Clinical

Scoring system predicts fibrosis in liver disease patients

Last Updated: 2007-04-11 10:10:21 -0400 (Reuters Health)

NEW YORK (Reuters Health) - A scoring system that incorporates six variables can accurately identify patients with nonalcoholic fatty liver disease (NAFLD) who also have advanced fibrosis, new research shows. Use of this system could greatly reduce the number of patients requiring liver biopsy, the authors believe.

NAFLD patients with advanced fibrosis are at increased risk for progressing to end-stage liver disease, which can have treatment implications, lead author Dr. Paul Angulo, from the Mayo Clinic in Rochester, Minnesota, and colleagues note. However, at present, liver biopsy is the only reliable method of identifying such patients.

The goal of the present study was to come up with a noninvasive method of predicting advanced fibrosis with NAFLD. To do this, the researchers created a scoring system by examining the predictive ability of several routinely measured clinical and laboratory values in 733 patients with biopsy-confirmed NAFLD. The system was created using data from 480 of the patients and then was validated in the remaining 253.

The new findings appear in the April issue of Hepatology.

The researchers identified six independent predictors of advanced liver fibrosis: age, hyperglycemia, BMI, platelet count, albumin, and AST/ALT ratio. The scoring system derived from these variables had an area under the receiver operating characteristic curve of 0.88 and 0.82 in the system-generation and -validation cohorts, respectively.

Depending on the cutoff score employed, positive and negative predictive values of up to 90% and 93%, respectively, were achieved. Use of the scoring system could have prevented liver biopsy in 75% of all patients with accurate prediction in 90%, the authors state.

Further studies are needed to determine if the predictive value of the scoring system could be improved by including serum markers of fibrosis and imaging results, the researchers conclude.

Hepatology 2007;45:846-854.