Cardiac Morphology and Performance in Severe Obesity: Size Really Does Matter

Severe obesity produces hemodynamic alterations that predispose to changes in cardiac morphology and ventricular function that may lead to left and right heart failure [1,2,3]. Excessive adipose accumulation in association with increased fat-free mass contributes to an increase in total and central blood volume. In the absence of hypertension, peripheral vascular resistance is reduced, facilitating an increase in cardiac output (CO). Since heart rate changes little, if at all, augmentation of CO is due entirely to an increase in left ventricular (LV) stroke volume. The increase in CO predisposes to LV and, presumably, right ventricular dilatation. In uncomplicated (normotensive) obesity, LV dilatation ostensibly leads to eccentric LV hypertrophy (LVH) due to elevated LV end-diastolic wall stress. The presence of LVH predisposes to LV diastolic dysfunction. If LV wall stress remains chronically elevated (possibly indicating inadequate LVH), LV systolic function may ensue. These alterations may predispose to LV failure, pulmonary venous hypertension, pulmonary arterial hypertension and eventually right ventricular failure. Hypoxia from sleep apnea and obesity hypoventilation may contribute to the development of pulmonary arterial hypertension and right heart failure. This clinical syndrome is known as obesity cardiomyopathy. Hypertension occurs in up to 60% of severely obese patients [4]. CO and LV stroke volume remain elevated in such patients [4]. Peripheral vascular resistance is often characterized as ‘inappropriately normal', but is lower than expected for the degree of blood pressure elevation [3,4]. Obesity hypertension predisposes to a hybrid form of LVH. Previously described as eccentric-concentric hypertrophy, it is now classified as a form of concentric LVH [3,4]. LV diastolic dysfunction in patients with obesity hypertension tends to be more severe than in normotensive obese subjects [3,4]. In recent years, the effects of various neurohormonal and metabolic alterations on cardiac structure and function have received increasing investigative attention, mainly involving animal models of obesity [2,5]. These include increased renin-angiotensin-aldosterone system activation, increased sympathetic nervous system tone, insulin resistance with hyperinsulinemia, leptin resistance with hyperleptinemia, adiponectin deficiency, lipotoxicity and lipoapoptosis [2,5]. Whether and to what extent such neurohormonal and metabolic abnormalities contribute to altered cardiac structure and function in humans remains uncertain.Early studies tended to support the hypothesis that obesity predisposes to eccentric LVH [1,6]. However, several recent studies have challenged this concept by demonstrating that concentric LVH or remodeling occurs as frequently as or more frequently than eccentric LVH in class I and II obese patients [5,7,8]. However, only one of these studies adjusted for the presence of hypertension [7] and none accounted for relative severity and duration of obesity and hypertension. Other recent studies of obese patients demonstrate a predominance of eccentric LVH among those with LV remodeling [9,10]. Thus, the issue of LV geometry in mild-to-moderate obesity remains unsettled. In this issue of Cardiology, Antonini-Canterin et al. [8] report the results of a study of cardiac structure and function and insulin resistance in 160 morbidly obese and 38 superobese patients. The World Health Organization classifies morbid or class III obesity as a BMI ≥40 [1,2,3]. Superobesity is defined as a BMI ≥50 [3,8]. Although not specifically stated, it is presumed that the BMI range in the morbidly obese group in this study was 40.0-49.9. Cardiac structure and function was assessed using transthoracic echocardiography. The homeostasis model assessment index (HOMA-IR) was calculated to assess insulin resistance.Key echocardiographic parameters were analyzed as both categorical and continuous variables [8]. Mean LV internal dimensions in diastolic and systolic, mean LV end-diastolic and end-systolic volume indexes, mean interventricular septal and LV posterior wall thicknesses, mean LV mass/height2.7, mean left atrial volume index and mean E/e' ratio were significantly larger/greater in superobese subjects than in morbidly obese subjects. There was no significant difference in LV ejection fraction between groups. Relative wall thickness trended toward concentric LVH in both groups, more so in superobese than in morbidly obese patients. HOMA-IR was significantly greater in superobese than in morbidly obese subjects. BMI correlated positively and significantly with LV end-diastolic and end-systolic volume indexes, LV mass/height2.7, E/e' and left atrial volume index. There was no correlation between BMI and either LV ejection fraction or HOMA-IR. Previous studies assessing LV morphology in class I, II and III obese patients consistently demonstrated that LV mass, and to a lesser extent LV diastolic chamber size/volume or wall thickness, were significantly larger/greater than that of lean patients [1,2,3,4,5,6]. Moreover, there appears to be a linear relationship between BMI and LV mass in such patients [1,2,3,6]. The results of this study indicate that such observations can be extended to the superobese population.Although there was no significant difference in mean relative wall thickness between morbidly obese and superobese groups, there was a trend towards concentric LVH in both groups. This is not surprising, given that more than half of the patients in both groups were hypertensive. One of the major flaws in many of the studies reporting a high prevalence of concentric LV remodeling or hypertrophy in obese subjects is a failure to adjust for hypertension or to account for the relative duration and severity of hypertension and obesity. Other potential causes of concentric LV remodeling or hypertrophy in seemingly normotensive obese patients include underdiagnosis of hypertension, renin-angiotensin-aldosterone system activation, increased sympathetic nervous system tone and metabolic factors, such as hyperinsulinemia due to insulin resistance and hyperleptinemia due to leptin resistance [1]. In class I, II, and III obese patients, LV diastolic filling has consistently been reported to be impaired when compared to lean controls [1,2,3,5,10,11]. Some studies have reported a correlation of diastolic filling abnormalities with BMI independent of LV mass, while others have reported such abnormalities primarily in those with LV hypertrophy [1,2,3,5,10,11]. The results of this study suggest that there is a linear relationship between BMI and diastolic dysfunction in morbidly obese and superobese patients. Whether this is related to LV mass is uncertain because multivariate analysis was not performed. In most studies, LV ejection phase indexes in all classes of obesity have been reported to be normal or increased [1,2,3,5]. Thus, it is not surprising that there was no significant difference in mean LV ejection fraction values between morbidly obese and superobese groups.The results of this study indicate that insulin resistance becomes more pronounced as obesity becomes more severe, although not in a linear fashion. Insulin resistance has been shown to contribute to increased LV mass in obese humans [12]. The results of this study cannot be applied to a population of uncomplicated (normotensive) morbidly obese or superobese patients, due to the high prevalence of hypertension. The study population is, however representative of morbidly obese and superobese patients encountered in clinical practice. When viewed in the context of previous research, the results suggest that adverse alterations in cardiac morphology and LV diastolic function form a continuum of severity, beginning with class I obesity and progressing through class II to class III (morbid) obesity and superobesity. Similarly, insulin resistance is more pronounced in superobesity than in less severe forms of obesity.