The adoption of sodium-glucose cotransporter-2 inhibitors in patients with heart failure with left ventricular ejection fraction > 40% varied significantly across U.S. hospitals, according to a study.

In the cohort study, published in JAMA Cardiology, investigators used data from the Get With The Guidelines-Heart Failure registry to examined the outcomes of 158,849 patients across 557 U.S. hospitals.

The study population (median age 76 years [interquartile range [IQR] = 66–85) included 89,816 women (56.5%) and 69,033 men (43.5%). The racial/ethnic composition comprised 71.2% White, 16.9% Black, 6.6% Hispanic, 2.5% Asian, and 2.9% other races/ethnicities.

The study population presented with multiple comorbidities: hypertension (89.3%), hyperlipidemia (65.1%), atrial fibrillation/flutter (49.3%), diabetes (47.5%), and chronic lung disease (39.6%). The median body mass index was 31.5 (IQR = 26.0–39.1).

The investigators found that prescription rates increased from 4.2% in July to September 2021 to 23.5% in July to September 2023.

Laboratory values at admission showed:

• Median eGFR: 56.2 mL/min/1.73m² (IQR = 39.1–78.0)
• BNP: 487.0 pg/mL (IQR = 238.0–932.0)
• NT-proBNP: 2913 pg/mL (IQR = 1,289–6,044)
• Hemoglobin: 11.5 g/dL (IQR = 9.9–13.0).

Among the 22,126 patients (13.9%) who received sodium-glucose cotransporter-2 (SGLT2) inhibitors prescriptions at discharge, concurrent medications included:

• β-Blockers: 74.4%
• ACEI, ARB, or ARNI: 41.5%
• MRA: 19.8%
• Double therapy: 9.5%
• Triple therapy: 7.7%.

Among 518 hospitals with ≥ 10 eligible discharges:

• 11 hospitals (2.1%) prescribed SGLT2 inhibitors to ≥ 50% of eligible patients
• 275 (53.1%) prescribed to 10% to 50% of patients
• 232 (44.8%) prescribed to < 10% of patients
• 15 hospitals (2.9%) reported no SGLT2 inhibitor use.

The median hospital prescription rate was 10.9% (IQR = 7.0%–17.6%). After adjustment for patient characteristics, the median odds ratio was 2.12 (95% confidence interval = 2.02–2.25).

Hospitals with higher SGLT2 inhibitor prescription rates were more likely to be larger facilities (≥ 75 beds), teaching hospitals, and have cardiac catheterization and heart transplant services.

The median length of hospital stay was 4.0 days (IQR = 3.0–7.0). Admission vital signs showed median systolic blood pressure of 143.0 mmHg (IQR = 125.0–164.0) and median heart rate of 82.0 beats/min (IQR = 70.0–96.0).

Study exclusions encompassed 14,508 patients on dialysis, 16,633 with eGFR < 20 mL/min/1.73m², and 8,020 who were discharged to hospice or left against medical advice.

The study demonstrated that prescription rates of SGLT2 inhibitors increased substantially over the 2-year period following the EMPEROR-Preserved trial, with marked variation across U.S. hospitals.

Conflict of interest disclosures can be found in the study.