How common is pelvic floor muscle atrophy after vaginal childbirth?

Transcription

1 Ultrasound Obstet Gynecol 2014; 43: 3 Published online 2 December 2013 in Wiley Online Library wileyonlinelibrary.com. DOI: /uog How common is pelvic floor muscle atrophy after vaginal childbirth? P. DIXIT, K. L. SHEK and H. P. DIETZ Sydney Medical School Nepean, University of Sydney, Penrith, New South Wales, Australia KEYWORDS: childbirth; levator ani; muscle atrophy; pelvic floor muscle; ultrasound ABSTRACT Objective To determine if there is evidence of levator ani atrophy in primiparous women. Methods This was a prospective observational cohort study of 202 primiparous women recruited between November 2006 and March 200. Translabial ultrasound volumes were obtained at 36 3 weeks gestation and at a mean of 4.5 months postpartum. Peripartum changes in bladder neck elevation and reduction of anteroposterior hiatal diameter on pelvic floor muscle contraction PFMC and changes in muscle thickness were analyzed. Results Of the 202 participants enrolled, 15 7% completed the study. There was a significant reduction in bladder neck elevation P = and change in anteroposterior hiatal diameter P = 0.03 on PFMC when comparing antenatal and postnatal results, the latter being significantly associated with delivery mode P = No significant changes were detected in muscle thickness P = Conclusions There is a reduction in sonographic measures of pelvic floor function after childbirth, but muscle atrophy is unlikely to be a significant factor. Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. INTRODUCTION The pelvic floor is an essential part of the abdominal envelope. It is composed of nerves, connective tissue, the levator ani puborectalis, pubococcygeus and iliococcygeus and the coccygeus muscles. The female pelvic floor is a compromise between several competing priorities including reproduction, urinary and fecal continence, and pelvic organ support. During vaginal childbirth the pelvic floor is subjected to substantial distension during crowning of the fetal head, resulting in alterations in morphology and function. A substantial minority of women experience macroscopic trauma avulsion injury 1 4 and/or traumatic overdistension 5 of the levator ani muscle, especially of the puborectalis muscle 6, and there may also be trauma to fascial components of the pelvic floor, although such fascial changes are much less well defined. This is associated with later morbidity, especially female pelvic organ prolapse 7,, particularly of the anterior and central compartments 9. There is also evidence for delivery-related impairment of pelvic floor innervation, in particular of the pudendal nerve 10,11, although it remains unclear as to whether neuropathic changes are permanent and whether they play a role in the pathogenesis of clinical conditions. It seems likely that there may be an impact on the innervation of the urethral rhabdosphincter, since there is some evidence of denervation of this structure in stress incontinent women 12 and evidence of an association between parity and urethral function 13. However, such evidence is missing for any link between denervation injury and levator ani function or the likely pathophysiological manifestation of levator ani impairment, i.e. female pelvic organ prolapse. If there was major permanent denervation of the levator ani muscle one would expect atrophic changes, i.e. a reduction in muscle mass, and impaired function. Thus, the aim of this study was to determine whether there is any significant reduction in the thickness of the levator ani muscle on average 4.5 months after a first vaginal delivery compared to values obtained in the late third trimester. In addition, we also studied peripartum changes in pelvic floor muscle function as determined by pelvic floor ultrasound imaging 14. METHODS This was a prospective observational cohort study using translabial ultrasound volume datasets obtained Correspondence to: Prof. H. P. Dietz, Sydney Medical School Nepean, University of Sydney, Penrith, New South Wales, Australia hpdietz@bigpond.com Accepted: 1 June 2013 Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER

2 4 Dixit et al. Figure 1 The plane of minimal hiatal dimensions horizontal white line is bound by the symphysis pubis S anteriorly and the levator ani L posteriorly as seen in this mid-sagittal image a. Once this plane is identified, an image in the axial plane is obtained b for measurement of hiatal diameter and pelvic floor muscle thickness. A, anal canal; R, rectal ampulla; U, urethra; V, vagina. Figure 2 Determination of anteroposterior AP hiatal diameter reduction and levator ani muscle thickness. a The axial plane as in Figure 1b, with the vertical line indicating the AP diameter of the hiatus. b Reduction of this diameter on pelvic floor muscle contraction. c Muscle thickness measured at four locations oblique lines lateral to vagina and anorectum. A, anal canal; L, levator ani; S, symphysis pubis; U, urethra; V, vagina. in