Objective: The association between SGLT2i and peripheral artery disease (PAD) remains unclear for high-risk patients. The aim of this study was to test the hypothesis that SGLT2i is associated with a composite PAD outcome compared to DPP4i.
Methods: A retrospective cohort combined Veterans Administration, Medicare, and National Death Index databases. New users of SGLT2i or DPP4i were followed from the prescription fill until a composite PAD event, death, or end study. The PAD event was defined as the earliest surgery: amputation, peripheral stent placement, or revascularization. Covariate adjusted Cox models compared PAD event between SGLT2i users and DPP4i users in a propensity score weighted cohort allowing for events to occur up to 90 days or 360 days after stopping SGLT2i.
Results: The weighted cohort included 36,190 SGLT2i vs 36,254 DPP4i episodes. Median age was 69 years, HbA1c was 8.40 [7.60, 9.30] and diabetes duration 9.3 (5.7, 13.7) years. There were 319 and 286 PAD events among SGLT2i and DPP4i users, respectively, for an event rate of 12.3 (11.0, 13.6) and 10.4 (9.6, 11.4) per 1000 person-years (aHR 1.21 [1.05, 1.41]). When allowing for PAD events for 360 days after SGLT2i use ended, the aHR was 1.27 [1.13, 1.44]).
Conclusions: Use of SGLT2i for diabetes treatment was associated with increased amputation and revascularization surgery outcomes compared to DPP4i.
K.E. Griffin: None. K.D. Snyder: None. A. Javid: None. A.J. Hackstadt: None. R.A. Greevy: None. C.G. Grijalva: Advisory Panel; Merck Sharp & Dohme Corp. C. Roumie: None.
VA Clinical Science research and Development investigator initiated grant CX000570-12 (Roumie)
