Does CMV Status Affect Morbidity and Survival Following Heart Transplantation? A Large Multicentre Retrospective Study

Abstract

<h3>Purpose</h3> Cytomegalovirus (CMV) infection following heart transplant is associated with increased morbidity and mortality. We evaluated the impact of CMV status on long term survival at all centres in the UK, each employing different approaches to managing CMV. <h3>Methods</h3> Adult patients undergoing heart transplantation and surviving >30 days between 1987 and 2018 were included. Data obtained from NHS Blood and Transplant included CMV donor (D) and recipient (R) status, R age, R sex, centre, treated rejection episodes and survival. Patients grouped by CMV status: D-ve R-ve (DN), D+ve R-ve (PM), D+ve R+ve (DP), D-ve R+ve (RM). 5 year survival, adjusted for age, sex and centre, was compared between groups. Adjusted survival between centres was also compared. Survival data was compared using Cox-proportional hazards. Treated rejection rates between CMV groups were compared using Chi² test. <h3>Results</h3> 2675 patients were included from 6 UK transplant centres; 2 use a pre-emptive strategy, 4 use valganciclovir in PM patients and high risk R+ve patients for 3-6 months. Five year survival per group CMV DN 629/750 (83.9%), PM 440/563 (78.1%) RM 655/773 (84.7%) DP 467/592 (78.8%). Compared to DN, both PM (HR 1.43 p=0.005) and DP (HR 1.34 p=0.025) associated with significantly higher mortality. There was no survival difference between centres, p=0.127 and no difference between centres for PM (p=0.69) or DP (0.23) groups. Rates of treated rejection at 12 months were higher in PM and DP patients (p=0.0033). <h3>Conclusion</h3> This is the first multicentre evaluation of the impact of CMV status on longer term survival following heart transplant. CMV mismatch and double positive statuses were associated with higher rates of rejection and reduced survival, irrespective of CMV prevention strategy. This should be considered when counselling patients, and in adjusted survival analyses. No differences in outcomes were seen between centres, although we were unable to account for changes in individual centre practice over time.