Diagnosing autoimmune hepatitis in nonalcoholic fatty liver disease: is the International Autoimmune Hepatitis Group scoring system useful?

BACKGROUND: There are no surrogate serum markers for autoimmune hepatitis (AIH) and nonalcoholic fatty liver disease (NAFLD). An AIH scoring system was reported by the International Autoimmune Hepatitis Group; however, the criteria did not focus on making the distinction between AIH and NAFLD. We examined the effectiveness of using the AIH score for diagnosing AIH in NAFLD patients. We also identified the prevalence of autoimmune phenomena, in terms of various auto-antibodies, including antinuclear antibodies (ANA), to determine whether these markers had any clinicopathological significance, and whether they were related to the patients' clinical courses.

METHODS: We studied 212 patients (103 males and 109 females) with biopsy-proven NAFLD. The AIH score of each patient was calculated without including the liver biopsy results. The patients were divided into three groups based on their clinicopathological features: the overlap group (those with clinical and histological features of both NAFLD and AIH), the systemic group (those with systemic autoimmune disease other than AIH), and the "other" group (patients with no autoimmune disease). To evaluate the clinicopathological significance of ANA in NAFLD patients, those without autoimmune diseases (the "others" group) were classified according to their ANA positivity and ANA titer.

RESULTS: Seventy patients (33.0%) were positive for ANA. Among the female patients, 106 patients (97.2%) had an AIH score of 10 or more. Of the 103 male patients, 21 (20.4%) had an AIH score of 10 or more. However, after liver biopsy, only 1 patient (0.5%) could be classified as "definite AIH." In the NAFLD patients without autoimmune diseases ("other" group), multivariate logistic regression analysis found that female sex was an independent predictor of the presence of ANA (P = 0.029). In contrast, multivariate logistic regression analysis found that severe obesity (body mass index [BMI], >or=30 kg/m2) was the only independent predictor of the presence of an ANA titer of 1 : 80 or more (P = 0.026).

CONCLUSIONS: The AIH score without liver biopsy findings was not useful for diagnosing AIH in NAFLD patients. In patients with elevated ANA titers and risk factors for NAFLD, it is very important to perform a liver biopsy to make a definitive diagnosis before treatment.