Geographical Distance From Transplant Center and Impact on Waitlist Outcomes and Healthcare Utilization Prelisting.

Abstract

Decompensated cirrhosis has a median survival of 2 years without liver transplantation (LT). This study investigates whether distance from LT center affects waitlist mortality and receipt of LT. The study population was generated from the transplant database in London, Ontario, Canada. Adult patients on the waitlist for LT between January 1, 2012, and December 31, 2021, were included. Data were linked to the Institute for Clinical Evaluative Sciences to examine clinically relevant outcomes, using ≤150 km vs >150 km to stratify descriptive analysis. Multivariate time-to-event analyses were conducted to evaluate the hazards of increasing distance from LT center on waitlist mortality and receipt of LT. Of the 552 patients meeting study criteria, 394 (71.4%) received LT in an overall predominantly male cohort (n = 390, 70%), with a median age of 59 years (interquartile range [IQR] 52-64) and median distance from the LT center of 110 km (IQR 59-191). There were no significant differences between patients living ≤150 km (n = 362) vs >150 km (n = 190) from the LT center. In liver disease etiology-alcohol-related liver disease remained the most common (32.9% vs 33.2%; P = 0.95) across both categories, with no difference in median Model for End Stage Liver Disease-Sodium scores between those who did and did not receive transplant (17 [IQR 9-25] vs 18 [IQR 10-27]; P = 0.12). On multivariable analysis, distance to the LT center did not affect receipt of LT, waitlist mortality, or postlisting ED visits and hospitalizations. Model for End-Stage Liver Disease-Sodium at listing was a significant predictor of increased waitlist mortality (hazard ratio 1.12; confidence interval 1.09-1.16; P < 0.01), whereas hepatocellular carcinoma diagnosis was associated with reduced waitlist mortality (hazard ratio 0.13; confidence interval 0.04-0.45; P < 0.01). Patients further from the LT center had a higher median number of hospitalizations (2 vs 1; P = 0.02) and emergency department (ED) visits (3 vs 2; P < 0.01) in the year before LT listing, and significantly ED utilization within 90 days postlisting (0 [IQR 0-2] vs 0 [IQR 0-1]; P < 0.05), albeit this was not consistent on multivariable analysis. Geographical distance does not significantly affect LT waitlist mortality or receipt of LT. However, differences in healthcare utilization suggest disparities may still manifest with a negative impact on patients in the pre-LT setting.