Model for predicting survival in liver patients has many applications but could be refined further
A review of the studies on the Model for End-Stage Liver Disease (MELD) found that it is an accurate predictor of survival of patients with a variety of liver diseases, is particularly useful in allocating organs for liver transplants, and can also be used to help determine the course of treatment in certain cases. However, it is possible to improve the accuracy of the model and efforts at refining will continue.
This review appears in the March 2007 issue of Hepatology, the official journal of the American Association for the Study of Liver Diseases (AASLD). Published by John Wiley & Sons, Inc., Hepatology is available online via Wiley InterScience.
Model for End-Stage Liver Disease uses three widely available laboratory tests in a mathematical formula to arrive at a score that predicts survival for patients with liver disease. It was initially created by researchers at the Mayo Clinic in Rochester, MN to predict survival following TIPS, a procedure to treat complications of liver disease that involves the placement of a shunt, but its use has been expanded. Most notably, for the past five years, MELD has become the standard for prioritizing patients awaiting liver transplants, replacing the previous system where patients who were on the list for the longest period of time received transplants first, regardless of how ill they were.
In the current review, Patrick S. Kamath, M.D. and W. Ray Kim, M.D. of the Mayo Clinic who were instrumental in creating and validating the model, reviewed how MELD is applied and assessed its strengths and limitations. They found that using MELD for organ allocation in liver transplants led to an immediate reduction in the number of patients awaiting a liver transplant (12 percent decrease in 2002) and also led to a reduction in mortality on the waiting list of almost 15 percent. However, MELD has not been shown to be useful in predicting mortality following a liver transplant, probably because other factors besides liver dysfunction play a role in transplant success. In addition, they found that use of healthcare resources has not increased since the implementation of MELD because the sickest patients, who were previously a drain on the healthcare system, are being transplanted earlier.
Model for End-Stage Liver Disease, along with the traditional Child Turcotte Pugh (CTP) scoring system, predicts long-term survival in patients with late-stage cirrhosis, but MELD has other applications as well. In single center studies, it accurately predicts survival in patients with variceal (esophageal) bleeding and hepatitis B, and predicts mortality in patients with alcoholic hepatitis and infections leading to kidney failure. In addition, MELD can shed light on outcomes for liver cancer treatments and different types of surgery other than liver transplants.
Despite its accuracy, MELD has some limitations. For patients awaiting liver transplants, it should be used only after reversible complications, such as bacterial infections, have been treated. Also, the values used to determine the MELD score may be variable depending on how they are measured. In order to refine MELD, the authors conducted a study on how changing MELD scores affect mortality. They found that the current MELD score was the most important predictor of survivor, regardless of how it was reached.
"In conclusion, based on its ability to rank patients with cirrhosis according to their short term mortality, MELD has been recognized as a major contributor to the daily practice of hepatology," the authors state. "Successful implementation of MELD-based liver allocation in the U.S. has been followed by widespread adoption of the system globally, attesting to its validity." It is also a useful tool in a wide spectrum of disease severity and variety. However, since it is by no means a perfect system, the authors conclude, efforts to refine it must continue.