Early postoperative ultrasound after open pyeloplasty in children with prenatal hydronephrosis helps identify low risk of recurrent obstruction

Purpose: Prediction of successful pyeloplasty can be challenging, particularly in asymptomatic children treated for worsening prenatally diagnosed hydronephrosis. We evaluated early noninvasive objective predictors of success in this population. Materials and Methods: We reviewed patients who underwent pyeloplasty for asymptomatic, prenatally detected, worsening hydronephrosis treated between 2000 and 2010 with followup greater than 1 year. For all patients renal pelvis anteroposterior diameter in a mid transverse view and Society for Fetal Urology hydronephrosis grade were evaluated preoperatively and 3 to 4 months postoperatively. Aside from subjective evaluation based on change in hydronephrosis grade, we estimated the percentage of improvement in anteroposterior diameter (preoperative diameter - postoperative diameter/preoperative diameter). Repeat intervention after pyeloplasty arbitrarily defined failure. Patients were categorized into 3 groups, ie no failure on followup ultrasound (group 1), no failure with postoperative nuclear scan to document success (group 2) and failed pyeloplasty (group 3). ROC curves were plotted to correlate 4 variables with failure, namely preoperative anteroposterior diameter, postoperative anteroposterior diameter, percent improvement in anteroposterior diameter and subjective change in grade. Results: Of 229 patients 192 (84%) who met inclusion criteria had ultrasound at 3 to 4 months postoperatively available. Mean age at surgery was 19 months and mean ± SD followup was 49.9 ± 27.7 months. Percent improvement in anteroposterior diameter and postoperative anteroposterior diameter measure were the most reliable variables to predict failure (AUC 0.88 and 0.86, respectively, p <0.0001), whereas preoperative anteroposterior diameter and subjective changes in grade were not good predictors (AUC 0.52, p >0.05). A percent improvement in anteroposterior diameter of 38% or greater or postoperative anteroposterior diameter of 11.5 mm or less was associated with success, with sensitivity of 100% and specificity of 61% and 55%, respectively. Conclusions: Percent improvement in anteroposterior diameter and postoperative pelvic anteroposterior diameter measure can provide objective guidance as to which patients need closer monitoring during followup, and can help select children at low risk for repeat intervention after pyeloplasty.