Routine Surveillance for Rejection Greater Than Two Years after Heart Transplant is Not Cost-Effective

Abstract

Introduction Endomyocardial biopsies (EMB) and gene expression profiling (AlloMap) are performed after heart transplant (HT) to evaluate for asymptomatic rejection though acute rejection is rare after year 2. The cost-effectiveness of routine surveillance studies is unknown. Hypothesis Routine surveillance for acute rejection is not cost-effective in patients greater than 2 years from HT. Methods A Markov model was constructed to compare 3 surveillance strategies: (1) Stopping routine surveillance studies 24 months after HT (baseline), (2) routine EMB every 6 months from 24-60 months, and (3) routine AlloMap every 6 months from 24-60 months. In all strategies, patients underwent an EMB for signs/symptoms of rejection. The model cycle length was 30 days. Hospitalization costs for rejection were based on the Healthcare Cost and Utilization Project and costs for tests and procedures from the Medicare & Medicaid Services Physician fee schedule. Patients were assumed to be treated as an inpatient for symptomatic acute cellular rejection (ACR) and all antibody-mediated rejection (AMR) and as an outpatient for asymptomatic ACR detected on routine surveillance. Probabilities used in the analysis were derived from patients transplanted at the University of Michigan from 2007 until January 2016 and who survived at least 2 years. Utilities for this analysis were based on literature review, when available. The analysis was performed with TreeAge Pro software. Results Probabilities were derived from 159 patients undergoing routine surveillance for rejection (4,645 cycles). After a routine EMB, AMR occurred with a probability of 0.005 and ACR never occurred. Probability of EMB after a routine AlloMap was 0.06 with a subsequent probability of ACR of 0.04 and no AMR. Compared to a baseline strategy of no surveillance, surveillance EMB was dominated and AlloMap had an incremental cost effectiveness ratio (ICER) $1.57 million/Quality-adjusted life year (QALY) (Table). Sensitivity analyses supported these findings; AlloMap remained cost prohibitive at a threshold of $100,000/QALY, and EMB was dominated up to a 25% risk of death for ACR and 15% risk for AMR. Conclusions Acute rejection is rare after post-HT year 2 and routine surveillance for rejection is not cost-effective.