Obesity is associated with several cardiac structural, functional, hemodynamic, and neurohormonal changes, resulting in increased risk for cardiovascular diseases (CVDs). Particularly, obesity is associated with greater risk for heart failure, hypertension, coronary heart disease, atrial fibrillation, and more recently, pulmonary arterial hypertension. Once such diseases have been diagnosed, however, patients with obesity present a more favorable prognosis compared to leaner patients. Such a phenomenon has been defined as “the obesity paradox.” The mechanisms leading to improved outcomes in obese patients are still largely unknown; however, most of the studies investigating the role of obesity on the development and progression of CVD were limited by the use of the body mass index, which does not distinguish between different body composition (BC) compartments, which have distinctive effects on the CV system.
In this review, we will discuss the increased risk of CVD in patients with obesity and the mechanisms through which obesity and particularly BC compartments (e.g., skeletal muscle mass) may confer protection once CVDs have been diagnosed. We will then review the impact of cardiorespiratory fitness in modulating the effects of obesity in patients with CVD and the role of unintentional and intentional weight loss in such conditions. Finally, we will discuss the potential beneficial effects of therapeutics targeting BC compartments in determining prognosis in patients with established CVD.
Obesity has reached epidemic proportions and is an independent risk factor for several noncommunicable diseases, including cardiovascular diseases (CVDs) (1). In established CVD, however, the relationship between obesity and clinical outcomes has not been as straightforward, such that an obesity paradox has been reported (2). The obesity paradox is particularly relevant in heart failure (HF) and coronary heart disease (CHD), in which the role of obesity has been investigated the most (2); however, it is also evident in other CVDs, such as hypertension (HTN), pulmonary arterial HTN (PAH), and atrial fibrillation (AF).
In this article, we will review the role of obesity and body composition (BC) compartments (i.e., fat mass [FM], fat-free mass [FFM], lean mass [LM], and skeletal muscle mass [SMM]) on cardiometabolic risk factors, and we will discuss the CVD in which the obesity paradox has been demonstrated more consistently. Finally, we will review the role of cardiorespiratory fitness (CRF) and weight loss in determining the prognosis in patients with obesity and established CVD.
BODY COMPOSITION CHANGES ASSOCIATED WITH OBESITY
Obesity is defined as excess body fat that impairs health (3). The most commonly used tool to define obesity is a body mass index (BMI) ≥ 30 kg · m−2 (2,3). The BMI correlates fairly well with FM in healthy individuals, at least at the population level (4). At an individual level and in specific physiologic and pathophysiologic conditions, BMI fails to accurately estimate BC, resulting in nutritional status misclassifications (5,6). An assessment of BC should be performed, when possible, to overcome such limitations. The simplest BC model divides body weight into 2 compartments, FM and FFM, which assumes that both remain constant during the course of our lives and/or in the setting of pathophysiologic conditions, and that the composition of FFM does not change over time. Fat-free mass is the largest BC compartment, and it includes total body water, bones, and SMM; the latter and total body water compose LM (Figure 1) (7–9). For this reason, LM is considered the best surrogate for SMM.
While the assessment of FM is important to determine excess adiposity, hence to define obesity, assessing SMM or its surrogates has been recently recognized as a major component of the nutritional status assessment. Lower SMM, particularly in the extremities (i.e., appendicular SMM), is associated with poor functionality and exercise capacity (10,11), and is defined as sarcopenia (12), a condition associated with poor prognosis in several chronic diseases. Importantly, obesity and sarcopenia can also coexist in the presence of excess adiposity and reduction in SMM and its functionality (i.e., sarcopenic obesity) (13), associated with even further less favorable outcomes.
In addition to the lack of BC assessment, BMI also fails to assess FM distribution. Excess visceral adiposity, indirectly measured as waist circumference (WC) or waist-to-hip ratio (WHR), has been considered a stronger cardiometabolic risk factor than BMI alone (14). Although the exact mechanisms are still unclear, visceral adipose tissue is responsible for the synthesis of detrimental pro-inflammatory cytokines described below. Despite its limitations, BMI in population studies in primary prevention is a potent predictor of prognosis (15). In secondary prevention of CVD, as discussed below, BMI generally predicts higher LM and SMM and, therefore, a better prognosis.
OBESITY, INCREASED CVD RISK, AND THE OBESITY PARADOX
Adipose tissue can secrete adipokines and pro-inflammatory cytokines (e.g., leptin, adiponectin, resistin, interleukin [IL]-1, IL-18, tumor-necrosis factor [TNF]-α) (16–18) that can affect cardiac structure and function. The release of these molecules is particularly relevant in understanding the increased HF risk conferred by obesity, but also the progression of atherosclerosis, providing evidence of detrimental effects of obesity in primary prevention for CHD. In fact, pro-inflammatory cytokines have not only been directly involved in the development of atherosclerotic plaques, but also in their stability and risk for rupture (19).
In addition to excess FM, the typical obesity phenotype, in absence of sarcopenia or sarcopenic obesity, assumes that the increase in body weight results in both increase in FM and FFM (20). While FFM is typically protective in chronic diseases, when highly augmented, it can negatively affect cardiac structure and function (Figure 2) (21). Obese patients have a greater cardiac output resulting from an increased stroke volume due to excess central blood volume and volume expansion (22,23). This increase in volume has been deputed to the increased blood flow requirements of FFM (24), which is, on the other hand, extremely low for FM alone to account for the significant increase seen in obesity. The consistent increase in cardiac output as a result of a greater preload can result in a progressively increased cardiac workload, inducing cardiac structural remodeling, which is often, but not exclusively, concentric with concomitant concentric left ventricular (LV) hypertrophy (LVH) (25). Such structural and hemodynamic changes result in cardiac dysfunction (26), more often diastolic, therefore increasing the risk for HF, especially HF with preserved LV ejection fraction (HF-PEF) (27,28), which can occur independently of other risk factors, such as HTN (29). Of note, cardiac dysfunction can also develop in the absence of LVH (30–32), suggesting the presence of alternative mechanisms of obesity-induced cardiac dysfunction. Obesity is also associated with greater LV-end diastolic pressure, right atrial pressure, and pulmonary wedge pressure (33–35). Finally, obesity is associated with an unhealthy diet rich in saturated fatty acids and sugars, which have shown in preclinical models to induce cardiac systolic and diastolic dysfunction, further increasing HF risk (36,37).
Obesity clearly increases the risk of HF with reduced ejection fraction (HF-REF), in addition to HF-PEF, as described above. However, once both forms of HF are diagnosed, overweight and obese patients present a better longterm prognosis compared to those who are normal weight and underweight (Figure 3) (38). Several hypotheses have been proposed and previously described to explain the obesity paradox in HF (2,21,39,40). The inaccuracy of BMI in differentiating BC compartments is a major limitation. Lean mass as a surrogate for SMM is a major predictor of CRF in HF (10,27), and because greater BMI is typically associated with an increase in FM paralleled by increased LM and SMM, the increased BMI may be protective because of greater LM (Figure 2). Obese patients with excess adiposity and reduced LM do not do as well as those with an increase in both compartments (6,41).
Regarding the relationship between visceral adiposity and HF prognosis, recent studies suggest that WC and WHR are associated with a worse prognosis (14); however, prior findings suggested opposite results (42). Finally, the obesity paradox in HF disappears in those with relatively good CRF (Figure 4) (43,44), suggesting that independent of their BMI and BC, improving CRF should be prioritized when treating patients with HF.
Obese patients have lower levels of brain natriuretic peptide (BNP) and N-terminal pro-BNP, resulting from greater clearance as well as reduced synthesis of natriuretic peptides (NPs), which induce natriuresis, diuresis, and vasodilation (45,46). The lower levels in obesity may accentuate the fluid expansion seen in patients with obesity. Natriuretic peptides are markers of cardiac wall stress and useful tools to diagnose HF. Because greater levels of NPs are associated with a worse prognosis in HF (47), we cannot exclude the possibility that lower levels may result in improved outcomes and potentially contribute to explaining the obesity paradox.
Obesity is an independent risk factor for HTN (48,49). The adipose tissue hosts components of the renin-angiotensin aldosterone system (RAAS) (50), perhaps explaining the increased activation of the RAAS seen in obese patients, finally resulting in glomerular injury, reduced natriuresis, increased fluid retention, and HTN (48). Obesity, particularly visceral adiposity, is also associated with increased activation of the sympathetic nervous system (SNS), which is, in turn, linked to HTN (48). Fat mass around the kidney and accumulated within the kidney can also induce a physical renal compression due to an intra-abdominal pressure increase, resulting in increased blood pressure (48).
The role of obesity in patients with diagnosed HTN has been less investigated. However, overweight and obesity have been shown to confer some degrees of protection once HTN was diagnosed, especially when concomitantly associated with CHD (51). The obesity paradox was initially also attributed to other confounding factors, such as smoking and increased alcohol intake (51). A recent meta-analysis including 489,222 participants with HTN found that for each 5-unit increment in BMI, there was an 8% relative risk reduction of all-cause mortality, and there was a significant dose-dependent association between BMI and decreased risk for CVD mortality (52). The mechanisms through which obesity exerts beneficial effects in HTN are largely unknown; however, considering that greater levels of renin are typically associated with worse outcomes, the relatively lower levels of renin reported in obese patients may be responsible, at least in part, for this protection.
