Display Settings:

Format

Send to:

Choose Destination
See comment in PubMed Commons below
Pathol Res Pract. 2005;201(8-9):565-72.

Cholangitis: a histologic classification based on patterns of injury in liver biopsies.

Author information

  • 1Department of Pathology, University of Massachusetts Medical School, Worcester, MA 01655, USA.

Abstract

Inflammatory disorders of the biliary tract present difficult diagnostic problems in liver needle biopsies. The aim of this study was to perform a detailed histologic analysis of liver biopsies from patients with biliary tract disorders, classify them by pattern of inflammation, and determine the accuracy of the histologic classification by clinical follow-up. Percutaneous liver needle biopsies from the surgical pathology files of UmassMemorial Healthcare (UMMHC) from 2000 to 2003 with a diagnosis suggesting a biliary tract process (n = 32) and four biopsies from cases with systemic non-biliary tract disorders were analyzed for multiple histologic features and classified as one of five patterns: acute cholangitis/pericholangitis (ACP), lymphocytic cholangitis (LC), granulomatous (G), ductopenia (D), or non-specific (NS). When compared to the "gold standard" diagnosis based on all clinical data, the concordance between the histologic classification and the clinical diagnosis was: 50% for ACP and bile duct obstruction; 77% for LC and immune-mediated cholangitis NOS; 100% for G and G cholangitis; 100% for D and idiopathic adulthood D; and 50% for NS and non-biliary tract disorders. Our findings suggest that classifying biopsies by pattern of injury is helpful in guiding the subsequent clinical work-up. ACP pattern correlates with bile duct obstruction, infection, and ischemia. LC correlates with serologic studies supporting immune-mediated processes. G pattern suggests further work-up for PBC, drug, tuberculosis, or sarcoidosis. D pattern establishes the clinical diagnosis. NS pattern includes cases of primary sclerosing cholangitis, which cannot be diagnosed by biopsy alone.

PMID:
16259109
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk