Standardizing Rezūm training with high-fidelity simulation: is technical performance associated with age?

INTRODUCTION

Benign prostatic hyperplasia (BPH), which typically manifests as lower urinary tract symptoms (LUTS), represents one of the most prevalent chronic conditions affecting ageing men and a major driver of healthcare utilization worldwide. Contemporary burden estimates from the Global Burden of Disease (GBD) project confirm a substantial and growing population impact of BPH across regions, paralleling demographic ageing and increasing diagnostic and treatment demand^[1]. Current international guidelines accordingly place BPH/LUTS among the core conditions requiring scalable, safe, and reproducible pathways of evaluation and intervention^[2]. Although medical therapy remains the recommended first-line treatment for most patients with BPH/LUTS, long-term symptom control, treatment adherence, and medication-related adverse effects remain important challenges, contributing to the increasing interest in minimally invasive surgical therapies (MIST) as alternatives for appropriately selected patients^[2].

Over the last decade, surgical management of BPH has expanded beyond conventional transurethral resection (TURP) and anatomical enucleation of the prostate (AEP), to include a wide spectrum of MIST, designed to reduce perioperative morbidity and preserve sexual function while offering clinically meaningful symptom relief^[3]. Although several of these technologies are “less invasive” by design, they are not “less technical”: outcomes remain highly dependent on correct endoscopic orientation, consistent respect for anatomical safety margins, and strict adherence to device-specific procedural algorithms. In this setting, training quality becomes a primary determinant of both short-term safety and long-term durability, especially during dissemination phases when techniques transition from expert centers to broad clinical practice.

As modern urological training faces structural constraints that may amplify inter-operator variability, recent European surveys describe heterogeneous exposure to operative endourology during residency and highlight a growing reliance on simulation to compensate for reduced case volumes and variability among training programs^[4]. Pan-European initiatives aimed at standardizing surgical education, including competency-based frameworks and structured curricula, have been introduced to address these gaps and improve equity in training access and assessment^[5-7]. These dynamics create a particularly relevant generational question: whether younger surgeons, often with fewer years of independent endoscopic practice but potentially greater familiarity with digital platforms, can achieve technical performance comparable to senior colleagues when training is delivered through a standardized, metrics-driven pathway.

Simulation-based education provides a pragmatic response to this challenge. Modern high-fidelity simulators can capture objective, detailed performance metrics (e.g., spatial accuracy, procedural timing, error patterns) and thereby transform training from an apprenticeship model with potentially subjective assessments into a reproducible and auditable process^[8]. Importantly, proficiency-based progression (PBP) approaches, where trainees advance only after meeting predefined objective benchmarks, have been associated with fewer errors and improved technical performance compared with conventional simulation or unstructured practice, supporting the concept that “measured proficiency” can replace “time served” as the core criterion of competence^[9]. In urology specifically, recent evidence syntheses indicate that simulation can improve educational outcomes and, in selected domains, translate into measurable improvements in operative performance, although heterogeneity in study design and endpoints remains a limitation of the field^[10].

Against this background, the purpose of the present study was to compare objective technical performance between younger (≤ 35 years) and older (> 35 years) urologists during a fully standardized high-fidelity convective water-vapor thermal therapy (Rezūm) simulation pathway, thereby informing the design of effective, equitable, and safe training programs.

METHODS

RESULTS

All enrolled participants successfully completed the five-scenario simulation pathway. Therefore, no participants were excluded from the final analysis. The final cohort comprised 460 urologists from 46 institutions nationwide [Table 1]. Mean age was 45.6 years (range 27-74). Most participants were male (406/460, 88.3%). For the primary comparison, participants were stratified a priori by chronological age into two groups: Junior (≤ 35 years, n = 120; mean age 32.9 years) and Senior (> 35 years, n = 340; mean age 50.1 years).

Performance across the five scenarios was summarized as mean normalized score (0-100) with 95%CI for each group, and between-group comparisons were assessed with two-sided P values.

Across the five scenarios, mean scores in the Junior group ranged from 84.98 to 93.85, whereas mean scores in the Senior group ranged from 80.69 to 95.55. Scenario-specific results are reported below and in Figure 1.

