Subjects and design
This was a randomized controlled trial (RCT) with an investigator-blinded parallel randomization (1:1) design (control and SESSP groups). The present study was performed at an obstetrics and gynecology clinic in Seoul, Korea, from August 2018 to November 2018. The sample size was determined a priori (version 3.1.3; University of Trier, Trier, Germany) in a pilot study with three participants in each group (experimental and control groups). The sample size was calculated according to a power of 0.80, alpha level of 0.05 and effect size f of 0.917. The analysis indicated that more than six subjects were required. The women were recruited via advertisements included provided telephone contact information for the study. After the initial contact, visits were scheduled to review the participant suitability according to the inclusion and exclusion criteria of the study, and incontinence severity was checked via the Ingelman–Sundberg scale, as administered during an interview .
Inclusion criteria  were 1) SUI diagnosed by a urogynecologist, 2) a leakage episode occurring more than once a week, 3) generally healthy (apart from SUI), 4) body mass index < 30 kg/m2, 5) age between 30 and 60 years, 6) non-smoker, 7) not addicted to alcohol or drugs, and 8) successfully completed medical screening questionnaire. Exclusion criteria  were 1) urogenital prolapse grade III or higher, 2) cardiac pacemakers, 3) devices implanted in the pelvis or hip joint, 4) pregnancy/planning to get pregnant, 5) pelvic or abdominal surgery within the last 6 months, 6) cognitive impairment, 7) concomitant treatment for SUI during the trial period, 8) neurological or psychiatric disease, and 9) urinary tract infection.
A total of 36 participants who met the inclusion criteria were allocated to control and SESSP groups through a list of random numbers generated by online randomization software (www.randomization.com) (Figure 1). Before the study, all procedures were explained to the participants, who then signed informed consent forms approved by the Institutional Review Board (1041849-201806-BM-056-01).
Surface electrical stimulation device
An SESSP device (EasyK7) with three cutaneous electrodes placed in perivaginal (two electrodes) and sacral regions (one electrode) was used to stimulate the PFM and surrounding structures. Three cutaneous electrodes were inserted into the bottom and back of the SESSP device. This allowed for stimulation, via direct contact, of both the perivaginal and sacral regions, and created an electromagnetic field that stimulated the PFM while the participant sat on the EasyK7 (Figure 2). The initial stimulation amplitude was set by the physical therapist to determine the most comfortable level for each participant. The EasyK7 delivered biphasic, asymmetric impulses of 25 Hz, and controlled the pulse (11 s) and rest (11 s) durations. The mean intensity among all participants was 19.13 ± 5.47 mA (range: 2.5 to 30 mA). Each EasyK7 session was 15 minutes in duration.
Participants in the SESSP group were provided with an EasyK7 device, underwent their first EasyK7 session on site, and were educated on proper device usage, management, and cleaning. Individuals who experienced the SESSP sensation to be aversive were excluded from the study. Participants were instructed to use the device once a day (15-min session) 5 days a week for 8 weeks. All participants participated an EasyK7 session designed to determine the maximum stimulation amplitude that they could tolerate.
The Control group underwent walking for 10 min and restricted PFMT regarding the PFM or abdominal muscle contraction. We gave and educated the EasyK7 to control group as a reward for participating in the experiment after 8 weeks. Both groups were assessed at baseline and after 8 weeks with repeated QOL questionnaires, as well as with ultra-short perineal pad tests and measurements of PFM functions performed using a perineometer.
The outcome of QOL was determined by patient self-assessment. Testers were blinded to the participant’s responses to the questionnaires. The I-QOL evaluated incontinence-specific QOL . The I-QOL consists of three health-related QOL factors: avoidance and limiting behaviors (eight items), psychosocial impact (nine items), and social embarrassment (five items); the Korean version was used in this study . The I-QOL has 22 items, each scored using a five-point scale. Three subscale scores and a total score are calculated. Scores for the 22 items were summed and then adjusted according to a 0–100 scale, with higher scores indicating higher QOL .
The objective status of incontinence was indicated by the ultra-short perineal pad test results and PFM power, strength, endurance, and TRMP was measured by perineometer. The ultra-short perineal pad test  was performed to assess urinary leakage status, as follows. The participants were allowed to empty their bladder, and any remaining urine was removed using a catheter without anesthetic gel. Then, the bladder was filled with 300 mL sterile water by a gynecologist using a catheter. For the pad test, the participants were provided with a pre-weighed sanitary diaper to place inside their underwear. Participants were instructed to perform a standardized physical activity for 1 minute at a submaximal level: jumping with feet apart and then together, jogging in place with high knees, and jumping up and down; 20 repetitions of each exercise and additional jogging in place were performed until 1 minute had passed . After 10 minutes, the total amount of urine leakage was determined by weighing the pad.
The assessments of PFM functions were performed in a hook-lying position for all participants using a vaginal pressure measurement device. The VVP-3000 perineometer (QLMED Ltd., Gyeonggi-do, Korea), which is a vaginal probe 24 mm in diameter and 115 mm in length with an active surface measurement length of 66 mm, was used. The vaginal probe is connected to a microprocessor with latex tubing, allowing for transmission of pressure readings when the inserted part is compressed by external pressure. The baseline pressure value was recorded in mmHg with no voluntary PFM contraction, and the device was then zeroed. Participants were asked to contract their PFM and squeeze with maximum effort for 2–3 seconds. They were instructed to pull their PFM in and up as much as possible, and with no use of extra-pelvic muscle contraction. The measurement was taken before the first contraction and was displayed as a flat curve, after the participants had been instructed to relax and given time to slowly breathe in and out . PFM strength was given by the difference between the resting and peak pressure measurement, and reported as the mean of two maximal pressures (in mmHg) . TRMP was measured from the onset of contraction until maximal pressure was reached, and recorded in seconds. Measurement of muscle power required the participants to perform the contraction at maximal pressure as fast as possible. The present study defined PFM power as PFM strength/TRMP (mmHg/s). Local muscular endurance is the ability of a muscle to sustain sub-maximal or maximal force, assessed as the time over which a person is able to maintain a maximal static or isometric contraction . PFM endurance was measured as the mean contraction pressure for 10 s, during one attempt (mmHg).
All statistical analyses were conducted using SPSS software (ver. 18.0; SPSS Inc., Chicago, IL, USA). A p value of 0.05 was used to indicate statistical significance. Kolmogorov–Smirnov Z-tests were used to verify that the data were normally distributed. Analysis of covariance (ANCOVA) was used to compare the groups before and after the intervention, with the baseline values used as covariates. Data are presented as mean ± standard deviation. The effect sizes and confidence interval (CI) for the primary outcomes were calculated to determine the clinical significance of the data. An effect size (r) is constrained between 0 (no correlation) and 1 (perfect correlation); in this study, 0 ≤ r < 0.1 was classified as no effect, 0.1 ≤ r < 0.3 as a small effect, 0.3 ≤ r < 0.5 as a moderate effect, and r ≥ 0.5 as a large effect.