An index to predict hospital length of stay after liver transplantation could address unmet clinical needs. Length of stay is an important surrogate for hospital costs and efforts to limit stays can preserve our healthcare resources. Here, we devised a scoring system that predicts hospital length of stay following liver transplantation. We used univariate and multivariate analyses on 73 635 adult liver transplant recipient data and identified independent recipient and donor risk factors for prolonged hospital stay (>30 days). Multiple imputation was used to account for missing variables. We identified 22 factors as significant predictors of prolonged hospital stay, including the most significant risk factors: intensive care unit (ICU) admission (OR 1.75, CI 1.58-1.95) and previous transplant (OR 1.60, CI 1.47-1.75). The length of stay (LOS) index assigns weighted risk points to each significant factor in a scoring system to predict prolonged hospital stay after liver transplantation with a c-statistic of 0.75. The LOS index demonstrated good discrimination across the entire population, dividing the cohort into tertiles, which had odds ratios of 2.25 (CI 2.06-2.46) and 7.90 (7.29-8.56) for prolonged hospital stay (>30 days). The LOS index utilizes 22 significant donor and recipient factors to accurately predict hospital length of stay following liver transplantation. The index further demonstrates the basis for a clear clinical recommendation to mitigate risk of long hospitalization by minimizing cold ischemia time.
Keywords: hospital length of stay; liver transplantation; predictive index.
© 2017 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.