Reliability of an Electronic Uterine Diagram to Standardize Intrauterine Adhesion Scoring.

Study objective: To study the inter-rater reliability of an electronic complementary uterine diagram (electronic American Fertility Society [eAFS] freely available at https://bit.ly/eAFS-score) in comparison with the conventional American Fertility Society (cAFS) intrauterine adhesion (IUA) scoring system. As a secondary aim, we wanted to assess the intra-rater reliability of these tools.

Design: This was an observational cross-over study with a nested test-retest design.

Setting: Self-identified gynecologists who performed diagnostic hysteroscopy as part of their routine practice were asked to score IUA on 7 preselected videos using both cAFS and eAFS tools in 2 separate rounds. The order of use of tools was determined at random. Gynecologists who agreed to be involved in a third round were asked to use the tool they were allocated to in the first round to assess intra-rater variability.

Patients: No patients were recruited for the purpose of this study.

Measurements and main results: Inter-rater agreement between evaluators when using eAFS and cAFS for IUA scoring. Overall, the inter-rater agreement was fair (0.25; confidence interval [CI], 0.17-0.34) for cAFS and moderate (0.53; CI, 0.48-0.58) for the eAFS tool. The biggest difference between both tools was seen in the "extent of cavity involved" component, which was fair (0.28; CI, 0.21-0.35) vs substantial (0.71; CI, 0.64-0.78) for cAFS vs eAFS, respectively. Moreover, agreement coefficients were comparable among "expert" and "non-expert" evaluators with the use of the eAFS tool. In addition, intra-rater perfect agreement was higher with the use of eAFS than cAFS (38% vs 32.5%).

Conclusion: The current AFS IUA scoring system has fair inter-rater agreement. However, using a complementary electronic uterine diagram increased this agreement from fair to moderate overall and from fair to substantial in the "extent of adhesions" domain of the tool. The use of the electronic diagram also improved inter-rater agreement among non-experts making it comparable with that of experts.