Background: The reduction of liver transplant wait list mortality remains a priority for transplant programs and depends on the accurate stratification of patients by mortality risk. Although estimation of 90-day mortality by Model for End-Stage Liver Disease (MELD) score has improved wait list survival, it is unclear how contemporary wait list mortality can best be diminished given the preponderance of listed patients with low MELD scores and long wait times.
Methods: In this intention-to-treat analysis of 289 consecutively listed patients with over 5 years of follow-up, we aimed to determine the contribution of late mortality to overall wait list outcome and identify clinical predictors that would help discriminate long-term survivors from fatalities.
Results: Seventy percent of wait list deaths occurred in patients listed with MELD scores less than 20, and 40% of deaths occurred in patients waiting longer than 1 year. Hypoalbuminemia at listing was a significant predictor of late mortality in all patients in both univariate and multivariate analyses, and it was most discriminatory among patients with MELD scores of 20 or less.
Conclusion: Our data suggest that hypoalbuminemia at listing reveals a vulnerable population of low MELD patients who are underserved by their MELD score over time. Such patients comprise almost 40% of the contemporary wait list and contribute substantially to list mortality given their poor access to transplantation. Targeting these at-risk patients with grafts from living or extended criteria donors may thus significantly diminish overall list mortality, and future initiatives to decrease overall wait list mortality must focus on improved risk stratification for low MELD patients.
Seventy percent of liver transplant waiting list deaths occur in patients with a low MELD. Low serum albumin at listing in those with a low MELD score predicts late waitlist mortality.
1 Department of Surgery, Division of Transplantation, The University of Pennsylvania Philadelphia, PA.
Received 9 December 2013. Revision requested 3 January 2014.
Accepted 8 May 2014.
The authors declare no conflicts of interest.
E.B. received funding from Harrison Department of Surgical Research.
P.M.P. authored the article, performed the data analysis, and directed the project planning. E.B. participated in data acquisition, in the intellectual development of the project, preliminary data analysis, and contributed to article preparation. M.t.H. participated in data acquisition and in the intellectual development of the project.
Correspondence: Paige Porrett, M.D., Ph.D., 3400 Spruce Street, One Founders Pavilion, Philadelphia, PA 19104. (email@example.com).
Accepted July 22, 2014