SEARCH

SEARCH BY CITATION

Keywords:

  • 3D ultrasound;
  • birth trauma;
  • bladder neck;
  • paravaginal defects;
  • prolapse;
  • translabial ultrasound

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Objective

It is assumed that support of the female urethra and bladder is maintained by paraurethral and paravaginal fascial structures, with hypermobility resulting from delivery-related trauma. This study used three-dimensional translabial ultrasound to assess these structures and document peripartal changes.

Design

A clinical observational pilot study was performed on 26 nulliparous women recruited in the third trimester of pregnancy. They underwent translabial two- and three-dimensional ultrasound. Twenty-three women were again seen 2–5 months postpartum. The assessor was blinded against two-dimensional ultrasound and delivery data. Vaginal tenting was rated as being present, indeterminate or absent at each of three levels, and was correlated with bladder neck descent (BND) and urethral rotation on Valsalva maneuver.

Results

Tenting was visible at all levels in 21 of 26 women antepartally. In three women tenting was absent on one level; in two cases tenting was rated indeterminate. There was no significant difference in BND between women with visible tenting and those without. The BND range for women with intact tenting was 5.4–41.6 mm. Twenty-one of the 26 women were included in the postpartum analysis. Of these, obvious peripartal changes were documented in five. Loss of tenting did not correlate significantly with changes in BND.

Conclusions

Most nulliparous women showed evidence of intact paravaginal support structures. Tenting occurred in women with widely varying BND, implying that excess bladder neck mobility may be due to increased fascial compliance. Postnatally, fascial disruption was suspected in a minority of patients only. In some women delivery-related changes may be due to attenuation rather than disruption of structures. Copyright © 2003 ISUOG. Published by John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

It is thought that the support of the urethra and bladder in the female depends largely on the integrity of paraurethral and paravaginal musculofascial structures, variously termed pubourethral ligament, pubovesical muscle and paravaginal and/or endopelvic fascia1. These structures attach the urethra, bladder neck, trigone and posterior bladder wall to the symphysis pubis and pelvic sidewall, thereby guaranteeing their intrapelvic position and continence of urine. Hypermobility or prolapse of urethra and bladder are assumed to be due to traumatic damage to the support structures, most likely sustained during childbirth2, 3. Consequently, it is assumed that surgical repair of such defects should cure prolapse and incontinence, and this hypothesis has found widespread acceptance among gynecological surgeons since its formulation in the 1970s3–6.

Recent advances in imaging technology have facilitated assessment of pelvic floor structures including the levator muscle and paravaginal and paraurethral tissues. While assessment of paravaginal defects via transabdominal ultrasound7, 8 has been claimed to be possible, this appears questionable due to the absence of clear reference points and the confounding effect of bladder and rectal filling as well as uterine position and size. A recent attempt at validating a transabdominal assessment technique9 concluded that it was without clinical value.

Magnetic resonance imaging (MRI) is increasingly used to assess pelvic floor structures. The main advantage of this method is that the acquired volume of data allows access to any arbitrarily defined plane. The first paper published in this field demonstrated that the attachments of urethra, bladder neck and vagina to the levator ani and pelvic sidewall can be visualized by axial or transverse MRI, and that retropubic suspensions recreate the ‘tenting’ appearance lost in some symptomatic women10. However, while paravaginal tissues are generally visible on the transverse planes generated to image the levator ani, they have so far received little attention, with all recently published studies focusing on muscle rather than fascia11–14.

Translabial ultrasound has been used extensively for pelvic floor assessment15–19, but generally only the midsagittal plane is imaged in clinical practice and research applications since this plane provides a convenient point of reference, the symphysis pubis. Parasagittal planes have not been described. Direct assessment of paravaginal defects appears impractical although their effect (urethral and bladder neck hypermobility and descent of a cystocele) can be defined19.

The advent of three-dimensional (3D) ultrasound made it possible to image paravaginal support structures by giving access to transverse planes similar to those employed in MRI. So far, only intracavitary techniques such as transvaginal and transrectal imaging have been used to document the state of paravaginal and paraurethral support structures20, 21.

