Comparison of different scoring systems in predicting short-term mortality after liver transplantation

Many scoring systems have been used in predicting the outcomes of liver transplantations. The aim of this study was to compare between 4 scoring systems-Sequential Organ Failure Assessment (SOFA), Model for End-Stage Liver Disease, Acute Physiology and Chronic Health Evaluation II, and Child Turcotte-Pugh -among patients who underwent living-donor liver transplantation (LDLT) seeking to evaluate the best system to correlate with post-operative outcomes.

This study retrospectively reviewed the medical records of 53 patients who had received LDLT in a tertiary care hospital from January 2005 to December 2010. Demographic, clinical, and laboratory data were recorded. Each patient was assessed by use of 4 scoring systems before transplantation and on post-operative days 1 to 7 and at 3 months.

The overall 3-month survival rate was 64%. The pre-transplant SOFA score had the best discriminatory power; moreover, the SOFA score on post-operative day 7 had the best Youden index (.875). The survival rate at 3-month follow-up after liver transplantation differed significantly (P = .00023, highest area under the receiver operator characteristic curve = .952) between patients who had SOFA scores <8 and those had SOFA score >8 on post-liver transplant day 7. This study also demonstrated that respiratory rate (P = .017) and serum bilirubin level (P = .048) and duration of intensive care unit stay (P = .04) are significant risk factors related to early mortality after LDLT.

The pre-transplant SOFA score was a statistically significant predictor of 3-month mortality; SOFA score on post-liver transplant day 7 had the best discriminative power for predicting 3-month mortality.