Utilising Synergism Between the Transverse Abdominal and Pelvic Floor Muscles at Different Postures in Nulliparous Women: a Randomized Case Control Study

Background: To determine the effects of the pelvic oor muscle training (PFM-T) in combination with transverse abdominal muscle (TRA) activation (cPFM-T) in female urinary incontinence. Methods: We enrolled nulliparous women in supine (SUG) (n = 22), sitting (SIG) (n = 19) and control (COG) (n = 14) groups. We performed the 8-week cPFM-T programme. We examined the effect of training on the parameters with the Kruskal–Wallis test, and the pairwise comparisons with the Mann-Whitney U-test and the Wilcoxon-rank test with the Bonferroni correction. Results: Before training, 15 participants reported occasional urinary leakage. After cPFM-T seven participants reported that urinary leakage had disappeared. Maximal isometric contraction of the pelvic oor muscles (PFM) until fatigue improved signicantly in the SUG (p < 0.001) and SIG (p = 0.015) and not signicantly in the COG (p = 0.499). Holding time increased in the SUG (p = 0.972) and the SIG (p = 0.717), and decreased in the COG (p = 0.132). The dynamic endurance of the PFM improved signicantly in the SUG (p < 0.001), but not in the SIG (p = 0.798) and the COG (p = 0.153). The number of maximal fast contractions within 1 minute increased in both the SUG (p < 0.001) and the SIG (p=0.813) and decreased in the COG (p = 0.257). Relaxation improved signicantly in the SIG (p = 0.011). TRA thickness increased in both training groups. Conclusions: Slow-twitch bres of the PFM can be trained effectively with PFM-T in both the body positions.

The effectiveness of Kegel's PFM-T exercises-classi ed by the International Consultation on Incontinence as level A evidence in Evidence Based Medicine [3,4] has been proven in numerous randomised controlled trials.
Several research groups have reported a co-contraction between the deep abdominal muscles and the PFM [5,6,7,8].
Together, these results suggest the necessity of involving a coordinated approach of the PFM and the deep abdominal muscles in the effective treatment of UI.
Several studies have revealed signi cantly higher resting PFM activity in unsupported sitting as compared to supported sitting and in a standing posture as compared to a supine position [8,9,10,11].
Chmielewska [11] reported a signi cant a long-lasting contraction during unsupported sitting, due to the superior recruitment by the sitting posture of the sensorimotor control system to that by the supine position. They identi ed different coactivation patterns of the PFM and abdominal muscles during sitting to those during standing and lying.
Su cient evidence that regular PFM-T improves the symptoms of incontinence has been reported [3,12,13]. However, only a few studies support the effect of the coordination of the diaphragm, deep abdominal muscles and PFM on incontinence [14,15].
Current evidence on skeletal muscle training and other factors (UI, supervised PFM-T programme, e.t.c.) are known to have an impact on women's participation in and adherence to PFM-T [3].
Here we aimed to investigate whether-based on trunk muscle synergism-the condition and functioning of the PFM would improve in the sitting and supine postures or in the control group during PFM-T with forced exhalation. We hypotized: the training position affect the muscle gymnastic performance.