This study used 3D translabial ultrasound in an attempt to define the extent and nature of traumatic damage to pelvic floor structures during delivery.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

This observational clinical study was conducted as a pilot project as part of a larger study of female pelvic floor function in pregnancy and puerperium. Originally, 200 nulliparous women were recruited in the antenatal clinic of a large tertiary hospital. Twenty-six of the 200 women were enrolled in this pilot project at the time of their second appointment for the main study (at 33–38 weeks of gestation). Inclusion criteria were participation in the main study and oral consent for the performance of a 3D ultrasound examination. Prior to the commencement of the pilot project, several leading manufacturers of ultrasound equipment were approached. Toshiba Australia agreed to supply both an ultrasound system (Toshiba Powervision 8000 v 3.0 with 3.75-MHz multifrequency convex transducer) and the necessary hard- and software for analysis of volume data. The ultrasound system was made available on three separate occasions and for approximately 2 weeks on each occasion, allowing the assessment of women both ante- and postnatally. Equipment availability therefore affected inclusion in the study. Ethics committee approval had been obtained for the main study (SESAHS EC 99/184); this study extension was deemed exempt from formal approval due to the minimal technical alteration in data acquisition.

After a standardized interview, paper-towel test and bladder emptying with flowmetry, routine translabial two-dimensional (2D) imaging was performed as described elsewhere19. Amongst other parameters, the descent of the internal urethral meatus or bladder neck on maximal Valsalva maneuver was defined relative to the inferoposterior margin of the symphysis pubis. The degree of urethral rotation on Valsalva maneuver was defined as the increase in angle between the central transducer axis and the axis of the proximal urethra17. The best of at least three maximal Valsalva maneuvers was used for evaluation.

Three-dimensional volume datasets were then acquired in both the anteroposterior and transverse directions. All imaging was performed by one of the authors (H.P.D.). In total between three and 16 volumes per patient were stored. The lack of a position sensor required repeated attempts in a large number of acquisitions due to asymmetrical movement of the transducer or faulty positioning. Both symmetry of the sweep (i.e. geometrical accuracy) and lateral reach were confirmed immediately after data acquisition and the process repeated as often as necessary to ensure inclusion of paravaginal areas and both levator muscles between the symphysis pubis and the anal canal. The same procedure was followed at the time of the return visit approximately 3 months postpartum.

Analysis of volume datasets was performed, with the assessor (H.P.D.) blinded against delivery details and all parameters of anterior vaginal wall descent, on a desktop computer (Micron Clientpro) supplied by Toshiba Australia (North Ryde, NSW), using proprietary software (Powerview), several months after data acquisition.

Identification of structures was easiest in the transverse plane and in ‘opacity mode’, a method of image reconstruction that renders pixels semitransparent. In order to orient 2D transverse images along the plane of the levator ani muscle, a pivot of about 10–20° was used (dorsocaudal to cranioventral tilt). Of the most technically satisfactory datasets, printouts were obtained at three levels, demonstrating the levator ani and the paravaginal structures in the lower, middle and upper vagina. The bladder neck was used as a point of reference, with images reconstructed at approximately mid-urethral level, just below the level of the internal urethral meatus, and approximately 1.5–2 cm above the bladder neck. This should result in planes approximately 2, 4 and 6 cm above the hymen. Medilink Australia (Silverwater, NSW) provided a Codonics NP 1660 M Medical Printer and the media required to produce medical quality printouts of approximately 500 transverse plane images.

Tenting of the vagina, i.e., the presence of ventrolateral vaginal sulci, was described at each of the three levels and scored as ‘present’, ‘indeterminate’ or ‘absent’ for both right and left sides (six assessments per volume dataset). Identification of sides was made possible by a consistent direction of acquisition: the sweep direction was always from the patient's right to the patient's left since pulling on the transducer cord gives a smoother sweep. This results in the right side of any resulting image representing the patient's left side and vice versa. The view obtained is therefore equivalent to a view of the urogenital diaphragm from below. Figure 1 shows transverse views of the levator hiatus on MRI and 3D volume ultrasound. Only the ultrasound image was obtained as part of this particular study.

thumbnail image

Figure 1. Magnetic resonance (a) and three-dimensional volume ultrasound (b) images of the levator hiatus in the axial or transverse plane (different patients). Urethra, vagina and rectum are visible between parts of the levator ani bilaterally. White arrows indicate lateral vaginal sulci (‘tenting’). The magnetic resonance image is courtesy of Dr O. A. Adekanmi and Mr R. M. Freeman, Urogynaecology Unit, Derriford Hospital, Plymouth, UK.

