Development of a quantitative donor risk index to predict short-term mortality in orthotopic heart transplantation.

BACKGROUND: No standard index based on donor factors exists for predicting mortality after orthotopic heart transplantation (OHT). We utilized United Network for Organ Sharing (UNOS) data to develop a quantitative donor risk score for OHT. METHODS: We examined a prospectively collected open cohort of 22,252 patients who underwent primary OHT (1996 to 2007). Of the 284 donor-specific variables, those associated with 1-year (year) mortality (exploratory p-value < 0.2) were incorporated into a multivariate (MV) logistic regression model. The final model contained donor factors that improved the explanatory power (by pseudo-R2, area under the curve and likelihood ratio test). A quantitative donor risk score was created using odds ratios (ORs) from the final model. For external validity, a cross-validation strategy was employed whereby the score was generated using a randomly generated subset of cases (n = 17,788) and then independently validated on the remaining patients (n = 4,464). RESULTS: A 15-point scoring system incorporated 4 variables: ischemic time; donor age; race mismatching; and blood urea nitrogen (BUN)/creatinine ratio. Derivation and validation cohort scores ranged from 1 to 15 and 1 to 12, respectively (mean 4.0 ± 2.1 for each). Each increase of 1 point increased the risk of 1-year death by 9% (OR = 0.09 [1.07 to 0.12]) in the derivation cohort and 13% (OR = 0.13 [1.08 to 0.18]) in the validation cohort (each p < 0.001). The odds of 1-year mortality by increments of 3 points were: 0 to 2 points (reference); 3 to 5 points (OR = 0.25 [1.12 to 0.40], p < 0.001); 6 to 8 pts (OR = 0.77 [1.56 to 2.02], p < 0.001); and 9 to 15 points (OR = 1.92 [1.54 to 2.39], p < 0.001). Donor risk score was predictive for 30-day mortality (OR = 0.11 [1.08 to 0.14], p < 0.001) and 5-year cumulative mortality (OR = 0.11 [1.09 to 0.13], p < 0.001). CONCLUSIONS: We present a novel donor risk index for OHT predicting short- and long-term mortality. This donor risk score may prove valuable for donor heart allocation and prognosis after OHT.

Duke Scholars

Author Stuart Dean Russell Medicine, Cardiology