Increased Prevalence of Cardiac Allograft Donors with Improved LV Systolic Function Utilizing an Out-of-Hospital Treatment and Recovery Center

Abstract

Background Published data from UNOS demonstrated that donor hearts with transient LV dysfunction can be used without increasing the risk of poor transplant outcomes, but account for less than 3% of allocated donors. Our organ procurement organization (OPO) utilizes a centralized treatment and recovery facility allowing for aggressive protocol driven management and frequent assessment of organ function early after brain death declaration. Methods We retrospectively reviewed our OPO experience with cardiac allograft donors from 2013-2017. Clinical data (patient demographics, cause of death, medical management, biomarkers, and serial echocardiographic imaging) were extracted from the OPO EMR for all 168 allocated cardiac donors during this time period. Improved LV systolic function was defined as an initial LVEF ≤40% which later improved to ≥50% consistent with the definition used in the prior UNOS analysis. Results We found that 23% (38/168) of allocated cardiac allografts from our OPO had transient LV dysfunction with an EF ≤40% on the initial echocardiogram. This rate was consistent over the years examined. Moreover 55% of donors had an initial echocardiogram with an EF of less than 50%. Comparison of donors with an initial EF ≤40% to donors with normal LV function on initial examination revealed minimal differences. There was no difference in the average age or sex between the two groups. No differences were found in the dose of vasopressors used either at the time of arrival to the facility or at organ harvest. The average amount of blood transfused was not different between the two groups. Although peak ALT was significantly higher in donors with transient LV dysfunction (402 vs 175 Units/L), there was no differences found in peak AST, troponin, lactate, or creatinine. Conclusion We found a markedly higher percentage of cardiac donors with transient LV dysfunction in allografts coming from our OPO facility compared to the UNOS database. The extent to which this reflects increased recognition versus increased organ yield is unknown at this time. These donors did not differ significantly from donors with normal LV function on initial examination. Coupled with prior data that showed equivalent outcomes in donors with recovered LV function, these data suggest that additional efforts to recover cardiac allografts initially found to have LV dysfunction could be an important means of increasing the donor pool and further work is ongoing to understand the most important factors in the recovery process.