Download figure to PowerPoint

In order to assess interobserver variability, another of the authors (A.B.S.), a gynecologist trained in pelvic floor ultrasound imaging and blinded against all other results, assessed 25 randomly selected volumes. The assessment was again performed in three planes per volume (75 frames) for tenting in both the left and the right paraurethral/paravaginal space (150 assessments); identical classifications were obtained in 134 of these (89%). Of the 16 assessments that showed discrepancies, 11 (69%) affected the lowermost plane.

To determine the relevance of tenting for overall support of the anterior vaginal wall, presence or absence of tenting was tested against the main parameter of anterior vaginal wall support, bladder neck descent (BND)17, 19, 22. To define peripartum traumatic change, antepartum and postpartum tenting at the three levels were compared. Women with reduced tenting at any level were compared to those without change, with BND and urethral rotation being used as parameters of anterior vaginal wall support. Reduced tenting was also tested against delivery mode.

In a second attempt to define peripartum change in paravaginal support, ante- and postpartum images at all three levels were compared directly (Figures 2–4), rated unchanged, better, worse or indeterminate, and the above analysis repeated. Comparative statistics were performed after confirmation of normal distribution by histogram assessment and Kolmogorov–Smirnov test. All tested quantitative continuous variables were normally distributed. A two-tailed t-test was used for analysis of peripartal changes in BND and urethral rotation in women with intact and absent tenting as well as for comparison of visual change vs. these two parameters.

thumbnail image

Figure 2. Antepartum (a) and postpartum (b) three-dimensional ultrasound images of transverse sections at Level II. In this patient no obvious change was detected. Arrows indicate tenting.

Download figure to PowerPoint

thumbnail image

Figure 3. Antepartum (a) and postpartum (b) three-dimensional ultrasound images of transverse sections at Level II. Changed paravaginal support is evident on the patient's right (left on image). Arrows indicate tenting.

Download figure to PowerPoint

thumbnail image

Figure 4. Antepartum (a) and postpartum (b) three-dimensional ultrasound images of transverse sections at Level II showing marked bilateral alteration in paravaginal support. Arrows indicate tenting.

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Twenty-six nulliparous women were examined on average at 35 (range, 33–37) weeks' gestation. Their mean age was 30.5 (range, 18–41) years. Twenty-three were able to return at an average of 3 (range, 2.3–4.1) months postpartum; one had moved interstate, one did not attend despite three reminders, and in one case the 3D equipment was unavailable. Sixteen of the 23 women delivered vaginally (14 normal vaginal deliveries, two ventouse deliveries) and seven underwent a Cesarean section (one before the onset of labor, five in the first stage of labor and one in the second stage).

At the antepartum examination, tenting was visible at all levels in 21 of the 26 women. In three cases one of the three levels was rated indeterminate and in one of these another level had an absence of tenting. In a further two there was an absence of tenting at one level, giving a total of three cases with absence of tenting at one of the three levels; there were no significant differences in anterior vaginal wall mobility between these three and the remaining 23 women. In the 23 women with intact/indeterminate tenting, the BND was 21.2 (SD, 10.6) mm and in the three women with absence of tenting at one of the levels it was 27.8 (SD, 14.1) mm (P = 0.5). Intact tenting was seen with a wide range of anterior vaginal wall mobility (BND range, 5.4–41.6 mm).

Two sets of postpartum volumes were found to be of inferior quality due to an acquisition error, leaving 21 sets of postpartum volumes for comparison with the antepartum ones; the subsequent analysis is limited to these 21 volume pairs. Table 1 gives data on changes between antepartum and postpartum tenting. Five women showed a loss of tenting in at least one of the three planes; of these, four had a vaginal delivery and one was delivered by Cesarean section after 10 h in first-stage labor, with the fetal head palpable at the ischial spines.

Table 1. Peripartal change in tenting as imaged in the transverse plane (n = 21)
Change in tentingPlane
123
  1. In five women tenting was lost in at least one plane.

