Donor risk analysis and validation in heart transplants: a single-centre experience

A heart transplant (Htx) remains the gold standard treatment for patients with advanced heart failure. Considering the limited availability of organs, donor risk scores might improve organ selection and allocation. The objective of the study was to compare United Network for Organ Sharing, RADIAL and Eurotransplant scoring models in calculating post-Htx outcomes in an Italian Htx population.

Between January 2000 and December 2017, a total of 461 adult patients underwent Htxs. United Network for Organ Sharing, RADIAL and Eurotransplant scores were calculated. Clinical features and donor risk scores were tested to identify preoperative, intraoperative and postoperative risk variables and eventually validate the scores on our population.

Early graft failure was detected in 16.1% (74/461). Post-Htx extracorporeal life support was used in 11.1% (51/461). Of the donor-related factors, the use of noradrenaline (P = 0.015) negatively influenced early outcomes, whereas an ischaemic time >240 min (P = 0.037) influenced early graft failure occurrence. The Eurotransplant donor score did not impact outcomes; the RADIAL score significantly influenced both early and late mortality; and the United Network for Organ Sharing score influenced only late mortality. On the multivariable analysis, after adjustment of scores per cohort, noradrenaline infusion was the main independent predictor of in-hospital mortality for the donors, whereas age of the recipient [odds ratio (OR) 1.003, 1.003-1.081; P = 0.032] and use of preoperative extracorporeal membrane oxygenation (OR 3.320, 1.124-9.805; P = 0.030) were the main independent predictors for the recipients.

None of the validated donor scoring systems fully behave as reliable predictors of transplant outcomes. According to our 'local only' graft selection, specific donor and recipient risk variables should be monitored in order to predict early and late outcomes satisfactorily.