FROM SURGERY
Racial and regional disparities in liver transplant allocation persist despite multiple allocation schemes as identified in a large dataset spanning at least 30 years, according to a study.
The Model for End-Stage Liver Disease (MELD) score was put forth in 1998 by the Organ Procurement and Transplantation Network under the guidance of the Centers for Medicare & Medicaid Services and the Department of Health & Human Services, to indicate that organs should be allocated to the sickest patients as interpreted by the MELD score. Since then, four separate liver allocation systems are being followed across the country owing to disagreements over a universal allocation scheme. These include MELD score, the initial three-tier UNOS (United Network for Organ Sharing) Status, Share 35, and now MELD Na.
Unfavorable supply and demand ratios for liver allograft allocation persist across the country with differential and limited access to care, which leads to decreased wellness, lower life expectancies, and higher baseline morbidity and mortality rates in some areas. Furthermore, a short cold storage capacity of liver allografts limits greater allocation and distribution schema.
Dominique J. Monlezun, MD , PhD, MPH, and his colleagues at the Tulane Transplant Institute at Tulane University, New Orleans, evaluated the effect of MELD and other allocation schemes on the incidence of racial and regional disparities in a study published in Surgery . They performed fixed-effects multivariate logistic regression augmented by modified forward and backward stepwise-regression of transplanted patients from the United Network for Organ Sharing Standard Transplant Analysis and Research database (1985-2016) to assess causal inference of such disparities.
“Significant disparities in the odds of receiving a liver were found: African Americans, odds ratio, 1.12 (95% confidence interval, 1.08-1.17); Asians, 1.12 (95% CI, 1.07-1.18); females, 0.80 (95% CI, 0.78-0.83); and malignancy 1.18 (95% CI, 1.13-1.22). Significant racial disparities by region were identified using Caucasian Region 7 (Ill., Minn., N.D., S.D., and Wisc.) as the reference: Hispanic Region 9 (N.Y., West Vt.) 1.22 (1.02-1.45), Hispanic Region 1 (New England) 1.26 (1.01-1.57), Hispanic Region 4 (Ok., Tex.) 1.23 (1.05-1.43), and Asian Region 4 (Ok., Tex.) 1.35 (1.05-1.73).” Since the transplantation rate in Region 7 closely approximated the sex and race-matched rate of the national post–Share 35 average, it was used as a reference in the study.
“Although traditional disparities as with African Americans and [whites] have been improved during the past 30 years, new disparities as with Hispanics and Asians have developed in certain regions,” stated the authors.
They acknowledged the limitations in the observational nature of the study and those of the statistical analyses, which could only approximate, rather than perfectly replicate, a randomized trial. Big Data tools such as artificial intelligence–based machine learning can provide real-time analysis of large heterogeneous datasets for patients across different regions.
The authors reported no conflicts of interest.
SOURCE: Monlezun DJ et al. Surgery. 2018 doi: 10.1016/j.surg.2017.10.009.
