Effect of pelvic floor muscle electrical stimulation on lumbopelvic control in women with stress urinary incontinence: randomized controlled trial

ABSTRACT Introduction The pelvic floor muscle (PFM) plays a role not only in lumbopelvic stabilization, but also in incontinence and sexual function. Objective This study aimed to determine the effectiveness of PFM training by electrical stimulation (ES) on urinary incontinence, PFM performance (i.e. strength and power), lumbopelvic control, and abdominal muscle thickness in women with stress urinary incontinence (SUI). Methods Participants were randomized into ES and control groups. The ES group underwent PFM ES for 8 weeks, whereas the control group underwent only a walking program. The impact of urinary incontinence on quality of life was assessed by the Incontinence Impact Questionnaire (IIQ)-7. PFM strength and power were measured using a perineometer. Lumbopelvic control was measured by one and double-leg-lowering tests. Abdominal muscle thickness was measured by sonography. Results The ES group showed significantly improved IIQ-7 scores and PFM performance, and had significantly higher values in both one and double-leg lowering tests (p < .05) after 8 weeks of training, indicating significant improvement from pre-session values (p < .005). There were no significant between- or within-group differences at rest in abdominal muscle thickness. Conclusion PFM ES could improve lumbopelvic control and PFM performance, and reduce subjective symptoms of urinary incontinence in women with SUI.

For lumbopelvic control and stabilization, strengthening programs targeting the abdominal muscles focus on deep abdominal muscle contraction, improvement of motor control, and LBPP reduction (Ehsani et al., 2020;Hwang et al., 2020a;Richardson, Jull, Hodges, and Hides, 1999). Additionally, PFM training is beneficial for improving LBPP at both end of treatment and 3-month follow-up (Mohseni-Bandpei, Rahmani, Behtash, and Karimloo, 2011). However, there have been no reports based on functional tests demonstrating that PFM training directly affects lumbopelvic control. The cause-and-effect relationship between lumbopelvic control and LBPP is still unclear (Jung et al., 2020). However, decreased lumbopelvic control causes faulty lower limb movement, and when persistent, can cause LBPP (Desai and Marshall, 2010;Vasseljen, Unsgaard-Tøndel, Westad, and Mork, 2012).
Electrical stimulation (ES), a form of PFM training, has mainly been used to reduce urinary leakage, and improve female sexual function and PFM contraction strength by facilitating volitional PFM contraction (Castro et al., 2008;Correia, Pereira, Hirakawa, and Driusso, 2014;Min et al., 2017). Additionally, we demonstrated that PFM ES could improve PFM performance (i.e. strength and power) (Hwang et al., 2020b). However, whether PFM training via ES can affect and promote lumbopelvic control should be confirmed. This study aimed to examine the effectiveness of 8 weeks of ES for improving the subjective symptoms of urinary incontinence, PFM performance, lumbopelvic control, and abdominal muscle thickness in women with SUI. We hypothesized that PFM ES would improve the subjective symptoms of urinary incontinence assessed using the Incontinence Impact Questionnaire-7 [IIQ-7], PFM performance (i.e. muscle strength and power), and lumbopelvic control (i.e. one and double-leg lowering tests).

Subjects and design
This study was performed between August and December 2018 in a urogynecology clinic and laboratory in Seoul, Korea. Subjects were randomized into ES and control groups, and the investigators were blinded to the group assignments. The required sample size was 16 subjects per group, using power, effect size, and α values of 0.80, 0.682, and 0.05, respectively, which was calculated a priori using G*Power (version 3.1.3; University of Kiel, Kiel, Germany), with reference to pilot data (n = 3 subjects per group) and with the main outcome of lumbopelvic stability (one-leg lowering test). Subjects were recruited by advertisements that provided a telephone contact; all volunteers were invited and evaluated in terms of the inclusion and exclusion criteria shown in Table 1. The participants were then asked to sign a written informed consent. Table 2 shows the characteristics of the subjects. A total of 33 subjects who met the inclusion criteria were divided into the control and ES groups using a list of random numbers (www.randomization.com) (Figure 1). Before the start of the study, all procedures were extensively explained, and all the subjects signed a written informed consent form, which included the entire protocol, its rationale and objectives, and safety considerations.
The study protocol was approved by the Yonsei University Mirae Campus Institutional Review Board (approval no. 1041849-201806-BM-056-02). Because the Clinical Research Information Service (CRIS) registration process is time-consuming, the university's review board allows trial commencement in a case-to-case basis, with permission granted by the institution. The first participant was recruited based on the ethical board approval (granted on July 4, 2018). The study protocol was registered with CRIS (KCT0003357) (granted November 11, 2018) and the registration timing was retrospective. The authors confirmed that all ongoing and related trials for this intervention were registered.

