Vaginal delivery has been established as the most important risk factor for developing symptomatic pelvic organ prolapse1 and stress urinary incontinence2 later in life. The levator ani muscle plays a significant role in supporting the pelvic organs.3 The change in support caused by vaginal delivery is often explained by changes in the muscle morphology of the levator ani surrounding the levator hiatus, which allows for the passage of the urethra, vagina, and rectum. Both overdistension and defects of the muscle seen after vaginal delivery have been related to the pathogenesis of pelvic organ prolapse.4 To protect the pelvic floor, caesarean delivery has been suggested as an alternative method, but cesarean delivery does not seem to fully protect against stress urinary incontinence or symptomatic pelvic organ prolapse.1,2,5 One explanation could be that pregnancy itself changes the support system.
Few studies have monitored the pelvic floor during pregnancy. One study has shown a decrease in pelvic organ support using pelvic organ prolapse quantification in nulliparous pregnant women throughout pregnancy.6 Ultrasonographic data indirectly support an effect of pregnancy on the pelvic organ support, because women delivering by cesarean have been found to have a smaller levator hiatus area after childbirth than in late pregnancy.7,8 Moreover, an increase in mobility of the bladder neck and urethra during pregnancy has been shown to persist 4 months postpartum in women undergoing cesarean delivery.9
To date, there is a lack of longitudinal studies monitoring the levator ani muscle and the levator hiatus during pregnancy. The primary aim of this study was to investigate whether pregnancy affects the levator hiatus dimensions and the position and mobility of both the bladder neck and levator ani muscle in nulliparous pregnant women using three-dimensional and four-dimensional transperineal ultrasonography. A second aim was to study whether measurements at 21 weeks of gestation were associated with possible changes.
MATERIALS AND METHODS
This prospective cohort study was performed at Akershus University Hospital, Lørenskog, Norway, from January 2010 to July 2011. All nulliparous pregnant women were invited to participate when they attended their routine second-trimester ultrasonographic examination at 18–22 weeks of gestation. Women with a previous pregnancy of more than 16 weeks duration, ongoing multiple pregnancy, or serious illness were excluded. Scandinavian-speaking women older than age 18 years were included. Exclusion criteria during the study were miscarriage, stillbirth, or delivery before the second examination at 37 weeks of gestation. The Regional Ethics Committee (REK Sør-Øst D 2009/170) and the Norwegian Social Science Data Service (2799026) approved the study, and all participants gave their written informed consent to participate.
Demographic data including age, body mass index (calculated as weight [kg] / height2 [m]), gestational age, marital status, and education were collected from the women's electronic medical records and from a questionnaire exploring additional background data. Transperineal ultrasonography was performed twice, between 18 weeks and 22 weeks of gestation and in the late third trimester, by two trained examiners.10 The ultrasonographic volumes were acquired with the women in the supine lithotomy position after voiding using the GE Kretz Voluson E8 system with a 4- to 8-MHz curved-array three-dimensional and four-dimensional ultrasonographic transducer at rest, during contraction, and during Valsalva maneuver using the methodology previously described.10 Before the ultrasonographic acquisition, all participants were instructed by a physical therapist in how to perform a correct pelvic floor muscle contraction, which was defined as a cranioventral shift of the levator ani. Correct contraction was verified by observation of inward perineal movement (classified as yes or no) and vaginal palpation.11,12
Valsalva maneuver was performed for at least 6 seconds,13 correct Valsalva maneuver being defined as a caudodorsal shift of the levator ani. Each maneuver was recorded three times.
The ultrasonographic images were stored offline by anonymous code numbers and analyzed using four-dimensional software. Of the three contractions recorded, the volume with the best contraction, defined as the one with the shortest anteroposterior diameter from the posteroinferior margin of the symphysis pubis to the rectal sling in the midsagittal plane, was chosen for analysis. For Valsalva maneuver, the volume with the largest anteroposterior diameter and the most caudal displacement of the bladder neck and levator plate in the midsagittal plane was chosen for analysis.