Same (n)171617
Better (n) 0 0 0
Worse (n) 2 4 3
Indeterminate (n) 2 1 1

On the assumption that loss of paravaginal support should be evident as a peripartum increase in urethral rotation and BND, an analysis was carried out comparing these parameters in women with and without loss of tenting; results are shown in Table 2. While there was an apparent difference between the two groups, the SDs are so large that a sample of approximately 270 observations would be required to provide 80% power to prove this difference in either BND or urethral rotation to be statistically significant.

Table 2. Peripartal change in tenting on transverse imaging and change in parameters of anterior vaginal wall mobility
 Tenting lost (n = 5)Tenting intact (n = 16)P*
  • *

    t-test. BND, bladder neck descent.

Change in BND (mm, mean (SD))− 16.4 (21.2)− 9.36 (11.3)0.5
Change in urethral rotation (°, mean (SD))38 (41.5)23.1 (36.6)0.5

Independent of the plane-by-plane identification of vaginal tenting, we also carried out direct visual comparison of ante- and postpartum images derived from 3D volumes. Table 3 gives results for changes in BND and urethral rotation in women who appeared unchanged and in those who appeared to have suffered a deterioration in paravaginal support. In 12 cases there seemed to be no change, in six there was apparent deterioration (not necessarily a loss of tenting), and in two cases postpartum images seemed to show better support. No significant differences were documented for BND or urethral rotation.

Table 3. Peripartal change in overall paravaginal support (by direct visual comparison) and change in parameters of anterior vaginal wall mobility
 Paravaginal support ante/postpartum
Same (n = 12)Worse (n = 6)P*
  • *

    t-test. BND, bladder neck descent; NS, not significant.

Change in BND (mm, mean (SD))− 9.14 (10.3)− 9.8 (17.5)NS
Change in urethral rotation (°, mean (SD))20.8 (34.5)25 (37.2)NS

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

This pilot project was designed to document peripartal change in anterior vaginal wall support in prospectively recruited nulliparous women, using translabial 3D ultrasound imaging. Due to limited equipment availability, only a small subgroup of the original study population was assessed by 3D ultrasound. As a result, the power of this pilot project to demonstrate significant peripartum changes is limited. Another limiting factor is the lack of quantification due to the lack of a position sensor, an issue that would have been addressed by the use of systems that use automatic volume acquisition. Furthermore, clinical examinations were not performed to increase patient compliance although it is acknowledged that such assessments would have been desirable in order to validate the method.

Interobserver variability testing showed a high degree of concordance between observers. However, these assessments were performed on the same volumes, resulting in an incomplete assessment of reproducibility. Bladder and rectal filling as well as head engagement may conceivably influence appearances. We are currently undertaking studies to address the issues of both reproducibility and validity.

The main advantage of 3D or ‘volume’ ultrasound is the fact that it allows access to the axial or transverse plane, i.e., the plane in which both levator hiatus and paravaginal support structures can be visualized. Until now, only rarely has the transverse plane been assessed by ultrasound. In 1999 Wisser et al.20 first described ultrasound imaging of the paravaginal fascia using a transrectal 3D probe, and a similar system had previously been used for the planning of pediatric pelvic floor surgery23. More recently, the former group published another small series assessing antenatal findings in 14 primigravid women21, attempting to define ‘normal’ anatomy.

The disadvantages of invasive methods such as transrectal or transvaginal imaging are obvious and include distortion of tissues and patient discomfort. However, it has to be pointed out that an intraluminal transducer, withdrawn at a defined rate (e.g., motorized), makes external position sensors redundant and simplifies image acquisition as in the reports mentioned above.

In the present study 3D volume data was used to reconstruct transverse views at a tilt of 10–20°, allowing optimal demonstration of the levator hiatus. Volumes were assessed at three levels, the mid-urethra, near the bladder neck, and 1.5–2 cm above the bladder neck. In all but three women the paraurethral/paravaginal fascial suspension of urethra and vagina to the pelvic sidewall was clearly evident. The only study to assess this parameter in nulliparous women was published by Ochsenbein et al. in 200121, showing tenting to be visible in all 14 of the patients examined. The authors used transrectal ultrasound to obtain images of the paravaginal support structures; no attempt was made to assess these at different levels.

