Multicenter comparison of the interobserver agreement of standard EUS scoring and Rosemont classification scoring for diagnosis of chronic pancreatitis

Abstract

Background: EUS has less than optimal interobserver agreement for the diagnosis of chronic pancreatitis. The newly developed Rosemont consensus scoring system includes weighted criteria and stricter definitions for individual features.

Objective: The primary aim was to compare the interobserver agreement of standard and Rosemont scoring.

Setting: Multiple tertiary-care institutions.

Intervention: Fifty EUS videos were interpreted by 14 experts. Each expert interpreted the videos on two occasions: First, the videos were read by using standard scoring (9 criteria). Second, after viewing a presentation of the Rosemont classification, the same experts re-read the videos by using Rosemont scoring.

Main outcome measurements: Fleiss' kappa (K) statistics are reported with 95% confidence intervals (CI).

Results: The interobserver agreement was "substantial" (K = 0.65 [95% CI, 0.52-0.77]) for Rosemont scoring and "moderate" (K = 0.54 [95% CI, 0.44-0.66]) for standard scoring; however, the difference was not statistically significant (P = 0.12).

Limitations: The sample size does not allow detection of differences in K of <0.25.

Conclusion: Use of the Rosemont classification did not significantly increase interobserver agreement for EUS diagnosis of chronic pancreatitis compared with standard scoring.