Participants
We performed our study at the Faculty of Health Sciences and Social Studies, University of Szeged, Hungary between October 1, 2016 and December 1, 2016. We enrolled 58 healthy, young (mean±SD: 21.27±1.46 years), nulliparous women in this trial. We recruited them through an online advertisement and then selected them for the training groups and the control group by T.M.(Additional le 3). The participants had no previous experience of PFM-T. We divided them into two study groups depending on the strength of the PFM and treated them with cPFM-T in both the supine position and the sitting position. The assessment comprised one phase: 3 repetitions of sustained 5-second voluntary PFM contraction with 10-second relaxation. We calculated the mean of 3 contractions and set the obtained values in ascending order. The 22 participants with lower muscle strength (under 60 mV) comprised the SUG and the 22 participants with higher muscle strength (over 60 mV) formed the SIG. Only 19 participants, however, completed the programme in the SIG (Additional le 3). We enrolled these 19 participants with higher PFM tension (over 60 mV) in the SIG because the PFM must be stronger to resist gravity [16]. We created a control group (n = 14) (COG) comprising seven persons with a PFM tension under 60 mV and seven individuals with a PFM tension over 60 mV. The COG did not change their lifestyles and did not undergo PFM-T. We included in the study groups participants willing to participate in the study and able to contract the pelvic oor and TRA muscles correctly. Participants were required to maintain their everyday activities (attending lessons, sport activities, and so on). The local ethics committee (National Healthcare Service Center) approved the study (019234/2014/OTIG) and we received written, informed consent from all participants. Exclusion criteria were known neurological or rheumatological diseases and previous vaginal or abdominal surgery.

Subjective measurements
In the study, before and after the training programme, we used a self-administered questionnaire (Additional le 7), based on three validated questionnaires (the King's health questionnaire, the Incontinence impact questionnaire and the Urogenital distress inventory) [17]. We included in the questionnaire risk factors for UI (obstetric history, height and weight, stress, physical activity level, sport, vaginal and abdominal surgery) and questions relating to the urinary tract and anal canal (involuntary urinary leakage, cystitis, constipation) and sexual activity (orgasm problems).

Objective measurements
Vaginal surface electromyography (vsEMG) We measured changes in PFM activity with a vaginal surface electromyographic (vsEMG) instrument (FemiScan™ MultiTrainer™, Mega Electronics, Finland), which measures the electronic signals of PFM activity, using a sterile Periform™ intravaginal probe with vsEMG electrodes [7]-a pear-shaped device 8 cm in length and 3.4 cm in medial-lateral diameter at its peak width, but tapering at the introitus, with nickel detection surfaces on both sides and an indicator to help patients to perform the tasks correctly. A correct voluntary contraction of the PFM contributes to downward (posterior) movement of the indicator [18]. We positioned a reference electrode on the patient's left forearm.

Procedure
We tested the participants in a lying position with the hips and knees exed, feet resting on a plinth and knees supported to allow the hips and PFM to relax [7]. We instructed the volunteers on the proper placement of a FemiScan™ vsEMG probe in situ within the vagina. The same physiotherapist (T.M.) assessed all participants twice (at 0 weeks and at 8 weeks), at which times PFM activities were measured in a supine position. Before each measurement, the therapist instructed the participants how to perform PFM tasks. The participants were not allowed to move the hips and lumbar spine. We displayed the vsEMG data as line graphs, thus providing visual feedback for the participants, and recorded the values in microvolts.

Page 5/17
The participants performed three tasks:

Transabdominal ultrasoundmeasurement
We performed the TRA measurements at the same time as the vaginal measurements and PFM tasks. We measured the degree of the change in the muscle thickness. We measured the thickness of the TRA because Madill and McLean [7] found that the synergistic coactivation between the TRA and the PFM was stronger than that between the rectus abdominis muscle (RA), the external oblique muscle (EO) and the PFM. Thus we decided that it was su cient to measure the thickness of the TRA only, assessing it by acquiring images of the TRA and the measurement techniques have been previously described [19]. She placed the centre of the transducer in the transverse plane just superior to the iliac crest in line with the mid-axillary line, performing the measurement of muscle thickness on the left side of the participant at rest and during the isometric tasks before and after cPFM-T. The participants were not allowed any movement of the hip and lumbar spine. We acquired the following recordings: (1) the thickness of the TRA during PFM relaxation, and (2) the TRA thickness during maximal isometric contraction till PFM fatigue. We measured the thickness of the TRA at the point where the muscle was at its thickest.
Pelvic oor muscle training A physical therapists (T.M.) supervised the training, and carried out the evaluation. We divided the participants into two study groups depending on the strength of the PFM and treated them with cPFM-T in both supine (n = 22) (SUG) and sitting (n = 19) (SIG) positions. We also set up one control group (n = 14) (COG). The treatment for the SUG and the SIG comprised 8 sessions, with a 1-hour cPFM-T session each week in a group and 15 minutes of individual home training, six times a week for a total of 8 weeks of treatment. Before the cPFM-T, we instructed all participants in the anatomy of the PFM and the lower urinary tract, the mechanism of continence and unsupported sitting posture. We discouraged the slumped sitting posture since unsupported sitting postures require greater PFM activity than supported sitting postures [9]. All training sessions comprised warming-up, gradual muscle strengthening and relaxation exercises. We found palpation and visualisation to be very powerful tools for re-educating muscles and very helpful, especially in the training of invisible muscles [20,21]. During the training sessions, the physiotherapist taught awareness through palpation and visualisation. Any woman can feel both TRA and PFM activity at the medial anterior superior iliac spina (ASIS), because the initial intravaginal pressure is predominated by PFM activity and the later increase in pressure (up to 70% maximum pressure) is produced by the combined activation of the PFM, the RA, the internal oblique muscle (IO) and the TRA [22]. It is possible for anybody to feel the different levels of PFM contractions by TRA/PFM cocontraction at the medial ASIS (through the abdominal wall). The RA muscle had to be relaxed.
I. In the rst 4 weeks of cPFM-T all study groups (n = 41) did exercises in the supine position with the hip and knees exed, feet resting on the plinth, a hip width apart.
II. In the second 4 weeks the participants performed exercises, but while the SUG (n = 22) did exercises in the supine position with the hips and knees exed, feet resting on the plinth, the SIG (n = 19) did exercises while sitting upright without support, feet resting on the plinth.
The Additional le 8 provides a detailed exercise regimen (based on the Sapsford's method [23]) and progression, including prerequisites, exercise position, instructions, feedback and a home exercises set for each week.

Statistical analysis
We report our data as mean±SD or median [1st quartile-3rd quartile] and sample size for each parameter and study group. We examined the effect of training on the parameters with the Kruskal-Wallis test. We performed the pairwise comparisons with the Mann-Whitney U-test and the Wilcoxon-rank test with the Bonferroni correction. We performed all statistical analyses using R statistical programme (version 3.5.1, R Foundation for R statistical computing). We considered values of p < 0.05 to be statistically signi cant. A statistician (M. SZ) performed the evaluation.

Questionnaire results
The participants were aged between 18 years and 25 years, with a body mass index (BMI) of 19.77 kg/m 2 -23.32 kg/m 2 . In Additional le 1 presents the questionnaire form and we list the characteristics of participants in a table (Table 1). Women with a sexual partner (n = 31) reported positive changes after the training programme in their sexual life. 20 participants (10 SUG, 8 SIG and 2 COG participants) complained of air ow to the vagina in unusual positions. After the training programme only 5 participants (2 SUG, 1 SIG and 2 COG participants) reported unchanged conditions. 6 participants (3 SUG and 3 SIG participants) suffered from constipation, which was in all cases resolved by the end of the training programme.

Electromyographic and ultrasound results
We present the data in Table 2 and Fig. 2. The maximal isometric contraction of the PFM till fatigue improved signi cantly in both study groups; however, after cPFM-T, this increase was higher in the SUG. The tonic contraction of the PFM improved in the COG, but the improvement was not signi cant (p = 0.499) (Fig. 2. A). Although in both study groups, the holding time of the maximal isometric contraction of the PFM till fatigue and the thickness of the TRA during the maximal isometric contraction of the PFM till fatigue both increased, these changes were not signi cant (Fig. 2. B). The latter parameter decreased signi cantly in the COG (Fig. 2. C) ( Table 2). On the other hand, while the strength of the maximal fast contractions of the PFM within one minute increased signi cantly in the SUG, it decreased in the SIG and the COG; this decrease, however, was not signi cant (Fig. 3. A). In both study groups, the number of maximal fast contractions within a minute increased, but the increase was signi cant only in the SUG. This parameter decreased in the COG ( Table 2, Fig. 3. B).
In both study groups, the values for relaxation of the PFM decreased but the decrease was signi cant only in the SIG (p = 0.011). This value increased in the COG (Fig. 4. A). The thickness of the TRA during relaxation of the PFM decreased in the SUG (p = 0.422) and the COG (p = 0.209), but not in the SIG (p = 0.717). Neither of these changes were signi cant ( Fig. 4. B) ( Table 2).

