Development of the prediction model for negative outcomes after primary laparoscopic pyeloplasty in children: a retrospective study of 535 patients

Abstract

Background: Laparoscopic pyeloplasty (LP) has been widely accepted as an effective method to treat ureteropelvic junction obstruction (UPJO) in pediatric patients. Limited studies focused on the relationship between risk factors and negative outcomes. Our study aimed to seek independent risk factors for negative outcomes and construct a prognostic nomogram to assist clinical decision-making and improve outcomes.Methods: A total of 535 patients with UPJO treated with primary LP between January 2016 and December 2020 were retrospectively reviewed. Negative outcomes were defined as restenosis requiring reoperation and grade III and IV complications based on the Clavien-Dindo grading system. Univariate and multivariate logistic regression analyses were used to select risk factors for negative outcomes after LP and developed the prediction model. The model was internally validated by the parametric bootstrapping method.Results: Among the 535 patients, 33 (6.2%) developed negative outcomes. Ten patients developed ureteropelvic junction (UPJ) restenosis and underwent secondary surgery. UPJ leakage (two cases), difficulty of urinating after the removal of the Double-J (DJ) stent (two cases), and ileus (two cases) were the most common grade IIIa complications, while distal ureteral stricture (five cases), hernia formation (three cases), and delayed wound healing around the fistula (two cases) were the most common grade IIIb complications.After univariate and multivariate logistic analyses, the patient's weight, preoperative anteroposterior pelvic diameter (APD), and difficulty of DJ stent insertion were independent risk factors for negative outcomes, and they were used to fit the prediction model. The Brier score was 0.048. The model was relatively wellcalibrated. The area under the receiver operating characteristic curve was 0.831 (95% CI: 0.756-0.906). Decision curve analysis illustrated good clinical utility.Conclusions: Primary LP is a safe and effective method for pediatric patients with UPJO. The patient's weight, preoperative APD, and difficulty of DJ stent insertion were independent risk factors for negative outcomes after LP. We established and validated a predictive model for negative outcomes after LP. With the help of this model, clinicians can make better decisions and improve patient outcomes.

Risk factors for adverse outcomes after pediatric pyeloplasty: A retrospective cohort study

“…Even in these studies, the identified risk factors have been the subject of controversy. [5][6][7][8][9] The objective of this research was to identify the risk factors for adverse outcomes after pediatric pyeloplasty.…”

“…However, very few authors have looked for preoperative risk factors that could influence postoperative outcomes. Even in these studies, the identified risk factors have been the subject of controversy 5–9 . The objective of this research was to identify the risk factors for adverse outcomes after pediatric pyeloplasty.…”

Prediction model for serious complications after primary laparoscopic pyeloplasty in children: is it a useful tool?

“…We have read with great interest the contribution done by Li et al , in their study: “ Development of the prediction model for negative outcomes after primary laparoscopic pyeloplasty in children: a retrospective study of 535 patients ” ( 12 ).…”

“…Secondly, seven surgeons performed operations separately, so bias was unavoidable because these physicians had slightly different preferences and approaches to some details during the surgery. Thirdly, it was a retrospective study ( 12 ). It is a bit contradictory to consider these limitations or weaknesses of the study: on the one hand being a single center, but on the other hand having multiple surgeons.…”

“…Thus, the pediatric urologist could improve his surgical strategy (adapted to the characteristics of the patient) and the success of the results. However, we consider relevant to include other possible risk factors in the univariate and multivariate model described by Li et al ( 12 ). Those factors could have also distorted the general results of this work.…”