10. Pdv Solis Herruzo

Since the modifications introduced by Menghini in 1957 (1,2) regarding the technique for liver biopsy collection, the use of this procedure has become widespread, which has allowed a better understanding of the pathology and course of liver diseases, and the basing of our diagnoses on objective, pathologic grounds. Presently, liver biopsy is considered a procedure indicated: a) to establish the origin of abnormal liver tests; b) to assess activity extent and stage in chronic hepatitis; c) to evaluate alcoholrelated liver disease; d) to study fever of unknown origin; e) in the diagnosis of multisystemic infiltrating and granulomatous diseases; f) to evaluate cholestatic conditions; g) in the diagnosis of neoplasm; h) in the assessment of drug-induced liver lesions; i) in the diagnosis of hereditary metabolic disease; j) in the assessment of response to therapy; k) in the evaluation of the liver following a liver transplantation; and l) to evaluate obscure jaundice, acute hepatitis, and hepatomegaly (3). The etiologic identification of liver conditions using noninvasive methods has advanced a lot in recent years, and markers to reveal the extent and stage of liver disease have been sought. Such achievements have begun to question the absolute need for liver biopsy before treatment onset. Indeed, most liver conditions are defined by histologic lesions; however, considering that the material obtained with this technique usually represents as little as 1/100,000 to 1/30,000 of the whole organ, representativity is arguable. Classic studies demonstrated that the information obtained from such biopsies was not reliable in the diagnosis of liver cirrhosis. For instance, Maharaj et al. (4) failed to identify this disease in 50% of patients. Abdi et al. (5) were more successful and recognized cirrhosis in 80% of cases. Poniachik et al. (6) histologically identified cirrhosis in only 68% of laparoscopic cirrhoses, and detected this disease in only 0.8% of cases overlooked by laparoscopy. The uncertainty of histologic diagnosis is also apparent when results obtained from two biopsies in one liver, one each from the right and left lobules, are compared, with 25-33% of cases showing relevant differences (7,8). A great part of diagnostic limitations regarding liver biopsy result from the small size of samples obtained by percutaneous puncture. Several studies demonstrating that diagnostic reliability depends on sample size have been reported in recent years. In a study including 161 liver biopsies, Colloredo et al. (9) found that the shorter or thinner the sample fragment, the more benign were inflammation extent and fibrosis stage interpreted to be. In the opinion of these authors, a correct assessment of histologic lesions required 11 portal spaces at least, and this was achieved only when biopsy samples were 2-cm long and 1.5-mm wide. Conclusions reached in the study by Badossa et al. (10) raised sample length to 2.5 cm, and Hohlund et al. (11) reduced it to 1.5 cm when using a 18g needle. Difficulties derived from assessment subjectivity add to those mentioned on representativity. Various studies have shown great variability in lesion interpretation among pathologists (interobserver variability) and even in one single pathologist when assessing one sample at two different times (intraobserver variability). In a recent study, Petz et al. (12) showed that interobserver variabiliCurrent indications of liver biopsy