Coronary Heart Disease
Excess adiposity, especially visceral compared to subcutaneous, has been associated with increased atherosclerotic plaque load, characterized by greater content of macrophage infiltration and plaque erosion (53,54). Although the underlying mechanisms are not fully understood, the chronic systemic low-grade inflammation characteristic of obesity seems to be a major player. Obese patients present an exaggerated activation of the macromolecular complex NACHT, LRR, and PYD domains-containing protein (NLRP)-3 inflammasome, responsible for the production of IL-1 and IL-18 (19,55). Such pro-inflammatory cytokines can be produced by several tissues in the body, including cardiac cells and FM, but also by the cells of the innate immune system in response to stimuli such as free fatty acids, particularly saturated, and cholesterol crystals (19,55), both of which are increased in the setting of obesity. Moreover, obesity is associated with major risk factors for CHD, such as type 2 diabetes mellitus and dyslipidemia.
In patients with CHD, the obesity paradox has also been confirmed (56,57). The role of BC has been rarely investigated; however, greater LM and FM have been associated with a more favorable prognosis (58), contrary to the previous notion that LM would represent the only protective BC compartment. On the other hand, lower amounts of LM and FM were associated with worse prognosis (Figure 5). Interestingly, improved in-hospital mortality has also been reported in obese elderly patients after myocardial infarction (59).
Similar to what was discussed for HF, the obesity paradox in CHD remains to be fully explained. Cardiorespiratory fitness is an independent prognostic factor in CHD, to a greater degree than BMI and WC, especially in those with greater CRF (Figure 6) (60,61). Therapeutics aimed at improving CRF, such as cardiac rehabilitation and increased physical activity, should be implemented more often and perhaps given priority compared to caloric restriction-induced weight loss strategies not associated with exercise training (62). Cardiac rehabilitation improves BC by increasing LM or preventing its loss, increases physical activity and muscle strength (63), as opposed to dieting alone, which typically results in FM as well as LM losses (64), with the risk of further reducing the amount and functionality of LM in patients with sarcopenia, sarcopenic obesity, or those at high risk for such conditions.
Pulmonary Arterial Hypertension
Obesity and a chronic insulin-resistant state typically resulting from excess adiposity have been recently associated with an increased risk of PAH (34,65,66). Prevalence of obesity in PAH is not clear, but recent studies suggest that it can reach up to 30% (67). The mechanisms driving obesity to PAH are still largely unknown. Obesity is highly associated with chronic obstructive sleep apnea (68), which is in turn associated with increased risk of PAH (65). Patients with obesity are also characterized by pulmonary arterial hypertrophy and hypoventilation resulting in alveolar hypoxia, the major cause of pulmonary vasoconstriction. Of note, PAH has been recently associated with chronic systemic low-grade inflammation (69,70), which is also characteristic of obesity, and it has been targeted in pilot studies to improve exercise capacity in this population (71).
Data assessing the effects of obesity on prognosis are limited, but suggest a neutral or perhaps a protective role of obesity. In a recent study, obesity was not associated with worse prognosis in patients with PAH, except for those who were morbidly obese and less than 65 years of age (67). However, in a large registry including >18,000 patients with PAH, obesity was associated with the lowest risk for inhospital mortality (72), further confirming an obesity paradox in this population. Another interesting study in 267 patients with PAH found that being overweight was associated with the lowest mortality, followed by obesity and morbid obesity, with the worst survival found in normal-weight and underweight patients (73).
Recent evidence supports a causative role of obesity in increasing the risk for AF (74). Obesity and especially severe obesity are associated with left atrial enlargement, conduction abnormalities, interstitial atrial fibrosis, and reduced posterior left atrial endocardial voltage, which are, in turn, associated with greater risk for AF (74,75). Excess adiposity can also accumulate in the pericardium (i.e., pericardial fat) (76) and epicardium (i.e., epicardial fat) (77), which are both positively associated with AF. In addition to the effects on cardiac structure and composition induced by obesity, the high prevalence of HTN and HF in this population also further increases the risk of AF. Of note, chronic systemic low-grade inflammation associated with the NRLP3 inflammasome pathway activation has also been recently involved in the pathophysiology of AF (78).
Despite clear evidence supporting the increased risk for AF in obese patients, once AF is diagnosed, obese patients present an improved long-term prognosis. In about 2,500 patients with AF, overweight and obesity were associated with the lowest risk for all-cause mortality as well as CVD mortality during a mean follow up of 3 years (79). Not only does increased BMI seem to exert protective effects, but also increased visceral adiposity indirectly assessed using WC (80). In almost 18,000 patients with AF, the ones who were defined obese using BMI, but also WC, presented an improved prognosis (80).
