While there is strong evidence for pelvic floor muscle (PFM) training (PFMT) as an intervention for women with stress urinary incontinence (SUI) (1) the mechanisms through which PFM exercise is effective at reducing the signs and symptoms of SUI in women are not known (2). The aims of this study were to determine (i) if any, aspects of PFM function, measured through manual assessment, are associated with the severity of SUI in women and (ii) whether improvements in certain aspects of PFM function are associated with improvements in SUI signs or symptoms. We hypothesized that PFM strength, power and endurance would be negatively associated with SUI severity and that improvements in these parameters would be correlated with reductions in symptoms.

This secondary analysis of data was planned apriori within a prospective interventional cohort study that included females with SUI who underwent a 12-week PFMT program. Females with symptoms of SUI were recruited from surgical wait lists for mid-urethral sling insertion and from new referrals to participating, local pelvic floor physiotherapists. Before the intervention period, SUI symptom severity was determined subjectively through the International Consultation on Incontinence Questionnaire -Female Lower Urinary Tract Symptoms Urinary Incontinence subscale score (ICIQ-FLUTS-UI) which participants completed at home. Both SUI severity, measured objectively using a standardized 30-minute pad test (30MPT), and PFM function were measured at a laboratory session conducted by an experienced (>5 years) pelvic floor physiotherapist.
The 30MPT was conducted once bladder volume was between 300 and 500mL as confirmed through transabdominal ultrasound imaging. Participants donned a pre-weighed incontinence pad and performed a standardized circuit of activities over 30 minutes, after which the pad was removed and reweighed. 
PFM function was evaluated through vaginal palpation using the PERFECT scheme, which is a clinical assessment approach used to guide intervention around specific impairments identified during assessment (3). The overall PERFECT scheme score was computed, and the first four items of the PERFECT scheme were also considered separately: Performance(P) was quantified using the Modified Oxford Scale  (MOS; range 0-5) as a measure of PFM force generated through maximal voluntary contraction (MVC);  Endurance(E) was the time in seconds (up to 10) that the MVC force could be held before it was deemed (through palpation) to have  reduced by half or more; repetitions(R) was the number of times (up to 10) that the MVC force could be achieved and held for the duration of the endurance time (determined under E); and Fast contractions(F) was the number of times the MVC force (determined under P) could be repeatedly achieved at one-second intervals. These four items can be interpreted in light of PFM function as described by the International Continence Society (ICS) (2): P=MVC force, E=local static muscle endurance at MVC, R=capacity for dynamic repetition of local static muscle endurance at MVC, and F=a combination of power and motor control. 
After the baseline assessment, women underwent a 12-week physiotherapist-supervised PFMT intervention. They attended six sessions through which they learned to perform a proper PFM contraction (manual and biofeedback) and learned PFM exercises and functional PFM contractions (“The Knack”). Women were instructed to complete a series of prescribed PFM exercises daily. The ICIQ-FLUTS questionnaire, the 30MPT and the PERFECT scheme assessments were repeated after the intervention period.
To detect a significant moderate correlation (r=0.6) between changes in aspects of PFM function and signs and symptoms of SUI, a required sample size of n=20 was estimated using G*Power software. Changes in the individual aspects of PFM function and in SUI severity were determined through separate paired t-tests, while associations among aspects of PFM function and SUI severity at baseline, and associations among changes in aspects of PFM function and SUI severity were evaluated using Spearman’s rho (α=0.05).

Baseline data from 98 females with SUI were available from the main study, from which follow-up data were available from 75 females who completed the PFM intervention program and returned for follow-up assessment; all data were retained to maximize study power. The mean age was 50 (±10) years, mean body mass index was 28.36 (±7.32) kg/m2) and the vast majority (96.5%) of participants were parous. Because aspects of PFM function measured using the PERFECT scheme may be constrained by minimum and maximum scores, the frequency distributions of all PERFECT scheme outcomes were inspected to ensure that floor or ceiling effects were not evident; all data were normally distributed. 
At baseline, there was a significant positive correlation (ρs=0.203, p=0.045, n=98) between urine leakage on the 30MPT and the number of repetitions of the local PFM static endurance contraction (R) women could perform, while there were no significant associations between ICIQ-FLUTS UI and any measured aspect of PFM function (Table 1). All aspects of PFM function improved significantly over the intervention period, as did subjective and objective measures of SUI severity, as shown in Table 2. There were no associations between Improvements in subjective (ICIQ-FLUTS UI) nor objective (30MPT) measures of SUI severity and improvements in the overall PERFECT score nor the individual aspects of PFM function (P,E,R or F). Indeed, while correlation coefficients were low, in some instances the Spearman’s correlation coefficients were counter to what was expected. For example, greater improvements in PFM MVC measured by the Modified Oxford Scale (ρs=0.234, p=0.045, n=75) and greater capacity to repeat PFM MVCs (ρs=0.229, p=0.049, n=75) were associated with smaller improvements on the 30MPT after the intervention. Similarly, greater gains in capacity to repeat PFM static endurance contractions were associated with less improvement on the FLUTS-UI subscale score (ρs=0.289, p=0.012, n=75).

Consistent with the literature (1), all measured aspects of PFM function assessed through the PERFECT scheme were improved after a 12-week PFMT intervention, and women’s symptoms (ICIQ-FLUTS-UI) and signs (30MPT) of SUI concurrently improved. Yet the severity of women’s signs and symptoms of SUI were not associated with impairments in any aspect of their PFM function measured by the PERFECT scheme. Further the extent of improvements in PFM function (PERFECT overall or P,E,R,F0 were not associated with improvements in women’s signs and symptoms of SUI. 
These results suggest that improvements in SUI derived from PFMT may be largely through mechanisms other than improvements in PFM strength or endurance, such as changes to levator ani muscle morphology or tissue mechanics, co-ordinated recruitment of the PFMs with the urethral sphincters, or improved urethral sphincter function. Because manual palpation of PFM strength is discrete and subjective, it is possible that associations between PFM function and SUI severity do exist, but that the PERFECT scheme is not sensitive enough to detect them.

The importance of the PFMT in the treatment and prevention of female SUI is already well established in the literature. Yet the results of this study suggest that there is a need for greater understanding of the role of the PFMs in SUI and the mechanisms through which PFMT is an effective intervention. Further study with more sensitive measures of PFM mechanics and function, such as those obtained through intravaginal dynamometry, and measures of PFM motor control, such as those obtained through electromyography, are required to more fully evaluate these relationships.

Dumoulin, C., Cacciari, L. P., & Hay-Smith, E. J. C. (2018). Pelvic floor muscle training versus no treatment, or inactive control treatments, for urinary incontinence in women. Cochrane Database of Systematic Reviews.
Bo, K., Frawley, H. C., Haylen, B. T., et al. (2016). An International Urogynecological Association (IUGA)/International Continence Society (ICS) joint report on the terminology for the conservative and nonpharmacological management of female pelvic floor dysfunction. Neurourology and Urodynamics, 36(2), 221–244. doi:10.1002/nau.23107
Laycock, J., & Jerwood, D. (2001). Pelvic Floor Muscle Assessment: The PERFECT Scheme. Physiotherapy, 87(12), 631–642. doi:10.1016/s0031-9406(05)61108-x