Render mode14 around the plane of minimal dimensions was used when measuring the levator hiatus dimensions: anteroposterior and transverse diameter and levator hiatus area (Fig. 1). Volume rendering is a technique used to display a two-dimensional projection of a three-dimensional structure. In this case, ultrasonography of the levator ani muscle in the midsagittal, the axial, and the coronal two-dimensional planes are rendered into a semitransparent “rendered” representation of the pelvic floor. A standard rendered image of the levator hiatus results in an axial image that corresponds to observing the patient's pelvic floor from below.15
Overdistension of the levator ani was defined as enlargement of the hiatal area during Valsalva maneuver of more than 20%16 between two examination points. Ballooning was defined as the levator hiatus area during Valsalva maneuver of more than 25 cm2.17
In the midsagittal plane, the positions of the bladder neck and levator plane were determined using a coordinate system with a horizontal reference line (x-axis) through the inferior symphyseal margin18 and a vertical line (y-axis) at an angle 90° on the horizontal line with the intersection at the tip of the symphysis pubis (Fig. 2). The position of the bladder neck and levator plate were presented on the x-axis and y-axis, and a change having a negative value represents a more posterior or caudal position than was observed at 21 weeks of gestation. Mobility of the bladder neck and levator plate was calculated as the hypotenuse of a right-angle triangle (displacement=√[Δx2+Δy2]) as described by Peschers.19
Four trained investigators analyzed the images. The two most experienced investigators, in which interrater data have been found to be good to very good,10 were used as the gold standard. Interobserver values between investigators were calculated pairwise for both the axial and sagittal measurements using 50 data sets before assessing the images in the present study. We aimed for intraclass correlation coefficients of 0.8 or more; however, 0.6 was accepted if reached after 50 data sets. The assessors were blinded to clinical data and images were analyzed in random order to avoid pairwise analysis, because it was not possible to blind the assessors to gestational weeks owing to visibility of the fetal head on most images taken at 37 weeks of gestation.
Statistical analysis was performed using SPSS 15.0. A power calculation was performed using results from a previous study20 in which a sample size of 47 was required to detect a 5% change in levator hiatus area at rest with a two-sided α of 0.05 and a power of 80%. To test interobserver agreement, intraclass correlation coefficients, bias, and 95% limits of agreement for the four examiners were calculated. Demographic data and ultrasonographic measurements at 21 weeks and 37 weeks of gestation are reported as mean values with standard deviation (SD) and frequencies with percentages. Normal distribution was found for all variables. Independent-sample t test and χ2 test were used to assess differences between groups. Changes in ultrasonographic measurements from 21 weeks to 37 weeks of gestation were calculated as mean differences and SD using a paired-sample t test. Pearson correlations coefficient was used to calculate correlations between findings at 21 weeks of gestation and changes in measurements to 37 weeks of gestation. P<.05 was considered significant.
Three hundred pregnant nulliparous women were included in the study. At follow-up late in the third trimester, 17 had delivered before their study appointment, six chose not to continue, two were excluded owing to intrauterine fetal death, and one data set at 37 weeks of gestation was missing, leaving 274 paired data sets for comparison. Mean gestational age at first examination was 20.9 weeks (SD 1.4) and mean gestational age for the second examination was 37.0 weeks (SD 0.7). Demographic data are presented in Table 1. No differences in background variables were found between the women who completed the study and the women who did not. Our study sample was comparable to the total population of nulliparous women scheduled to deliver at Akershus University Hospital during the inclusion period (n=2,621) with respect to age and marital or cohabitation status, but more women in the sample had a university or college education (76.3% compared with 50.8%, P<.001).
Interobserver reliability was calculated for the analysis of ultrasonographic images by the four investigators. Intraclass correlation coefficients were found to be very good for all values (range 0.80–0.98), except for transverse diameter of the levator hiatus at rest and during Valsalva maneuver in which they were found to be good (range 0.62–0.89).
Mean values of levator hiatus dimensions at 21 weeks and 37 weeks of gestation are presented in Table 2. We found significant increases for all levator hiatus dimension measurements from 21 weeks to 37 weeks of gestation (Fig. 3, Video 1, available online at http://links.lww.com/AOG/A402). The most marked change was found for levator hiatus area at rest and during Valsalva maneuver with increases in mean area of 17.1% and 21.4%, respectively. Forty-eight percent (n=131) of the women were found to have increased their levator hiatus area during Valsalva maneuver from 21 weeks to 37 weeks of gestation by more than 20%, fulfilling the criterion for overdistension.16 Ballooning was found in 8% at 21 weeks of gestation and in 15% at 37 weeks of gestation. No significant differences in background variables were found between the groups with and without overdistension or ballooning, except for the women with ballooning at 37 weeks of gestation. They were 1.5 years older (P=.03) than the women without ballooning (data otherwise not shown).
Table 3 shows mean values of the position of the bladder neck and levator plate in the sagittal plane. All measurements for the bladder neck, except bladder neck on contraction, showed a more posterocaudal position at rest and during Valsalva maneuver at 37 weeks of gestation as compared with 21 weeks of gestation. The posterocaudal shift was most marked during Valsalva maneuver. The mobility of the bladder neck from 21 weeks to 37 weeks of gestation was found to be significantly increased at the end of the pregnancy, but for the levator plate, this was only true for mobility seen from rest to contraction (Table 4).
