A Markov Analysis of Screening for Late-Onset Cytomegalovirus Disease in Cytomegalovirus High-Risk Kidney Transplant Recipients.

Abstract

Management strategies are unclear for late-onset cytomegalovirus infection occurring beyond 6 months of antiviral prophylaxis in cytomegalovirus high-risk (cytomegalovirus IgG positive to cytomegalovirus IgG negative) kidney transplant recipients. Hybrid strategies (prophylaxis followed by screening) have been investigated but with inconclusive results. There are clinical and potential cost benefits of preventing cytomegalovirus-related hospitalizations and associated increased risks of patient and graft failure. We used decision analysis to evaluate the utility of postprophylaxis screening for late-onset cytomegalovirus infection. We used the Markov decision analysis model incorporating costs and utilities for various cytomegalovirus clinical states (asymptomatic cytomegalovirus, mild cytomegalovirus infection, and cytomegalovirus infection necessitating hospitalization) to estimate cost-effectiveness of postprophylaxis cytomegalovirus screening strategies. Five strategies were compared: no screening and screening at 1-, 2-, 3-, or 4-week intervals. Progression to severe cytomegalovirus infection was modeled on cytomegalovirus replication kinetics. Incremental cost-effectiveness ratios were calculated as a ratio of cost difference between two strategies to difference in quality-adjusted life-years starting with the low-cost strategy. One-way and probabilistic sensitivity analyses were performed to test model's robustness. There was an incremental gain in quality-adjusted life-years with increasing screening frequency. Incremental cost-effectiveness ratios were $783 per quality-adjusted life-year (every 4 weeks over no screening), $1861 per quality-adjusted life-year (every 3 weeks over every 4 weeks), $10,947 per quality-adjusted life-year (every 2 weeks over every 3 weeks), and $197,086 per quality-adjusted life-year (weekly over every 2 weeks). Findings were sensitive to screening cost, cost of hospitalization, postprophylaxis cytomegalovirus incidence, and graft loss after cytomegalovirus infection. No screening was favored when willingness to pay threshold was <$14,000 per quality-adjusted life-year, whereas screening weekly was favored when willingness to pay threshold was >$185,000 per quality-adjusted life-year. Screening every 2 weeks was the dominant strategy between willingness to pay range of $14,000-$185,000 per quality-adjusted life-year. In cytomegalovirus high-risk kidney transplant recipients, compared with no screening, screening for postprophylactic cytomegalovirus viremia is associated with gains in quality-adjusted life-years and seems to be cost effective. A strategy of screening every 2 weeks was the most cost-effective strategy across a wide range of willingness to pay thresholds.