The Decreasing Predictive Power of MELD in an Era of Changing Etiology of Liver Disease

Elizabeth L. Godfrey; Tahir H. Malik; Jennifer C. Lai; Ayse L. Mindikoglu; N. Thao N. Galván; Ronald T. Cotton; Christine A. O'Mahony; John A. Goss; Abbas Rana


American Journal of Transplantation. 2019;19(12):3299-3307. 

In This Article

Abstract and Introduction


The field of liver transplantation has shifted considerably in the MELD era, including changing allocation, immunosuppression, and liver failure etiologies, as well as better supportive therapies. Our aim was to evaluate the predictive accuracy of the MELD score over time. The United Network for Organ Sharing provided de-identified data on 120 156 patients listed for liver transplant from 2002–2016. The ability of the MELD score to predict 90-day mortality was evaluated by a concordance (C-) statistic and corroborated with competing risk analysis. The MELD score's concordance with 90-day mortality has downtrended from 0.80 in 2003 to 0.70 in 2015. While lab MELD scores at listing and transplant climbed in that interval, score at waitlist death remained steady near 35. Listing age increased from 50 to 54 years. HCV-positive status at listing dropped from 33 to 17%. The concordance of MELD and mortality does not differ with age (>60 = 0.73, <60 = 0.74), but is lower in diseases that are increasing most rapidly—alcoholic liver disease and non-alcoholic fatty liver disease—and higher in those that are declining, particularly in HCV-positive patients (HCV positive = 0.77; negative = 0.73). While MELD still predicts mortality, its accuracy has decreased; changing etiology of disease may contribute.


Liver transplantation has changed considerably as a field in the modern era. Part of these changes relate to policy; allocation policies have undergone a number of changes in the modern era. The Model for End-Stage Liver Disease (MELD) score was first introduced in 2002 for prioritization of liver transplant candidates. Since then, a variety of changes have been made to the exception policy for hepatocellular carcinoma, standardized exceptions have been introduced for a variety of other syndromes, and geographic sharing programs have been introduced and changed several times.[1] A particularly significant change has been the incorporation of serum sodium into the MELD score in 2016.[2]

Demographics and clinical advances have also contributed to the shifting landscape of liver transplantation. The US population has and continues to shift dramatically in terms of ethnic, national, and economic makeup, which has led to downstream changes in liver disease etiologies and care considerations. Medical management of liver disease has improved substantially, as well: new immunosuppression and better supportive therapies are enabling sicker and higher-risk patients to be eligible for and survive to receive transplants.[3,4] Crucially, antiviral treatments are reducing viral liver disease, permitting non-alcoholic fatty and alcoholic liver disease, including alcoholic hepatitis in recent years, to emerge as the leading indications for listing.[5–7]

Throughout this period, the MELD score has predicted mortality for a broad variety of liver diseases, particularly at the 90-day timepoint as it was designed, and allocation algorithms utilizing it have demonstrated reduced waitlist mortality and improved patient survival.[8–12] Various modifications, mostly based on the MELD, have been proposed over the years to address the MELD's shortcomings in certain conditions such as acute liver failure, acute-on-chronic liver failure, hepatorenal syndrome, bacterial infection, encephalopathy, and more.[10,13–16] However, the MELD itself and its overall performance over time have not been evaluated in the context of the changing demographics, etiologies, and medical therapies for liver disease.

Our aim was to evaluate the predictive accuracy of the MELD score and its successor, the sodium MELD (MELD-Na), over time for mortality in this landscape of changing population and practices.