Discussion
Our study demonstrated that an eight-week period of cPFM-T with forced exhalation, performed by young nulliparous women in both the supine posture and the sitting posture, using trunk muscle synergies, is effective in improving cases of incontinence. Assessments of the PFM can be used to determine which structural or functional features are de cient, and to inform the design of subsequent training regimens to address these dysfunctions. A diversity of exercises, possibly tailored to the abilities of each woman, may be used and proposed training includes raising the number of repetitions of contractions [24]-a recommendation endorsed by our results.
Sapsford [23] advocated a new approach to the rehabilitation of urinary incontinence-motor relearning for diaphragmatic, deep abdominal muscles and the PFM rather than selective muscle strengthening.
Thompson et al. [25] suggested careful monitoring of Sapsford's complex rehabilitation training, because abdominal muscles are more active than PFMs in symptomatic women. These results suggest that a coordinated approach involving both deep abdominal muscles and PFM is necessary.
The correct breathing technique is very important in PFM-T. The diaphragm is a respiratory muscle participating in the stabilisation of the lumbar spine by enhancing abdominal pressure [26], which in turn stabilises the lumbar spine. Hodges et al. [27] reported synergism between the diaphragm and the TRA. We also utilised this synergism by forced exhalation in our study.
Neumann and Gill [6] suggested that the activation of deep abdominal muscles is essential for an effective contraction of the PFM, because their continent participants were unable to contract the PFM effectively while maintaining relaxation of the TRA and the IO. Similarly, Madill and McLean [22] found that the patterns of abdominal muscle activity appear to occur due to voluntary PFM contractions in healthy continent women.
As a representation of vaginal closure force, the isometric contraction was considered to be greater in the supine than in the standing position. Subsequent studies, however, have determined that women are able to perform equally strong PFM contractions in either body position [7,8,11].
The PFM is a striated muscle, with two thirds of its bres are type I (slow-twitch bres), responsible for the resting tone of the levator muscle and one third of its bres, type II (fast-twitch bres), responsible for sudden, fast but powerful contractions. The activity of the slow bres is necessary for the resting potential and the retention of urine and stool, while the fast bres are responsible for resistance during sudden abdominal pressure [28]. We monitored the activity of these two bre types (retention and rapid function) in our study with vsEMG measurements and trained both types in weaker and stronger PFM-T.
The fact that SUI happens most frequently in the upright position informed our study of the functional (static and dynamic) parameters occurring in everyday life. The holding function of the PFM is important for the patients-that is, they need to be able to get to the toilet in time before their urine starts leaking. The quality of breathing plays an important role in PFM training. It is necessary to teach participants the correct abdominal breathing technique which activates the TRA. In our study, we activated the TRA and voluntary contraction of the PFM with strong exhalation techniques during both measurements and training. Using biofeedback (TRA ultrasound imaging and vsEMG curves of PFM) we visualised the cocontraction of the two muscles with the participants. Our vsEMG ndings indicate that the static isometric contraction force of the PFM increased signi cantly in both study groups-and the holding time of static isometric contraction of the PFM increased slightly-with forced exhalation. In the COG, the static strength of the PFM increased slightly, but this increase was not signi cant and was characterised by a short retention time. The thickness of TRA during maximal isometric contraction of the PFM till fatigue improved in both training groups, but in neither group was this improvement signi cant. The COG claimed that this parameter decreased signi cantly because, during our study, they had spent much more time in a sitting position and were therefore in an enforced inactive lifestyle. Another study obtained similar results for musculus multi dus [29].
Hung et al. [14] 2010 The fast activity of the PFM is responsible for resistance during sudden abdominal pressure. In the COG, dynamic strength and repetition decreased and did not manifest signi cant changes.
