Trimester-Specific Anterior Posterior Diameter and Percent Change Over Time to Predict Postnatal Surgery in Neonates with Antenatal Hydronephrosis
Kendrick M. Campbell, BA1, Shilin Zhao, PhD2, Gabriella L. Crane, MD3, Mark C. Adams, MD4, Douglas B. Clayton, MD4, John C. Pope, IV, MD4, Stacy T. Tanaka, MD4, Abby S. Taylor, MD4, Chevis N. Shannon, MBA, MPH, DrPH5, John W. Brock, III, MD4, John C. Thomas, MD4.
1Vanderbilt University School of Medicine, Nashville, TN, USA, 2Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, TN, USA, 3Division of Diagnostic Imaging, Vanderbilt University Medical Center, Nashville, TN, USA, 4Division of Pediatric Urology, Vanderbilt University Medical Center, Nashville, TN, USA, 5Division of Pediatric Neurosurgery, Vanderbilt University Medical Center, Nashville, TN, USA.
Background:
Antenatal hydronephrosis (ANH) is common. Literature suggests that antenatal anterior posterior diameter (APD) of the renal pelvis determined on ultrasound (US) may be an early predictor for needing postnatal surgical treatment. We hypothesize that earlier APD values will best predict postnatal surgical intervention and evaluate whether percent change of APD from maximum US to first postnatal US add value to prediction accuracy.
Methods:
We retrospectively reviewed 130 patients evaluated for ANH at our institution from 2010 to 2017. ANH was defined as at least one APD value greater than 4 mm in the 2nd trimester or greater than 7 mm in the 3rd trimester. Surgical intervention was recorded until patients reached 5 years of age. Patients were excluded if there was no antenatal or postnatal US or care at this institution. The maximum APD value was taken in the 2nd and 3rd trimester. 2nd and 3rd trimester APD percent changes (APD-delta) were calculated using the largest APD and the first ipsilateral postnatal APD after 48 hours of life. We utilized the area under the curve (AUC) to determine the ability of APD and APD-delta in the 2nd and 3rd trimesters to predict postnatal surgical intervention. An AUC greater than or equal to 0.80 is an accepted cut-off that represents good predictability of a test. We also identified the lowest threshold APD and APD-deltas that were predictive of postnatal surgery in the 2nd and 3rd trimester.
Results:
The cohort was predominantly male (71%). 12% of patients ultimately had surgery at a mean age of 8.3 ± 8.8 months. The most common procedure was pyeloplasty. There were no statistical differences between 2nd trimester and third 3rd trimester APD values (AUC=0.89 and AUC=0.84 respectively), with corresponding APD threshold values of 10.9 mm and 12 mm. The predictive power of APD-delta was higher in the second trimester (AUC=0.86) than the third trimester. A percent change greater than or equal to 12% in the second trimester was predictive of the need of surgical intervention.
Conclusions:
We found that both 2nd and 3rd trimester APD were similar in their ability to predict the need for post-natal surgery. Increases in 2nd trimester APD-delta greater than or equal to 12% may be a useful adjunct when counseling families about need for close follow-up, and future study with a larger cohort could confirm the significance of this additional parameter.
Back to 2019 Abstracts