Importance: Clinical guidelines for patients with heart failure with reduced ejection fraction (HFrEF) strongly recommend treatment with a sodium-glucose cotransporter-2 inhibitor (SGLT2i) to reduce cardiovascular mortality or HF hospitalization. Nationwide adoption of SGLT2i for HFrEF in the US is unknown.
Objective: To characterize patterns of SGLT2i use among eligible US patients hospitalized for HFrEF.
Design, setting, and participants: This retrospective cohort study analyzed 49 399 patients hospitalized for HFrEF across 489 sites in the Get With The Guidelines-Heart Failure (GWTG-HF) registry between July 1, 2021, and June 30, 2022. Patients with an estimated glomerular filtration rate less than 20 mL/min/1.73 m2, type 1 diabetes, and previous intolerance to SGLT2i were excluded.
Main outcomes and measures: Patient-level and hospital-level prescription of SGLT2i at hospital discharge.
Results: Of 49 399 included patients, 16 548 (33.5%) were female, and the median (IQR) age was 67 (56-78) years. Overall, 9988 patients (20.2%) were prescribed an SGLT2i. SGLT2i prescription was less likely among patients with chronic kidney disease (CKD; 4550 of 24 437 [18.6%] vs 5438 of 24 962 [21.8%]; P < .001) but more likely among patients with type 2 diabetes (T2D; 5721 of 21 830 [26.2%] vs 4262 of 27 545 [15.5%]; P < .001) and those with both T2D and CKD (2905 of 12 236 [23.7%] vs 7078 vs 37 139 [19.1%]; P < .001). Patients prescribed SGLT2i therapy were more likely to be prescribed background triple therapy with an angiotensin-converting enzyme inhibitor/angiotensin receptor blocker/angiotensin receptor-neprilysin inhibitor, β-blocker, and mineralocorticoid receptor antagonist (4624 of 9988 [46.3%] vs 10 880 of 39 411 [27.6%]; P < .001), and 4624 of 49 399 total study patients (9.4%) were discharged with prescriptions for quadruple medical therapy including SGLT2i. Among 461 hospitals with 10 or more eligible discharges, 19 hospitals (4.1%) discharged 50% or more of patients with prescriptions for SGLT2i, whereas 344 hospitals (74.6%) discharged less than 25% of patients with prescriptions for SGLT2i (including 29 [6.3%] that discharged zero patients with SGLT2i prescriptions). There was high between-hospital variance in the rate of SGLT2i prescription in unadjusted models (median odds ratio, 2.53; 95% CI, 2.36-2.74) and after adjustment for patient and hospital characteristics (median odds ratio, 2.51; 95% CI, 2.34-2.71).
Conclusions and relevance: In this study, prescription of SGLT2i at hospital discharge among eligible patients with HFrEF was low, including among patients with comorbid CKD and T2D who have multiple indications for therapy, with substantial variation among US hospitals. Further efforts are needed to overcome implementation barriers and improve use of SGLT2i among patients with HFrEF.