ECMO Salvage for Primary Allograft Failure Following Cardiac Transplantation

Abstract

Introduction: Primary allograft failure (PAF) is the leading cause of death within 30 days of cardiac transplantation. The aim of this study was to evaluate the outcome of patients with PAF managed with veno-arterial ECMO. Methods: Data were reviewed on 53 consecutive heart transplants performed at our institution between 1 June 2005 and 31 January 2008. Nine patients (17%) who required ECMO in order to wean from cardiopulmonary bypass (group 1) were compared with the remaining patients in the cohort (group 2). Results: There were no significant differences in preoperative recipient characteristics. Group 1 patients were more likely to receive hearts with abnormal ventricular function on echocardiography prior to retrieval (4/5 vs. 4/38, p= 0.033). Eight (89%) patients requiring ECMO were weaned off support following a mean of 91.3 h, and the Ross procedure between 1994 and 2005 (8.9± 4.7 years post-Ross operation). Subjects performed a bicycle exercise test, had 2D-echocardiography and cardiac MR imaging. 76% of subjects had aortic stenosis (including 12 cases of biscuspid aortic valve), 14% had aortic regurgitation and 10% had mixed disease. In 67% of subjects a valvotomy was performed prior to the Ross procedure. Freedom from re-do aortic and/or pulmonary valve surgery was 74%. The majority of subjects were classed as NYHA I (74%) with the remainder being NYHA II. On exercise testing, the mean predicted exercise capacity achieved was 85± 22%. On echocardiography the peak velocity across the aortic valve and homograft was 1.3± 0.4m/s and 2.5± 0.6m/s respectively. Cardiac MR imaging identified only trivial aortic valve and homograft regurgitation (6.3± 8.3% and 6.6± 9.5%, respectively). Biventricular systolic function was normal (LV EF 62± 7% and RV EF 61± 6%). The indexed LV mass was 83± 19 g/m2. Of note, in 30% of cases there was evidence of late gadolinium enhancement within the left ventricle. In the main this was noted in the subendocardium of the interventricular septum. The mean aortic sinus root dimension was 40± 8mm. There is excellent autograft and homograft functionmany years post Ross procedure. The finding of septal wall infarctionmaybe related to coronary re-implantationduring the procedure.