Who is too healthy and who is too sick for liver transplantation: external validation of prognostic scores and survival-benefit estimation

Objective. Thresholds for when a patient should be considered too healthy or too sick to undergo liver transplantation (LT) have been pursued, but have undergone little external validation and may differ between centers and countries. Material and methods. We investigated the ability of the Model for End-stage Liver Disease (MELD), D-MELD, Donor Risk Index (DRI) and Balance of Risk (BAR) scores to predict 1-year graft survival, and determined the 1-year survival-benefit of LT, compared with conservative management, according to MELD score and graft quality among 538 adult LT recipients with underlying chronic non-malignant liver disease. Results. One-year graft survival rates showed small, but statistically significant variation according to MELD (p = 0.002) and D-MELD score (p = 0.04), and among LTs after year 2000 also according to BAR score (p = 0.01), but not according to DRI. Diagnostic accuracy of these scores was poor; area under the curve was 0.50-0.65 depending on the score. A 1-year survival-benefit of LT emerged at MELD scores >= 15, but also at lower MELD scores when using high-quality grafts (DRI < 1.075). Conclusions. The performance of various prognostic scores in the Finnish setting was poor. Careful clinical evaluation is imperative when deciding on the timing of LT in the course of chronic liver disease.