- Email: [email protected]
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY USING 3-DIMENSIONAL COMPUTERIZED TOMOGRAPHY: INITIAL INSIGHTS
0022-5347/01/1664-1444/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 166, 1444 –1449, October 2001 Printed in U.S.A.
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY USING 3-DIMENSIONAL COMPUTERIZED TOMOGRAPHY: INITIAL INSIGHTS ANDREW A. STEC,* HARPREET K. PANNU, YOUSEF E. TADROS, PAUL D. SPONSELLER, ELLIOT K. FISHMAN AND JOHN P. GEARHART From the Division of Pediatric Urology, Departments of Urology, Radiology and Orthopaedics, The Johns Hopkins School of Medicine, Baltimore, Maryland
ABSTRACT
Purpose: We present the pelvic floor anatomy of the major pelvic floor musculature in classic bladder exstrophy, including the levator ani, obturator internus and obturator externus. By improving our knowledge of pelvic floor anatomy we hope to understand better the relationship of the pelvic floor to the bony anatomy as well as the role of osteotomy in changing pelvic floor anatomy to enhance urinary control after surgery. Materials and Methods: 3-Dimensional computerized tomography was done in 6 boys and 1 girl, including 5 patients 2 days to 5 months old (mean age 7 months) undergoing primary closure and 2 who were 4 and 8 years old undergoing repeat closure. The pelvic floor musculature, including the levator ani, obturator internus and obturator externus, in these cases was compared to that in 26 age and sex matched controls. Results: The levator ani musculature encompasses a significantly wider area of 9.5 cm.2 in patients with classic bladder exstrophy than in controls. The anterior segment of the levator ani was shorter (1.2 cm.) and the posterior segment of the levator ani was longer (2.5 cm.) than in controls. The degree of divergence of the levator ani in classic exstrophy was significantly more outwardly rotated (38.8 degrees) than controls. In addition, the transverse diameter of the levator hiatus was 2-fold that in our control group and in that of published controls, while the length of the hiatus was 1.3-fold that in normal controls. There was also significant flattening, involving a 31.7 degree decrease in steepness between the right and left halves of the levator ani, of the puborectal sling in classic bladder exstrophy versus controls. Because of these findings, there is more anterior superior rotation in the pelvic floor in exstrophy cases. The obturator internus was more outwardly rotated (15.1 degrees) in exstrophy and the obturator externus also showed more outward rotation (16.9 degrees) than in controls. Conclusions: This study provides better understanding of the pelvic floor anatomy in classic bladder exstrophy. Significant differences have been documented in the pelvic floor in classic bladder exstrophy cases and controls. Hopefully these differences may have a pivotal role in providing new insight into long-term issues, such as urinary and fecal incontinence, and pelvic organ prolapse, in classic bladder exstrophy. KEY WORDS: osteotomy, urinary incontinence, bladder exstrophy, abnormalities, bladder
In 1912 Symington proposed that the musculature of the pelvic floor should be studied by examining series of gross pathology sections.1 It was not until 1953 that the levator ani muscles were first imaged by Berglas and Rubin, who directly injected contrast medium into the musculature and then performed radiography.2 In 1990 Ruskin and Pearl discussed the need for and development of an anatomically accurate 3-dimensional (D) model of the pelvic floor, deeming it necessary for academic study and learning.3 They proposed that a 3-D model of the pelvic floor was the next logical step after 2-D scans and cadaver dissection for learning about the pelvic floor. In 1995 in the initial preliminary study of the bony pelvic anatomy in the exstrophy complex Sponseller et al reported 30% shorter length of the pubic bones in the exstrophy pelvis,4 generating interest in the true anatomical nature of this entity. The pelvic floor musculature is extremely important in the question of why some children with exstrophy achieve some degree of continence even before bladder neck repair, while others cannot. Surprisingly to our
knowledge there have been no previous studies of the pelvic floor musculature in children with classic bladder exstrophy and no published descriptions of this specific anatomy. Thus, while operative treatment of bladder exstrophy has markedly improved in the last 2 decades, investigation of the pelvic floor is a potential avenue for continued surgical improvement. In the past even when osteotomy was performed and the pelvic ring was closed, the levator muscle group did not receive attention since it did not originate from or insert into adjacent bony structures connected by a joint.5 A major issue arises, in that the levator ani has an important role in providing urinary and fecal continence, supporting the bladder and uterus, and preventing pelvic organ prolapse.6 These issues are broad enough to include other pelvic floor muscles, such as the obturator internus, which acts as an intermediary for the attachment of the pelvic floor muscles to the pelvis and as a framework for the pelvic diaphragm attachment to bone.7 In this initial study of children with classic bladder exstrophy we performed computerized tomography (CT) with 3-D modeling to investigate the pelvic floor musculature. While magnetic resonance imaging (MRI) is acknowledged to be the best means of evaluating soft tissue, CT with 3-D
Accepted for publication May 25, 2001. Supported by a grant from the American Foundation for Urologic Disease/American Urological Association Research Scholar Program. * American Foundation for Urologic Disease Research Scholar. 1444
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
reconstruction has been shown to be an adequate diagnostic tools for soft tissue abnormalities and it has been performed for years for anorectal imaging.8 –11 MATERIALS AND METHODS
