The present priority scheme for the allocation of donor hearts based on patient acuity and waiting time contributes to the escalating costs of heart transplantation, ignores the potential outcome advantages of prospective tissue matching, and is vulnerable to manipulation. Costs have trebled in recent years, as recipients frequently spend weeks before transplantation as inpatients in intensive care units and become more susceptible to nosocomial complications. The findings from an international cooperative study suggest that patient survival is correlated with the level of histocompatibility (ie, human lymphocyte antigen [HLA]) matching. We observed a similar inverse association between retrospective fortuitous HLA matching and the risk of rejection in 39 patients undergoing heart transplantation over a 29-month period (p = 0.03 by nonparametric analysis). These observations prompted us to consider the feasibility of donor heart allocation based on the degree of HLA matching and waiting time alone. Current methods permit the accurate determination of HLA type in a matter of hours using donor peripheral blood alone. Human lymphocyte antigen typing, therefore, could be performed locally before organ harvesting, making issues of donor heart preservation irrelevant. We evaluated the extent of HLA matching that might be achieved practically. Forty-seven patients on our waiting list during calendar year 1991 were tested retrospectively for HLA matching with all geographically accessible 1991 heart donors identified by the United Network for Organ Sharing for all donors from hospitals east of the Mississippi River.(ABSTRACT TRUNCATED AT 250 WORDS)