Abstract Book - 26th ESPU Congress, Joint Meeting with ICCS + SPU + AAP/SOU + AAPU + SFU

16:28 - 16:31

S19-5

(PP)

INTERRATER RELIABILITY OF FETAL SONOGRAPHY VS MRI

David CHALMERS

, Mariana MEYERS

, Katie BRODIE

, Claire PALMER

and Jeffrey CAMPBELL

1) Maine Medical Center, Urology, South Portland, USA - 2) Children's Hospital Colorado, Radiology, Auror, USA - 3)

Children's Hospital Colorado, Urology, Aurora, USA - 4) Children's Hospital Colorado, Biostatistics, Epidemiology,

Research and Design, Aurora, USA

PURPOSE

Sonography is commonly used to screen for antenatal fetal anomalies, including obstructive uropathy. Fetal Magnetic

Resonance Imaging (MRI) is an emerging tool to further define complex pathology. We hypothesized that the superior

anatomic detail of fetal MRI would improve the inter-rater reliability of antenatal hydronephrosis grading.

MATERIAL AND METHODS

After IRB approval, we retrospectively reviewed a database of paired fetal sonography and MRI studies from a single

institution. MRI and sonography were performed within 24hrs. 2 pediatric urologists and 1 pediatric radiologist

independently graded all renal units using three grading systems: anterior-posterior diameter (APD), Society for Fetal

Urology (SFU) hydronephrosis grading system, and the Urinary Tract Dilitation (UTD) grading system. Cohen's kappa

statistic was calculated to determine the inter-rater reliability of each scoring method and imaging modality

combination. Intraclass correlation was used to assess the consistency of APD measurements.

RESULTS

A total of 49 patients with 87 renal units were analyzed. The overall kappa values for the SFU grading system indicated

"moderate" inter-rater reliability for both sonography (R=0.55/L=0.43) and MRI (R=0.52/L=0.56). The overall kappa

values for the UTD grading system indicated "substantial" inter-rater reliability for both sonography (R=0.80/L=0.67)

and MRI (R=0.66/L=0.65). The intraclass correlation coefficient of the APD was significantly improved with the use of

MRI (R=0.93/L=0.6) vs sonography (R=0.50/L=0.56)

CONCLUSIONS

Fetal hydronephrosis grading, and the body of evidence supporting it, has been established using sonography. As the

role of fetal MRI is further defined, the interpretation of fetal hydronephrosis must be understood in this new context.

Our results indicate that imaging modality does not significantly affect inter-rater reliability using the SFU or UTD

grading systems. The consistency of APD measurement is improved with the use of fetal MRI.