Sudden cardiac death after heart transplantation: can ICD prevent SCD?

Abstract

Advances in surgical technique, immunosuppressive therapy, and patient monitoring have improved overall outcomes in heart transplantation over the last few decades. However, long-term results continue to be suboptimal. In the 2012 International Society for Heart and Lung Transplantation report on adult heart transplantation, the median posttransplant survival was slightly over 10 years in the current era. 1 Transplant-related deaths were largely due to rejection, graft failure, coronary allograft vasculopathy (CAV), infection, renal failure, and malignancy. The contribution of each varied in relation to time since transplantation. Although the incidence of sudden cardiac death (SCD) after heart transplantation appears relatively low, the true incidence has been difficult to quantify to date. Small case series and single institutional studies report a wide range of rates of SCD as well as a variety of significant predictors. In this issue of HeartRhythm, Vakil et al 2 perform a retrospective analysis of SCD data obtained from the largest cohort of heart transplant recipients to date. Their analysis of the United Network of Organ Sharing (UNOS) database included more than 37,000 recipients who were transplanted in the United States over the last 25 years, with more than 17,000 deaths reported. Approximately 4% of the total cohort experienced SCD (10% of all reported deaths) during a mean follow-up period of 6.5 years. While the accuracy of this analysis may be of some concern because of inherent limitations to the UNOS database with respect to reporting and lack of adjudication on the cause of death, the fact that cause of death was reported by the treatment teams may improve reliability of these data. Certainly, there should be enough reliability of the database to provide a general assessment of the incidence of SCD after transplantation and to allow trending of SCD data over time. Now that we have an idea of the incidence of SCD after transplantation, are there any significant predictors that could indicate risk? Previous studies have demonstrated posttransplant SCD to be associated with ventricular dysfunction (left ventricular ejection fraction r40%), severity and number of rejection episodes, CAV, higher donor age, bradycardia, and longer ischemic times. 1,3,4 Vakil et al report that only left ventricular ejection fraction r40%, pharmacologically treated allograft rejection, and higher donor age were independently associated with an increased risk of SCD in the UNOS cohort. Ventricular dysfunction was associated with the highest risk with a 43-fold increase. However, the specific cause of ventricular dysfunction could not be discerned from the UNOS data. Interestingly, the presence of CAV was a univariate predictor of SCD in the UNOS cohort but not in the multivariate analysis. This finding may be related to an underestimate of the presence of CAV according to the UNOS definition of angiographic coronary stenosis of 450%. This definition would not capture those recipients with microvascular angiopathy who could cause significant ventricular dysfunction and subsequent graft failure. As such, a proportion of recipients found to have ventricular dysfunction may also have had a microvascular form of CAV that goes undetected by conventional angiography. The incidence of CAV reported in the total UNOS cohort was similar to that reported in other studies at about 50%. Prior studies have also shown that bradyarrhythmias are important causes of death after heart transplantation. 5,6 Vaseghi et al 5 reported on the terminal rhythms in heart transplant recipients who died suddenly. Asystole was documented in 34%, pulseless electrical activity in 20%, and ventricular fibrillation in 10%. A large retrospective study subsequently demonstrated that long-term survival was better in heart transplant recipients with permanent pacemakers than in those without. 6 Vakil et al could not reproduce these findings in the UNOS cohort. In a subgroup analysis of the 8% of the cohort with permanent pacemakers, there was no significant difference in time to SCD between those implanted with permanent pacemakers and those not implanted. This finding would suggest that bradyarrhythmias or asystole was not a significant mechanism for SCD after transplantation.