Distinct blood volume and left ventricular adaptation to severe obesity in middle-aged adults at risk for heart failure.

AIMS Obesity is commonly hypothesized to lead to the development of heart failure (HF) in part due to increases in blood volume (BV) and left ventricular (LV) remodelling. Whether adiposity and obesity severity are associated with BV expansion and subsequent LV remodelling in middle-aged individuals at increased risk (IR) prior to the onset of HF is unknown. METHODS AND RESULTS We analysed data from 96 middle-aged (40-64 years) non-obese (25.8 [23.6-28.6] kg/m2) controls (CON) and 126 IR middle-aged adults (elevated cardiac biomarkers plus established risk factors). IR adults were stratified based upon body mass index class: (1) <30 kg/m2, IRNon-Obese (n = 28, 28.2 [24.6-29.9] kg/m2); (2) Class I >30-35 kg/m2, IRClass-I (n = 39, 33 [31.9-33.6] kg/m2); and, (3) Class II/III >35 kg/m2, IRClass-II/IIII (n = 59, 41.2 [37.1-43.8] kg/m2). BV (carbon monoxide rebreathing), body composition (hydrodensitometry or dual-energy X-ray absorptiometry), and LV structure and function (echocardiography) were assessed. Fat mass was independently associated with BV (β = 0.17, p < 0.001) which was independently associated with LV end-diastolic volume (LVEDV) index (β = 0.54, p < 0.001). BV was lower in CON (5046 ± 1123 ml) than all IR groups (IRNon-Obese: 5622 ± 1137; IRClass-I: 6033 ± 1237; IRClass-II/III: 6548 ± 1153 mL; all p < 0.05). IRClass-II/III had greater erythrocyte volume compared to CON (p < 0.005), even after normalization to fat-free mass (CON: 36.2 ± 4.6; IRClass-II/III: 39.9 ± 5.1 ml/kg fat-free mass; p < 0.001). Only IRClass-II/III had an enlarged LV end-diastolic volume when normalized to body surface area compared to both CON and IRNon-Obese (both, p < 0.05). CONCLUSIONS While lean mass is the primary determinant of BV, fat mass is independently associated with BV expansion and larger LVEDV. IR adults with class II/III obesity display distinct LV enlargement that is disproportionate to body size (i.e. LVEDV index) and may represent a physiologically distinct subgroup of obesity as opposed to a simple continuum of disease severity.