Electrical stimulation
The ES device (EasyK7, Alphamedic Co., Ltd., Daegu, Korea) employed three transcutaneous electrodes, which were placed in both the perivaginal (two electrodes) and sacral regions (one electrode) of the subject to create an electromagnetic field that stimulates the PFM and surrounding structures upon sitting on the device (Hwang et al., 2019b). The three electrodes create an electromagnetic field that stimulates the PFM over a wide area when the subject sits on the device ( Figure 2). ES was applied as biphasic asymmetric  impulses at 25 Hz, with on and off times of 11 seconds each. The mean current intensity was 17.63 ± 7.47 mA (range: 2.5-30 mA). Each ES session lasted for 15 min.

Intervention
Subjects in the ES group underwent their first session in our laboratory, where they were provided with an ES device and were taught how to use, manage, and clean it. Subjects were asked to use the device for 15 min once a day (i.e. 15 min session, as set by the manufacturer), for 5-6 days a week for 8 weeks, according to the previously outlined training frequency (Hwang et al., 2019b). In addition, the ES group determined the minimum electrical amplitude enough to feel the inward displacement and lift of perineum in the initial session as individually tolerated by increasing amplitude. We guided the subjects on utilizing above the measured electrical amplitude threshold, and monitored them by telephone calls twice weekly. The control group walked > 20 min daily. Adherence to this schedule was checked by telephone calls twice weekly. Both groups were assessed before and after 8 weeks of training.

Incontinence Impact Questionnaire-7 (IIQ-7)
The IIQ-7 was designed to assess the impact of urinary incontinence on health-related quality of life (Uebersax, Wyman, Shumaker, and McClish, 1995). The IIQ-7 is a seven-item life-impact assessment instrument specific to urinary incontinence, and is composed of three domains: 1) physical activity; 2) travel; and 3) social/relationships and emotional health. The subjects graded how much they experience impaired daily function of urinary incontinence  in the previous week using four response choices (0, "not at all"; 1, "slightly"; 2, "moderately"; and 3, "greatly") (Uebersax, Wyman, Shumaker, and McClish, 1995). The mean score was multiplied by 100/3 to modify it to a 0-100 scale. Higher scores indicated a greater impact on daily life (Uebersax, Wyman, Shumaker, and McClish, 1995).

PFM performance
PFM assessments were performed by a urogynecologist using a vaginal pressure measurement device with all participants in the hook-lying position (Tennfjord, Engh, and K, 2017). We used a VVP-3000 perineometer (QLMED Ltd., Gyeonggi-do, Korea) and a 115 mm long and 24 mm thick vaginal probe (active surface measurement length 66 mm). The baseline value (i.e. no voluntary PFM contraction) was recorded in mmHg and the device was then zeroed at rest. PFM strength was measured from baseline to peak effort > 2 s, and was reported in mmHg as the mean pressure rise during two maximal voluntary contractions (MVCs) (Tennfjord, Engh, and K, 2017). To measure PFM power, all subjects were asked to contract the PFM as rapidly as possible. PFM power was defined as peak pressure/time to MVC (mmHg/s) (Hwang et al., 2019b).

