Factors affecting reflex pelvic floor muscle contraction patterns in women with pelvic floor disorders

Authors

  • J.-M. Yang,

    1. Department of Obstetrics and Gynecology, Taipei Medical University – Shuang Ho Hospital, Taipei, Taiwan
    2. Department Obstetrics and Gynecology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author
  • S.-H. Yang,

    1. School of Nutrition and Health Sciences, Taipei Medical University, Taipei, Taiwan
    Search for more papers by this author
  • W.-C. Huang,

    Corresponding author
    1. Department Obstetrics and Gynecology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
    2. Department of Obstetrics and Gynecology, Cathay General Hospital, Taipei, Taiwan
    • Department of Obstetrics and Gynecology, Taipei Medical University – Shuang Ho Hospital, Taipei, Taiwan
    Search for more papers by this author
  • C.-R. Tzeng

    1. Department Obstetrics and Gynecology, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
    2. Department of Obstetrics and Gynecology, Taipei Medical University Hospital, Taipei, Taiwan
    Search for more papers by this author

Correspondence to: Dr W.-C. Huang, Department of Obstetrics and Gynecology, Cathay General Hospital 280, Section 4, Ren-Ai Road, Taipei, 106, Taiwan, R.O.C. (e-mail: huangwc0413@hotmail.com)

ABSTRACT

Objective

To explore factors affecting the presence of two reflex pelvic floor muscle contraction (PFMC) patterns in women with pelvic floor disorders.

Methods

This was a retrospective analysis of pelvic floor ultrasonography and urodynamic data for 667 consecutive symptomatic women with pelvic floor disorders. We identified on ultrasonography the presence or absence of two reflex PFMC patterns, anorectal lift (ARL) and inward clitoral motion (ICM), preceding or occurring during coughing, and evaluated their associations with possible factors affecting reflex PFMC reactivity, including patient demographics, pelvic organ prolapse stages, ultrasonography findings and urodynamic data.

Results

Of the 667 women, 560 (84.0%) clearly demonstrated reflex ARL and 536 (80.4%) demonstrated ICM. There were significant differences in age (P < 0.001), parity (P = 0.033) and menopausal status (P = 0.005) between women with and those without reflex ICM before or during coughing. The multivariable logistic regression model showed that age was the only independent factor associated with presence of reflex ICM (odds ratio, 0.93 (95% CI, 0.88–0.99), P = 0.017). In contrast, no significant differences were noted between women with and without reflex ARL.

Conclusions

Increasing age is negatively associated with the presence of reflex ICM during coughing in symptomatic women with pelvic floor disorders. Copyright © 2013 ISUOG. Published by John Wiley & Sons Ltd.

INTRODUCTION

Defined by the International Continence Society (ICS)[1] as voluntary and involuntary contractions and relaxations, normal pelvic floor muscles (PFM) respond to changes in intra-abdominal pressure in a timely manner via stretch reflex contractions to stabilize the pelvic structures[2-4]. Normal pelvic floor function therefore requires an intact neuromuscular system. Defective PFM can result in pelvic floor disorders, leading to significant quality-of-life issues. Reflex pelvic floor muscle contraction (PFMC) ensures appropriate pelvic floor function. Studies have found that reflex PFMC during coughing is significantly weaker in women with mild pelvic organ prolapse than it is in those without this condition[5] and that loss of reflex PFMC is associated with severe incontinence and impaired quality of life[6]. To evaluate a woman's ability to counteract any sudden increase in intra-abdominal pressure, an assessment of reflex PFMC is therefore required[5].

Methods to assess PFMC include inspection, digital palpation, electromyography, dynamometry, perineometry and ultrasonography[1]. Ultrasonography provides a considerable amount of dynamic data regarding different pelvic structures that can be difficult to assimilate visually on observation, particularly during quick events such as coughs[7-10]. Two valid and reliable ultrasonographic measures for reflex PFMC are the anorectal lift (ARL) and inward clitoral motion (ICM) preceding or during coughing[3, 7, 8]: movement of the anorectal junction can be used to analyze deep PFM because the sling of levator ani muscles wraps around the anorectal junction and its displacement is closely associated with PFMC[8, 11]; clitoral motion can be used to analyze superficial PFM because displacement of the clitoris is controlled by the ischiocavernosus and bulbospongiosus muscles and muscles inferior to the perineal membrane, which surround the vagina and the urethra[12]. These two reflex PFMC patterns appear to be involved in constricting the genital hiatus and stabilizing pelvic floor structures[4, 13].

