Risk of Hydroceles Following Pyeloplasty

Overview

This population-based study will use provincial health administrative data to examine the risk of hospital admission for hydrocelectomy among males who underwent laparoscopic pyeloplasty, compared with males from the general population with similar baseline health characteristics. We will use health administrative data from Ontario, Canada, to identify males who had a laparoscopic pyeloplasty between 1992 and 2024. We will calculate a propensity score based on sociodemographic characteristics, comorbidities, and health care utilization. We will then match laparoscopic pyeloplasty males to non-pyeloplasty male controls (1:4) based on propensity score, age, and index date. The primary outcome is hospital admission for hydrocelectomy. The secondary outcomes are a hydrocele diagnosis and receipt of a scrotal ultrasound.

\*Background\* Previous reports among male living kidney donors have documented postoperative testicular pain and/or scrotal swelling on the side of nephrectomy. Many affected donors were subsequently diagnosed with hydroceles and required surgical intervention (hydrocelectomy). However, existing studies were generally small, had limited follow-up, and lacked appropriate comparison groups. To address these limitations, a large population-based study of living kidney donors was conducted. Male donors had a substantially higher incidence of scrotal surgery than matched nondonors: 7.8% of donors (70/898) underwent scrotal surgery compared with 0.2% of nondonors (19/8,980), corresponding to 8.3 versus 0.2 events per 1,000 person-years. The hazard ratio was 38.8 (95% CI, 22.1-67.9; P \< 0.001), and the 20-year cumulative incidence was 13.8% in donors versus 0.7% in nondonors. In an exploratory secondary analysis, additional surgical cohorts were examined, and similar signals were observed among non-donors undergoing other kidney surgeries. For example, among patients who underwent pyeloplasty, the incidence of scrotal surgery was 1.8 events per 1,000 person-years following laparoscopic procedures and 0.8 events per 1,000 person-years following open procedures. Although these rates were lower than those observed among donors, they exceeded those expected in the general population. Additionally, higher rates were observed in the laparoscopic surgery group compared to the open surgery group. These analyses were descriptive and unadjusted for confounding but suggested a potential mechanistic link. One hypothesis is that surgical manipulation of the renal pelvis and/or ureter may disrupt lymphatic or venous drainage to the scrotum, contributing to hydrocele formation. This proposed mechanism has prompted further investigation into whether similar postoperative scrotal complications occur following other urological procedures involving manipulation of the collecting system, such as pyeloplasty. Primary Objective: To evaluate whether males who underwent laparoscopic pyeloplasty have a higher risk of a hospital admission for hydrocelectomy compared to males with similar indicators of baseline health selected from the general population. \*Study Setting and Data Sources\* This study will be conducted at ICES (ices.on.ca). ICES is an independent, non-profit research institute with legal authority under Ontario's health information privacy legislation to collect and analyze health care and demographic data without individual consent for the purposes of health system evaluation and improvement. Data use for this project is authorized under Section 45 of Ontario's Personal Health Information Protection Act (PHIPA), which does not require Research Ethics Board approval. Data from multiple linked databases will be used to identify the cohort, establish baseline characteristics, and define outcomes. These databases include Ontario's Registered Persons Database (RPDB), the Ontario Health Insurance Plan (OHIP), and Canadian Institute for Health Information's (CIHI) Discharge Abstract Database (DAD), National Ambulatory Care Reporting System (NACRS), and Same Day Surgery (SDS). Given the nature of the data sources, minimal missingness is expected across the study variables. This retrospective cohort study will rely entirely on existing administrative health data available at ICES. To promote research transparency and reproducibility, this study protocol will be registered on ClinicalTrials.gov, including the design and statistical analysis plan, prior to initiating any outcome analyses. \*Study Population\* Those who underwent pyeloplasty will be identified using hospitalization records and admissions in CIHI-DAD and SDS from 1992 to 2024. Only the first pyeloplasty that occurs will be examined. Three cohorts will be constructed: laparoscopic pyeloplasty, open pyeloplasty, and non-pyeloplasty controls. \*Baseline Characteristics and Matching\* Baseline variables will be assessed at the index date, defined as the date of the first pyeloplasty surgery. For general population comparators who did not undergo pyeloplasty, an index date will be randomly assigned based on the distribution of index dates in the pyeloplasty group. Baseline characteristics will be summarized using descriptive statistics and standardized differences. To reduce confounding, 1:4 propensity score matching will be performed, pairing each pyeloplasty patient with up to four controls using greedy nearest-neighbour matching without replacement. Matching variables will include propensity score, age, and index date. Post-matching balance will be assessed using standardized mean differences. \*Outcomes\* The primary outcome is hospital admission and receipt of surgery for hydrocele excision (i.e., hydrocelectomy). To ensure accurate outcome ascertainment, evidence of both a hospital-based procedural code (CCI codes: 1QH87LA, 1QH87LB, 1QH52HA, 1QH52LA, 1QH80LA, 1QG52HA, 1QG52LA; CCP codes: 731, 730, 7391, 7339) and a surgeon fee-for-service code (OHIP fee codes: S611, S630) is required, with each recorded in separate healthcare databases within 30 days of one another. The date of hospital admission will be recorded as the outcome date. Observation time will be censored at death, emigration, or the maximum follow-up date (March 31, 2025). Secondary outcomes include receipt of a scrotal ultrasound and diagnosis of hydrocele. \*Statistical Analysis\* Three primary cohorts will be defined: (1) laparoscopic pyeloplasty, (2) open pyeloplasty, and (3) non-pyeloplasty controls. For all pyeloplasty patients, the index date will be the date of surgery. For controls, index dates will be assigned by bootstrapping from the distribution of pyeloplasty index dates. The primary analysis will compare laparoscopic pyeloplasty with matched non-pyeloplasty controls. Incidence rates (per 1,000 person-years), rate differences, and hazard ratios (HRs) will be calculated using Cox proportional hazards models with robust variance to account for matching. If the proportional hazards assumption is violated, stratified log-rank tests will be used, and restricted mean survival times will be estimated at 20 years. Cumulative incidence will be estimated using Aalen-Johansen methods to account for the competing risk of death and reported at 1, 5, 10, 15, 20, and 25 years. In subgroup analyses, we will stratify by age (\<18, 18 to \<70, 70+) and cohort entry year (1992 to 2002, 2003 to 2013, 2014 to 2024), with primary outcome analyses repeated within each subgroup. \*Additional Analyses\* In additional analyses, the primary analysis will be repeated comparing different cohorts. These additional comparisons include: 1. Laparoscopic pyeloplasty vs. open pyeloplasty (primary and secondary outcomes) 2. Open pyeloplasty vs. non-pyeloplasty controls (only primary outcome) For comparisons of surgery types (e.g., laparoscopic vs. open), inverse probability of treatment weighting (IPTW) using propensity scores will be applied. All estimates will be reported with 95% confidence intervals. Hierarchical testing will be applied, with significance testing stopping after the first non-significant result (α = 0.05 per test); all remaining estimates will be presented without p-values.