The conundrum of high-grade hydronephrosis with non-obstructive drainage on diuretic renography

Patients with high grade hydronephrosis (HN) and non-obstructive drainage on mercaptoacetyltriglycine (MAG-3) diuretic renography (renal scans) can pose a dilemma for clinicians. Some patients may progress and require pyeloplasty; however, more clarity is needed on outcomes among these patients.

Our primary objective was to predict which patients with high-grade HN and non-obstructive renal scan, (defined as T ½ time <20 min) would experience resolution of HN. Our secondary objective was to determine predictors for surgical intervention.

Patients with prenatally detected HN were prospectively enrolled from 7 centers from 2007 to 2022. Included patients had a renal scan with T ½<20 min and Society for Fetal Urology (SFU) grade 3 or 4 at last ultrasound (RBUS) prior to renal scan. Primary outcome was resolution of HN defined as SFU grade 1 and anterior posterior diameter of the renal pelvis (APD) < 10 mm on follow-up RBUS. Secondary outcome was pyeloplasty, comparing patients undergoing pyeloplasty with patients followed with serial imaging without resolution. Multivariable logistic regression was used for analysis.

Of the total 2228 patients, 1311 had isolated HN, 338 patients had a renal scan and 129 met inclusion criteria. Median age at renal scan was 3.1 months, 77% were male and median follow-up was 35 months (IQR 20-49). We found that 22% (29/129) resolved, 42% of patients had pyeloplasty (54/129) and 36% had persistent HN that required follow-up (46/129). Univariate predictors of resolution were age≥3 months at time of renal scan (p = 0.05), T ½ time≤5 min (p = 0.09), SFU grade 3 (p = 0.0009), and APD<20 mm (p = 0.005). Upon multivariable analysis, SFU grade 3 (OR = 4.14, 95% CI: 1.30-13.4, p = 0.02) and APD<20 mm (OR = 6.62, 95% CI: 1.41-31.0, p = 0.02) were significant predictors of resolution. In the analysis of decision for pyeloplasty, SFU grade 4 (OR = 2.40, 95% CI: 1.01-5.71, p = 0.04) and T ½ time on subsequent renal scan of ≥20 min (OR = 5.14, 95% CI: 1.54-17.1, p = 0.008) were the significant predictors.

Patients with high grade HN and reassuring renal scan can pose a significant challenge to clinical management. Our results help identify a specific candidate for observation with little risk for progression: the patient with SFU grade 3, APD under 20 mm, T ½ of 5 min or less who was 3 months or older at the time of renal scan. However, many patients may progress to surgery or do not fully resolve and require continued follow-up.