Brief Summary:
\*Summary\*
Background: Some male living kidney donors experience scrotal pain and swelling on the same side as their nephrectomy following donation, occasionally requiring surgical intervention for relief.
Methods: In a population-based cohort study, we will identify male living kidney donors who underwent laparoscopic nephrectomy in Alberta, Canada, between 2002 and 2020. Each donor will be matched 1:10 to male non-donors from the general Alberta population with similar baseline health characteristics. Matching variables will include age, cohort entry date, rural versus urban residence, income quintile, prior vasectomy, and history of inguinal hernia repair. Donors and non-donors will be followed using linked provincial healthcare administrative databases in Alberta, Canada, until March 31, 2021. The primary outcome will be hospital admission for hydrocelectomy. The secondary outcome will be the receipt of a scrotal ultrasound.
Study Objective: This study aims to determine whether male living kidney donors who undergo laparoscopic nephrectomy have a higher risk of hydrocelectomy compared to matched male non-donors with similar baseline health profiles.
\*Background\*
Understanding the risks associated with living kidney donation is essential for ensuring informed consent, guiding donor follow-up, and identifying potential opportunities to reduce harm. While most donors recover well, emerging evidence suggests that some male donors may experience postoperative scrotal pain and swelling on the side of their nephrectomy. In this years following donation, these symptoms can lead to surgical intervention such as hydrocelectomy.
\*Study Design and Setting\*
We propose a population-based matched cohort study using linked administrative health data from the Alberta Kidney Disease Network (AKDN), which includes datasets from Alberta Health, the provincial health ministry. Over 99% of Alberta residents are registered for provincial health coverage, ensuring near-complete population capture of hospital care and physician services.
We are publishing the study protocol a priori to conducting statistical analyses including outcome analyses to strengthen methodological rigour and transparency. This study adheres to the STROBE and RECORD reporting guidelines and has received research ethics approval from the Universities of Alberta and Calgary, with a waiver of informed consent due to the use of de-identified administrative data.
\*Data Sources\*
Baseline characteristics and outcomes are derived from AKDN-linked datasets, including physician claims, hospital discharge abstracts, provincial laboratory data, and pharmacy records. Alberta adopted the Canadian Classification of Health Interventions (CCI) in April 2002, which defines the start of the accrual period. Laboratory data includes serum creatinine and urine albumin from 2002 onward. These databases have been validated and widely used in prior health services research.
\*Study Population\*
We will include all male Alberta residents who underwent living donor nephrectomy between April 1, 2002, and March 31, 2020. Donors will be identified using the algorithm of an ICD-10 code for kidney donation (Z52.4) and a CCI procedural code for kidney procurement or excision (1.PC.58, 1.PC.89, or 1.PC.91). The date of donation nephrectomy will be the index date. Donors will be excluded if they are under 18, out-of-province, have incomplete demographic data, died on or before the nephrectomy date, or had prior dialysis, organ transplant, or scrotal procedures (excluding vasectomy or inguinal hernia repair), scrotal ultrasound. For the primary analysis we will concentrate on males who had a laparoscopic donor nephrectomy and exclude those who underwent an open donor nephrectomy, as in prior studies scrotal complications more prevalent after laparoscopic nephrectomy compared to open nephrectomy.
The non-donor comparator group will be drawn from the general population of all adult males in Alberta. Among this cohort, we will exclude those with comorbidities that would contraindicate donation, those without at least one physician visit in the previous two years, and those with prior scrotal conditions. The non-donor index date will be randomly assigned to all males in Alberta based on the distribution of the donors' nephrectomy dates (2002-2020). Each donor will be matched to 10 non-donors based on age (±1 year), cohort entry year (±3 years), rural/urban residence, income quintile, prior vasectomy, and prior inguinal hernia repair.
\*Baseline Characteristics\*
Baseline variables will be assessed at the index date (i.e. the nephrectomy date in the donor group or simulated index date for non-donors). Demographic data (age, sex, rurality, income quintile) will be obtained from Alberta Health records and census files. Clinical characteristics such as comorbidities, estimated glomerular filtration rate (eGFR), and albuminuria will be obtained from hospital discharge records, outpatient laboratory data, and physician claims data preceding the index date.
\*Outcomes\*
Participant observation time will be censored at the time of death, emigration, or until the maximum follow-up date (March 31, 2021). The primary outcome is a hospital admission for hydrocelectomy confirmed through both a receipt of a physician billing code (AHCIP: 73.0, 73.0A, 73.1, 73.1A, 73.1B, 73.9, 73.91, 75.0) and a hospital procedural code (CCI: 1QH87LA, 1QH87LB, 1QH52HA, 1QH52LA, 1QH80LA, 1QG52HA, 1QG52LA) within 30 days of one another. The secondary outcome will be the receipt of a scrotal ultrasound. Data on laterality of the nephrectomy is not available, but \~90% of donor nephrectomies are expected to be left-sided. Data on the laterality of the hydrocelectomy is not available.
\*Statistical Analysis Plan\*
Baseline characteristics will be compared using standardized mean differences, with \>10% indicating a meaningful difference. We will use Cox proportional hazards models with robust variance estimation to account for matched sets and estimate hazard ratios for our outcomes. Proportional hazards assumptions will be evaluated using Kolmogorov-type supremum tests; if violated, stratified log-rank tests and differences in 20-year restricted mean survival time will be used. Outcomes will be described as incidence rates per 1000 person-years and cumulative incidence. These results will be obtained using the Aalen-Johansen estimator, counting the first event and treating death as a competing event, where donors and nondonors will be compared visually with graphical curves and at prespecified time points.
Prespecified secondary analyses will include a comparison of donors who underwent laparoscopic versus open nephrectomy. Inverse probability of treatment weighting (IPTW) from propensity scores will be used to balance covariates across the groups. All statistical tests will be two-sided, using a hierarchical testing framework. We will use R version 4.2.3 to conduct analyses, with significance defined as p \< 0.05.
\* P values (significance testing) and confidence intervals \*: We will perform hierarchical statistical tests to assess our hypotheses in the pre-defined order described below. The 2-sided significance level will be 0.05 for each tested hypothesis. After a non-significant test no further significance testing will be performed and no further P values will be reported. All estimates (with and without P values) will be reported as point estimates with 95% confidence intervals. We acknowledge the interval widths of the estimates without P values are not adjusted for multiple testing and that those inferences may not be reproducible. The order:
(i) the primary outcome (hydrocele surgery) in donors vs. nondonors (ii) the secondary outcome (scrotal ultrasound) in donors vs. nondonors (iii) primary outcome in donors who underwent a laparoscopic vs. open procedure (iv) secondary outcome in donors who underwent a laparoscopic vs. open procedure.