We did not find any correlation between levator hiatus measurements at 21 weeks of gestation and the change of levator hiatus size occurring between the first and second visits (from 21 weeks to 37 weeks of gestation). We did find a strong correlation between levator hiatus measurements in the axial plane at 21 weeks and 37 weeks of gestation (r=.8, P<.01). For measurements in the sagittal plane, strong correlations were found for the anteroposterior position of the levator plate (r=.7, P<.01) and bladder neck during Valsalva maneuver (r=.6, P<.01) between mid- and late pregnancy. The remaining measurements in the sagittal plane showed moderate correlations (r=.45, P<.01).
None of the variables listed in Table 1 were found to correlate with the change between time points seen in levator hiatus dimensions or bladder neck or levator plate position during pregnancy.
A systematic search on MEDLINE (English language; 1966–2013; search terms “pelvic floor,” “ultrasonography,” “ultrasound,” and “pregnancy”) revealed no prospective studies presenting data on morphologic changes of the levator hiatus dimensions during pregnancy.
The large change from 21 weeks to 37 weeks of gestation found in the present study implies that pregnancy in itself is an important factor to be taken into account when discussing risk factors for pelvic organ prolapse. Overdistension of the levator ani muscle has been suggested to be another sign of injury to the muscle in line with the more visible detachment of the muscle from the pubic bone detected after delivery.16 Overdistension has been found in up to 28% of women 4 months postpartum.16 In our study population, distension of the levator hiatus area of more than 20% during Valsalva maneuver was found in almost half of the women before delivery, suggesting a different cause of the change than birth trauma.
Ballooning has been defined as an independent risk factor for symptoms and signs of pelvic organ prolapse.4 Our findings of an increasing rate of ballooning during the study period could indicate that pregnancy itself is a risk factor for developing ballooning. However, it is uncertain whether predelivery ballooning and overdistension are reversible after delivery. A study comparing nulliparous women with primiparous women after cesarean delivery suggests that approximately 70% of the change in levator hiatus area found in late pregnancy was irreversible.9 These findings could imply that some of the irreversible change takes place during pregnancy and that pregnancy by itself is a risk factor for developing pelvic organ prolapse, but further longitudinal studies with long follow-up periods are needed for confirmation.
The increase in mobility of the bladder neck is in accordance with previous findings21 and has been shown to persist after delivery.9 Increase in bladder neck mobility and urethral closing pressure has been found to be associated with postpartum stress urinary incontinence.22 This has been suggested to partly explain why cesarean delivery does not fully protect against stress urinary incontinence and might also partly explain why approximately one-third of pregnant nulliparous women have some degree of stress urinary incontinence during their pregnancy.23 Conflicting results on this matter have been published,24,25 however, and the association between stress urinary incontinence and bladder neck mobility still remains unclear.
The correlations found between measurements at 21 weeks and 37 weeks of gestation could indicate that there is no compensatory mechanism for women with small levator hiatus to become more enlarged than women with large levator hiatus, but rather that change is constant independent of levator hiatus size at 21 weeks of gestation.
Strengths of the study are that the clinical examinations were done within a low variation of time and there was a low rate of women lost to follow-up. A major emphasis was put on achieving high interobserver reliability among the four investigators before starting the study. Very good to good intraclass correlation coefficients ensured reliable ultrasonographic data.10
Limitations of the study were that we were unable to monitor the patients from Day 1 of their pregnancies for practical reasons, and it is therefore impossible to comment on whether we have captured most of the pregnancy-induced changes in levator hiatus dimensions and bladder neck and levator plate mobility.
When comparing our measurement of the levator hiatus area in late pregnancy with findings in other studies, there is generally good accordance8,9 despite variation among the studies, which cannot be explained by demographic data. It is unclear whether the images were analyzed using the same methodology. This makes it difficult to compare existing data and underlines the necessity of precise methodologic descriptions of the recording and analyzing processes as well as the need for prospective longitudinal studies including women earlier in pregnancy.
In conclusion, we found an increase in all levator hiatus dimensions and bladder neck mobility from 21 weeks to 37 weeks of gestation. This may contribute to preparing the pelvic floor muscles for the considerable distension during vaginal delivery. The results indicate that the changes in pelvic organ support are not solely caused by delivery, but also by physiologic changes during pregnancy. Our findings support the idea that cesarean delivery cannot protect the pelvic floor from being exposed to changes that might cause pelvic organ prolapse or other pelvic floor symptoms later in life.
1. Gyhagen M, Bullarbo M, Nielsen TF, Milsom I. Prevalence and risk factors for pelvic organ prolapse 20 years after childbirth: a national cohort study in singleton primiparae after vaginal or caesarean delivery. BJOG 2013;120:152–60.
2. Gyhagen M, Bullarbo M, Nielsen TF, Milsom I. The prevalence of urinary incontinence 20 years after childbirth: a national cohort study in singleton primiparae after vaginal or caesarean delivery. BJOG 2013;120:144–51.
