The Toronto nomogram: A Bayesian meta-regression derived prenatal ultrasound index to predict lower urinary tract obstruction and prune belly syndrome

Prenat Diagn. 2024 Feb;44(2):117-123. doi: 10.1002/pd.6384. Epub 2023 May 25.

Abstract

Introduction: A nomogram for predicting the diagnosis of lower urinary tract obstruction (LUTO) based on an antenatal ultrasound index generated from a Bayesian Meta-regression analysis has been in development and noted with superior diagnostic accuracy compared to the keyhole sign (KHS). We aim to assess the accuracy of the nomogram in expanded diagnostic utilization to predict LUTO.

Methodology: The validation of the nomogram for expanded diagnostic utilization was based on data from a prospective institutional antenatal clinic database between January 2020 and June 2022. Diagnostic accuracy indices were determined for confirmed postnatal diagnosis of LUTO or prune belly syndrome (PBS). Receiver operating characteristics (ROC) curves were generated to compare the area under the curve (AUC) of the nomogram versus KHS.

Results: Based on 84 male fetuses with antenatal ultrasound of moderate-severe hydronephrosis (PUV n = 15, PBS n = 4), the KHS had 26.3% (95%CI 9.1-51.2) sensitivity and 100% (95%CI 94.4%-100%) specificity, with 14 false-negatives. The nomogram showed a 84.2 (95%CI 60.4%-96.6%) sensitivity and 95.4 (95%CI 87.1%-99%) specificity with three false-positives. The nomogram also had a superior AUC compared to KHS (0.98 vs. 0.63).

Conclusion: The nomogram can be used as a valuable tool to trigger further postnatal screening and provide individualized risk assessments to families during prenatal counseling.

MeSH terms

  • Bayes Theorem
  • Female
  • Humans
  • Male
  • Nomograms
  • Pregnancy
  • Prospective Studies
  • Prune Belly Syndrome*
  • Ultrasonography, Prenatal
  • Urinary Tract*