Comparative cardiovascular and renal effectiveness of sodium-glucose co-transporter 2 inhibitors and glucagon-like peptide 1 receptor-agonists: Scandinavian cohort study.

Aims To assess the comparative cardiovascular and renal effectiveness of SGLT2 inhibitors vs GLP-1-receptor-agonists in routine clinical practice. Materials and methods Cohort study of nationwide registers from Sweden, Denmark and Norway, 2013-2018, including 87525 new users of SGLT2 inhibitors and 63921 new users of GLP-1-receptor-agonists, analyzed intention-to-treat. Coprimary outcomes, analyzed intention to treat, were major adverse cardiovascular events (MACE; myocardial infarction/stroke/cardiovascular death), heart failure (hospitalization for heart failure/heart failure death) and serious renal events (renal replacement therapy/hospitalization for renal events/death from renal causes). Results Use of SGLT2 inhibitors vs GLP-1-receptor-agonists, was associated with higher risk of MACE (adjusted incidence rate: 15.2 vs 14.4 events per 1000 person-years; HR 1.07 [95% CI 1.01-1.15]), similar risk of heart failure (6.0 vs 6.0 events per 1000 person-years; HR 1.02 [0.92-1.12]) and lower risk of serious renal events (2.9 vs 4.0 events per 1000 person-years; HR 0.76 [0.66-0.87]). In as-treated analyses, the HR (95% CI) was 1.11 (1.00-1.24) for MACE, 0.88 (0.74-1.04) for heart failure and 0.60 (0.47-0.77) for serious renal events. In secondary outcome analyses, use of SGLT2 inhibitors vs GLP-1-receptor-agonists was not associated with statistically significant differences in risk of myocardial infarction (HR 1.09 [95% CI 1.00-1.19]), cardiovascular death (HR 0.97 [95% CI 0.84-1.12]), death from renal causes (HR 0.75 [95% CI 0.41-1.35]) or any cause death (HR 1.01 [95% CI 0.94-1.09]) while risk of stroke was higher (HR 1.14 [95% CI 1.03-1.26]) and risk of renal replacement therapy (HR 0.74 [95% CI 0.56-0.97]) and hospitalization for renal events (HR 0.75 [95% CI 0.65-0.88]) was lower among users of SGLT2 inhibitors. Conclusions Use of SGLT2 inhibitors vs GLP-1-receptor-agonists was associated with a similar risk of heart failure and lower risk of serious renal events while use of GLP-1-receptor-agonists vs SGLT2 inhibitors was associated with a slightly lower risk of MACE. In as-treated analyses, the associations with MACE and serious renal events increased in magnitude and the HR for heart failure tended towards a protective association for SGLT2 inhibitors. This article is protected by copyright. All rights reserved.