Stress urinary incontinence (SUI) is defined by the International Continence Society as an involuntary loss of urine during effort situations. Some theories have been shown to explain the urinary continence mechanisms. Although these main theories present some different points of view, both reinforce the importance of the pelvic floor muscles (PFM) on urethral closure, favoring the female urinary continence maintenance. 
Previous studies [1] have already found a significant difference in the PFM strength between continent and incontinent women. However, it is worth noting that PFM function involves several characteristics that go beyond muscular strength. In this context, we aimed to investigate and compare the PFM morphology and function of continent and predominant SUI women, as well as to verify which PFM morphological and functional parameters better predict SUI. We hypothesized that both PFM anatomy and function are impaired in predominant SUI women.

A cross-sectional study was carried out after its approval by the local Human Research Ethics Committee. Initially, 302 women were recruited, however, 122 of them were excluded for not meeting the study’s eligibility criteria, resulting in a final sample of 180 women. The continence status was ascertained using the International Consultation on Incontinence Questionnaire Urinary Incontinence – Short Form (ICIQ UI-SF), resulting in 43 continent women and 137 predominant SUI women. The target population of this study was multiparous women who became pregnant once or more. We excluded nulliparous women (n=39), women who had current urinary tract infection (n=3), other types of incontinence rather than predominant SUI (like urgency urinary incontinence alone or mixed urinary incontinence with predominant urgency incontinence) (n=22), physical or neurological disorders that would hinder their participation in the evaluation (n=3), previous oncology treatment as brachytherapy or neovagina (n=6), both SUI or/and pelvic organ prolapse surgery history (n=15), any pelvic organ prolapse exceeding the vaginal opening (n=8) and PFM contraction grade either zero or one according to the Modified Oxford Grading Scale (n=26).
PFM morphology was assessed by 4D translabial ultrasound (4D TLUS) through offline analyzes of levator avulsion, levator plate angle, symphysis-levator distance, anorectal angle, bladder neck position, hiatal area and puborectal muscle thickness at rest. PFM function was assessed by digital palpation (following the PERFECT scheme as well as assessing the PFM coordination, reflex PFM contraction and use of accessory muscles), surface electromyography at rest and during PFM contraction and some ultrasound displacement measurements from rest to PFM contraction as the change in levator plate angle, change in symphysis-levator distance, change in anorectal angle, bladder neck elevation, hiatal area narrowing, puborectal strain and change in puborectal muscle thickness. 
Logistic regression adjusted for age, number of vaginal delivery and menopause was used to compare these variables between groups. Multivariate logistic regression was used to identify the best predictors for SUI using the stepwise criteria for variable’s selection. A significance level of 5% was adopted.

Most of the participants included in this study were white (78.9%), married (72.8%) and with higher education level (49.4%). Mean age at examination was 48±10 years old (range from 22 to 70) and mean body mass index was 26±5.3 kg/m2 (range from 15.9 to 46). Regarding previous obstetric data, the participants’ median number of pregnancies was 2.00 (range from 1 to 6). 42.6% of these women experienced cesarian section exclusively, 40.3% experienced vaginal delivery exclusively and 17% of these women experienced both route of birth, without significant difference between groups (p=0.06). Of those who had delivered vaginally (101/180), 19.4% had undergone forceps delivery and 50% had experienced an episiotomy. The incontinent women presented predominant SUI (mean ICIQ UI-SF score: 13.7±3.7) and most of them had not remembered when the urinary loss started (58.4%) as well as they reported not using any kind of absorbent to contain urinary loss (59.9%).
A significant difference regarding age (p=0.006), number of vaginal deliveries (p=0.005) and menopause (p=0.01) was found between groups. Based on the contribution of these variables in the development of urinary incontinence, all the following statistical analyzes were adjusted for them.
When compared the ultrasound pelvic floor morphological parameters between groups, we found that incontinent women presented bigger genital hiatus area (p=0.007) and symphysis-levator distance (p=0.01), bigger anorectal angle (p<.0001), smaller levator plate angle (p=0.03), lower position of the bladder neck in relation to the pubic symphysis (p=0.0006) and thinner superior puborectal muscle thickness (right: p=0.004 and left: p=0.007). No difference was found in the presence of levator avulsion (p=0.07) and in the inferior puborectal muscle thickness (right: p=0.5 and left: p=0.9) between groups.
Regarding PFM function, incontinent women presented less PFM power of contraction (p<.0001), endurance (p=0.006) and resistance (p=0.02) assessed by digital palpation as well as most of them used some accessory muscle during PFM contraction (p<.0001), had less coordination (p=0.0003) and absence of PFM reflex contraction (p<.0001). Incontinent women also presented an increased PFM electromyography activity at rest (p=0.03). No significant differences were found regarding the number of fast PFM contractions (p=0.3) and PFM electromyography activity during contraction (p=0.03). Besides that, no significant differences in the dynamic ultrasound biometric parameters were found between groups, except for the change in the symphysis-levator distance (p=0.02), since incontinent women performed a greater reduction in this distance during PFM contraction.
Then, we composed a multivariate logistic regression analyses including the PFM functional and morphological parameters. We found that the use of accessory muscles during PFM contraction, PFM coordination, anorectal angle at rest, bladder neck position at rest, superior PRM thickness and the change in anorectal angle are the best predictors for urinary incontinence (Table 1).

In the present study, SUI women presented significant alterations in the pelvic floor morphology, which seems related to pelvic floor laxity. This condition has already been reported in SUI women, resulting in a greater bladder neck descent during effort situations, when compared to continent women [2].
Similarly, SUI women also presented altered PFM function. PFM function is directly linked to its ability to perform an efficient muscle contraction that results in bladder neck elevation and genital hiatus narrowing [3]. In this way, alterations in PFM power of contraction, endurance, resistance, coordination and timing of contraction may lead to urinary incontinence since it does not result in an effective contraction over the bladder neck and genital hiatus.
Thus, based on the multivariate logistic regression analysis, the impairment of both morphological and functional parameters seems to be related to SUI in women, predicting it in 42.2%. Despite we have controlled the difference in the number of vaginal deliveries (as well as age and menopause status) between groups, the alterations in these predicting parameters can be explained by possible damages in neuromuscular, fascial and ligament support, incurred during successive vaginal deliveries.

Both PFM morphology and function are impaired in predominant SUI women, being consistent with our stated hypothesis. Thus, to discriminate between continent and incontinent women, the assessment of PFM should not be restricted to PFM strength, since the use of accessory muscles during PFM contraction, PFM coordination, anorectal angle at rest, bladder neck position at rest and change in anorectal angle are related to SUI.

Chamochumbi CCM, Nunes FR, Guirro RRJ, Guirro ECO. Comparison of active and passive forces of the pelvic floor muscles in women with and without stress urinary incontinence. Rev Bras Fisioter 2012;16(4):314-9.
Hoyte L, Schierlitz L, Zou K, Flesh G, Fielding JR. Two- and 3-dimensional MRI comparison of levator ani structure, volume, and integrity in women with stress incontinence and prolapse. Am J Obstet Gynecol. 2001;185(1):11-9.
Albrich S, Steetskamp J, Knoechel SL, Porta S, Hoffmann G, Skala C. Assessment of pelvic floor muscle contractility: digital palpation versus 2D and 3D perineal ultrasound. Arch Gynecol Obstet 2016;293(4):839–843.