Are Body Mass Index, Quantitative Whole Body Fat Mass, and Whole Body Fat Free Mass Predictors of Post Heart Transplantation Outcomes?

Abstract

<h3>Purpose</h3> Current heart transplantation (HT) guidelines support weight loss to achieve a body mass index (BMI) ≤35kg/m² prior to listing; however, there are conflicting data on whether pre-transplant obesity is associated with worse outcomes post transplant. A more specific marker of human body composition, whole body fat mass (WBFM), derived from fat segmentation using abdominal computed tomography (abdCT), has been shown to predict outcomes in chronic illness. We sought to assess whether WBFM, whole body fat free mass (WBFFM), and BMI were predictive of 1-yr post-transplant outcomes. <h3>Methods</h3> Adults who underwent HT from 2014-2019 at a single tertiary medical center with interpretable abdCT obtained within 3 months prior to HT were retrospectively assessed. Patients were excluded if abdCTs were unavailable, acquired >3 months from transplant date, or had poor image quality. Tissue characterization/fat and muscle segmentation were performed using Slice-O-Matic software v5.0 (TomoVision, Canada). WBFM and WBFFM were derived. Statistical analysis was performed with R statistical software (v4.1.1). <h3>Results</h3> A total of 69 pts were analyzed, of which 56 were male (81.2%), 39 Caucasian (56.5%), and 24 (34.8%) with pre-transplant diabetes mellitus (DM). Median BMI [IQR] was 29.2 kg/m² [23.7,31.7], median WBFM [IQR] 27.9kg [22.6, 34.2], and median WBFFM [IQR] 51.9kg [46.5,57.9]. In univariate Poisson regression, BMI (p<.01) and WBFM (p=0.01) were predictive of index hospital length of stay (LOS). However, BMI, WBFM, and WBFFM were not predictive of composite of infection, rejection, graft failure, or death in the 1st year. In univariate logistic regression, BMI, WBFM, WBFFM were not predictive of rejection or graft failure. In multivariate Poisson regression adjusting for age, male sex, pre-transplant DM, and CKD history, BMI was only predictive of index LOS (IRR 1.02, p<0.01, CI 1.01, 1.03). <h3>Conclusion</h3> Consistent with prior studies, BMI is not associated with worse 1-yr outcomes after HT. While quantitatively derived WBFM and BMI are predictive of index LOS, we found that overall BMI, WBFM, and WBFFM are not predictive of infection, rejection, graft failure, or death within 1 year. Limitations include small sample size and retrospective nature. Future studies should aim to include larger sample size and to prospectively assess long-term outcomes.