Six-month prophylaxis is cost effective in transplant patients at high risk for cytomegalovirus infection

Abstract

The risk of late-onset cytomegalovirus (CMV) infection remains a concern in seronegative kidney and/or pancreas transplant recipients of seropositive organs despite the use of antiviral prophylaxis. The optimal duration of prophylaxis is unknown. We studied the cost effectiveness of 6- versus 3-mo prophylaxis with valganciclovir. A total of 222 seronegative recipients of seropositive kidney and/or pancreas transplants received valganciclovir prophylaxis for either 3 or 6 mo during two consecutive time periods. We assessed the incidence of CMV infection and disease 12 mo after completion of prophylaxis and performed cost-effectiveness analyses. The overall incidence of CMV infection and disease was 26.7% and 24.4% in the 3-mo group and 20.9% and 12.1% in the 6-mo group, respectively. Six-month prophylaxis was associated with a statistically significant reduction in risk for CMV disease (HR, 0.35; 95% CI, 0.17 to 0.72), but not infection (HR, 0.65; 95% CI, 0.37 to 1.14). Cost-effectiveness analyses showed that 6-mo prophylaxis combined with a one-time viremia determination at the end of the prophylaxis period incurred an incremental cost of $34,362 and $16,215 per case of infection and disease avoided, respectively, and $8,304 per one quality adjusted life-year gained. Sensitivity analyses supported the cost effectiveness of 6-mo prophylaxis over a wide range of valganciclovir and hospital costs, as well as variation in the incidence of CMV disease. In summary, 6-mo prophylaxis with valganciclovir combined with a one-time determination of viremia is cost effective in reducing CMV infection and disease in seronegative recipients of seropositive kidney and/or pancreas transplants.

Figure 1.

Kaplan-Meier event-free survival curves for (A) CMV infection and (B) CMV disease.

Figure 2.

(A) Decision tree with probability for various clinical outcomes and (B) Markov transitional model with four main health states and subsequent substates.

Figure 3.

(A) Two-way sensitivity analysis for ICER with varying costs in medication and in hospital admission: shaded area represents the dominant effect of 6-mo CMV prophylaxis. ★, Baseline value for hospital and medication cost. (B) One-way sensitivity analysis for ICER with varying incidence in CMV disease in the 6-mo CMV prophylaxis group. ▴, Actual CMV disease rate in the 6-mo group; *, no protocol-driven viremia determination was performed in the 6-mo group.