Lumbopelvic control: one and double-leg lowering tests
The one and double-leg lowering tests were used to measure lumbopelvic control during lower limb movement (Haladay, Denegar, Miller, and Challis, 2015;Haladay, Miller, Challis, and Denegar, 2014;Hwang et al., 2020a;Jung et al., 2020;Richardson, Jull, Hodges, and Hides, 1999). In the supine position, the subject flexed the hips and knees in a 90° angle. A Smart KEMA pressure sensor (KOREATECH Co., Ltd., Seoul, Korea) was set to 40 mmHg and placed below the lordotic curvature of the spine between L1 and S1 ( Figure 3) (Hwang et al., 2020a;Jung et al., 2020). Using its strap, the Smart KEMA motion sensor (KOREATECH Co., Ltd.) was attached to the thigh between the greater trochanter and knee joint. During performance of the abdominal drawing-in maneuver, the pressure on the sensor was increased by 10 mmHg. Subjects were asked to hold the lumbopelvic position by contracting the abdominal muscles while slowly lowering one or both legs to the supporting surface to maintain a preset pressure of 50 mmHg (Hwang et al., 2020a). One and double-leg lowering (i.e. hip extension) angles were measured with a motion sensor, and lumbopelvic control was defined as the moment when the pressure sensor reading decreased below 50 mmHg ( Figure 3). As the core muscles are necessary for lumbopelvic control and stabilization during leg motion, larger leg-lowering angles indicated greater lumbopelvic control (Jung et al., 2020). Lumbopelvic control was assessed by measuring the success of leg lowering while maintaining lumbar curvature (Haladay, Denegar, Miller, and Challis, 2015;Jung et al., 2020;Sahrmann, 2001). Quantifying lumbopelvic control is difficult because previous leg lowering tests evaluated lumbopelvic control using an ordinal scale correlating to success. Both leg-lowering tests in our study quantified lumbopelvic stability by measuring the hip extension angle instead of performance success (Hwang et al., 2020a(Hwang et al., , 2019aJung et al., 2020Jung et al., , 2018.

Abdominal muscle thickness
A real-time ultrasound scanner (A35; Samsung Medison, Seoul, Korea) was used to measure the thickness of the TrA, internal oblique abdominis (IO), and external oblique abdominis (EO) muscles on the right side of the abdominal wall in M-mode, using a 4.5 cm, 3-16-MHz linear probe (LA3-16A) connected to a screen that showed the image. Calipers in centimeters were used to measure muscle thickness. Three trials were performed for each task.
To obtain resting thickness measurements, all subjects were placed in supine position, with the examiner on the subject's dominant side, which was confirmed by performing a ball-kicking test. To standardize the location of the transducer, the hyperechoic interface between the TrA and the thoracolumbar fascia was positioned on the dominant side of the ultrasound image. All images were taken at the end of expiration.

Statistical analysis
All statistical analyses were performed using SPSS (ver. 18.0; SPSS Inc., Chicago, IL, USA). In all analyses, p < 0.05 was set to be statistically significant. The Kolmogorov-Smirnov Z-test confirmed data normality. Two-way repeated measures analysis of variance was used to examine time × group interaction effects on PFM performance, lumbopelvic control, and abdominal muscle thickness at rest. Whenever a significant interaction was observed, the paired t-test was used to determine within-group differences, and the independent t-test was used to determine differences between the ES and control groups.

Results
There were no significant differences between the groups in all baseline parameters (IIQ-7, PFM strength and power, one and double-leg lowering test, TrA, IO, and EO thickness at rest). Table 3 shows the within-groups comparison for IIQ-7, PFM performance, lumbopelvic control and abdominal muscle thickness for both groups.

Incontinence Impact Questionnaire-7 (IIQ-7)
There were significant time × group interactions for IIQ-7 (p = 0.000) as well as effects of group (p = 0.038) and time (p = 0.000). The ES group showed significantly lower post-training values than the control group for IIQ-7 (p = 0.000). IIQ-7 (p = 0.000) was significantly decreased after 8 weeks of training in the ES group (Figure 4). In contrast, there were no significant differences between pre-and post-training PFM performance in the control group.

PFM performance
There were significant time × group interactions for PFM strength (p = 0.012) and power (p = 0.000). For PFM strength, there was no significant effect based on group (p = 0.145), but a significant difference based on time (p = 0.006). There were mainly significant effects of groups (p = 0.035) and time (p = 0.002) for PFM power. The ES group showed significantly greater post-training values than the control group for PFM strength (p = 0.033) and power (p = 0.002). PFM strength (p = 0.009) and power (p = 0.001) significantly increased after 8 weeks of training in the ES group (Figure 4). In contrast, there were no significant differences between pre-and post-training PFM performance in the control group.

