Observer variability in the histopathologic diagnosis of microscopic colitis and subgroups

Area of Science:

• Gastroenterology and Pathology
• Diagnostic Histopathology
• Colitis Research

Background:

• Microscopic colitis (MC) diagnosis relies on histology, encompassing collagenous colitis (CC) and lymphocytic colitis (LC).
• Incomplete MC (MCi) describes cases with chronic diarrhea, normal endoscopy, and non-definitive histologic findings.
• Accurate classification of MC subtypes is crucial for patient management and understanding disease spectrum.

Purpose of the Study:

• To evaluate intraobserver and interobserver agreement for diagnosing MC subtypes (CC, LC, MCi).
• To assess the ability to differentiate MCi from normal histology and inflammatory bowel disease.
• To identify areas for improvement in MC subtype diagnostic criteria.

Main Methods:

• Three pathologists independently reviewed blinded H&E stained biopsy specimens from five groups: CC, LC, MCi, inflammatory bowel disease, and normal.
• Specimens were re-evaluated after 4 months to assess intraobserver agreement.
• Agreement was quantified using kappa (κ) statistics.

Main Results:

• High intraobserver agreement (κ 0.70–0.83) and interobserver agreement (κ 0.60–0.75) were observed across five diagnostic groups.
• Agreement significantly improved when simplifying to three groups (MC/MCi vs. non-MC), with intraobserver κ 0.88–0.96 and interobserver κ 0.81–0.89.
• Discrimination between MCi and CC/LC showed lower agreement, suggesting diagnostic ambiguity.

Conclusions:

• High diagnostic agreement exists for differentiating microscopic colitis (MC) and incomplete MC (MCi) from non-MC conditions.
• The ability to distinguish between specific MC subtypes (MCi, CC, LC) is less consistent.
• A consensus and revision of histologic criteria for MC subtypes are recommended to improve diagnostic accuracy.