Two- and three-dimensional ultrasonography and sonohysterography versus hysteroscopy with laparoscopy in the differential diagnosis of septate, bicornuate, and arcuate uteri

Study objective: To estimate the diagnostic accuracy and to compare the diagnostic value of 3-dimensional sonohysterography (3D-SIS), 3-dimensional transvaginal ultrasonography (3D-TVS), 2-dimensional sonohysterography (2D-SIS), and 2-dimensional transvaginal ultrasonography (2D-TVS) (initial and expert diagnosis) in the differential diagnosis of septate, bicornuate, and arcuate uteri.

Design: Prospective clinical study (Canadian Task Force II-2).

Setting: University hospital and private hospital and clinic.

Patients: A total of 117 women with a history of recurrent abortions or infertility and a 2D-TVS initial diagnosis of a septate, bicornuate, or arcuate uterus.

Interventions: Expert 2D-TVS, 3D-TVS, 2D-SIS, and 3D-SIS performed by experienced examiners and hysteroscopy with laparoscopy to establish the final diagnosis.

Measurements and main results: Hysteroscopy performed in conjunction with laparoscopy (HL) detected 23 arcuate, 60 septate, 22 bicornuate, and 12 normal uteri. 3D-SIS showed perfect diagnostic accuracy (100.0%) in general detection of uterine abnormalities, compared with initial 2D-TVS (77.8%), expert 2D-TVS (90.6%), 2D-SIS (94.0%), and 3D-TVS (97.4%). In the overall diagnosis of uterine anomalies, all of the diagnostic methods had statistically significantly better diagnostic value than initial 2D-TVS (p < .001), whereas 3D-SIS was the only method that was better than expert 2D-TVS (p < .001).

Conclusions: Although 3D-SIS was identical to HL, with the highest accuracy, there was no significant difference in diagnostic value between 3D-TVS with 2D-SIS and 3D-SIS or between expert 2D-TVS and 3D-TVS with 2D-SIS. The high diagnostic value of these ultrasonographic tools questions the need for endoscopy in the differential diagnosis of the most common congenital uterine anomalies.