Lumbopelvic control: one and double-leg lowering tests
There were significant time × group interactions for the one-leg lowering test in the right (p = 0.002) and left legs (p = 0.001), and for the double-leg lowering test (p = 0.000). For the one-leg lowering test, there was a significant difference due to time (right, p = 0.001; left, p = 0.000), but not due to group (right, p = 0.240; left, p = 0.180). For the double-leg lowering test, there was a significant difference due to time (p = 0.000), but not due to group (p = 0.674). The ES group showed a significantly greater hip angle in the one-leg lowering test for the right (p = 0.012) and left legs (p = 0.004), and in the double-leg lowering test (p = 0.041), than the control group after training, as well as significant increases in values for the one-leg lowering test in the right (p = 0.002) and left legs (p = 0.001), and for the double-leg lowering test (p = 0.000). In contrast, the control group showed no significant differences between pre-and post-training values. Table 3. Impact of urinary incontinence, pelvic floor muscle performance, lumbopelvic control, abdominal muscle thickness, and contraction ratio values based on comparing pre-and post-training for both groups (mean ± SD).

Discussion
We hypothesized that PFM training using ES would improve impact of urinary incontinence, PFM performance, lumbopelvic control, and abdominal muscle thickness. The results of the present study demonstrated that PFM training using ES was beneficial for improving not only PFM performance, but also lumbopelvic control in women with SUI. In addition, we measured the abdominal muscle thickness to confirm whether the effect of PFM training using ES on lumbopelvic control was caused by the improvement of PFM performance, increase of abdominal muscle thickness or both. Previous studies suggested that voluntary pelvic floor contraction indicated co-contraction of the abdominal muscles (Madill and McLean, 2006;Smith, Coppieters, and Hodges, 2007). However, this study did not confirm significant differences between pre and post-PFMtraining abdominal thickness, although directly comparing PFM ES effects with those from voluntary contraction is difficult. Further studies are warranted to determine the effect of PFM training using ES on neural adaptation of abdominal muscle during lumbopelvic control. This study illustrated the ability of PFM training to improve lumbopelvic control in women with SUI. Although the cause-and-effect relationship between lumbopelvic stability and LBPP is controversial, PFM training using ES appears to improve lumbopelvic control in women with SUI by improving PFM performance in the absence of abdominal muscle adaptations.
PFM training is the first guideline in SUI intervention to decrease urethral hypermobility, and to increase PFM performance and urethral pressure. PFM ES could reduce urinary loss and increase intravaginal pressure and PFM strength by facilitating the ability to volitionally contract the PFM (Castro et al., 2008). Although the present study used a different method, previous RCTs compared with no treatment (control), ES could significantly improve quality of life in women with SUI (Castro et al., 2008;Correia, Pereira, Hirakawa, and Driusso, 2014;Pereira, Bonioti, Correia, and Driusso, 2012). In this study, we also found a statistically significant improvement in the quality of life after PFM ES.
A possible reason for improvement of the quality of life after PFM ES is similar to the results of previous studies. With improved PFM performance after PFM ES, the mechanism of urethral closure could improve during situations of increasing intraabdominal pressure. Requirement of the PFM performance for preventing urinary leakage is increased during lifting, jumping, and running (Correia, Pereira, Hirakawa, and Driusso, 2014). Additionally, comparing between the relative changes in symptoms of urinary incontinence and PFM performance, the relative changes in urine loss [(post-treatment value/pre-treatment value) × 100] decreased as relative change in PFM performance increased after eight weeks of PFM training using ES (R 2 = 12.5%, β = −0.35) (Hwang et al., 2020b). Thus, the decrease in IIQ-7 after eight weeks of PFM training using ES is possibly due to increased PFM performance during increased intra-abdominal pressure by SUI.