Figure 1. Scenario-specific normalized simulator scores by age cohort. Comparisons were performed between Junior (≤ 35 years; n = 120) and Senior (> 35 years; n = 340) participants within each individual scenario. For each scenario, mean scores, MD and their relative 95% confidence intervals (square brackets), with corresponding P values, are reported. MD: Mean differences.

· Scenario 1: the mean score was 93.38 (95%CI: 92.32-94.44) in the Junior group and 93.30 (95%CI: 92.70-93.89) in the Senior group (P = 0.89). The mean difference (Junior -Senior) was +0.08 points.

· Scenario 2: the mean score was 89.59 (95%CI: 88.34-90.84) in the Junior group and 89.21 (95%CI: 88.47-89.96) in the Senior group (P = 0.61). The mean difference was +0.38 points.

· Scenario 3: the mean score was 84.98 (95%CI: 83.04-86.92) in the Junior group and 80.69 (95%CI: 79.17-82.21) in the Senior group (P = 0.001). The mean difference was +4.29 points.

· Scenario 4: the mean score was 93.55 (95%CI: 92.36-94.74) in the Junior group and 91.56 (95%CI: 90.68-92.43) in the Senior group (P = 0.008). The mean difference was +1.99 points.

· Scenario 5: the mean score was 93.85 (95%CI: 92.56-95.14) in the Junior group and 95.55 (95%CI: 94.95-96.15) in the Senior group (P = 0.019). The mean difference was -1.70 points.

DISCUSSION

In this retrospective, simulator-based analysis, chronological age (≤ 35 years vs. > 35 years) was not a consistent determinant of technical performance during a standardized, high-fidelity Rezūm pathway. Mean scores were high across all five scenarios, and between-group differences were scenario-dependent. Performance was essentially identical in the first two tasks. The Junior cohort achieved higher scores in the large bilobar prostate scenario (Scenario 3; 84.98 vs. 80.69, P = 0.001) and in the altered outlet geometry configuration (Scenario 4; 93.55 vs. 91.56, P = 0. 008), whereas the Senior cohort showed higher performance in the small-volume precision task (Scenario 5; 95.55 vs. 93.85, P = 0.019). Collectively, these findings suggest that, within a structured and objectively scored curriculum, generational differences, often presumed to reflect variability in prior endoscopic exposure, do not translate into systematic gaps in simulator-derived competence. Instead, task design and cognitive and technical demands appear to influence where performance dispersion emerges. Although statistically significant in selected scenarios, between-group differences were modest in magnitude within an overall high-performing cohort, suggesting limited practical separation in simulator-derived competence.

The scenario-dependent pattern observed in the present study can be interpreted as a curriculum signal rather than a demographic label.

Although chronological age has sometimes been considered a surrogate for differences in endoscopic experience or familiarity with digital technologies, these variables were not directly measured or controlled for in the present study. Therefore, the observed differences should not be interpreted as evidence of causal relationships, but rather as hypothesis-generating findings that warrant further investigation.

In structured simulation, between-group differences often emerge in tasks that increase cognitive load (e.g., extended injection planning, multi-lobe sequencing, and outlet geometry changes) or impose higher visuomotor demands under time and irrigation constraints. The Junior cohort’s advantage in Scenarios 3 and 4 may be consistent with differences in visuomotor performance in a technology-mediated environment under higher cognitive load. However, because these factors were not directly measured or controlled for, this interpretation should be considered speculative and hypothesis-generating. In contrast, the Senior cohort’s trend toward higher performance in Scenario 5 may reflect consolidated endoscopic heuristics and more conservative landmark discipline in constrained anatomy. A standardized, objective, simulation-based pathway may function as a “competency equalizer,” limiting the practical impact of heterogeneous prior exposure and producing reproducibly high technical execution across age groups.

From a clinical and implementation standpoint, these results are relevant because BPH/LUTS remains a high-burden condition with expanding demand for procedural solutions, and contemporary guidelines increasingly position MISTs as scalable options between long-term pharmacotherapy and more invasive transurethral surgery^[1-3]. The proliferation of device-based therapies creates a training bottleneck, as procedural reproducibility depends on consistent anatomical targeting, disciplined adherence to stepwise device algorithms, and avoidance of seemingly minor technical deviations that may have outsized consequences for retreatment patterns and adverse events. In this setting, training is not merely preparatory; it functions as a clinical governance tool to standardize quality as technologies disseminate beyond expert centers.

