Diagnostic Accuracy of Ultrasound for Adenomyosis-A Blinded, Retrospective, Single-Expert Validation Study of the 2022 MUSA Guidelines.

STUDY

OBJECTIVE

To examine the predictive value of the 2022 Morphological Uterus Sonographic Assessment ultrasound markers for the diagnosis of adenomyosis.

DESIGN

Retrospective cohort study.

SETTING

Large academic medical center.

PATIENTS: Patients who underwent a hysterectomy procedure in 2022 and had an ultrasound within the prior year at a specialized gynecologic ultrasound center.

INTERVENTIONS: Hysterectomy and preoperative transvaginal ultrasound.

MEASUREMENTS AND MAIN

RESULTS

The images were reviewed by an obstetrician-gynecologist specialized in gynecologic ultrasound who was blinded to the pathology results. Presence or absence of each Morphological Uterus Sonographic Assessment marker was documented and compared to the pathology results. The sensitivity, specificity, and area under the curve (AUC) were calculated for each individual finding and for composite measures. Our sample included 26 hysterectomies. The histologic prevalence of adenomyosis was 46.2%. The composite direct markers had a sensitivity of 100% (95% CI: 100.0-100.0) and a specificity of 35.7% (95% CI: 10.6-60.8) with an AUC of 65.4 (95% CI: 47.1-83.7). The marker with the highest AUC was echogenic subendometrial lines and buds with a sensitivity of 83.3 (95% CI: 62.2-100.0), and a specificity of 64.3 (95% CI: 39.2-89.4), and AUC of 73.1 (95% CI: 56.0-90.1). The direct finding with the lowest sensitivity but highest specificity was myometrial cysts with a sensitivity of 25% (95% CI: 0.5-49.5) and a specificity of 85.7 (95% CI: 67.4-100.0) and AUC of 57.7 (95% CI: 38.7-76.7).

CONCLUSION

In this blinded validation study of recently published guidelines on the ultrasound findings indicative of adenomyosis, we found a high sensitivity for the composite direct findings and all findings. The finding with the highest AUC was subendometrial lines and buds. Although this is a small study, it is a promising look at the value of ultrasound markers in predicting adenomyosis.