Context: Models that discriminate among the sickest cohorts in liver transplantation are important to optimize outcomes. We introduced the Survival Outcomes Following Liver Transplantation (SOFT) score in September 2008 (1); designed to predict survival in liver recipients at 3 months post-transplant with a C statistic of 0.70 (confidence interval [CI], 0.69 to 0.71).

Objective: We applied the SOFT score to cohorts of the sickest transplant candidates and the poorest quality allografts.

Methods: The United Network for Organ Sharing provided de-identified patient-level data. We analyzed cohorts of the sickest candidates and poorest quality donor allografts from August 1, 2006 to October 1, 2010 defined as candidates admitted to the ICU at the time of transplant (n=2670); recipient having a laboratory MELD > 40 (n=1500); recipient age > 70 (n=545); cold ischemia time > 14 hours (n=1779); donor age > 70 (n=1010); Donor Risk Index (DRI) > 2.0 (n=4241); and macrosteatosis > 30% on donor biopsy (n=536). Model discrimination was further assessed using the area under the receiver operating characteristic (ROC) curve.

Results: When applied to these high-risk cohorts, the SOFT score projected survival with a C-statistic of 0.65 (CI .62-.67) for candidates admitted to the ICU at the time of transplant; 0.67 (CI .63-.71) for recipient having a laboratory MELD > 40; 0.70 (.60-.80) for donor age > 70; 0.67 (62-.71) for cold ischemia time > 14 hours; 0.70 (.63-.77) for donor age > 70; 0.68 (.65-.72) for Donor Risk Index (DRI) > 2.0; and 0.70 (.62-.79) for macrosteatosis > 30% on donor biopsy.

Conclusion: The SOFT score was informative amongst cohorts of the sickest transplant candidates and the poorest quality allografts.

« Back to 2013 American Transplant Congress