Cost‐effectiveness of transfusing virus‐inactivatedplasma instead of standard plasma

Publisher Website
Google Scholar
Abstract
BACKGROUND: Virus inactivation of plasma intended for transfusion avoids the transmission of hepatitis B virus, hepatitis C virus, and HIV. However, because most plasma recipients also receive other blood components concomitantly, the procedure reduces but cannot eliminate the risk of transfusion‐transmitted infection. As virus‐inactivated plasma has just been licensed in the United States and other countries, a cost‐effectiveness analysis is pertinent.STUDY DESIGN AND METHODS: A Monte Carlo simulation of a Markov model representing the possible outcomes of plasma recipients was used to derive costs and utilities of transfusing virus‐inactivated plasma instead of standard plasma. Probability distributions for patients' age and sex and for the number of blood components transfused per case were determined in 924 plasma recipients in a tertiary‐care hospital. Other values were obtained from the medical literature. Results of the baseline and sensitivity analyses are the mean (± SD) of 10 simulations with 107 patients per simulation.RESULTS: In the baseline analysis, transfusing virus‐inactivated plasma instead of standard plasma prolonged the quality‐adjusted survival by 1 hour and 11 minutes per patient, at a cost‐effectiveness ratio of $2,156,398 ± $257,587 per quality‐adjusted life year gained. Cost‐effectiveness was most sensitive to the patients' mean age, the incremental cost per unit of virus‐inactivated plasma, the HIV and hepatitis C virus transmission rates, and the short‐term mortality of plasma recipients due to their underlying diseases.CONCLUSIONS: Compared to most accepted medical procedures, the transfusion of virus‐inactivated plasma produces little health benefit at a very high cost. This poor cost‐effectiveness ratio is due to the low current risk of infection with transfusion‐transmitted viruses and to the greater age and poor short‐term prognosis of most plasma recipients.