Aims Oral glucose-lowering medications are associated with excess risk of heart failure (HF). Given the absence of comparative data among drug classes, we performed a retrospective study in 32 Health Services of 16 Italian regions accounting for a population of 18 million individuals, to assess the association between HF risk and use of sulphonylureas, DPP-4i, and glitazones. Methods and results We extracted data on patients with type 2 diabetes who initiated treatment with DPP-4i, thiazolidinediones, or sulphonylureas alone or in combination with metformin during an accrual time of 2 years. The endpoint was hospitalization for HF (HHF) occurring after the first 6 months of therapy, and the observation was extended for up to 4 years. A tota...