the context of a perinatal trial investigating pelvic floor structure and function before and after the first childbirth. Unrelated data obtained in this trial have been published previously 5,15,16. Pregnant nulliparous women were recruited from the antenatal clinic of a tertiary obstetric unit. Participants were invited to attend an appointment between 36 and 3 weeks gestation and again at a mean of 4.5 months after delivery. After an interview and clinical examination, ultrasound volume datasets were acquired in the supine position after bladder emptying at rest, on Valsalva and on pelvic floor muscle contraction PFMC at both the antepartum and postpartum appointments as previously described 17. Acquisition was undertaken by staff trained by the senior author for a period of at least 3 months, and great care was taken to obtain an optimal pelvic floor muscle contraction, e.g. to avoid co-activation of abdominal musculature. Visual biofeedback was used to optimize the effect of a PFMC on bladder neck lift and hiatal reduction, and ultrasound operators were all trained in detecting coactivation of the abdominal muscles by direct observation of the patient s torso and by observing the effect of a voluntary contraction on pelvic floor structures. If there was a caudad shift of these structures on PFMC the patient was instructed to place a hand on her abdomen, to keep the abdominal wall soft, and to optimize the effect of a contraction by observing its effect on the ultrasound monitor. Offline analysis of volume datasets for muscle thickness and measures of levator ani function was performed using the software 4D View version 10.0 GE Kretz Ultrasound, Zipf, Austria by the first author blinded to all clinical data including delivery mode. Pelvic floor muscle bulk was characterized by quantifying muscle thickness in single axial planes after identification of the plane of minimal hiatal dimensions, i.e. the shortest distance between the pubic bone anteriorly and the levator ani muscle posteriorly in the mid-sagittal plane Figure 1. Muscle thickness was measured in four locations, i.e. right paravaginal lateral to the central aspect of the vagina, right pararectal lateral to the central aspect of the anorectal junction, left paravaginal and left pararectal as previously described Figure 2 1. Muscle thickness was measured on volumes acquired on PFMC. To ensure high repeatability, muscle thickness was measured in the axial plane within 1 cm of this plane. Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 3.

3 How common is pelvic floor muscle atrophy after childbirth? 5 Figure 3 The mid-sagittal plane at rest a and on pelvic floor muscle contraction PFMC b. The distance between the inferoposterior margin of the symphysis pubis * and the bladder neck ** is increased on PFMC. The difference between vertical measurements in a and b represents bladder neck lift on PFMC. A, anal canal; B, bladder; L, levator ani; R, rectal ampulla; S, symphysis pubis; Ut, uterus. To assess pelvic floor muscle function, we measured the anteroposterior AP diameter of the hiatus in the axial plane of minimal hiatal dimensions Figure 2 and bladder neck position in the mid-sagittal plane Figure 3 both at rest and on PFMC 17,1. The difference in AP diameters of the hiatus taken at rest and on PFMC indicates the reduction in hiatal diameter on PFMC AP diameter REST AP diameter PFMC. The difference in bladder neck positions at rest and on PFMC is a measure of bladder neck elevation bladder neck PFMC bladder neck REST. Levator avulsion was diagnosed as previously described, using tomographic imaging with a reference plane at the level of the plane of minimal dimensions and an interslice interval of 2.5 mm, requiring both the reference plane and those 2.5 and 5 mm above this plane to be abnormal for the diagnosis of a full avulsion 19. Pairwise comparisons were employed to identify peripartum changes of bladder neck elevation, reduction in AP diameter and muscle thickness. Minitab V13 Minitab Inc., State College, PA, USA was used for statisticalanalysis.allmeasuresusedtoassessmuscle function were normally distributed. Two-sample t-tests were used to study the association between peripartum changes in bladder neck elevation and reduction in AP diameter with delivery mode vaginal/abdominal delivery. ANOVA was used to carry out comparisons between multiple groups. We did not undertake power calculations due to the absence of suitable data in the literature, and since this was a sub-analysis of a parent study powered for different outcome measures. A test re-test series of 21 datasets was performed between P.D. and K.L.S.. Repeatability of all measured ultrasound parameters was tested using intraclass correlation coefficients ICCs single measures, absolute agreement definition using the SPSS V17 software SPSS Inc., Chicago, IL, USA after confirming no or minimal systematic bias by comparing means. The repeatability of measurement of bladder neck position and hiatal AP diameter