Abstract 14871: Variable Impact of Lean Body Mass and Body Fat on Mortality in Non-Obese Patients

Abstract

Background_In several cardiac diseases and in the elderly population, an inverse relationship between obesity, as determined by body fat (BF), and subsequent prognosis has been demonstrated (the “obesity paradox”). However, some have postulated that it is mainly the underweight who carry higher mortality, and that body fat (BF) is, in fact, detrimental.

Patients and Methods_We retrospectively assessed 21637 patients, classified as nonobese by BMI ,with a preserved left ventricular ejection fraction (LVEF; ≥50%). We calculated BF as determined by the Jackson Pollock equation and LMI as (1-BF) x BMI kg/m². The population was divided based on age and gender adjusted BF classification [underweight (n=774), normal (n=4833) and overweight (n=16030)] and a gender adjusted LMI classification. The population was analyzed by total mortality over average 3.1-year follow-up by National Death Index.

Results_ In univariate analysis, the overweight had a slight trend towards worse survival when compared with normal BF (9% vs 10%; p 0.16). (Figure 1). In proportional hazard analysis, after adjusting for age, gender, relative wall thickness, LVEF and left ventricular mass index, higher BMI was associated with lower mortality (0.92 CI 0.90-0.93). Both higher BF category (HR 0.89; CI 0.80-1.00) and higher lean mass category were protective (HR 0.71; CI 0.64-0.80). The protective effect of BF, however, disappeared when the underweight were excluded (HR 1.00; CI 0.89-1.13).

Conclusions_In non-obese patients with preserved LVEF, LMI is protective, whereas BF only appears protective when considering the negative impact of low BF in the underweight. Figure 1. Multivariate model showing event rates based on body fat after adjusting for age, gender, relative wall thickness, LVEF, lean mass and left ventricular mass index.