Cost prediction in liver transplantation using pretransplant donor and recipient characteristics

Background. Liver transplantation is a costly procedure and its cost is likely driven by both donor and recipient factors. Recently, the recipient model for end-stage liver disease (MELD) score has been correlated with increased posttransplant cost; however, other factors have not been identified. We sought to identify if other donor and recipient factors are associated with increased cost. Methods. One hundred sixty-six liver transplants performed at our center from January 2004 through February 2006 were included in the estimation sample, and the subsequent 75 transplants were used as a validation cohort. To determine whether donor factors influenced cost, two latent class linear regression models were created from the estimation sample: one considering only recipient variables (model A) and a second incorporating both donor and recipient factors (model B). The resultant models were then validated in the second group of patients and compared with the best single-segment linear regression models. Results. Model A predictors include pretransplant intensive care unit (ICU) stay, age x body mass index, and calculated MELD. In model B, significant predictors are calculated MELD, age, age x pretransplant ICU stay, and donor age more than 40 as significant variables. In validation, only model A remained predictive of cost. Conclusions. Although marginal donor factors are recognized to influence clinical outcome, they did not factor significantly in cost modeling. In addition to MELD, the recipient factors of pretransplant ICU stay, age, and body mass index are pretransplant variables correlated mostly with posttransplant cost across broad populations.