Against this backdrop, competency-based education and simulation are increasingly central to modern surgical training. Contemporary evidence syntheses in urology indicate that simulation-based education can improve technical performance and support structured assessment, particularly when curricula are standardized and grounded in objective metrics rather than purely time- or volume-based exposure^[8-10]. PBP provides a particularly strong educational rationale, as meta-analytic data across randomized and prospective studies show that PBP reduces procedural errors and improves performance compared with conventional simulation or unstructured practice^[9]. In BPH surgery specifically, a recent scoping review highlights the rapid expansion of endoscopic BPH simulators and underscores a key gap in the field: the need for robust validation pathways and stronger linkage between simulator metrics and clinically meaningful endpoints^[11].

Learning-curve effects are well described across endourology and BPH surgery, particularly for procedures that combine complex psychomotor control with intraoperative planning. AEP has been characterized by a substantial learning curve before efficiency and outcomes stabilize, reinforcing the value of structured mentorship and objective assessment during adoption^[12]. Similar experience-related effects have been reported for GreenLight photovaporization, where stepwise, proctored programs have been proposed to support safe progression and reduce early variability^[13]. Importantly, even when MISTs are engineered to simplify treatment delivery, measurable learning curves can still be observed: Aquablation shows progressive improvement across multicenter experiences, and prostatic urethral lift (PUL) has a relatively short but discernible learning trajectory, with performance and efficiency influenced by operator experience and case complexity^[14-16]. Collectively, these observations argue that a “minimally invasive” label should not be conflated with “training-light,” particularly when broad dissemination is anticipated.

In contrast, despite the rapid diffusion of Rezūm as an outpatient-oriented option for BPH, the evidence base on Rezūm-specific training remains comparatively sparse. A recent European Association of Urology (EAU) Endourology systematic review focusing on simulation-based training for Rezūm and UroLift concluded that the overall literature is limited and that key validity domains (including predictive validity and standardized performance endpoints) remain underexplored^[17]. Empirically, only a small number of studies have specifically evaluated Rezūm simulator performance metrics and validity in trainee populations, leaving a clear knowledge gap regarding how simulator-derived proficiency relates to operator characteristics and, ultimately, to clinical execution^[18].

Rezūm is particularly suitable for simulation-based training because procedural effectiveness depends on several discrete, measurable elements, including accurate identification of the bladder neck and verumontanum, correct needle deployment depth, appropriate number and spacing of injections by lobe, and tailored management of median lobe obstruction. These steps map naturally to objective simulator endpoints (accuracy, procedural accuracy, and adaptability), supporting the role of high-fidelity simulation as part of a preclinical competency assessment prior to broader clinical implementation. Clinically, Rezūm is supported by randomized, sham-controlled evidence demonstrating durable 5-year improvements in LUTS and flow parameters with low retreatment rates, while preserving sexual function in appropriately selected patients^[19]. More recent clinical evidence has further confirmed the favorable safety and tolerability profile of Rezūm across different patient populations, supporting its role as a reliable minimally invasive treatment option for BPH^[20]. Real-world and expanded-indication cohorts further suggest feasibility in challenging settings such as catheter-dependent retention and larger prostates, although outcomes vary by population and highlight the importance of appropriate anatomical strategy^[21-23]. In this context, large-scale, standardized simulator datasets may be particularly valuable for clarifying how operator factors relate to objective performance and for informing transparent credentialing frameworks. More broadly, these developments are consistent with the emerging paradigm of intelligent surgery, in which artificial intelligence, high-fidelity simulation, and data-driven performance assessment are increasingly integrated to support personalized surgical training, objective competency evaluation, and continuous performance improvement^[24].

From a training-design perspective, our findings support a pragmatic model for MIST implementation. Training should begin with standardized didactic instruction and then employ high-fidelity simulation with objective performance metrics to verify competence prior to clinical practice. Educational resources should be primarily directed toward scenarios that show the greatest performance dispersion (here, the large bilobar prostate scenario and the altered outlet geometry), because these tasks impose higher cognitive load and spatial-planning demands and are most likely to expose variability in execution. This framework is consistent with competency-based progression and supports credentialing based on predefined, measurable proficiency thresholds rather than demographic characteristics or loosely defined “experience”^[9].