Improved Statistical Tool To Rank Sickest Patients Waiting For Liver Transplants

Researchers found that including serum sodium concentration in the statistical model now used could reduce by 7 percent the number of patients (as many as 50 people) who die each year while waiting for a liver transplant. Serum sodium levels can be measured with a blood test.

The study will be published in the Sept. 4, 2008, issue of the New England Journal of Medicine.

"The subset of patients with the lowest sodium blood levels is not served well by the current system," says W. Ray Kim, M.D., liver transplant physician at Mayo Clinic and lead researcher on the study. Low serum sodium occurs in patients with advanced liver disease, who frequently have a large amount of ascites (fluid in abdomen), a serious complication of liver cirrhosis.

In the United States, the United Network for Organ Sharing (UNOS), a nonprofit, scientific and educational organization, administers the distribution of organs from deceased donors to patients waiting for transplants.

In 2001, Mayo Clinic researchers developed the formula used to rank patients waiting for livers from deceased donors. Called the Model for End-Stage Liver Disease (MELD), it was adopted by UNOS in 2002. The highest MELD scores indicate the sickest patients, who are placed at the top of the transplant waiting list.

The MELD score is based on three standard laboratory tests:

• Total serum bilirubin concentration, indicated by jaundice
• International normalized ratio for the prothrombin time, which measures the blood's ability to clot
• Serum creatinine concentration, an indicator of renal (kidney) function

"The addition of sodium to MELD would be a relatively modest improvement," says Dr. Kim. "Since low serum sodium does not affect the majority of patients, we didn't pick it up in our original work. However, we have discovered that in those affected, low sodium has a substantial impact on mortality. This impact is particularly large in patients with low MELD, who would be placed low on the waiting list under the current system."

How the study was done. In this observational study, researchers analyzed patient data collected upon UNOS registration for liver transplantation. In 2005, there were 6,769 patients placed on the list. Of those, 422 died and 1,781 underwent liver transplant within 90 days. Based on these data, researchers created a new statistical model that incorporates serum sodium and MELD.

The new model was verified by using waiting-patient data from 2006, when 477 patients died within three months of placement on the waiting list. "By factoring in sodium, researchers predicted that about 7 percent of deaths of waiting patients would have had considerably higher priority for transplant, and it may have prevented their deaths," says Dr. Kim.

More than 400 patients receive liver transplants at Mayo Clinic's sites in Minnesota, Florida and Arizona each year. Mayo Clinic is the most experienced liver transplant center in the nation, with some of the highest survival rates in the world.

Other Mayo Clinic researchers involved in this study include Walter Kremers, Ph.D.; Russell Wiesner, M.D.; Patrick Kamath, M.D.; Joanne Benson; and Terry Therneau, Ph.D.