Outcomes Associated with Surgical and Pharmacologic Treatment of Obesity in Heart Failure

The prevalence of overweight/obesity and cardiovascular disease are steadily increasing in the United States.¹ Dysfunctional adiposity and elevated body mass index (BMI) are associated with incident heart failure, which conveys substantial morbidity and economic burden.^1,2 Bariatric surgery is associated with lower risk of incident heart failure, atrial fibrillation, stroke, and myocardial infarction among patients with obesity.³ Given the relationship between cardiovascular-kidney-metabolic (CKM) syndrome and adverse cardiovascular outcomes, treatment of obesity is a promising intervention for heart failure.^2,4 However, there are limited data describing the effect of weight loss therapies on clinical outcomes in patients with existing heart failure.

In this issue of Circulation: Heart Failure, Mentias and colleagues describe trends and outcomes associated with anti-obesity therapeutics in a real-world, observational cohort of Medicare-enrolled patients with heart failure and obesity.⁵ The authors use a propensity-score matched sample to examine associations between bariatric surgery and all-cause mortality, hospitalizations for heart failure, and incident atrial fibrillation among adults in the Medicare cohort aged less than 75 years old with a BMI ≥35 kg/m². In secondary analyses, the authors assess the association between pharmacotherapies known to cause weight loss with the same outcomes, as well as trends in bariatric surgery utilization among patients with heart failure.

There are several key findings highlighted by this paper, including that bariatric surgery was significantly associated with decreased all-cause mortality, decreased hospitalization for heart failure, and decreased incidence of atrial fibrillation. However, patients of lower socioeconomic status were less likely to undergo bariatric surgery, and patients who had fewer comorbidities were more likely to have surgery. Treatment of obesity with bariatric surgery increased significantly over the study period, from 0.2% in 2013 to 1.0% in 2020, but remained low overall. The authors also included analyses focusing on pharmacotherapies known to induce weight loss (including semaglutide, liraglutide, naltrexone-bupropion, or orlistat). These medications were significantly associated with decreased all-cause mortality and decreased hospitalization for heart failure, but not decreased incidence of atrial fibrillation. Overall, the findings related to both bariatric surgery and pharmacotherapy were consistent between patients with heart failure with reduced ejection fraction (HFrEF) and preserved ejection fraction (HFpEF). Finally, there was a substantial delay in the initiation of anti-obesity treatment following the index hospitalization for heart failure when considering both bariatric surgery and pharmacotherapy.

This study expands prior work examining the role of intentional weight loss in improving cardiovascular outcomes in heart failure and provides a real-world view of bariatric surgery utilization among older patients with heart failure. The authors conclude that bariatric surgery is a safe and effective therapy for weight loss in patients with heart failure, and potentially conveys a mortality benefit. While observational data based on administrative data cannot provide the same level of evidence as clinical trials, the findings described in this study build on and clarify prior observational work. This is particularly relevant to findings on the role of weight loss in heart failure management, as prior work has shown that unintentional weight loss is associated with worse outcomes in heart failure. These findings are likely attributed to sarcopenia and frailty from advancing chronic disease; however, observational data can only provide a snapshot of complex physiologic relationships in heart failure.

There are several limitations to approaching the study questions with administrative data. These include risk of selection bias, misclassification bias, immortal time bias, confounding by indication, and residual confounding. Studies relying on administrative data have limited capacity to account for other clinical features, and follow-up data is often lacking. Additionally, administrative data preclude the examination of outcomes by distinct clinicopathologic phenotypes. In the setting of obesity and cardiovascular outcomes, the distribution of adiposity is a key component of risk stratification.⁴ Visceral adiposity is strongly associated with poor cardiovascular outcomes, possibly through secondary myocardial injury from inter-organ crosstalk, extracellular vesicles, or inflammatory cascades.^6–8 Therefore, additional questions remain surrounding the differential effects of intentional weight loss in heart failure patients with specific patterns of adiposity.

Prior work has demonstrated distinct clinicopathological phenotypes of HFpEF, including obesity-related HFpEF, which appears to be particularly responsive to weight loss therapy.^9–11 In this study, the authors were limited to focusing on differences between HFpEF and HFrEF. The consistency of the findings across patients with HFpEF and HFrEF may indicate that the benefits of weight loss are ubiquitous across the spectrum of heart failure in patients with obesity. The most attractive hypothesis is that weight loss leads to decreased visceral adiposity and broadly improved CKM health.^4,12 However, alternative explanations include misclassification of ejection fraction due to clinician miscoding,¹³ or a lack of statistical power to detect a difference in subgroups when only one-third of the study population had HFpEF.

The findings presented by Mentias et al. must be interpreted in the context of the study temporality. The study years were 2013–2020 for the bariatric surgery analyses, and 2015–2020 for the pharmacotherapy analyses. The glucagon-like peptide-1 receptor agonist (GLP1-RA) semaglutide was approved for the treatment of type 2 diabetes in 2017 and was not approved for the treatment of overweight/obesity until 2021. Furthermore, by the end of the study period, optimal guideline-directed medical therapy (GDMT) was updated to include sodium-glucose cotransporter-2 inhibitors (SGLT2i) for all forms of heart failure, and subsequently GLP1-RAs for HFpEF – these two medication classes have been shown to have powerful cardioprotective effects.¹ Very few of the patients included in the analyses of pharmacotherapy for obesity were prescribed optimal GDMT, as indicated by a low mean modified Heart Failure Collaboratory Consortium score (less than 20, out of a possible 100).¹⁴ The proportion of patients on optimal GDMT was unknown in the bariatric surgery analyses, due to limitations of the data source. Thus, the effect estimates may be attenuated among patients on optimal GDMT, including an SGLT2i and GLP1-RA.

The authors note that patients with lower socioeconomic resources were less likely to undergo bariatric surgery, despite the entire study population being Medicare insured. There are multiple potential factors underlying this finding, including provider implicit biases, economic burden of surgery, and limited access to care. Focused attention on mitigating and reversing disparities in access to optimal GDMT and advanced anti-obesity therapeutics is required to avoid perpetuating the treatment trends described in this study. Furthermore, the study highlights the substantial delay between initial hospitalization for heart failure and initiation of obesity treatment (approximately 17 months for bariatric surgery and 9.4 months for pharmacotherapy). Further work is needed to investigate drivers of disparities in obesity treatment and reduce delays in care. Implementation approaches that leverage community engagement to identify local, scalable solutions may reduce the burden of disease in socioeconomically disadvantaged populations.

What are the implications of this study for clinical practice? The consistency of the findings across obesity treatment modalities reinforces the burgeoning role of subspecialists in addressing weight loss.^12,15 The recent American Heart Association Presidential Advisory paper on CKM syndrome provides further support for the role of weight loss in preventing cardiovascular disease progression.² Optimal treatment of heart failure should include a patient-centered discussion on weight management to improve cardiovascular outcomes.^2,15

In conclusion, these observational data support the role of obesity treatment in reducing mortality and rehospitalization among older adults with heart failure. The findings also highlight the low utilization of bariatric surgery in clinical practice and notable socioeconomic disparities in access to anti-obesity therapies. Intentional and equitable implementation of evidence-based strategies targeting obesity is needed to prevent further widening disparities in cardiovascular health at the population level. Emerging cardioprotective therapies, including obesity treatments, have the potential to transform heart failure outcomes.