OBJECTIVES: To investigate the accuracy and reliability of 3-dimensional (3D) transvaginal sonography in classifying submucous fibroids using the International Federation of Gynecology and Obstetrics PALM-COEIN (polyp, adenomyosis, leiomyoma, malignancy and hyperplasia, coagulopathy, ovulatory dysfunction, endometrial, iatrogenic, and not yet classified) classification and protrusion (percent) compared to 2-dimensional (2D) transvaginal sonography, 2D saline infusion sonography, and 3D saline infusion sonography, using hysteroscopy as a reference test.
METHODS: A prospective cohort pilot study was performed among 14 consecutive patients undergoing hysteroscopic surgery, preceded by routine sonography (2D transvaginal sonography, 2D saline infusion sonography, 3D transvaginal sonography, and 3D saline infusion sonography).
RESULTS: The intraclass correlation coefficient (ICC) for 2D transvaginal sonography versus hysteroscopy was 0.69 (95% confidence interval [CI], 0.06, 0.90) compared to 0.94 (95% CI, 0.83, 0.98) for 2D saline infusion sonography. The ICCs for 3D transvaginal sonography versus hysteroscopy were 0.69 (95% CI, 0.03, 0.90 [investigator A]) and 0.55 (95% CI, -0.48, 0.86 [investigator B]). The ICCs for 3D saline infusion sonography versus hysteroscopy were 0.94 (95% CI, 0.81, 0.98 [investigator A]) and 0.87 (95% CI, 0.60, 0.96 [investigator B]). Interobserver agreement of 3D transvaginal sonography was 0.81 (95% CI, 0.43, 0.94) compared to 0.86 (95% CI, 0.56, 0.96) for 3D saline infusion sonography.
CONCLUSIONS: In these preliminary data, 3D transvaginal sonography was not as accurate as 2D or 3D saline infusion sonography and was not more accurate than 2D transvaginal sonography. There was moderate interobserver agreement for 3D transvaginal sonography. There might be room for improvement, as 3D transvaginal sonography is more accurate when endometrial thickness increases. Further study is warranted to evaluate in which patients saline infusion sonography eventually can be obviated.