Antenatal hydronephrosis is a common clinical entity and is identified on up to 5% of prenatal ultrasounds.¹ Most children with isolated hydronephrosis are expected to resolve over time²; however, for patients with suspected ureteropelvic junction obstruction (UPJO), surgical intervention may be indicated to prevent deterioration of renal function.³ Identifying these patients can be challenging and for patients with suspected UPJO, mercaptoacetyltriglycine (MAG-3) diuretic renal scans are commonly used to investigate obstruction.⁴ We previously investigated the Hydronephrosis Outcome Prediction (HOP) score, which is a simple scoring system that combines 3 parameters from the initial ultrasound, and showed that it could predict which patients with prenatal hydronephrosis would resolve.⁵ Here, we aimed to evaluate the HOP score at the initial ultrasound and to determine its association with obstruction on MAG-3 scan.

Patients with prenatally detected and postnatally confirmed isolated hydronephrosis were prospectively enrolled from 7 centers between 2007 and 2022 through Institutional Review Board–approved protocols at each center. Baseline ultrasounds were included if performed between 3 days and 6 months of life and if Society for Fetal Urology (SFU) grade, the anterior posterior diameter of the renal pelvis (APD), and renal lengths were available for analysis. Exclusion criteria were vesicoureteral reflux, primary obstructive megaureter, other urological anomalies, and follow-up of less than 3 months. Three ultrasound parameters are used to calculate the HOP score: SFU grade, APD in mm, and the absolute percentage difference in renal lengths between the affected kidney and the contralateral side (see Table). The HOP score was developed as a part of a study by Li et al⁶ and has scores ranging from 0 to 12 points. For example, a patient with SFU grade 3, APD of 12 mm, and renal length difference of 15% would have a HOP score of 7. Obstruction on MAG-3 was defined using 2 endpoints: (1) T½ time ≥20 minutes and function less than 40% on the affected kidney, or (2) simply T½ time ≥20 minutes. Area under the curve and logistic regression were used for determining the association between HOP scores and MAG-3 results.

Of 601 patients with isolated hydronephrosis and ultrasound data available to calculate a HOP score, 180 met inclusion criteria with MAG-3 results available. Median HOP score was 8.0 (IQR 6-10) and median follow-up time was 34 months (IQR 22-48). Using the first endpoint of function <40% and T½ time ≥20 minutes, 19 (11%) patients had obstruction on MAG-3 with a median HOP score of 9.0. The remaining 161 patients had median HOP score of 8.0. HOP score 9 or higher was the optimum threshold for predicting obstruction with AUC 0.67 (95% CI: 0.55-0.77). On logistic regression, patients with HOP score 9 or higher at initial ultrasound were significantly more likely to have obstruction on MAG-3 (OR = 3.8, 95% CI: 1.3-11, P = .01). Using the second endpoint of T½ time ≥20 minutes, 89 (49%) patients had obstruction on MAG-3 and patients with HOP score 9 or higher were also significantly more likely to have obstruction on MAG-3 (OR = 2.6, 95% CI: 1.4-4.8, P = .002). Overall, 92 (51%) of patients underwent pyeloplasty, while 25 (11%) resolved and 63 (35%) were under continued follow-up.

The findings of this study demonstrated that patients with a HOP score of 9 or higher at the initial ultrasound were more than 2.5 times as likely to show obstruction on subsequent MAG-3 as patients with HOP scores less than 9. Although SFU grade and APD are typically used to assess ultrasound results,¹ the HOP score provides a wider range of values (0-12 points) for assessment than SFU grade, combines important elements of the ultrasound together into 1 score, and potentially could be used to look at changes in HOP scores over time. Obstruction on MAG-3 was selected as our primary outcome because using surgery as an outcome is not ideal due to the various factors involved with the decision to proceed to surgery. In a previous analysis we conducted on over 600 patients from the same cohort, we found that HOP scores of 4 or less predicted resolution of hydronephrosis.⁵ Here, we show that the HOP score is an accessible tool that aids in predicting the outcomes of hydronephrosis from the initial postnatal ultrasound. Indeed, all 3 components of the HOP score (SFU grade, APD, and renal length difference) can be easily calculated from most ultrasound reports. Although new methods, like deep learning, are on the horizon with the objective to identify patients with obstruction more accurately,^7,8 these methods are currently not widely available for clinical use. Our study demonstrates that using a combination of 3 easily accessible ultrasound parameters in the form of the HOP score can be valuable for patient management and can help predict the outcome of obstruction on diuresis renography from the first postnatal ultrasound.

Acknowledgements

The authors acknowledge our coinvestigators on this project from the Society for Pediatric Urology Hydronephrosis Taskforce. Our multicenter collaboration includes centers from across the United States and Canada and new sites are welcome to join.