Results
This study included a total of 22,450 OHT recipients. Of these, 13,389 (59.6%) were male and received a heart from a male donor (MtoM), 3,660 (16.3%) were female and received a heart from a female donor (FtoF). A total of 2,292 (10.2%) recipients were female and received a heart from a male donor (MtoF), and 3,019 (13.4%) were male and received a heart from a female donor (FtoM). Baseline recipient, donor, and transplant characteristics are presented in Table 1 . Median age was highest in the FtoM cohort, and this cohort had both the highest percentage of patients with ischemic cardiomyopathy and lowest percentage of patients with nonischemic cardiomyopathy (all P<0.001).
Distributions of donor-to-recipient sizing are presented inTable 2 . For the weight and BMI metrics, a cutoff of > 20% donor-to-recipient discrepancy, or a donor-to-recipient ratio < 0.8 were used to define “undersized”. A > 15% donor-to-recipient discrepancy, or a donor-to-recipient ratio < 0.85 were used to define undersizing using the pHM metric. These thresholds were based on prior publications6,8,9. FtoM recipients were least commonly undersized based on weight and BMI sizing metrics (both P<0.001). When using the pHM sizing metric, FtoM recipients were most commonly undersized (48.1% undersized, P<0.001).
Following OHT, the FtoM cohort experienced the highest incidence of renal failure requiring dialysis (Table 3 ) (P=0.039). Rates of posttransplant stroke and pacemaker placement were equivalent. Thirty-day mortality was highest in the FtoM cohort and lowest in the MtoF cohort (4.6% vs 2.9%, P=0.003). Drug-treated acute graft rejection within the first year of transplantation was highest in the FtoF cohort and lowest in the MtoM cohort (16.5% vs 9.7%, P<0.001).
Median follow up time was 2.98 years (IQR 1.00 to 5.63 years). At one year, actuarial survival was greatest in the MtoF cohort (92.0%). Actuarial survival was lowest in the FtoM cohort (90.0%, P=0.0169). Kaplan Meier survival comparison of all donor-recipient sex pairs are displayed in Figure 1 .
A multivariable analysis was performed to identify independent risk-adjusted predictors of one-year mortality following OHT. This analysis adjusted for multiple independent predictors of mortality including race, heart failure etiology, body mass index, pretransplant mechanical ventilation or use of intra-aortic balloon pump, ventricular assist device, donor age, and graft cold ischemia time. Predictors are presented in Table 4 . In this analysis, both FtoF (HR 1.29, 95% CI 1.14 to 1.62, P=0.001) and FtoM (OR 1.17, 95% CI 1.01 to 1.36, P=0.034) donation were associated with increased odds for mortality (in relation to MtoM).
Because the FtoM cohort was found to have the lowest unadjusted 30-day and one-year mortality, we investigated the impact of various donor-recipient sizing metrics on posttransplant outcomes. In a univariable analysis, donor weight undersizing was associated with a 1% increase in odds for one-year mortality for each 1% undersized in relation to the recipient, though these findings did not reach statistical significance (OR 1.01, 95% CI 1.00 to 1.01, P=0.055). This relation was also observed with undersizing based on pHM, and did not meet statistical significance (OR 1.01, 95% CI 1.00 to 1.02, P=0.170). Undersizing by BMI was found to have a significant association with one-year mortality (per 1% undersized, OR 1.01, 95% CI 1.00 to 1.01, P=0.008).
Multivariable logistic regression was performed to identify independent risk factors of one-year posttransplant mortality in the FtoM cohort. In this model, donor-recipient sizing metrics were adjusted for recipient age, year of transplantation, heart failure etiology, total bilirubin and serum creatinine, pretransplant mechanical ventilation, ventricular assist device, and waitlist time. Full models are displayed inSupplemental Tables 1 – 3 . In these models, donor heart undersizing was significantly associated increased odds of mortality for all three sizing metrics (Table 5 ). For each sizing metric, a 1% increase in the degree of undersizing was associated with a 1% increase in odds of one-year mortality. A large change in donor-recipient sizing ratio (50% reduction, or recipient value twice that of the donor value) was associated with significant increases in odds of mortality. Of sizing metrics, a 50% reduction in pHM ratio had the highest odds of mortality (OR 3.74, 95 CI 1.25 to 11.16, P=0.018).
Donor sizing metrics were also analyzed as categorical variables with weight and BMI ratios of < 0.8 considered undersized, and with a pHM < 0.85 considered undersized. In these analyses after risk adjustment, both undersizing based on weight (OR 1.40, 95% CI 0.94 to 2.07, P=0.095) and BMI (OR 1.47, 95% CI 0.84 to 2.58, P=0.180) were not significantly associated with posttransplant mortality. Undersizing using the pHM metric was associated with a significant increase in mortality (OR 1.32, 95% CI 1.02 to 1.71, P=0.035).