Posttransplant Complications and Survival
Following transplantation, patients with ≥5% weight loss experienced the highest rate of renal failure requiring dialysis, while those with ≥5% weight gain experienced the lowest (11.2% vs 14.1% vs 10.3%, P<0.001). The ≥5% weight loss cohort also experienced the highest incidence of pacemaker requirement (2.9% vs 3.7% vs 2.3%, P=0.044) and drug-treated acute rejection at one year (11.0% vs 13.4% vs 12.1%, P=0.005) (Table 2 ). Rates of stroke, hospital length of stay, and 30-day mortality were comparable.
Ninety-day and one-year survival were greatest in recipients with ≥5% weight gain and lowest in those with ≥5% weight loss (94.7% vs 92.8% vs 95.8%; and 91.6% vs 89.0% vs 92.3%, P=0.008) (Figure 1 ). In multivariable analysis controlling for both BMI and ventricular assist device, ≥5% weight loss was associated with 26% increased hazards for one-year mortality (HR 1.26, 95% CI 1.07 to 1.48, P=0.005) and ≥5% weight gain was associated with 17% decrease in mortality (HR 0.83, 95% CI 0.69 to 1.00, P=0.047) compared to the stable weight cohort (Table 3 ). Decreasing weight modeled as a continuous variable was also associated with increased hazards for mortality (per 1% lost, HR 1.02, 95% CI 1.01 to 1.03, P<0.001) (Supplemental Table 1 ). Other factors associated with increased hazards for mortality included increasing BMI, congenital heart disease, restrictive heart disease, increasing total bilirubin and serum creatinine, pretransplant mechanical ventilation and extracorporeal membrane oxygenation, increasing donor age, female donor, and increasing graft cold ischemia time.
To model the effects of weight loss across initial BMI at waitlisting, B-spline regression was performed. Probability of one-year posttransplant mortality was plotted across initial waitlist BMI inFigure 2 . Weight loss ≥ 5% is shown to have increasing probability of mortality in waitlist BMI levels below 24 kg/m2. However, in patients with stable weight or weight gain, decreasing initial waitlist BMI did not appear to increase probability of one-year posttransplant mortality.