Capson et al. [8] and Chmielewska et al. [11] measured the relaxation tone of the PFM in the supine position and in the sitting position. They deduced that gravity forces increase the pressure on the PFM in the vertical position, increasing its tone and leading to a higher resting activity of the PFM in the sitting position than in the supine position. During our measures, the SIG participants found it easier to relax in the horizontal position. Based on our results, practising relaxation exercises can be bene cial not only in the horizontal position but also in the vertical position. Improvement in the relaxation ability of the PFM is also a bene cial result, because especially during urination, for a healthy, normal urination mechanism it is necessary to consciously relax the PFM. The relaxation state of the PFM improved signi cantly due to gravity forces only in the SIG.
We observed that the conditioning capabilities of the PFM improved in the SUG during the 8 week period of cPFM-T. The ability to sustain isometric contractions improved signi cantly, while holding time also improved, but this improvement was not signi cant. There was also a signi cant improvement in dynamic endurance and repetitions. Therefore we recommend a more intense strengthening of the TRA in the horizontal body position.
In the case of the SIG, the maximal isometric contraction of the PFM till fatigue also improved signi cantly, together with the holding time, but the latter improvement was not signi cant. The dynamic endurance decreased and the number of repetitions improved slightly, but these changes were not signi cant. The relaxation state of the PFM improved signi cantly due to gravity forces. The resting tone of the TRA remained unchanged after 8 weeks. Thus, even more intense TRA activation is required in the vertical position, as a reliance on the enhanced gravitational forces caused by the lumbopelvic posture or on the activated TRA induced by strong exhalation is insu cient to induce the necessary changes in the functioning of the TRA.
In the case of the COG, all parameters decreased-particularly the thickness of TRA during maximal isometric contraction of the PFM till fatigue, which decreased signi cantly-by reason of sedentary lifestyle.
We also established that young nulliparous women might also be affected by urinary leakage (27%) as con rmed by the results of Haslam et al. [1].
In the case of the SUG, both training exercises and measurements were performed in the supine position, whereas the SIG participants were measured in the lying position but performed all the exercises in the sitting position. Since SUI takes place most frequently in the vertical position, we should not measure and strengthen the static and dynamic functions of the PFM only in the supine position. We recommend that during cPFM-T, isometric and relaxation tasks should be performed both in the supine position and in the sitting position.
According to Sapsford et al. [5], the antigravity posture requires more intense PFM activity. Furthermore, Chmielewska et al [11] reported a signi cant difference between the sustained 1-minute contraction of the PFM in the supine position and that in sitting position, while we found that the maximal isometric sustained contraction of PFM till fatigue signi cantly improved both in the supine position and in the sitting position.
Group training and individual training, according to recent research, are equally effective, and group training is more cost-effective without limitation [30]. Furthermore, due to differences in knowledge and behaviour, physiotherapist-guided training can help and motivate patients in persistent PFM-T [31].

Conclusions
TRA relaxion is easier in the horizontal position and strengthening is more effective in the horizontal body position. Physiotherapist-guided group training is more e cient.
The authors suggest that during PFM training the isometric tasks should be performed both in the supine position and in the sitting position, the dynamic endurance tasks of the PFM should be performed at the Phone number: +36(1)8869329. E-mail: amd@ogyei.gov.hu. All participants provided written informed consent.

Consent for publication
Not applicable.
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
All nancial and non-nancial competing interests must be declared in this section.

Funding
The project has been supported by the European Union, co-nanced by the European Social Fund. EFOP-3.6.1-16-2016-00008. The role of the funding body in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript should be declared.

Authors' contributions
Each author's participation in the manuscript. All authors have read and approved the manuscript.