This prospective single-center cohort study investigated relationships between pelvic-genital anatomical measurements and postoperative pain following RIRS. The protocol received approval from Mersin University Clinical Research Ethics Committee (2025/553, May 14, 2025) per Declaration of Helsinki. All participants provided written informed consent. Between September 2024 and May 2025, 420 consecutive adult patients scheduled for RIRS were initially assessed for eligibility. Inclusion criteria comprised age ≥ 18 years, renal/ureteral calculi 4–20 mm, available preoperative CT imaging, and ability to complete three-week follow-up. Exclusion criteria included urogenital anomalies (horseshoe kidney, ectopic kidney, ureteropelvic junction obstruction, duplex collecting systems, significant ureteral strictures), previous pelvic surgery, bilateral stones, active infection, pregnancy, chronic pain syndromes, opioid dependency, previous DJ stent placement, ASA > III, and psychiatric disorders interfering with pain assessment. After applying exclusion criteria (n = 68 excluded: 23 anatomical anomalies, 18 previous pelvic surgery, 12 bilateral stones, 8 active infection, 7 other criteria) and accounting for incomplete follow-up (n = 32 lost to follow-up or incomplete data), 320 patients were deemed eligible and completed the full three-week assessment protocol. Sample size calculation (G*Power 3.1.9.7) for correlation analysis with r = 0.3, α = 0.05, power = 90% yielded 112 patients required. Target enrollment of 320 patients accounted for multiple comparisons (45 anatomical parameters), anticipated attrition, and stratified analyses.

Anatomical measurements were obtained from preoperative CT scans (≤ 2 mm slice thickness, standardized protocols) independently by two blinded radiologists showing excellent inter-rater reliability (ICC: 0.92–0.98). Comprehensive assessment encompassed 45 anatomical parameters including 37 pelvic measurements (pelvic inlet anteroposterior/transverse diameters, mid-pelvis anteroposterior/transverse diameters, pelvic outlet anteroposterior/transverse diameters, pelvic depth, sacral slope, pelvic tilt, pelvic incidence, ureter length, upper/middle/lower ureter diameters, bladder capacity, bladder wall thickness, pelvic floor thickness, levator ani thickness, perineal body length, pubococcygeal distance, and gender-specific organ measurements) and 8 renal/stone-related parameters (kidney length/width/thickness, renal pelvis diameter, infundibulopelvic angle, calyceal neck diameter, parenchymal thickness, stone size). Complete parameter specifications are provided in Supplementary Table S1.

All procedures were performed under general anesthesia (endotracheal intubation, sevoflurane maintenance) to eliminate variability from spinal anesthesia and residual neuraxial blockade effects on early postoperative pain assessment. Standardized multimodal analgesia was administered intraoperatively: paracetamol 1 g IV 30 min before incision, dexketoprofen 50 mg IV after ureteral access, and fentanyl 1–2 µg/kg IV as needed during surgery. Postoperatively, patients received scheduled paracetamol 1 g every 6 h and dexketoprofen 50 mg every 12 h for 48 h, with tramadol 50-100 mg as rescue analgesia (VAS ≥ 4). This standardized protocol ensured consistent analgesic baseline across the cohort for valid pain score comparisons.