The core muscles (i.e. diaphragm, IO, TrA, multifidus, and PFM) provide necessary control for lumbopelvic stabilization during limb movement (Marshall, Desai, and Robbins, 2011;Marshall and Murphy, 2010). To accurately assess the effects of interventions on core muscle performance, and to develop more effective exercise programs, clinicians require an objective and quantitative measure of core muscle performance and/or motor function (Jung et al., 2020). Leg-lowering tests show moderate to strong associations with rectus abdominis (RA) activity, and moderate associations with both IO/TA and EO and hip joint moments were required for counteracting hip joint moments to hold a stable pelvic position during the test (Haladay, Denegar, Miller, and Challis, 2015). Thus, we used the one and double-leg lowering tests for measuring lumbopelvic control, and demonstrated that lumbopelvic control could be improved by PFM ES.
The enhanced lumbopelvic control seen after PFM training using ES may be explained as follows. First, improved PFM performance could affect pelvic stability. A cadaveric study by Pool-Goudzwaard et al. (2004) indicated that simulated PFM tension significantly stiffened the sacroiliac joints by 8.5%, which also suggested that increased PFM activity may improve inadequate lumbopelvic stability and the ability to transfer load through the lumbopelvic region. Thus, improved PFM performance after 8 weeks of training may improve lumbopelvic stability and the ability to transfer leg lowering load through the lumbopelvic region. Second, because improved PFM performance could affect the urethral pressure more than the vesical pressure during periods of increased intraabdominal pressure, abdominal muscle recruitment may be increased in one and double-leg lowering tests. The RA, EO, and IO/TrA muscles are primarily responsible for pelvic control during the leg-lowering test (Haladay, Denegar, Miller, and Challis, 2015;Haladay, Miller, Challis, and Denegar, 2014). Since global muscles, such as RA and EO, could especially increase intraabdominal pressure, good contraction timing and high PFM strength are needed to ensure greater urethral pressure than vesical pressure (Bø, 2003). Third, lumbopelvic control may be improved through certain psychological and emotional effects, such as decreased fear of incontinence and avoidance of physical loading and exertion. Urinary loss occurs with increased intraabdominal pressure, such as during coughing, sneezing, or exertion, in women with SUI (Amaro, Moreira, Gameiro, and Padovani, 2005;Bø, 2003).
This study indicated that abdominal muscle thickness did not change significantly after PFM training using ES. This may be explained by differences in neuromuscular adaptation and skeletal muscle conditioning between artificial and voluntary contractions (Hortobágyi and Maffiuletti, 2011). Improved MVC could result from spinal or supraspinal neural adaptations (Hortobágyi and Maffiuletti, 2011). During voluntary compared to ES-evoked contraction, muscle activation is synergistic rather than targeted, and antagonist muscle activation is coordinated rather than uncoordinated (Hortobágyi and Maffiuletti, 2011). In addition, as we did not directly apply ES to abdominal muscles, the PFM training using ES would not have stimulated abdominal muscles sufficiently to increase muscle thickness. Therefore, the training could not have achieved a sufficient load to alter the abdominal muscle thickness.
However, co-contraction of the PFM and abdominal muscles could not be confirmed by electromyography (EMG) during one or double-leg lowering tests. Madill and McLean (2006) reported that, with a voluntary PFM contraction, there was an increase in maximum electrical activity of 9.61% ± 7.42% in the RA, 224.3% ± 47.4% in the TrA, 18.72% ± 13.33% in the EO, and 81.47% ± 63.57% in the IO based on surface EMG data (Madill and McLean, 2006). Additionally, the recruitment of abdominal muscle in association with voluntary contraction of PFM may be affected by the position of the spine . Thus, although the present study did not include any EMG measurements of the PFM or abdominal muscles, PFM ES may increase cocontraction of the PFM and abdominal muscles during one and double-leg lowering tests after 8 weeks of training.
The principal limitation of this study was the lack of EMG data to assess changes in PFM and abdominal muscle activation. Further studies are needed to determine the influence of improved lumbopelvic control on LBPP using PFM ES. In addition, although we included women with varying ages, including both preand postmenopausal women, studies that has larger samples with PFM dysfunction are required.

Conclusion
In conclusion, our results demonstrated that PFM performance and lumbopelvic control were significantly increased, and the impact of urinary incontinence on quality of life was significantly decreased after PFM ES in women with SUI. However, the abdominal muscle thickness was not significantly different before and after ES. PFM training using ES can be considered an option for improving lumbopelvic control, PFM performance, and impact of urinary incontinence in women with SUI.