3. Ashton-Miller JA, DeLancey JO. Functional anatomy of the female pelvic floor. Ann N Y Acad Sci 2007;1101:266–96.
4. Dietz HP, Franco AV, Shek KL, Kirby A. Avulsion injury and levator hiatal ballooning: two independent risk factors for prolapse? An observational study. Acta Obstet Gynecol Scand 2012;91:211–4.
5. Rortveit G, Daltveit AK, Hannestad YS, Hunskaar S, Norwegian ES. Urinary incontinence after vaginal delivery or cesarean section. N Engl J Med 2003;348:900–7.
6. O'Boyle AL, O'Boyle JD, Calhoun B, Davis GD. Pelvic organ support in pregnancy and postpartum. Int Urogynecol J Pelvic Floor Dysfunct 2005;16:69–72.
7. Shek KL, Dietz HP. The effect of childbirth on hiatal dimensions. Obstet Gynecol 2009;113:1272–8.
8. Toozs-Hobson P, Balmforth J, Cardozo L, Khullar V, Athanasiou S. The effect of mode of delivery on pelvic floor functional anatomy. Int Urogynecol J Pelvic Floor Dysfunct 2008;19:407–16.
9. Shek KL, Kruger J, Dietz HP. The effect of pregnancy on hiatal dimensions and urethral mobility: an observational study. Int Urogynecol J 2012;23:1561–7.
10. Siafarikas F, Staer-Jensen J, Braekken IH, Bo K, Engh ME. Learning process for performing and analyzing 3D/4D transperineal ultrasound imaging and interobserver reliability study. Ultrasound Obstet Gynecol 2013;41:312–7.
11. Bø K, Kvarstein B, Hagen R, Larsen S. Pelvic floor muscle exercise for the treatment of female stress urinary incontinence. II. Validity of vaginal pressure measurements of pelvic floor muscle strength and the necessity of supplementary methods for control of correct contraction. Neurourol Urodyn 1990;9:479–87.
12. Bø K, Hagen RH, Kvarstein B, Larsen S. Pelvic floor muscle exercise for the treatment of female stress urinary incontinence. I. Reliability of vaginal pressure measurements of pelvic muscle strength. Neurourol Urodyn 1990;9:471–7.
13. Orejuela FJ, Shek KL, Dietz HP. The time factor in the assessment of prolapse and levator ballooning. Int Urogynecol J 2012;23:175–8.
14. Dietz HP, Wong V, Shek KL. A simplified method for determining hiatal biometry. Aust N Z J Obstet Gynaecol2011;51:540–3.
15. Dietz HP. Pelvic floor ultrasound: a review. Am J Obstet Gynecol 2010;202:321–34.
16. Shek KL, Dietz HP. Intrapartum risk factors for levator trauma. BJOG 2010;117:1485–92.
17. Dietz HP, Shek C, De Leon J, Steensma AB. Ballooning of the levator hiatus. Ultrasound Obstet Gynecol 2008;31:676–80.
18. Dietz HP, Steensma AB. Posterior compartment prolapse on two-dimensional and three-dimensional pelvic floor ultrasound: the distinction between true rectocele, perineal hypermobility and enterocele. Ultrasound Obstet Gynecol 2005;26:73–7.
19. Peschers U, Schaer G, Anthuber C, Delancey JO, Schuessler B. Changes in vesical neck mobility following vaginal delivery. Obstet Gynecol 1996;88:1001–6.
20. Braekken IH, Majida M, Engh ME, Bo K. Morphological changes after pelvic floor muscle training measured by 3-dimensional ultrasonography: a randomized controlled trial. Obstet Gynecol 2010;115:317–24.
21. Dietz HP, Eldridge A, Grace M, Clarke B. Does pregnancy affect pelvic organ mobility? Aust N Z J Obstet Gynaecol 2004;44:517–20.
22. Delancey JO. Why do women have stress urinary incontinence? Neurourol Urodyn 2010;29:S13–7.
23. Hilde G, Staer-Jensen J, Siafarikas F, Engh ME, Braekken IH, Bo K. Impact of childbirth and mode of delivery on vaginal resting pressure and on pelvic floor muscle strength and endurance. Am J Obstet Gynecol 2013;208:50.e1–7.
24. Jundt K, Scheer I, Schiessl B, Karl K, Friese K, Peschers UM. Incontinence, bladder neck mobility, and sphincter ruptures in primiparous women. Eur J Med Res 2010;15:246–52.
© 2013 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
25. Shek KL, Dietz HP, Kirby A. The effect of childbirth on urethral mobility: a prospective observational study. J Urol 2010;184:629–34.