The literature is scarce regarding factors affecting reflex PFMC patterns. The objectives of this study were to investigate clinical differences, amongst women with pelvic floor disorders, between those with and those without each reflex PFMC pattern and to explore factors associated with the presence or absence of each reflex PFMC pattern in these women.

METHODS

This study was approved by the institutional review boards of the participating hospitals (MMH-I-S-198, TMU-JIRB 201011013 and 201112047). We performed a retrospective analysis of data from consecutive women who presented with lower urinary tract symptoms to the urogynecology clinics of two medical centers in an ongoing parent study investigating the effects of volitional and reflex PFMC on incontinence-related quality of life and sexual function from January 2006 to December 2011. We excluded women with cerebrovascular disease, dementia, overt neurological disease, diabetes mellitus and previous pelvic surgery; a total of 667 women were included in the study. Data collected at the time of evaluation included patient demographics, pelvic organ prolapse quantification (POP-Q) system stages[1], urodynamic study findings and pelvic floor ultrasonography findings.

To assess pelvic floor support, a split speculum was used with the patient in the dorsal lithotomy position and straining maximally. Site-specific analysis of pelvic floor support was performed using the POP-Q system[1].

Complete urodynamic studies included free uroflowmetry, filling and voiding cystometry and urethral pressure profilometry. We examined each urodynamic report for urodynamic stress incontinence (USI) and detrusor overactivity (DO). USI was defined as the involuntary leakage of urine during filling cystometry associated with increased intra-abdominal pressure in the absence of a detrusor contraction[1]. DO was defined as the occurrence of involuntary detrusor contractions during filling cystometry, and these contractions could be spontaneous or provoked, phasic or terminal and with a waveform of variable duration and amplitude on the cystometrogram[1].

Reflex PFMC was assessed by ultrasonography using a Voluson 730 or E6 (GE Medical Systems, Zipf, Austria) or a Philips HD 11 (Philips Medical Systems, Bothell, WA, USA) ultrasound machine and a 5.0–9.0-MHz transvaginal probe. Initially, pelvic floor ultrasonography was performed with patients in a supine position with a comfortably full bladder for both quantitative and qualitative analyses of the morphology of the lower urinary tract, anterior vaginal wall and genital hiatus, as described previously[6-8]. Ultrasonography parameters for genital hiatal position included the genitohiatal angle (angle between the anorectum–symphyseal line and the midline of the symphysis pubis at rest) and the genitohiatal distance (distance between the anorectal junction and the inferior border of the symphysis pubis at rest)[7, 8]. After the patient's bladder was emptied, the transvaginal probe was placed at the introitus and adjusted so that ultrasonography could capture the area from the symphysis pubis to the anorectal junction. The patients were given time to become accustomed to the transvaginal probe, to prevent it from stimulating a bulbocavernosus reflex. The women were then instructed to perform a series of hard coughs without fear of urine leakage, and the movement of the pelvic floor structures was assessed in both real-time and cine-loop modes and recorded simultaneously using a DVD recorder. Two patterns of reflex PFMC were recognized[6, 7]: ICM, defined as a caudal–dorsal or inferior–posterior motion of the clitoris during coughing and considered to represent the superficial PFM, and ARL, defined as a cranial–ventral or superior–anterior movement of the anorectal junction towards the symphysis pubis during coughing and considered to represent the deep PFM (Figure 1). Only PFMC patterns related to the strongest coughs were considered for analysis. A reflex ICM was considered present only when clitoral motion was a distinct movement along the inferior border of the symphysis pubis rather than a conjoined movement with the symphysis pubis. The latter was considered an artifact induced by body motion or the transvaginal probe during coughs. In some women, an ARL was concealed or obscured by prolapsing pelvic organs during increased intra-abdominal pressure, in which case the ARL was considered ‘indeterminate’. Our previous study[7] demonstrated substantial inter- and intraobserver agreement in assessing reflex PFMC using ultrasonography, with Cohen's kappa values ranging from 0.645 to 0.679.