Magnetic resonance imaging can also demonstrate paraurethral and paravaginal support structures in the axial or transverse plane10, these being evident as a U- or H-shaped vagina with anterolateral extensions towards the pelvic sidewall. Some of the more recent MRI studies on this subject did not have sufficient resolution to assess the issue of paravaginal support structures11. Others focused exclusively on the levator ani and the traumatic effects of childbirth on this structure12–14. Others assessed organ position only24, which, in our opinion, is much more easily achieved using 2D ultrasound19. No MRI studies to date have assessed peripartum changes in paravaginal support.

Despite the limitations of the method used in this study, it was possible to document marked changes to paravaginal support structures in some women, while in others no alterations were seen. Avulsion of paravaginal support tissues from the pelvic sidewall was suspected in five patients, four after normal vaginal delivery and one after a Cesarean section performed late in the first stage of labor, with the fetal head palpable at the level of the ischial spines. Somewhat contrary to expectations, however, such qualitative changes did not correlate with an increase in anterior compartment mobility. While women with loss of tenting did show increased BND and urethral rotation, the wide confidence intervals rendered this difference non-significant.

There are two potential explanations for this finding. Either our methodology lacked accuracy, resulting in insufficient power, or increased bladder neck mobility postpartum may not in fact equate to avulsion or tearing of fascial structures. It is conceivable that such peripartum change may be due to stretching or attenuation of support structures rather than outright tearing as has been suggested in the past25.

We would like to hypothesize that hypermobility as observed in parous women may be due to either weakening or tearing or, as observed above, to pre-existing congenital laxity or increased compliance of structures which nevertheless remain intact. Any combination of these three mechanisms appears possible in one site or at different sites in one particular patient, explaining the infinite variation in the presentation of female pelvic organ prolapse, and possibly also the fact that increased bladder neck mobility alone, while a strong predictor22, does not equate to genuine stress incontinence. An intact suburethral hammock may keep the patient dry, whether it is rigid or elastic, in contrast to an avulsed or torn structure.

This issue is not just of academic interest. A whole surgical paradigm—the site-specific defect or defect-specific approach4, 6, 26—assumes the existence of discrete fascial defects. Such circumscribed fascial trauma may well be absent in a significant proportion of patients with genuine stress incontinence and/or pelvic organ prolapse.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References

Most, if not all, nulliparous women show evidence of intact paravaginal support structures (tenting) on translabial 3D ultrasound imaging. Tenting is visible in women displaying a wide range of urethral and bladder neck mobility, implying that hypermobility can be due to increased compliance (laxity) of fascial support structures rather than being exclusively caused by delivery trauma. Such trauma, however, seems evident in a minority of patients.