RIRS procedures used standardized techniques: semi-rigid then flexible ureteroscopy, holmium laser lithotripsy, systematic stone extraction. All procedures were performed by experienced urologists (> 200 RIRS procedures annually) using standardized protocols to minimize operator variability. Flexible ureteroscopy was performed using reusable Karl Storz Flex-X2S digital flexible ureteroscope (7.5 Fr working shaft diameter, 3.6 Fr working channel, Karl Storz Endoskope, Tuttlingen, Germany) in 245 cases (76.6%), and single-use Boston Scientific LithoVue digital flexible ureteroscope (8.5 Fr distal tip, 3.6 Fr working channel, Boston Scientific, Marlborough, MA, USA) in 75 cases (23.4%). Ureteral access sheaths when used were 11.5/13F (Cook Medical, Bloomington, IN, USA) or 12/14F (Boston Scientific) depending on ureteral caliber. The scope-to-sheath diameter ratio ranged from 0.58 to 0.65, allowing adequate irrigation outflow around the ureteroscope shaft. Intrarenal pressure was measured via a purpose-modified ureteral access sheath system incorporating a side-channel pressure transducer (Stryker T5 IntraCompartmental Pressure Monitor System, Stryker Corporation, Kalamazoo, MI, USA). The pressure sensor was positioned within the renal pelvis via the access sheath side-port, providing direct measurement of true intrarenal pressure rather than irrigation inflow pressure. Mean pressure was calculated as time-weighted average across the entire operative duration; peak pressure represented maximum recorded value during the procedure. Pressure measurements were recorded at 10-second intervals. All procedures utilized the Lumenis Pulse 120 H Holmium: YAG laser system. Moses Technology pulse modulation was employed in 187 procedures (58.4%), with remaining cases using standard pulse mode. Settings were individualized based on stone characteristics: typical parameters included pulse energy 0.5–1.2 J and frequency 10–25 Hz for dusting fragmentation, versus 1.0–2.0 J and frequency 5–15 Hz for fragmentation. Long pulse width mode was preferentially used for larger, harder stones (Hounsfield units > 1000), while short pulse mode was used for softer stones to minimize retropulsion. Irrigation was delivered using a standardized protocol combining gravity-based flow (saline bag at 100–150 cm height) with controlled manual supplementation via hand-pump bulb syringe when needed for visibility. Target irrigation pressure was < 150 mmHg at the pump level, consistent with international practice patterns. Automated irrigation systems were not used. Comprehensive intraoperative data were prospectively collected including total laser energy (kilojoules), total lasing time (minutes), ureteral access sheath usage, irrigation volume, ureteral manipulation (dilation degree, scope passages), and complications graded by standardized criteria (mucosal injuries grade 0–3, perforation, bleeding, stone migration). DJ stent placement was surgeon-discretion based on operative duration > 60 min, ureteral trauma, residual fragments, or solitary kidney. Stents (4.8 or 6.0 French, polyurethane or silicone) remained 21 ± 7 days. Final population: 275 stented (85.9%), 45 non-stented (14.1%).

Pain assessment utilized Visual Analog Scale (0–10) at 2 h, 6 h, 24 h, 48 h, 7d, 14d, 21d postoperatively, plus 24 h and 7d post-stent removal by blinded staff. Stent-related symptoms were evaluated using validated Turkish USSQ (six domains, 1–7 scale) at 21 days [4]. Quality of life and satisfaction used 10-point scales.

Statistical analyses utilized SPSS 28.0 and R 4.3.1. Data are presented as mean ± SD or median (IQR). Between-group comparisons employed t-tests or Mann-Whitney U tests; categorical variables employed chi-square or Fisher’s exact tests. Effect sizes were calculated as Cohen’s d. Correlation analyses utilized Pearson or Spearman coefficients based on data distribution. Given extensive multiple testing (315 tests: 45 anatomical parameters × 7 outcomes), Bonferroni correction controlled family-wise error rate (α = 0.000159; 0.05/315). While controlling Type I error, Bonferroni increases Type II error risk. Post-hoc power analysis demonstrated that n = 320 with α = 0.000159 provided 90% power for r ≥ 0.23 and 80% power for r ≥ 0.20. To address Bonferroni’s conservatism, Holm stepwise correction was applied as sensitivity analysis, maintaining family-wise error control with greater statistical power. For procedural variables (intrarenal pressure, laser, access sheath, complications), separate analyses employed appropriate correction (5 parameters × 7 outcomes = 35 tests, α = 0.00143).

Three hierarchical regression models compared anatomical versus procedural predictive value: Model 1 (Anatomical) included parameters with |r|>0.10; Model 2 (Procedural) included intrarenal pressure, laser energy, access sheath, and complications; Model 3 (Combined) integrated both sets to assess incremental value (ΔR²). Multicollinearity was assessed using variance inflation factors (VIF > 5); model performance was evaluated using R², adjusted R², RMSE, and AIC. Propensity score matching (1:1, caliper = 0.2) addressed stent confounding, matching on stone size, stone location, operative time, mean intrarenal pressure, laser energy, and patient age. Model stability was assessed via 10-fold cross-validation with 1000 bootstrap iterations. Fisher’s z-transformation compared correlation coefficients between genders.