Application of Predicted Heart Mass Difference in a Canadian Cohort of Heart Transplant Recipients

Purpose Recipient-donor sex-mismatch is a predictor of 1-year mortality after heart transplant (HT). A more granular construct for size mismatch is the difference in heart mass. A retrospective analysis of the UNOS Database showed that only the lowest undersized septile (>19%) was associated with 1 year mortality. These findings have not been validated. We aimed to assess the prognostic value of donor and recipient predicted heart mass (pHM) difference at 2 large Canadian HT centers. Methods A retrospective study of consecutive adult HT recipients (2004 - 2014) was performed at Toronto General Hospital and the Ottawa Heart Institute. The pHM difference, defined as a percent difference between donor and recipient pHM was calculated as previously described: [(pHMrecipient -pHMdonor)/pHMrecipient]]*100. Multivariable Cox and logistic regression models were used to evaluate the association between pHM difference and 1-year mortality and moderate or severe primary graft dysfunction (PGD), adjusted for ischemic time and donor age. Due to the non-linear nature of pHM difference, we applied restricted cubic splines. Results 412 patients (mean age 50±13 years, 71% male) were analyzed. Average ischemic time was 3.7±1.1 hours. Median pHM difference was 3.3% (IQR-15.5-7.0%). 25 (6%) donors were undersized by >19%. There were 46 deaths (6%) and 55 (13%) cases of moderate or severe PGD at 1 year post-transplant. pHM difference was not associated with 1-year mortality (HR 0.99 per 1% increase in pHM difference, 95% CI 0.96-1.02; figure ) or PGD (OR 0.99 per 1% increase in pHM difference, 95% CI 0.96-1.01; figure ). Ischemic time was the only significant predictor of mortality and PGD: HR 1.34 per hour increase, 95% CI 1.08 to 1.68 for mortality and OR 1.64, 95% CI 1.28-2.11 for PGD). Conclusion In this cohort, there was no association between pHM difference and 1-year mortality or moderate/severe PGD. Longer ischemic times may account for the lack of association between pHM and 1-year outcomes