Figure 1.

Ultrasound images showing pelvic floor structures at rest (a) and during coughing (b). Right and left arrows indicate the direction of motion of the clitoris (cl) and anorectal junction (image), respectively, during coughing. ac, anal canal; bl, bladder; r, rectum; sp, pubic symphysis; u, urethra.

Statistical analysis

We assessed the relationship of patient demographic variables, POP-Q system stages, urodynamic study findings and genitohiatal parameters on pelvic floor ultrasonography with occurrence of reflex ARL and reflex ICM before or during coughing. Concordance between the presence of reflex ARL and reflex ICM (i.e. the presence or absence of each reflex PFMC pattern in the same patient) was assessed in subgroups with different urodynamic findings using kappa statistics. Associations between the clinical and sonographic factors and the presence of reflex ICM and reflex ARL were analyzed using chi-square tests. Univariable and multivariable logistic regression analyses were also performed to examine the association of the factors with the presence of reflex PFMC patterns, with calculation of odds ratios (OR) and 95% CIs. Listwise deletion of cases with missing data was used for each separate analysis. All analyses were performed using SPSS 17.0 for Windows (SPSS, Inc., Chicago, IL, USA), with P < 0.05 considered statistically significant.

RESULTS

The 667 consecutive women had a mean (±) age of 54 ± 11 years, body mass index (BMI) of 24.6 ± 3.3 kg/m2, and parity of 3 ± 1. Of the women with the following data available, 52.2% (220/421) were postmenopausal, 10.9% (45/414) had undergone at least one Cesarean section and 21.1% (119/565) had advanced pelvic organ prolapse (> Stage II according to the POP-Q system). Urodynamic studies showed that 44.2% (292/660) women had USI, 8.3% (55/660) had DO, and 4.4% (29/660) had both USI and DO. Ultrasonography showed that reflex ARL and ICM were clearly identifiable in 560 (84.0%) and 536 (80.4%) women, respectively (Table 1). Despite repeated reviews, the presence or absence of reflex ARL could not be determined in 15 (2.2%) women.

Table 1. Distribution of two reflex pelvic floor muscle contraction patterns in 667 symptomatic women with pelvic floor disorders
Variablen (%)
  1. a

    Anorectal lift activity obscured by prolapsing pelvic organs.

Involuntary anorectal lift 
  Absent92 (13.8)
  Present560 (84.0)
  Indeterminatea15 (2.2)
Involuntary inward clitoral movement 
  Absent131 (19.6)
  Present536 (80.4)

The concordance between presence of reflex ARL and reflex ICM was substantial in women with isolated DO (kappa = 0.615), fair in women with USI regardless of coexistence of DO (kappa = 0.286) and lacking in women with neither USI nor DO (kappa = 0.170).

Demographic and clinical characteristics (age, parity, BMI, history of Cesarean section(s), menopausal status, POP-Q system stages, urodynamic findings and genital hiatal position on ultrasonography) were similar between women with and those without reflex ARL (Table 2). In contrast, on comparing women with and without reflex ICM, it was found that women with ICM were significantly younger (P < 0.001), had lower parity (P = 0.033) and were more likely to be premenopausal (P = 0.005). There was no significant difference in BMI, history of Cesarean section(s), POP-Q system stages, urodynamic findings or genital hiatal position (Table 2).

Table 2. Clinical differences between symptomatic women with pelvic floor disorders with and those without either of two reflex pelvic floor muscle contraction patterns
VariableAnorectal liftInward clitoral motion
AbsentPresentPAbsentPresentP
  1. Data

    are presented as n (%); n values total less than 667 for each comparison, reflecting missing data for the relevant variables.
  2. a

    Including women with objective evidence of USI with/without detrusor overactivity (DO) on urodynamic studies.

  3. b

    Including women with isolated findings of DO on urodynamic studies.

  4. c

    Including women with neither USI nor DO on urodynamic studies. ICS POP-Q, International Continence Society pelvic organ prolapse quantification[1]; USI, urodynamic stress incontinence.

