A numerical scoring system was applied and compared with conventional histological classification to assess the histological outcome in 42 patients with chronic hepatitis B followed for 16 to 162 months (mean = 75 months). Four histological categories in the biopsies were assessed and scored: (i) piecemeal necrosis; (ii) lobular necrosis; (iii) portal inflammation, and (iv) fibrosis and cirrhosis. The sum of all four categories was defined as the "Histological Activity Index." Altogether, 102 liver specimens, including 2 to 4 repeats from each patient, were investigated. A good correlation was noted between a high value of the Histological Activity Index score and several liver histology as monitored by conventional terminology for chronic hepatitis. Among patients with HBeAg persistence, 8 of 14 (57%) deteriorated during follow-up as judged by an increase in the Histological Activity Index score compared to 3 of 13 (23%) of the patients with HBeAg seroconversion (0.5 < p < 0.1). Piecemeal necrosis has been postulated to be a predictive marker for the eventual development of cirrhosis. Here, we found that a low score for piecemeal necrosis in the initial liver biopsy was significantly less predictive of a high fibrosis score in the follow-up biopsy than was a high score for initial piecemeal necrosis (p < 0.001).
It is concluded that the scoring system used can be applied to monitor the histological long-term follow-up, especially when separated into its four constituent categories. It also offers a means of predicting a chronic hepatitis outcome.
Disease activity and thus histological changes may fluctuate during the course of chronic hepatitis B infection. The exacerbations may be due to clearance of HBeAg, to superinfection with hepatitis A, non-A, non-B or hepatitis D virus (HDV) (delta agent), or to changes in immune reactivity (spontaneous or due to immunosuppressive treatment) in the host (1-4). Among the hepatotropic viruses, HDV seems to be the most impor-