July 18, 2005
Predicting the Outcome of Drug-induced Liver Disease
Drug-induced liver disease (DILD), a potential complication seen with some medications, is usually not life-threatening, but may occasionally be more severe with a high mortality, requiring a liver transplant in selected cases. During the last decade, drug-induced liver injury has led to the withdrawal of a number of drugs from the market. Hy's rule, an observation by the late Dr. Hyman Zimmerman, states that the combination of high liver cell damage as measured by liver enzymes and jaundice induced by a drug is associated with a fatality rate of 10-50 percent. This rule has been advocated by the FDA for assessing the liver toxicity of newly developed drugs, but it has never been scientifically validated.
A study by Einar Björnsson, M.D., Ph.D. and Rolf Olsson, M.D., Ph.D. of the department of internal medicine at Sahlgrenska University Hospital in Gothenburg, Sweden and published in the August 2005 issue of Hepatology, analyzed reports of suspected drug-induced liver injury received by the Swedish Adverse Drug Reactions Advisory Committee (SADRAC) in order to evaluate the validity of Hy's rule and determine what factors might predict the outcome of different forms of DILD.Hepatology, the official journal of the American Association for the Study of Liver Diseases (AASLD), published by John Wiley & Sons, Inc. is available online via Wiley InterScience at http://www.interscience.wiley.com/journal/hepatology.
A total of 784 reports were included in the analysis, each with bilirubin greater than twice upper normal limits and ALT (a liver enzyme that when elevated indicates liver damage) greater than three times upper normal limits. Of these, 409 cases had hepatocellular or HC liver injury (damage to liver cells); while 206 had cholestatic (bile flow) injury and 169 had mixed liver injury. In 633 cases, one drug was suspected of causing the liver injury, while in 151 cases more than one drug may have been responsible. A total of 72 patients died from liver failure or underwent liver transplants. Patients with HC liver injury had a higher mortality than the other two forms of DILD. Those who died or had transplants were older than those who recovered and had higher bilirubin and AST (a liver enzyme also used as an indicator of liver damage) and ALT levels. In the cholestatic/mixed group, bilirubin and AST/ALT ratios were significantly higher in those who died or underwent transplants.
An analysis showed that AST and bilirubin were found to independently predict death or transplant in the general study population and within the HC group (as well as age in the HC group), while within the cholestatic and mixed injury groups, bilirubin was an independent predictor. "Our analysis is unique because it is performed in a large cohort of patients with severe DILD, giving the opportunity to elucidate the most important factors for outcome," the authors state. They note that adverse drug reactions are significantly underreported, which is a limitation of the present study. The study supports Hy's rule in that HC injury had a mortality rate of approximately 10 percent, but the rule seemed to be highly variable depending on the drug associated with the damage.
"We conclude that in accordance with Hy's rule, the HC jaundice has a high but variable mortality rate, depending on the drug involved," the authors state. "AST and bilirubin levels are the most important predictors of death or liver transplantation."
On The Web: