Can reference images improve interobserver agreement in reporting liver fibrosis?

Area of Science:

• Histopathology
• Liver Disease
• Medical Diagnostics

Background:

• Staging liver fibrosis in medical biopsies is subject to significant interobserver variability.
• Existing disease-specific scoring systems are valuable but may not fully address concordance issues.

Purpose of the Study:

• To evaluate the effectiveness of a generic, four-tiered fibrosis staging system in improving histopathologist concordance.
• To assess whether standardized reference images enhance agreement in liver fibrosis staging.

Main Methods:

• UK-based specialist and general histopathologists assessed virtual liver biopsy slides using a proposed four-tiered system.
• Reference images were created and distributed to the same pathologists for reassessment of a second set of slides.
• Agreement was measured using percentage agreement and Fleiss' Kappa statistic.

Main Results:

• The use of reference images eliminated category spread, reducing it from 15% to 0%.
• Overall percentage agreement improved from 67.79% to 70.08%.
• Interobserver agreement, measured by Fleiss' Kappa, increased from 0.55 to 0.59.

Conclusions:

• A standardized, generic fibrosis staging system, particularly when supplemented with reference images, can improve diagnostic concordance among histopathologists.
• This approach offers a practical method to reduce variability in liver fibrosis staging, enhancing diagnostic reliability.