide after adjusting for the most recently measured HbA(1c) value. Methods: Data for this analysis were from the aforementioned 52-week, randomized, double-blind, active-controlled study. The primary endpoint was confirmed hypoglycemia (i.e., symptomatic hypoglycemia confirmed with a concurrent fingerstick glucose <= 70 mg/dL [3.9 mmol/L]); the secondary endpoint was severe hypoglycemia (requiring medical or non-medical assistance or symptoms of neuroglycopenia). Complementary log-log regression random effects models with terms for treatment, most recently measured HbA(1c) value, time (i.e., days since randomization), gender, and age (< or >= 65 years) were used to assess adjusted subject-specific treatment effects. Results: Over the full range of HbA(1c) levels and follow-up time, the risk of confirmed hypoglycemic events was lower with sitagliptin compared with glipizide (31 vs. 448 events; adjusted hazard ratio [HR] = 0.05 [95% CI: 0.03, 0.09], p<0.001). The risk was also lower with sitagliptin in the younger (HR 0.06 [95% CI: 0.03, 0.12], p<0.001) and older (HR = 0.02 [0.01, 0.08], p<0.001) age groups compared with glipizide. For severe hypoglycemia events (2 vs. 22), the risk was lower with sitagliptin (HR = 0.08 [95% CI: 0.01, 0.47]; p = 0.005). Limitations: The actual time between the HbA(1c) measurement and the hypoglycemic event was variable and not controlled for in the analysis. Conclusion: In pre-specified analyses adjusting for the most recently measured HbA(1c) value, there was a substantial reduction in risk for confirmed hypoglycemia with sitagliptin compared to glipizide when added to ongoing metformin therapy in patients with T2DM. The risk of confirmed hypoglycemia was very low in younger and older patients treated with sitagliptin.