Efficacy of sodium–glucose cotransporter 2 inhibitors in heart failure patients with low body mass index
K Nakamoto, H Minamiguchi, Y Ichibori, N Mori, T Kanda, T Tsujimura, Y Okina, H Tsutui, M Matsumura, T Toyoshima, S Nakawatase, T Sakio, D Yoshii, Y Ikeuchi, O IidaAbstract
Background
Sodium-glucose cotransporter 2 (SGLT2) inhibitors are a cornerstone therapy for patients with heart failure (HF), improving clinical outcomes across the ejection fraction spectrum. However, in real-world practice, their use is often withheld in patients with low body mass index (BMI) due to concerns about potential risks, such as dehydration and worsening nutritional status. Evidence regarding the efficacy and safety of SGLT2 inhibitors in this specific, potentially vulnerable population remains limited.
Purpose
This study aimed to determine whether the clinical benefit of SGLT2 inhibitors on HF outcomes is consistent between patients with low BMI (BMI＜18.5) and those with normal or higher BMI (BMI≧18.5 kg/m²) in real-world practice.
Methods
We retrospectively analyzed 469 consecutive HF patients (mean age: 78 ± 12 years, male: 60.6%. chronic kidney disease: 61.6% mean left ventricular ejection fraction: 49 ± 17%, atrial fibrillation: 50.2%) hospitalized between January 2024 and July 2025. Patients were stratified by BMI: low-BMI (BMI＜18.5 kg/m²: 24.6%) and non-low-BMI (BMI≧18.5: 75.4%). The outcome measure was a composite of all-cause mortality or HF rehospitalization. A multivariable Cox model was used to identify associated factors with all-cause mortality or HF rehospitalization, and to test for an interaction effect between SGLT2 inhibitor use and baseline characteristics including BMI.
Results
The prescription rate of SGLT2 inhibitors was 52.8%. During a mean follow-up of 254 ± 201 days, the composite endpoint of all-cause mortality and/or heart failure rehospitalization occurred in 122 patients. Compared with the non–low-BMI group, patients in the low-BMI group were significantly older (82 ± 10 vs. 77 ± 13 years, P<0.001), less frequently male (46.3% vs. 65.1%, P<0.001), and less likely to receive SGLT2 inhibitors (43.5% vs. 55.8%, P=0.032). There were no significant differences between the groups in left ventricular ejection fraction, prevalence of atrial fibrillation, or chronic kidney disease. Kaplan–Meier analysis demonstrated a significantly higher incidence of the composite endpoint in the low-BMI group compared with the non–low-BMI group (P=0.031). In Cox proportional hazards models including an interaction term between SGLT2 inhibitor use and BMI category, no significant interaction was observed (P for interaction = 0.73). In contrast, a significant interaction was observed between SGLT2 inhibitor use and CKD status (P for interaction = 0.010).
Conclusion
The current study revealed that low BMI is a potent risk factor in patients with HF. Despite the overwhelming prognostic impact of baseline characteristics in a real-world setting, the efficacy of SGLT2 inhibitors was consistently maintained even in high-risk, underweight patients. These findings suggest that low BMI itself should not be a barrier to initiating this vital therapy.