Age  0.384  < 0.001
  < 40 years7 (17)34 (83) 5 (12)37 (88) 
  40–49 years14 (11)112 (89) 18 (14)109 (86) 
  50–59 years16 (12)121 (88) 31 (22)109 (78) 
  60–69 years11 (15)64 (85) 20 (26)56 (74) 
  > 69 years10 (22)36 (78) 21 (46)25 (54) 
Parity  0.139  0.033
  03 (43)4 (57) 3 (43)4 (57) 
  13 (10)26 (90) 4 (13)26 (87) 
  221 (13)142 (87) 27 (16)138 (84) 
  313 (11)102 (89) 28 (24)89 (76) 
  > 318 (17)87 (83) 32 (30)74 (70) 
Body mass index  0.439  0.388
  < 21.9 kg/m212 (16)63 (84) 16 (20)61 (80) 
  21.9–23.2 kg/m26 (9)59 (91) 9 (14)56 (86) 
  23.3–24.8 kg/m28 (11)73 (89) 16 (20)66 (80) 
  24.9–27.0 kg/m213 (18)58 (82) 20 (27)53 (73) 
  > 27.0 kg/m210 (14)60 (86) 13 (19)57 (81) 
Previous Cesarean section  0.809  0.496
  No52 (14)312 (86) 86 (23)283 (77) 
  Yes5 (11)39 (89) 7 (16)38 (84) 
Postmenopausal  0.930  0.005
  No27 (14)172 (86) 32 (16)169 (84) 
  Yes31 (14)185 (86) 61 (28)159 (72) 
ICS POP-Q system stage  0.0839  0.2184
  Stage 09 (28)23 (72) 9 (27)24 (73) 
  Stage I36 (14)229 (86) 47 (17)225 (83) 
  Stage II15 (11)122 (89) 27 (19)114 (81) 
  Stage III–IV15 (13)103 (87) 30 (25)89 (75) 
Urodynamic findings  0.128  0.630
  USIa36 (11)278 (89) 66 (21)255 (79) 
  Detrusor overactivityb11 (20)44 (80) 12 (22)43 (78) 
  Otherc44 (16)233 (84) 52 (18)232 (82) 
Ultrasonography findings      
 Genitohiatal angle  0.837  0.118
  112–141°19 (15)110 (85) 30 (23)102 (77) 
  142–147°16 (13)105 (87) 16 (13)109 (87) 
  148–153°19 (15)110 (85) 31 (23)102 (77) 
  154–161°22 (17)105 (83) 23 (18)104 (82) 
Genitohiatal distance  0.312  0.276
  31.3–43.3 mm22 (17)110 (83) 20 (15)113 (85) 
  43.4–47.0 mm21 (16)109 (84) 22 (17)109 (83) 
  47.1–50.4 mm12 (10)113 (90) 26 (20)101 (80) 
  50.5–54.6 mm14 (12)105 (88) 24 (19)101 (81) 
  > 54.6 mm22 (17)105 (83) 33 (25)98 (75) 

Univariable logistic regression analysis showed significant associations between the absence of reflex ICM and age (OR, 0.95 (95% CI, 0.93−0.97), P < 0.001), parity (OR, 0.81 (95% CI, 0.69−0.94), P = 0.007) and postmenopausal status (OR, 0.49 (95% CI, 0.31−0.80), P = 0.004). However, on multivariable logistic regression analysis, only age was found to be an independent factor, with the absence of reflex ICM being significantly more likely in older women (OR, 0.93 (95% CI, 0.88−0.99), P = 0.017).

DISCUSSION

Based on a large cohort of symptomatic women with pelvic floor disorders, this study examined the associations between age, parity and menopausal status and the presence of reflex ICM, finding age to be the only independent variable and suggesting that superficial PFMC may be lacking in older women. However, no such association with age was found for the presence of reflex ARL during coughing. Because the study subjects were all symptomatic, our findings cannot be extrapolated to asymptomatic populations.

We found that Cesarean delivery was not protective against the loss of reflex superficial PFMC. In addition, neither pelvic organ prolapse stratified according to POP-Q system stage nor genitohiatal parameters on ultrasonography was associated with reflex superficial PFMC. The investigation of such associations between clinical and sonographic factors and pelvic floor function has been reported rarely in the literature[5, 14, 15].

