Back to 2024 Abstracts
Renal Length Difference As An Independent Predictor Of Decision For Pyeloplasty: A Multicenter Analysis
Anne G. Dudley, MD1, Carol Davis-Dao, PhD
2, Joshua Chamberlain, MD
3, Shannon Cannon, MD
4, Gina Lockwood, MD
5, Nora Kern, MD
6, Kunj Seth, MD
7, Rebecca Zee, MD, PhD
8, Valre Welch, MSN, CPNP
8, Elias Webhi, MD
2, Sarah Williamson, MD
9, Antoine Khoury, MD
2, C. D. Anthony Herndon, MD
8, Luis Braga, MD, PhD
10.
1Connecticut Children's Medical Center, Hartford, Hartford, CT, USA, 2Children's Hospital of Orange County and University of California Irvine, Orange, CA, USA, 3Loma Linda University Health, Loma Linda, CA, USA, 4University of Wisconsin, Madison, WI, USA, 5University of Iowa, Iowa City, IA, USA, 6University of Virginia, Charlottesville, VA, USA, 7Stanford University, Palo Alto, CA, USA, 8Children's Hospital of Richmond at Virginia Commonwealth University, Richmond, VA, USA, 9Children's Hospital of the King's Daughters, Norfolk, VA, USA, 10McMaster University, Hamilton, ON, Canada.
Background:Prenatal hydronephrosis is a commonly detected condition, with a small percentage of children ultimately requiring postnatal intervention for ureteropelvic junction obstruction. Renal length varies across affected and unaffected renal units and may change in response to obstruction or with growth. We hypothesized that a greater difference in renal lengths, when comparing the affected and unaffected kidney, may be associated with decision for pyeloplasty.
Methods: Patients with unilateral, isolated hydronephrosis enrolled in the Society for Fetal Urology Hydronephrosis Registry were included in the analysis. Children were eligible if data from the initial, baseline, renal ultrasound was available, and was completed between 3 days and 6 months of life. Children with bilateral hydronephrosis, ureteral or bladder anomalies were excluded. Patients were studied in a case-control design comparing children with pyeloplasty to non-operative patients. Data was analyzed using Wilcoxon rank-sum test and multivariable logistic regression modelling.
Results: A total of 1186 patients met inclusion criteria, from a registry enrollment of 2770 patients. Of the included patients, children were followed median follow-up of 20 months (IQR 9-36 months). The median age at initial ultrasound was 1.8 months (IQR 0.9-3.5 months) and 78% were male (922/1186).
Children who underwent pyeloplasty (n=195) showed a median 19% (IQR 10-27%) difference in renal lengths (comparing the hydronephrotic (affected) kidney to the normal kidney), compared with 6% (IQR 2.8-11%) in the nonoperative group (n=991) (p<0.0001). Subgroup analysis of patients with high grade SFU 3-4 hydronephrosis (n=458) demonstrated persistent difference in operative versus non-operative group, (20% renal length difference, IQR 11-27%; versus 9%, IQR 4.4-15%, p<0.0001). Subgroup analysis for SFU grade 3, SFU grade 4 and laterality confirmed persistent significant differences in renal length percentages across groups (Table 1).
Multivariable logistic regression model showed for every 1% increase in difference in initial renal length (comparing the affected to the unaffected kidney length), the odds of surgery increased by 1.04-fold (p=0.0002) after adjusting for initial age, sex, circumcision status and initial SFU grade and initial AP diameter. For a 20% difference in initial renal length, the odds of surgery increased 4-fold (OR=4.1, 95% CI 2.5-6.9, p<0.0001) with multivariable adjustment (Table 2).
Conclusions: Greater difference in renal length at initial ultrasound was a strong, significant predictor of decision for pyeloplasty, after controlling for initial age, sex, initial SFU grade and AP diameter. At 20% percentage of difference in renal size, odds of intervention increased by 4-fold. This data is valuable for counseling families at the initial visit, to better select infants at high risk for progression to intervention.
Table 1: Comparison of Percentage Difference in Renal Lengths between Operative Patients (Pyeloplasty) and Non-Operative Patients
| | | | |
| Comparison of Absolute Percentage Difference in Renal Lengths | Cohort size | Operative (Pyeloplasty) GroupMedian % Difference in Renal Lengths | Non-Operative Group Median % Difference in Renal Lengths | P value |
| All patients | 1186 | 19% (IQR 10-27%)n=195 | 6.0% (IQR 2.8-11%)n=991 | <0.0001 |
| SFU 3-4 | 458 | 20% (IQR 11-27%)n=182 | 9.0% (IQR 4.4-15%)n=276 | <0.0001 |
| SFU 3 only | 290 | 15% (IQR 6.0-22%)n=64 | 8.0% (IQR 4.0-14%)n=226 | <0.0001 |
| SFU 4 only | 168 | 23% (IQR 15-30%)n=118 | 16% (IQR 9-23%)n=50 | 0.009 |
| Right only | 250 | 17% (IQR 8.0-25%) n=68 | 5.4% (IQR 2.1-10%) n=182 | <0.0001 |
| Left only | 841 | 20% (IQR 12-28%) n=126 | 6.6% (IQR 3.2-11%)n=715 | <0.0001 |
Table 2: Multivariable Model for Predicting Decision for Pyeloplasty | | |
| Predictor | Adjusted Odds Ratio (95% CI) | p value |
| Percentage Difference in Renal Lengths<20%≥20% | 1.0 (ref)4.1 (2.5-6.9) | <0.0001 |
| AP Diameter<20 mm≥20 mm | 1.0 (ref)5.6 (3.2-9.9) | <0.0001 |
| SFU Grade Grade 1-3Grade 4 | 1.0 (ref)9.0 (5.5-15) | <0.0001 |
| Age at Initial RBUS | 0.91 (0.79-1.04) | 0.17 |
| SexFemaleMale | 1.0 (ref)1.5 (0.95-2.3) | 0.08 |
| Circumcision statusCircumcisedUncircumcised | 1.0 (ref)1.0 (0.98-1.02) | 0.98 |
Back to 2024 Abstracts