was excellent for both rest and PFMC ICC between % CI and % CI The ICC for muscle thickness 4 measurements, was signifying substantial agreement. This study was an approved extension of a prospective study that was approved by the local Institutional Review Board SWAHS HREC RESULTS Two hundred and two participants were enrolled. They were seen at a mean gestational age of range, weeks. One hundred and sixty-five 2% women returned for the postpartum follow-up appointment at a mean of 135 SD, 47; range, 2 36 days after delivery. Of the 37 patients who did not attend, 26 were lost to follow-up, seven cancelled participation in the study, three moved location and one was pregnant again. In seven cases the ultrasound volume datasets could not be retrieved due to technical or clerical errors, leaving a total of 15 7% patients. All further analysis pertains to this number. The women s mean ± SD age was 26.2 ± 5.0 years and body mass index at first assessment was 25.6 ± 5.76 kg/m 2. Cesarean section was performed prelabor in 12 women, in the first stage of labor in 29 women and in the second stage in eight. Eightyeight women underwent normal vaginal delivery NVD, vacuum extraction was performed in 13 and forceps were used in eight. The mean gestational age at delivery was 40 range, to42+ 4 weeks and the mean birth weight was 343 g. The median length of the second stage of labor was 55 range, min. There were six levator avulsions five on the right, one bilateral. On ultrasound post-processing analysis, the average bladder neck elevation was 3.3 ± 3.2 mm at the antepartum assessment, compared with 2.3 ± 2.2 mm postpartum P = This reduction was not associated with delivery mode Figure 4a. With regards to the reduction in hiatal diameters on PFMC 7. ± 4.5 mm postpartum vs.6± 4.6 mm antepartum, P = 0.03, this was also less pronounced at the postpartum visit. This peripartum change in hiatal reduction was associated with delivery mode 0.5 mm SD 4.3 in Cesarean section vs 1.4 mm SD 4.5 in the vaginal delivery group; P = 0.013, but a breakdown of data according to delivery mode did not show any consistent pattern Figure 4b. On analyzing muscle thickness we found no evidence of significant change peripartum. Mean measurements at both time points were 10.1 mm P = 0.76, and this was irrespective of delivery mode peripartal change in mean muscle thickness after Cesarean section, Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 3.

4 6 Dixit et al. a Delivery mode N Mean SD Elective CS CS in 1 st stage CS in 2 nd stage NVD Vacuum Forceps Change in cm b Delivery mode N Mean SD Elective CS CS in 1 st stage CS in 2 nd stage NVD Vacuum Forceps Change in cm c Delivery mode N Mean SD Elective CS CS in 1 st stage CS in 2 nd stage NVD Vacuum Forceps Change in cm Figure 4 ANOVA graphs showing mean and 95% CI of peripartal change, relative to delivery mode, in: a bladder neck lift on pelvic floor muscle contraction PFMC; b reduction of anteroposterior hiatal diameter on PFMC; and c puborectalis muscle thickness n = 15. There were no significant associations. CS, Cesarean section. 0.4 mm SD 2.5 vs +0.1 mm SD 1.9 after vaginal delivery; not significant on two-sample t-test. There were no significant differences even when elective Cesarean section was analyzed separately. Figure 4c shows an ANOVA graph of peripartal change in muscle thickness relative to delivery mode. Virtually identical results were obtained when these three analyses were repeated after exclusion of the six patients diagnosed with levator avulsion. DISCUSSION Pelvic floor dysfunction encompasses a number of common urogynecological complaints that can have a significant negative impact on many aspects of a woman s life. The etiology of pelvic floor dysfunction is likely to be complex and multifactorial 20. Childbirth is one of the most studied potential etiological factors. Maternal expulsive efforts during labor and the force exerted by uterine contractions during descent of the fetal head may induce stretch and compression to the nerve supply of the pelvic floor which may cause nerve ischemia, neurapraxia and impaired nerve function 21. Neurophysiological studies have repeatedly demonstrated evidence of denervation injury of the pelvic floor after vaginal delivery 11,22. It is believed that nerve injury sustained during childbirth may lead to muscular atrophy and pelvic floor dysfunction hence altering pelvic floor function and morphology. Branham et al. 23 reported anatomical improvement of the appearance of the levator ani muscle when comparing serial magnetic resonance imaging MRI scans performed at 6 weeks and again at 6 months postpartum in primiparous women. The authors hypothesized a neuropathic mechanism of injury and subsequent recovery of muscle bulk 23. However, a recent study using a similar ultrasound methodology argued against the likelihood of muscle atrophy even in women with vesicovaginal/rectovaginal fistula, which is generally Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 3.

