Interobserver agreement in prenatal ultrasound assessment of fetal urinary tract dilatation.

BACKGROUND

To assess interobserver agreement in the evaluation of the fetal urinary tract in fetuses with urinary tract dilatation (UTD).

METHODS

We conducted a retrospective study evaluating interobserver agreement on renal tract features using third-trimester renal ultrasound images of fetuses with UTD. Two fetal medicine specialists and two pediatric nephrologists independently and systematically assessed ultrasound features of the fetal renal parenchyma and urinary bladder, as well as the Society for Fetal Urology (SFU) and the ONEN grading systems. Interobserver variability was analyzed using Cohen's kappa statistics and intraclass correlation coefficients (ICC).

RESULTS

A total of 1639 ultrasound images from 144 fetuses with antenatal UTD were analyzed. Slight to moderate agreement was observed for most renal ultrasound parameters, including cortico-medullary differentiation (κ = 0.28-0.50), renal echogenicity (κ = 0.18-0.45), and cortical cysts (κ = 0.02-0.15). A similar pattern was observed for the subjective assessment of bladder features. In contrast, both the SFU and the ONEN grading system demonstrated moderate to substantial agreement, with ICC values ranging from 0.61 to 0.73 and 0.55 to 0.66, respectively, and kappa values from 0.38 to 0.63 and 0.48 to 0.67.

CONCLUSION

Significant interobserver variability exists in the interpretation of commonly used prenatal ultrasound features for evaluating the fetal urinary tract and predicting postnatal outcomes in UTD. The SFU and the ONEN grading systems improved agreement, supporting the need for standardized ultrasound assessment tools to enhance prenatal counseling and antenatal management in fetuses with UTD.