We studied 6 boys and 1 girl with classic bladder exstrophy, including 5 who were 2 days to 15 months old (mean age 7 months) undergoing primary closure, and a 4 and an 8-year-old child undergoing repeat closure without previous osteotomy. The primary and repeat closure groups were compared with 16 and age and sex matched controls, respectively. In all cases spiral CT of the abdomen and pelvis was done before primary closure of the exstrophy bladder and osteotomy. Spiral CT involved 3 mm. collimation, 6 mm. table speed and 3 mm. reconstruction in the newborns, and 5 mm. collimation, 8 mm. table speed and 5 mm. reconstruction in the other pediatric patients. For comparison with the study group pelvic CT images of 26 age and sex matched children were obtained from the department of radiology. These patients were screened and the pelves were considered normal. The major reason for CT in the controls were abdominal trauma, Wilm’s tumor and neuroblastoma. The CT series of each patient was loaded onto an SGI O2 Workstation (Silicon Graphics, Mountain View, California) running 3-D Virtuoso CT/MR, R2.5.2, VA30, GMA (Siemens, Erlangen, Germany). This software allowed each series of images to be analyzed by multiplanar reformatting, providing anatomical views from the sagittal, coronal and axial planes. The software was used to render interactive 3-D models of the pelvic anatomy. It provided the advantage of establishing a horizontal axial plane as well as a vertical coronal plane. These established planes were configured to compensate for any rotation of the child on the CT table or any other delineating factors present at the time of CT, eliminating variability across all patient images. Thus, we obtained data using the established planes as a consistent reference for each patient. All measurements, angles, lengths, widths, relationships among muscles and overall muscle orientations were obtained directly from these anatomical models using the software package tools. Levator ani. The area encompassed by the puborectal sling was determined by reference to the medial cortices of the pubis and defined as the area between the right and left halves of the levator ani from their origin on the coccyx to their insertion points on the pubis. Insertion points are the sites of muscle attachment and we use the words origin and insertion as in anatomical texts. The anterior length of the levator ani was measured at the approximate level of the medial cortices of the pubis as the length from the anterior edge of the rectum to the levator insertion on the pubis. The posterior length of the levator ani was determined at the approximate level of the medial cortices of the pubis as the length along the levator from the anterior edge of the rectum to its origin on the coccyx. The width of the levator ani was considered its the maximum width in the coronal plane. The angle created between the right half of the levator ani and the midline of the body is an indicator of the outward rotation or divergence of the halves of the levator ani. This angle was also measured with reference to the medial cortices of the pubis, which standardized the level at which measurements were made in all patients. In the coronal plane the angle between the right and left levator ani halves was used to measure the steepness of the sides of the puborectal sling. The width and length of the levator hiatus was measured at the level of the pelvic floor. In the sagittal plane the pelvic floor angle was defined as the angle from vertical as the muscle extended from the coccyx. The angle between the obturator internus and levator ani was measured
1445
at the level of the triradiate cartilage, which is the medial growth center of the acetabulum, and the site at which the pubis, ischium and ilium unite and usually fuse at puberty. The angle is where the levator ani and obturator internus cross, allowing us to examine the spatial relationship of the location of these 2 muscles (fig. 1, E). Obturator internus. The length of the obturator internus was determined just below the level of the triradiate cartilage as the length along the muscle from its origin on the pubis to its insertion on the greater trochanter. The width of the obturator internus was considered its maximum width in the coronal plane. The axial divergence of the obturator internus was defined as the angle between a line drawn parallel to the body midline and one down the length of the obturator internus, allowing us to measure the outward rotation of the muscle (fig. 1, C and D). Obturator externus. At the approximate level of the medial cortices of the pubis, the length of the obturator externus was measured down the muscle from its origin on the pubis to its insertion in the trochanteric fossa. The width of the obturator externus was considered its maximum width in the axial plane. The axial divergence of the obturator externus was considered the outward rotation of the muscle, as measured by the angle between a line drawn parallel to the midline of the body and one from the insertion to origin point along the length of the muscle (fig. 2, D and E). RESULTS
We calculated p values by the 2-tailed t test assuming unequal variance with p ⬍0.05 considered significant (see table). Levator ani. We determined the area encompassed by the puborectal sling. The levator ani encompassed a mean area of 9.5 cm.2 in patients with classic bladder exstrophy and 4.9 cm.2 in controls. Thus, the levator ani in children with exstrophy supports an almost 2-fold greater body cavity area than the levator in normal children (fig. 1). No difference was noted in the overall length of the levator ani in exstrophy
FIG. 1. CT. A, body cavity area supported by levator ani (a) in normal 1-year-old girl. B, body cavity area supported by levator ani in 8-month-old girl with classic bladder exstrophy. C, axial view of lateral divergence of obturator internus in pelvic floor of child with classic bladder exstrophy. D, axial view of obturator externus (b) in pelvic floor of child with classic bladder exstrophy. E, our method of measuring angle between levator ani and obturator internus where 2 muscles overlap.