Cardiorespiratory fitness also plays a crucial role in concomitant obesity and AF (74,81). Greater CRF is associated with lower AF-related complications (i.e., recurrent arrhythmia) and augments the beneficial effects of intentional weight loss in this population (81). These data support the concept that increasing CRF should be the primary goal of lifestyle and perhaps also pharmacologic therapies in patients with obesity and AF.
THE EFFECTS OF WEIGHT LOSS ON CVD RISK FACTORS AND CVD
Weight loss in overweight and obesity remain the cornerstone for the primary prevention of metabolic diseases (i.e., type 2 diabetes mellitus) as well as CVD. Even if an obesity paradox in several CVD has been reported, patients with obesity would have perhaps not developed such conditions if they avoided FM gain in the first place. Particularly, weight loss, even when modest, is associated with improvement in systolic and diastolic blood pressure (82), likely resulting from improvement in insulin sensitivity and reduction in SNS activity, both of which are highly involved in the development and progression of HTN.
The long-term effects of weight loss in patients with established CVD has been less studied. Unintentional weight loss is considered a marker of poor prognosis (83), independent of initial BMI (84), in most chronic diseases (85), perhaps resulting from an underlying catabolic process. In fact, even in obese patients, unintentional weight loss is associated with less favorable outcomes compared to those who remain weight stable or even compared to those who gain weight, particularly in HF (86–88). However, the effects of intentional weight loss on long-term outcomes in HF remain to be determined, to the extent that major international societies do not recommend weight loss in this population and actually recommend avoiding unintentional weight loss (89), except in those with more severe obesity, in which FM loss may improve CRF (89). Improvements in CRF and quality of life induced by weight loss in HF have been, in fact, confirmed in both HF-REF (90) and HF-PEF (91), mostly as a result of improvement in BC compartments amount and functionality (92–94), but when greater degrees of weight loss are achieved, cardiac structure and function and hemodynamics can also improve significantly (95,96).
In a meta-analysis investigating the effects of weight loss, observational weight loss, likely resulting from unintentional weight loss, has been associated with greater mortality in CHD (97); however, when weight loss was intentional, resulting from lifestyle modifications, the number of events was significantly lower (97). Similar to what was described in HF, larger randomized controlled trials are highly encouraged to finally determine whether intentional weight loss should be implemented in all obese patients with CHD, and which weight class (i.e., overweight, obesity class I, class II) may benefit the most.
The role of weight loss in patients with concomitant obesity and PAH has not been investigated. Improvement in insulin sensitivity through pharmacologic strategies can reverse PAH in preclinical models by restoring fatty acid oxidation (98). Whether weight loss-induced improvements in insulin sensitivity can induce similar effects is unknown at this time.
In AF, weight loss is one of the most powerful therapeutics to improve the AF burden. In a randomized trial of 150 patients with AF and obesity, weight loss was associated with improved symptoms, AF recurrences, and improvements in cardiac remodeling (99). In another study investigating 355 patients, weight loss ≥10% of initial body weight resulted in a six-fold greater probability of arrhythmia-free survival compared to those with lower degrees of weight loss (100). This study also investigated the effects of weight fluctuation, a common phenomenon seen in obese patients attempting to lose weight and typically associated with worse outcomes related to worsening in BC (i.e., increased FM, reduced LM). A weight fluctuation between follow-up visits ≥5% was, in fact, associated with increased risk of arrhythmia recurrence (100). Finally, it is unclear whether improvements in AF-related outcomes and quality of life result in improved overall survival, calling for larger adequately powered randomized controlled trials in the obese AF population.
Obesity is an independent risk factor for CVDs and other CVD risk factors; however, once CVD is diagnosed, there is a strong obesity paradox (2,38,60). While unintentional weight loss is associated with poor outcomes, intentional weight loss resulting from lifestyle changes has the potential to improve clinical outcomes. Due to the lack of strong evidence supporting the recommendation of weight loss in patients with CVD, except for AF, in which weight loss results in improved AF burden, therapies aimed at improving CRF should be implemented (60,101–103), considering that greater CRF is associated with improved outcomes in most CVDs. Therapies such as dietary intervention and exercise training, particularly resistance training to increase LM (104) and aerobic training to improve its functionality, are promising in the setting of obesity and established CVD, although dedicated large clinical trials investigating the effects of these interventions on clinical outcomes are needed.
Conflicts of Interest and Source of Funding: Salvatore Carbone is supported by the VCU DOIM Pilot Project Grant Program 2017 and by the VCU Pauley Heart Center Pilot Project Grant Program 2017.
1VCU Pauley Heart Center, Virginia Commonwealth University, Richmond, VA 23298 USA
2John Ochsner Heart and Vascular Institute, Ochsner Clinical School—The University of Queensland School of Medicine, New Orleans, LA 70121 USA