Single-operator learning curve analysis of transvaginal fractional CO₂ laser treatment for female stress urinary incontinence: a subgroup study based on vaginal mucosal stratification

<title>Abstract</title> <p> <bold>Purpose:</bold> Stress urinary incontinence (SUI) is a prevalent condition that significantly impairs quality of life, and transvaginal fractional CO₂ laser has emerged as a minimally invasive treatment; however, its learning curve remains poorly characterized. This study aimed to investigate the learning curve of a single operator performing transvaginal fractional CO₂ laser for SUI, to conduct subgroup analysis stratified by vaginal mucosal status, and to introduce withdrawal precision and rotation precision as objective, quantifiable hand-stability metrics for assessing operator proficiency. <bold>Methods:</bold> This retrospective study included 80 female patients with SUI who underwent transvaginal fractional CO₂ laser therapy (wavelength 10600 nm) at a single center between January and June 2026. All treatments were performed by one operator (deputy chief nurse, 29 years of experience). Patients were stratified into Subgroup A (normal mucosa, n = 43) and Subgroup B (atrophic mucosa, n= 37) based on intraoperative mucosal assessment. Cumulative sum (CUSUM) control charts were used to analyze learning curves, with 5-case rolling average cure rates for cross-validation. Multivariate logistic regression identified independent predictors of cure. Pearson correlation analysis evaluated relationships between hand-stability metrics and case sequence. Clinical trial number: not applicable. <bold>Results:</bold> CUSUM inflection points occurred at case 30 (Subgroup A) and case 33 (Subgroup B). Cure rates improved significantly from the learning to proficiency phases (53.3% vs. 92.3%, Subgroup A; 51.5% vs. 100.0%, Subgroup B; both P &lt; 0.01). Case sequence, power setting, and treatment endpoint achievement were independent predictors of cure (P &lt; 0.05). Withdrawal deviation and rotation deviation showed strong negative correlations with case sequence (r = −0.906 to −0.807, all P &lt; 0.001), while hand-stability scores showed strong positive correlations (r = 0.926–0.942, P &lt; 0.001). All 7 adverse events (8.8%) occurred during the learning phase and were mild. <bold>Conclusion:</bold> Transvaginal fractional CO₂ laser for SUI demonstrates a quantifiable learning curve, with longer training required for patients with atrophic mucosa due to altered laser–tissue interactions. Withdrawal precision and rotation precision serve as promising prototype metrics for objective proficiency assessment. The identified thresholds (withdrawal deviation &lt; 0.1 cm, rotation deviation &lt; 2°) are proposed as competency-based entry criteria for future clinical trials. </p>