Participants

Participants were patients aged 18 to 80 years with symptomatic and objective SUI who wanted to pursue conservative management. Symptomatic SUI was defined as urinary leakage on exertion. Objective SUI was demonstrated by a positive cough stress test (CST) with a comfortably full bladder followed by a uroflow, urodynamic stress incontinence, or positive 24-hour pad weight test. If a participant was unable to demonstrate a positive CST, they were eligible to be included in the study with a positive pad weight test in the absence of urgency incontinence symptoms. The 24-hour pad weight test does not have a standardized abnormal value. Normal mean values across 4 studies ranged from 0.3 to 4 g. Hence, a result was considered positive with a 24-hour pad test >4 g. Participants had participated in or declined PFMT previously and had a negative urinalysis and a normal cervical screening test within the recommended time frame. Exclusion criteria included POP stage >2, pregnancy, previous surgery in the treatment area within 12 months, active genital infection, systemic steroid use in last 3 months, vaginal lubricant use within 7 days, recurrent urinary tract or genital herpes, transvaginal mesh or sling implant, and serious systemic disease or chronic condition that could interfere with study compliance.

Intervention

The intervention was performed using the FemTouch (Lumenis, Yokneam Illit, Israel) delivery system in conjunction with the AcuPulse CO ₂ laser system (Lumenis) and the AcuScan 120 Microscanner (Lumenis) to provide fractional treatment. Using the scanner, the laser energy is delivered by depression of a foot pedal in a fractional manner, from 7.5 to 12.5 mJ with a 10% to 15% density and a fixed penetration depth of 600 microns, starting from the proximal vagina and withdrawn at 1-cm intervals to cease 5 mm above the introitus distally. Participants underwent 3 treatments, 4 weeks apart, with increasing energy and density settings. Treatment 1 delivered 7.5 mJ at a density of 10%. Treatments 2 and 3 delivered 10 mJ and 12.5 mJ of energy, respectively, both with a density of 15%. Energy was delivered circumferentially to the entire vagina with 1 pass. An additional pass was performed at the 10-o’clock and 2-o’clock position, and additional 2 passes were performed at 12-o’clock position at each 1-cm interval. Sham treatment was performed with an identical protocol; however, an inactive pedal was used so no energy was delivered.

Outcomes

The primary outcome was subjective SUI, defined as a positive answer to “Do you leak when you cough or sneeze?” or “Do you leak when physically active/exercising?” on questions 6c or 6e of the International Consultation on Incontinence Questionnaire (ICIQ)-Urinary Incontinence Short Form (UI SF).

Secondary outcomes were assessed by the Patient Global Impression of Improvement (PGI-I) and change in ICIQ-UI SF, the ICIQ-Overactive Bladder (OAB), the Incontinence Impact Questionnaire Short Form (IIQ-7), the Pelvic Organ Prolapse/Urinary Incontinence Sexual Function Questionnaire (PISQ-12) short form, and EQ-5D scores. Furthermore, objective SUI and adverse events (pain during the procedure measured by a visual analog scale [VAS] score [0–10] and any participant complaints) were recorded.

PGI-I gives an overall indication of improvement for a specified condition, with 7 items ranging from “very much better” to “very much worse.” The ICIQ-UI SF has a scoring scale from 0 to 21 (with higher numbers indicating worse symptoms) and allows stratification of severity grade on the basis of the overall score. ^, Therefore, the participants were also assigned categories of slight (0–5), moderate (6–12), severe (13–18), and very severe (19–21) urinary incontinence. The minimally important difference has been estimated at 2.52. The ICIQ-OAB is a questionnaire evaluating OAB with a scoring scale of 0 to 16. More severe symptoms are indicated by a higher score. The IIQ-7 is a 7-item questionnaire that assesses the life impact of urinary incontinence. Each item receives a score from 0 to 3 for a scale from 0 to 21. A cutoff of 9 points has been suggested as indicating a higher impact of SUI on quality of life. The PISQ-12 assesses change in sexual function and has 12 items with a total score of 0 to 48, with higher scores suggesting greater dysfunction. The EQ-5D was used to measure generic health-related quality of life. EQ-5D offers 3 scoring levels between 1 and 3 for 5 domains (mobility, self-care, ability to perform usual activities, pain or discomfort, and anxiety or depression), with lower levels representing better overall quality of life. During the trial, the question “Do you think you received active treatment or sham treatment?” was added after approximately 40% of participants were recruited to measure the success of blinding.

The outcomes were collected at baseline, 1 month, and 3 months after the last treatment session. The primary outcome time point was at 3 months.

Sample size calculation

The initial sample size was calculated to detect a clinical improvement of 20% in the intervention arm, assuming a cure rate of 56% in the comparison (sham) arm that was based on the cure rate of supervised PFMT. This was adjusted during the study because of the publication of the first RCT comparing Er:YAG laser with sham treatment. On the basis of this RCT, we estimated a dry rate of 22% and 3% in laser and sham participants, respectively. This led to a sample size of 101 to provide 80% power (α=.05) to detect a difference between groups, assuming an attrition rate of 10%.

Randomization and blinding

Participants were randomized with equal probability to CO ₂ laser or sham treatment using a central computer-generated random sequence allocating participants using blocks of 4, 6, and 8. The sequence was unknown to the investigators and the randomization code was released to the treating doctor after the participant had been recruited.

Participants but not investigators (including those involved in treatment and data collection) were blinded to the allocation throughout treatment and follow-up. Participants were informed of their allocation after the 3-month review. The sham treatment used identical settings and treatment protocol, but an inactive pedal was used, which made a clicking noise but not an electronic beep.

Statistical analyses

All categorial outcomes were compared using the chi square test. The treatment effects in dichotomous outcomes are expressed as relative risk (RR) with 95% confidence intervals (CIs).

Change scores in ICIQ-UI SF, ICIQ-OAB, IIQ-7, and PISQ-12 were compared using analysis of covariance (ANCOVA), entering the baseline score as a covariate to the model. Missing cases were excluded.

Scores in EQ-5D health states, pad weight volumes, and mean participant pain scores were not normally distributed, and medians were compared using the nonparametric Mann–Whitney U test.

The assumption of homogeneity of variances was violated for the ICIQ-OAB score at 3 months, as assessed by Levene test for equality of variances. We performed a sensitivity analysis transforming the data to convert skewed data to normality. Moreover, there was 1 outlier (>3 standardized residuals from the mean) in the PISQ-12 score at 1 month. Data were checked again for accuracy before running ANCOVA testing with (primary analysis) and without (sensitivity analysis) the outlier to assess how the outlier affected the results.

All analyses followed an intention-to-treat principle. The statistical package IBM SPSS Statistics for Windows, version 26 (IBM Corp, Armonk, NY) was used for analysis.

Funding

Partial funding was obtained during the study from the Continence Foundation of Australia (M21005/3176612). The funder did not play any role in conducting the study or writing the paper. The CO ₂ laser was loaned to the research group by the manufacturer. The manufacturer did not have any involvement in recruitment, treatment, follow-up, analysis, or decision for publication. Participants were not remunerated for participation in the study, and they did not incur any costs.