5 How common is pelvic floor muscle atrophy after childbirth? 7 the result of severely obstructed labor 24,andthereisno published evidence of changes suggestive of levator ani denervation in symptomatic women. In fact, most women with marked functional impairment seem to suffer from levator avulsion, i.e. macroscopic trauma to the insertion of the puborectalis muscle on the inferior pubic ramus 25. In this study we have found a mild impairment of pelvic floor muscle function after childbirth, consistent with previously published data 26,27. There was less bladder neck elevation and less shortening in hiatal AP diameter on PFMC after childbirth. Delivery mode does not seem to matter much in this context. This may be due to the fact that ultrasound measures of pelvic floor muscle function determine displacement rather than strength or force. An increase in tissue distensibility after vaginal childbirth due to trauma to fascia or muscle may increase displacement for a given force, compensating for a reduction in force due to levator trauma, which affects 10 30% of primiparae after vaginal childbirth 1,2,2 31. On the other hand, a reduction in pelvic floor muscle function may also be due to denervation. This study has provided no evidence for such a hypothesis as there was no reduction in muscle bulk on comparing ante- and postpartum findings. There are some limitations to this study that are worthy of mention. The study population was largely Caucasian and hence the findings in this study may not be applicable to other ethnic groups. Furthermore, the final analysis was performed on only 7% of the original participants due loss of follow-up and missing volume data from technical or clerical errors which has reduced the power of the study. We were unable to obtain neurophysiological data to corroborate our findings, with concentric needle electromyography CNE preferable over pudendal nerve terminal motor latencies 32.However,evenifwehadhad the means of using CNE in our patients, this would very likely have impaired follow-up rates compared to a completely non-invasive approach. Furthermore, it is recognized that we determined muscle thickness only at the level of the puborectalis muscle, which might limit the sensitivity of our study in detecting atrophic changes. It is likely that some of the effects of levator activation such as bladder neck elevation are at least partly due to other components of the levator ani, such as the iliococcygeus muscle. The iliococcygeus is located more cranially and is also much thinner than the puborectalis, making it more difficult to assess. Evaluation of iliococcygeus atrophy would likely require MRI rather than ultrasound, and this method is not routinely available to us. It is possible that in some patients a degree of atrophy may have been masked by fat replacement or scar formation which may mimic unchanged muscle bulk. However, it is difficult to see how one would control for such confounders, except by obtaining multiple muscle biopsies. Levator avulsion could be another potential confounder, but exclusion of those patients resulted in virtually identical results. Finally, this study is comparing late pregnancy measurements to postnatal measurements. It is conceivable that pregnancy may have resulted in significant changes compared to the nonpregnant state, which would confound our analysis. The thickness measurements obtained by us mean of about 10 mm both before and after childbirth are somewhat higher than those obtained by other researchers in nulliparae 1,33 which may be due to our methodology that is, measurement of muscle thickness on pelvic floor muscle contraction, the population, or a pregnancy effect. However, any such confounding effect would strengthen rather than weaken the conclusions of this study. Clearly, if one were to postulate a transient effect of pregnancyassociated muscle hypertrophy, then it would be all the more reassuring that very similar measurements were obtained on average 4.5 months after childbirth. In conclusion, while we have found some impairment of pelvic floor muscle function at a mean follow-up time of 4.5 months after a first delivery, there was no significant change in sonographically determined levator ani muscle thickness. This finding argues against obstetric neuropathy as a major factor in the pathophysiology of pelvic floor disorders. ACKNOWLEDGMENT The study was partly funded by Australian Women & Children s Research Foundation OZWAC, Penrith, Australia. DISCLOSURE H.P. Dietz has acted as consultant for American Medical Systems, Continence Control Systems and Materna. He has accepted speaker s fees from Astellas, General Electric and American Medical Systems and has benefited from an educational grant by General Electric. None of the other authors has a conflict of interest. REFERENCES 1. Dietz HP, Lanzarone V. Levator trauma after vaginal delivery. Obstet Gynecol 2005; 106: Krofta L, Otcenasek M, Kasikova E, Feyereisl J. Pubococcygeus puborectalis trauma after forceps delivery: evaluation of the levator ani muscle with 3D/4D ultrasound. Int Urogynecol J 2009; 20: Cassado Garriga J, Pessarodona Isern A, Espuna Pons M, Duran Retamal M, Felgueros Fabregas A, Rodriguez-Carballeira M. Tridimensional sonographic anatomical changes on pelvic floor muscle according to the type of delivery. Int Urogynecol J 2011; 22: Shek K, Chantarasorn V, Dietz H. Can levator avulsion be predicted antenatally? Am J Obstet Gynecol 2010; 202: e Shek K, Dietz H. Intrapartum risk factors of levator trauma. BJOG 2010; 117: Svabik K, Shek KL, Dietz HP. How much does the puborectalis muscle have to stretch in childbirth? Ultrasound Obstet Gynecol 200; 32: DeLancey J, Morgan D, Fenner D, Kearney R, Guire K, Miller J, Hussain H, Umek W, Hsu Y, Ashton-Miller J. Comparison of levator ani muscle defects and function in women with Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 3.