We found no obvious correlation between a peripartal increase in bladder neck mobility and loss of tenting, suggesting that even peripartal changes may at times be due to stretching rather than disruption of support structures. Further research in this field should focus on standardization of imaging methodology, an aim that will be helped by recent technical advances allowing real-time volume acquisition without transducer movement.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. References
  • 1
    DeLancey JO. Anatomy. In Textbook of Female Urology and Urogynaecology, CardozoL, StaskinD (eds). Isis Medical Media: London, UK, 2001; 112124.
  • 2
    Gainey HL. Post-partum observation of pelvic tissue damage. Am J Obstet Gynecol 1943; 46: 457466.
  • 3
    Richardson AC, Lyon JB, Williams NL. A new look at pelvic relaxation. Am J Obstet Gynecol 1976; 126: 568573.
  • 4
    Richardson AC, Edmonds PB, Williams NL. Treatment of stress urinary incontinence due to paravaginal fascial defect. Obstet Gynecol 1981; 57: 357362.
  • 5
    Miklos JR, Saye WB. A randomized comparison of Burch colposuspension and abdominal paravaginal defect repair [Letter and comment]. Am J Obstet Gynecol 1997; 176: 255256.
  • 6
    Miklos JR, Kohli N, Lucente V, Saye WB. Site-specific fascial defects in the diagnosis and surgical management of enterocele. Am J Obstet Gynecol 1998; 179: 14181422; Discussion 1822–1823.
  • 7
    Ostrzenski A, Osborne NG. Ultrasonography as a screening tool for paravaginal defects in women with stress incontinence: a pilot study. Int Urogynecol J Pelvic Floor Dysfunct 1998; 9: 195199.
  • 8
    Ostrzenski A, Osborne NG, Ostrzenska K. Method for diagnosing paravaginal defects using contrast ultrasonographic technique. J Ultrasound Med 1997; 16: 673677.
  • 9
    Nguyen JK, Hall CD, Taber E, Bhatia NN. Sonographic diagnosis of paravaginal defects: a standardization of technique. Int Urogynecol J Pelvic Floor Dysfunct 2000; 11: 341345.
  • 10
    Klutke C, Golomb J, Barbaric Z, Raz S. The anatomy of stress incontinence: magnetic resonance imaging of the female bladder neck and urethra. J Urol 1990; 143: 563566.
  • 11
    Hjartardottir S, Nilsson J, Petersen C, Lingman G. The female pelvic floor: a dome—not a basin. Acta Obstet Gynecol Scand 1997; 76: 567571.
  • 12
    Tunn R, DeLancey JO, Howard D, Thorp JM, Ashton-Miller JA, Quint LE. MR imaging of levator ani muscle recovery following vaginal delivery. Int Urogynecol J Pelvic Floor Dysfunct 1999; 10: 300307.
  • 13
    Peschers UM, Forster A, Leib B, Heiss D, Lainsinger G, Dimpfl T. Changes in levator ani muscle magnetic resonance imaging in women with postpartum stress urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct 1999; 10: S5.
  • 14
    DeLancey JO, Speights SE, Tunn R, Howard D, Ashton Miller JA. Localized levator ani muscle abnormalities seen in MR images: site, size and side of occurrence. Int Urogynecol J Pelvic Floor Dysfunct 1999; 10: S20S21.
  • 15
    Koelbl H, Hanzal E. Imaging of the lower urinary tract. Curr Opin Obstet Gynecol 1995; 7: 382385.
  • 16
    Schaer GN. Ultrasonography of the lower urinary tract. Curr Opin Obstet Gynecol 1997; 9: 313316.
  • 17
    Dietz HP, Wilson PD. Anatomical assessment of the bladder outlet and proximal urethra using ultrasound and videocystourethrography. Int Urogynecol J Pelvic Floor Dysfunct 1998; 9: 365369.
  • 18
    Peschers UM, Fanger G, Schaer GN, Vodusek DB, DeLancey JO, Schuessler B. Bladder neck mobility in continent nulliparous women. Br J Obstet Gynaecol 2001; 108: 320324.
  • 19
    Dietz HP, Haylen BT, Broome J. Ultrasound in the quantification of female pelvic organ prolapse. Ultrasound Obstet Gynecol 2001; 18: 511514.
  • 20
    Wisser J, Schaer G, Kurmanavicius J, Huch R, Huch A. Use of 3D ultrasound as a new approach to assess obstetrical trauma to the pelvic floor. Ultraschall Med 1999; 20: 1518.
  • 21
    Ochsenbein N, Kurmanavicius J, Huch R, Huch A, Wisser J. Volume sonography of the pelvic floor in nulliparous women and after elective cesarean section. Acta Obstet Gynecol Scand 2001; 80: 611615.
  • 22
    Dietz HP, Clarke B, Herbison P. Bladder neck mobility and urethral closure pressure as predictors of genuine stress incontinence. Int Urogynecol J Pelvic Floor Dysfunct 2002; 13: 289293.
  • 23
    Stuhldreier G, Kirschner HJ, Astfalk W, Schweizer P, Huppert PE, Grunert T. Three-dimensional endosonography of the pelvic floor: an additional diagnostic tool in surgery for continence problems in children. Eur J Pediatr Surg 1997; 7: 97102.
  • 24
    Perez N, Garcier JM, Pin Leveugle J, Lhoste Trouilloud A, Ravel A, McLaughlin P, Viallet JF, Boyer L. Dynamic magnetic resonance imaging of the female pelvis: radio-anatomy and pathologic applications. Preliminary results. Surg Radiol Anat 1999; 21: 133138.
  • 25
    DeLancey JO. Anatomy and biomechanics of genital prolapse. Clin Obstet Gynecol 1993; 36: 897909.
  • 26
    Smith ARB. Role of connective tissue and muscle in pelvic floor dysfunction. Curr Opin Obstet Gynecol 1994; 6: 317319.