Clinical and biochemical expression of the histopathological lesions of primary biliary cirrhosis

Background/Aims: This study aimed to assess the relationships which may link elementary histological lesions with symptoms and biochemistries in primary biliary cirrhosis. Methods: We studied 103 patients with primary biliary cirrhosis who participated in a double-blind, placebo-controlled trial of UDCA treatment and in whom liver biopsy specimens obtained at entry were reassessed. Relationships between histological features, fatigue, pruritus and biochemistries were calculated by using exact tests for 2 ordinal variables. Results: The degrees of severity of fatigue and pruritus were significantly and exclusively related to the presence of florid interlobular bile duct lesions (p<0.01 and p<0.02, respectively). The only laboratory parameter associated with the presence of interlobular bile duct florid lesion was IgM level. The most discriminant biochemical test for interlobular bile duct paucity was gamma glutamyltranspeptidase activity. The degree of severity of both lymphocytic hepatocellular piecemeal necrosis and lobular inflammation and necrosis was mainly associated with increased gammaglobulin and IgG levels and to a lesser extent with increased IgM and aspartate aminotransferase levels. The extent of fibrosis was mainly associated with gammaglobulin levels and to a lesser degree with serum albumin, bilirubin and IgG levels. Conclusions: Symptoms and biochemistries classically used to assess primary biliary cirrhosis reflect in part the degree of severity of the main elementary histological lesions. We propose that the picture of primary biliary cirrhosis results from the clinical and biochemical expression of three distinct processes, e.g., bile duct inflammation and destruction, parenchymal inflammation and necrosis, and fibrosis. The various combinations of these processes may explain why the spectrum of primary biliary cirrhosis varies from typical primary biliary cirrhosis to mixed type of primary biliary cirrhosis and autoimmune hepatitis and suggests that the response to therapies may depend on the predominance of each process in a given patient.