Assessment of the Risk of Hospitalization for Heart Failure With Dipeptidyl Peptidase-4 Inhibitors, Saxagliptin, Alogliptin, and Sitagliptin in Patients With Type 2 Diabetes, Using an Alternative Measure to the Hazard Ratio

BACKGROUND: Saxagliptin statistically significantly increased the risk of hospitalization for heart failure compared with placebo in the clinical trial of SAVOR-TIMI 53. Neither the reason why only saxagliptin among several dipeptidyl peptidase-4 (DPP-4) inhibitors increased the risk, nor the clinical implication of the result has been explained. OBJECTIVE: To evaluate the risk of hospitalization for heart failure associated with DPP-4 inhibitors by using an alternative measure to the hazard ratio. METHODS: We used the difference in restricted mean survival time (RMST) between DPP-4 inhibitors and placebo to evaluate the risk of cardiovascular events, including hospitalization for heart failure associated with DPP-4 inhibitors. Three randomized clinical trials with cardiovascular events as a primary end point-EXAMINE (alogliptin), SAVOR-TIMI 53 (saxagliptin), and TECOS (sitagliptin)-were reevaluated by estimating the RMSTs for the DPP-4 inhibitors and placebo based on the reconstructed individual patient data for each time-to-event outcome from publicly available information. RESULTS: The differences of RMSTs (DPP-4 inhibitors minus placebo) for hospitalization for heart failure were -4 days [-6, -2] in the SAVOR-TIMI 53 (720 days follow-up), -3 days [-9, 3] in the EXAMINE (900 days follow-up), and 1 day [-5, 7] in the TECOS (1440 days follow-up). There were no substantial differences in the risk of other cardiovascular outcomes between DPP-4 inhibitors and placebo. CONCLUSIONS: There are no substantial clinically relevant differences in the risk of cardiovascular events, including hospitalization for heart failure, between 3 of the DPP-4 inhibitors and placebo.