Obesity is currently one of the most serious threats to public health, with deleterious consequences on patients and healthcare systems.1 In 2015, more than 2 billion adults (i.e. one third of the world’s population) were estimated to be overweight or obese, while the prevalence of the disorder is projected to increase dramatically over the next few years, ringing an alarm for an emerging epidemic.2 Besides the great impact on healthcare costs,3 obesity is associated with numerous comorbidities, including but not limited to coronary artery disease, atrial fibrillation and sleep apnoea, leading to impaired quality of life and ultimately reduced life expectancy.4–8 However, the relationship between body weight and mortality is not linear. Several observational studies have shown a J-curved or U-shaped relationship between body mass index (BMI) and death, giving rise to a phenomenon called obesity paradox.9,10 The obesity paradox has also been observed in patients with heart failure (HF), suggesting that overweight and mildly obese patients may have a better prognosis, compared to their lean counterparts.11,12 If this hypothesis is correct, several clinically important questions arise inevitably: should clinicians recommend weight loss in all obese and overweight patients with HF? What is the ideal BMI for HF patients?
In this issue of EJPC, Gentile et al. explored the prognostic value of BMI in patients with HF.13 A total of 5155 patients with either ischaemic or non-ischaemic HF were followed up for the end point of 5-year all-cause mortality. The vast majority of patients had heart failure with reduced ejection fraction (HFrEF; left ventricular ejection fraction (LVEF) <40%; 63%), while the rest were equally distributed between heart failure with mid-range EF (HFmEF, LVEF 40–49%; 18%) and heart failure with preserved ejection fraction (HFpEF, LVEF ≥50%; 19%). Patients were further stratified according to the World Health Organization classification of BMI: underweight: less than 18.5 kg/m2; normal weight: 18.5–24.9 kg/m2; overweight: 25–29.9 kg/m2; mild obesity: 30–34.9 kg/m2; moderate obesity: 35–39.9 kg/m2; and severe obesity: 40 kg/m2 or greater. Among these categories, the authors found that only underweight was significantly associated with an increased risk of all-cause mortality (adjusted hazard ratio (HR) 2.24, 95% confidence interval (CI) 1.66–3.03). Conversely, the risk of death was lower in overweight (adjusted HR 0.89, 95% CI 0.79–0.99), mildly (adjusted HR 0.72, 95% CI 0.61–0.84) and moderately obese patients (adjusted HR 0.65, 95% CI 0.49–0.86), compared to normal weight patients.
Gentile and colleagues13 should be congratulated for performing a large study with a comprehensive and robust statistical approach. The authors evaluated the independent association of different BMI categories with mortality, after adjusting for important demographic and clinical factors (i.e. age, gender, pharmacological treatment). Given the fact that BMI cannot distinguish lean from fat body mass, the authors estimated body fat percentage (BFP) in order to assess the impact of excess body fat on patient prognosis. Interestingly, patients in the highest (third) BPF tertile (BFP ≥33.1%) had a better prognosis, compared to patients in the first BFP tertile (BFP <25.1%). These findings are in agreement with previous studies.14,15–16 A meta-analysis of six studies and 22,807 patients with chronic HF showed that the risk of adverse outcomes (all-cause mortality, cardiovascular mortality, rehospitalisations) was highest in the low BMI group (<20 kg/m2) and lowest in the overweight group (BMI 25–29.9 kg/m2).15 Lavie et al. previously expanded on this observation by demonstrating an inverse relationship between body fat mass and mortality.14 Although the exact mechanisms of the obesity paradox in HF are poorly understood, this paradoxical relationship can be partially explained by the fact that HF is a catabolic state and obese patients usually have more muscular strength and thus more metabolic reserve compared to underweight patients.12
The current study by Gentile et al. significantly adds to the existing literature regarding HF and the obesity paradox.13 Although the patient population was heterogeneous, the authors tried to homogenise their data by performing subgroup analyses, according to LVEF (HFpEF, HFmEF, HFrEF) and HF aetiology (ischaemic, non-ischaemic) (Figure 1). Notably, underweight was an independent predictor of mortality in all subanalyses, while the prognostic value of higher BMI varied across the different subgroups of patients. In particular, overweight and obesity did not show any prognostic benefit when the analysis was restricted to patients with ischaemic HF, but mild obesity was associated with a better prognosis in all LVEF subgroups. These findings, however, should be interpreted with caution, because the study was not sufficiently powered for such subanalyses.
Desired BMI in HF. Schematic illustration of desired BMI in HF patients, according to EF and HF aetiology. BMI: body mass index; EF: ejection fraction; HF: heart failure.
Moreover, this study has some further limitations that merit discussion. First, the study findings are based on observational data and thus causality cannot be established. Cachexia may be just a marker of severe HF and not a risk factor for adverse events. The impact of important covariates (i.e. smoking, right ventricular function, functional mitral or tricuspid regurgitation) was not accounted for in the multivariate analysis, indicating the presence of residual confounding. Finally, patients were not re-evaluated for BMI changes during the follow-up period (median 40 months) and therefore misclassification bias may be of concern.
There is little doubt that BMI is an important therapeutic target for the prevention and treatment of HF. A BMI within the normal and mildly overweight range seems to be the perfect target for HF patients. Future studies should explore whether adopting a healthy lifestyle (i.e. physical activity, healthy diet) may change the clinical course of HF in underweight or obese patients.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Comments