6 Dixit et al. and without pelvic organ prolapse. Obstet Gynecol 2007; 109: Dietz HP, Steensma AB. The prevalence of major abnormalities of the levator ani in urogynaecological patients. BJOG 2006; 113: Dietz H, Simpson J. Levator trauma is associated with pelvic organ prolapse. BJOG 200; 115: Snooks SJ, Swash M, Mathers SE, Henry MM. Effect of vaginal delivery on the pelvic floora 5-year follow-up. BrJSurg1990; 77: Allen RE, Hosker GL, Smith AR, Warrell DW. Pelvic floor damage and childbirtha neurophysiological study. BrJObstet Gynaecol 1990; 97: Kenton K, Mueller E, Brubaker L. Continent women have better urethral neuromuscular function than those with stress incontinence. Int Urogynecol J 2011; 22: Dietz HP, Clarke B. The urethral pressure profile and ultrasound imaging of the lower urinary tract. Int Urogynecol J 2001; 12: Dietz HP. Pelvic floor ultrasound in incontinence: What s in it for the surgeon? Int Urogynecol J 2011; 22: Shek K, Dietz H. The effect of childbirth on hiatal dimensions: a prospective observational study. Obstet Gynecol 2009; 113: Shek K, Chantarasorn V, Dietz H. The effect of childbirth on urethral mobility. JUrol2010; 14: Dietz HP. Ultrasound imaging of the pelvic floor. Part I: two-dimensional aspects. Review Ultrasound Obstet Gynecol 2004; 23: Dietz H, Shek K, Clarke B. Biometry of the pubovisceral muscle and levator hiatus by three-dimensional pelvic floor ultrasound. Ultrasound Obstet Gynecol 2005; 25: Dietz H, Bernardo M, Kirby A, Shek K. Minimal criteria for the diagnosis of avulsion of the puborectalis muscle by tomographic ultrasound. Int Urogynecol J 2011; 22: Bump R, Norton P. Epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin 199; 25: Jou ILK, Shen C, Yamano Y. Changes in conduction, blood flow, histology and neurological status following acute nervestretch injury induced by femoral lengthening. J Orthop Res 2000; 1: Snooks SJ, Swash M. Abnormalities of the innervation of the urethral striated sphincter musculature in incontinence. Br J Urol 194; 56: Branham JTV, Jaffe T, Crockett M, South M, Jamison M, Weidner A. Levator ani abnormality 6 weeks after delivery persists at 6 months. Am J Obstet Gynecol 2007; 197: e Dietz HP, Habtemariam T, Williams G. Pelvic floor structure in women with vesicovaginal fistula. J Urol 2012; 1: Sarma S, Hersch M, Siva S, Dietz H, Moore K. Women who cannot contract their pelvic floor muscles: avulsion or denervation? Neurourol Urodyn 2009; 2: Peschers UM, Schaer GN, DeLancey JO, Schuessler B. Levator ani function before and after childbirth. BrJObstetGynaecol 1997; 104: Dietz H. Levator function before and after childbirth. Aust NZ J Obstet Gynaecol 2004; 44: Blasi I, Fuchs I, D Amico R, Vinci V, La Sala G, Mazza V, Henrich W. Intrapartum translabial three-dimensional ultrasound visualization of levator trauma. Ultrasound Obstet Gynecol 2011; 37: Chan S, Cheung R, Yiu A, Lee L, Pang A, Choy K, Leung T, Chung T. Prevalence of levator ani muscle injury in Chinese primiparous women after first delivery. Ultrasound Obstet Gynecol 2012; 39: Albrich S, Laterza R, Skala C, Salvatore S, Koelbl H, Naumann G. Impact of mode of delivery on levator morphology: a prospective observational study with 3D ultrasound early in the postpartum period. BJOG 2012; 119: Eisenberg V, Brecher S, Yodfat I, Bitman G, Achiron R, Schiff E, Alcalay M. Risk factors for levator avulsion trauma in women with obstetric anal sphincter injuries. Ultrasound Obstet Gynecol 2011; 3: Vodusek DB. Clinical neurophysiological tests in urogynaecology. Int Urogynecol J 2000; 11: Singh K, Reid WM, Berger LA. Magnetic resonance imaging of normal levator ani anatomy and function. Obstet Gynecol 2002; 99: Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2014; 43: 3.