An early study reported different innervations for deep and superficial PFM[16]. The deep PFM, such as the levator ani muscles, are innervated by the pelvic nerve, which originates from sacral nerve roots S3–5 and travels along the superior surface of the pelvic floor[16]. The superficial PFM, such as the urethral rhabdosphincter, ischiocavernosus, bulbospongiosus and transverse perineal muscles, are innervated by the perineal branch of the pudendal nerve[16]. However, another study showed that the urethral rhabdosphincter and levator ani muscles are innervated by fibers derived from the pudendal nerve[17]. Thus, damage to the pudendal nerve may result in dysfunction of the rhabdosphincter and levator ani muscles.

The neurophysiology of the pudendal nerve is affected by age and parity[18, 19]. The findings of our study indicate the independent impact of aging on pudendal nerve function, because parity and menopausal status are likely interdependent covariates of aging[20]. Aging affects PFM by changing the characteristics or proportions of isomyosin muscle fiber types, gradually decreasing the integrity and functional capacity of striated muscles, and causing cell shrinkage and fragmentation[4, 21-24].

We did not find any association between the factors examined and reflex ARL, which is indicative of reflex activities of deep PFM. Although the association between vaginal childbirth with levator muscle defect and subsequent pelvic floor dysfunction is well established[25, 26], an animal study showed that the canine levator ani muscle is not affected pathologically by vaginal delivery and seems to adapt to numerous successive pregnancies and births through type I fiber hypertrophy[27]. Our data support the findings of Trowbridge et al., who showed that levator ani muscles consist predominantly of slow twitch fibers, the number and diameter of which are largely unaffected by age[28].

We also observed lower concordance between the two reflex PFMC patterns in women with USI (kappa = 0.286). The timely activation and coordination of different PFM groups in response to stress is important for maintaining urinary continence[2, 3, 6, 29]. Appropriate coordination among different PFM groups has been reported in continent women[2]. During coughing, the levator ani muscles contract simultaneously with the diaphragm and abdominal wall muscles to build abdominal pressure, thus tensing the suburethral fascia, and thereby enhancing urethral compression[4]. In women with SUI, the response of the levator ani muscles to coughs is delayed[4, 29]. We hypothesized that because of anatomical proximity, the motion of the superficial PFM may reflect the recruitment activities of the striated muscles of the urethrovaginal sphincter and compressor urethra, which also contribute to the maintenance of urinary continence[4].

Although good concordance (kappa = 0.615) was noted between the two reflex PFMC patterns in women with isolated DO, no concordance (kappa = 0.170) was noted in symptomatic women with neither USI nor DO. Our study supports the finding that the loss of reflex activity of the PFM is not crucial to the development of SUI[15]. However, we have reported previously that women with urinary incontinence who did not demonstrate reflex superficial PFMC had greater severity of and worse effects from incontinence than did those who demonstrated reflex superficial PFMC[6]. Thus, reflex PFMC may be inherent but the coordination of different patterns of reflex PFMC may be a learned behavior. Furthermore, our study confirms that the pathogenesis of DO involves functional or structural changes in the unstable detrusor[30], rather than the reactivity of PFM to stress events.

Our study has several limitations. First, no standardized method was used to ensure cough intensity during ultrasound examination. Although we instructed the subjects to perform a series of strong coughs during the examination, some participants would nevertheless produce coughs with minimal effort because of the fear of urinary leakage. This confounding factor could be reduced by assessing reflex PFMC with an empty bladder. Second, variability in the increased intra-abdominal pressure may induce an unpredictable reflex response of the PFM. To standardize the provoking coughs, a pressure catheter may be placed in the vagina or rectum, thus measuring the increase in intra-abdominal pressure during coughing. Third, the study included only symptomatic women with pelvic floor disorders, and studying normal women might not lead to the same findings. Fourth, the timing, synchronization and grading of reflex PFMC were not considered in the analyses. Finally, the patients' hormonal milieu, which may affect pelvic floor function, was not examined. Despite these limitations, this study offers a promising method for ultrasound evaluation of reflex PFMC and shows clearly that aging is an independent factor affecting the occurrence of reflex superficial PFMC. When assessing reflex superficial PFMC, the age of the patient should, therefore, be taken into account.

ACKNOWLEDGMENTS

This study was supported by grants from the National Science Council (NSC 97-2314-B-195-012-MY3, NSC100-2314-B-038-005 and NSC 101-2314-B-038-021-MY3).

Ancillary