Predicting the likelihood of prolongation of half-time among infants with initially indeterminate drainage values: A single-institution retrospective study of 535 patients with ureteropelvic junction obstruction

Introduction and objective Prior studies have shown a broad half time (T1/2) interval on MAG3 diuresis renography (DR) that is indeterminate for obstruction. We aimed to refine and sub-divide the indeterminate range and associate it with clinically meaningful outcomes: pyeloplasty and pyeloplasty-free survival. Methods We identified patients <1.5 years-old at presentation with unilateral, isolated moderate to severe hydronephrosis who underwent DR from 2000 to 2016. A logistic regression model was created using T1/2 to predict surgery. An indeterminate range was defined based on patients with <90% probability of pyeloplasty or resolution. This group was subdivided into three T1/2 intervals: 5-20, 21-40, and 41-60 min. Endpoints were pyeloplasty and pyeloplasty free survival. Indications for surgery were loss of differential renal function (DRF), worsening T1/2, family preference, and/or pain. Results Among 2025 patients with DR, 704 met criteria (169 were lost to follow up). Of the remaining 535, 218 had pyeloplasties and 317 did not. The Pyeloplasty group had significantly worse DRF, T1/2 at initial DR, and exited the study earlier, at a median age 1.1years vs 2.3 years (p < 0.001). For all patients with antenatally detected unilateral UPJ obstruction, the odds of undergoing pyeloplasty at any time increased by 1.8 times (p < 0.001 [95% CI: 1.04, 1.08]) per 10 unit increase in T1/2 until T1/2 = 60. However, in patients with intermediate drainage, five year surgery-free survival probability for patients with T1/2 5-20, 21-40, and 41-60 min were 79.7%, 46.7% and 33.3% respectively (chi 2 = 41.2, P = <0.001). Discussion Previous efforts to define indeterminate drainage resulted in ranges for T1/2 that were too broad to be clinically useful. Within our endpoint-defined indeterminate range, our data show that there are significant step offs in 5-year surgery-free survival for patients with T1/2 < 20 min, 21-40 min, and 41-60 min. Although there is a steady decrease in surgery-free survival among patients with a T1/2 of 21-40 min over the first 5 years of life, half can be managed nonoperatively. These patients likely represent the true intermediate risk group and closer follow up is justified. Conclusions Initial T1/2 on DR is predictive of future surgery. When drainage is "indeterminate" for obstruction, sub-stratification allows for more accurate prognostication.