The value of three-dimensional ultrasound in identifying Mullerian anomalies at risk of adverse pregnancy outcomes

Abstract

Objective To examine the reliability of three-dimensional ultrasound (3 D-US) in the differentiation between subseptate and arcuate uteri, due to the different associated pregnancy outcomes; refine the existing 3 D-US parameters and evaluate the concordance between 3 D-US and MRI in diagnosing these anomalies. Study design This was a prospective cohort study of 455 women suspected of having a Mullerian anomaly. The diagnosis of subseptate, bicornuate or arcuate uterus was made by 3 D-US in 55 women. Two independent examiners manipulated the 3 D-US volume datasets and recorded the internal intercornual distance, indentation length, indentation tip angle, and myometrial wall thickness in the coronal plane of the uterus. Subsequently, 48 women underwent MRI which was used as the reference test for diagnosis. We calculated the degree of correlation between the two ultrasound assessors’ 3 D-US measurements using interclass correlation coefficient and as well as a Bland-Altman plot. The mean values of the four parameters were used to create receiver operating characteristic curves for determining the best cutoff values for differentiation between subseptate and arcuate uterui. We used the Cohen’s Kappa test to measure the level of agreement between 3 D-US and MRI. Results There was good interobserver agreement between the two 3 D-US assessors for all four parameters. There was a substantial level of agreement between 3 D-US and MRI in differentiating between bicornuate, subseptate and arcuate uteri with a kappa value of 0.727 (95% CI 0.443–0.856). Distinction between subseptate and arcuate uterus was improved when using an indentation length ≥12.5 mm (AUC 0.99) and indentation tip angle ≤89.25 degrees (AUC 0.97) as cutoffs for diagnosis but not the internal intercornual distance or myometrial wall thickness. Conclusion 3 D-US evaluation of the coronal view of the uterus can be relied upon to make a noninvasive, accurate differentiation between subseptate and arcuate uteri. The fundal indentation length and indentation tip angle cut offs of ≥12.5 mm and ≤88 mm, respectively were found to be most accurate for distinction. Thus, allowing for individualizing pre-pregnancy management plans and patient-informed healthcare choices. Highlights There are no agreed upon criteria for differentiating arcuate from subseptate uteri. Such differentiation is critical for counseling and management due to the substantial difference in pregnancy outcome. We aimed to propose cut off values for ultrasound measurements standardized against MRI diagnostic criteria for accurate differentiation between arcuate and subseptate uteri. We demonstrated substantial agreement between 3D-US and MRI in differentiating between bicornuate, subseptate and arcuate uteri. 3D-US evaluation of the coronal view of the uterus is reliable to make an accurate differentiation between subseptate and arcuate uteri. Using the indentation length ≥12.5 mm and indentation tip angle ≤89.25 degrees as parameters to be measured on the coronal view by 3D-US increases its diagnostic accuracy for distinction between arcuate and subseptate uteri.