1 Department of Surgery, University of Alabama at Birmingham, Birmingham, AL. Electronic address: Rmcannon@uabmc.edu.
2 Hiram C Polk Jr MD Department of Surgery, University of Louisville, Louisville, KY.
3 Department of Medicine, University of Miami, Miami, FL.
4 Department of Surgery, Emory University, Atlanta, GA.
5 Department of Surgery, University of Kentucky, Lexington, KY.
6 Department of Surgery, University of Alabama at Birmingham, Birmingham, AL.
BACKGROUND: Patients thought to be at greater risk of liver waitlist dropout than their laboratory MELD (lMELD) score reflects are commonly given MELD exceptions, where a higher allocation MELD (aMELD) score is assigned that is thought to reflect the patient's risk. This study was undertaken to determine whether exceptions for reasons other than hepatocellular carcinoma (HCC) are justified, and whether exception aMELD scores appropriately estimate risk.
METHODS: Adult primary liver transplant candidates listed in the current era of liver allocation in the UNOS database were analyzed. Patients granted non-HCC related MELD exceptions and those without MELD exceptions were compared. Rates of waitlist dropout and liver transplantation were analyzed using cause-specific hazards regression, with separate models fitted to adjust for lMELD and aMELD.
RESULTS: There were 29,243 patients, with 2,555 in the exception group. Nationally, exception patients were more likely to dropout (HR: 1.451.601.76; p<0.001) or undergo liver transplant (HR: 3.323.493.67; p<0.001) than their lMELD adjusted counterparts. Adjusting for aMELD, exception patients were less likely to dropout (HR: 0.700.770.85; p<0.001) and less likely to undergo liver transplant (HR: 0.720.760.80 ; p<0.001). Exception patients were not at significantly increased risk of waitlist dropout when adjusted for lMELD in 4 of 11 UNOS regions CONCLUSIONS: Despite appropriate utilization of non-HCC MELD exceptions on a national level, patients with non-HCC MELD exceptions were awarded inappropriately high priority for transplantation in many regions. This highlights the need to consider local conditions faced by transplant candidates when estimating waitlist mortality and determining priority for transplantation.