1446
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
FIG. 2. CT. A, angle (a) of axial divergence of levator ani muscle in control. B, angle of axial divergence of levator ani muscle in patient with exstrophy. C, our method of measuring levator ani length. Line at anterior edge of rectum separates anterior (a) and posterior (b) portion of levator ani in 10-month-old boy with classic bladder exstrophy. Note significant posterior location of levator ani. D, axial view of lateral divergence of obturator internus (a) in classic bladder exstrophy. E, axial view of lateral divergence of obturator externus (b) in classic bladder exstrophy.
Av. Exstrophy Av. Controls Levator ani: Area (cm.2) Total length (cm.) Posterior segment length (cm.) Width (cm.) Axial divergence (degrees) Angle between halves (degrees) Pelvic floor angle (degrees) Angle between obturator internus and levator ani (degrees) Obturator internus: Length (cm.) Width (cm.) Axial divergence (degrees) Obturator externus: Length (cm.) Width (cm.) Axial divergence (degrees)
9.5 3.76, 1.20 2.56 0.44 38.8 75.4 111.8 53.7
p Value
4.9 3.61, 1.74 1.87 0.44 23.3 43.7 127.3 51.3
0.015 0.721, 0.047 0.049 0.997 0.008 0.008 0.027 0.501
2.82 0.72 8.1
2.76 0.83 23.2
0.862 0.207 0.001
3.37 1.07 28.4
3.62 1.17 45.3
0.572 0.575 0.001
cases and controls (3.76 and 3.61 cm., respectively, p ⬎0.7). The anterior segment of the levator ani was 1.20 cm. in exstrophy cases and 1.74 cm. in normal controls. This significant difference in the length of the anterior segment of the levator ani indicates that in normal children 48% of the levator ani is located anterior of the rectum, whereas in children with exstrophy only 32% of the muscle is anterior (p ⬍0.05, fig. 3). In exstrophy the mean length of the posterior segment of the levator ani was 2.56 cm., significantly greater than 1.87 cm. in controls (p ⬍0.05). In children with exstrophy 68% of the puborectal sling is centered posterior to the urethra, and away from the underside of the bladder and urethra, whereas in normal children the levator is more centered with only 52% of the muscle located posterior (fig. 3). There was no significant difference in the maximum width of the levator ani muscle in patients with classic bladder exstrophy (0.44 cm.) and controls (0.44 cm.). We also determined the axial divergence of the levator ani. In children with exstrophy each half of the levator ani was outwardly rotated a mean of 38.8 degrees from the midline (fig. 2, B). In controls there was only 23.3 degrees of outward rotation. Comparing the 2 groups
showed that each half of the levator ani in exstrophy cases was an average of 15.5 degrees more outwardly rotated than normal, which was significant (p ⫽ 0.008). Coronal images of the levator ani enabled us to measure the angle between the right and left halves of the puborectal sling. In exstrophy and control cases the average angle was 75.4 and only 43.7 degrees, respectively. The 31.7 degree decrease in steepness between the right and left halves of the levator ani indicates significant flattening of the puborectal sling in exstrophy (p ⫽ 0.008, fig. 4). The mean transverse width of the hiatus (45 mm.) was 2-fold that (20 mm.) in age matched controls 45 versus 20 mm. (p ⫽ ⬍0.05). The mean length of the hiatus was 1.3-fold that in controls (33.2 versus 25 mm., p ⫽ ⬍0.05). The average puborectal sling in exstrophy made an angle of 111.8 degrees from vertical at the point where it came off of the coccyx, which was significantly shallower than the 127.3 degrees in controls (p ⬍0.03). The 15.5 degree more anterosuperior rotation of the pelvic floor in exstrophy causes the puborectal sling to have a more flattened and less defined conic shape than normal. The angle between the obturator internus and levator ani as they overlap in the pelvic floor was not statistically different in the exstrophy and control groups (53.7 and 51.3 degrees, respectively, p ⫽ 0.05). Obturator internus. The mean length of the obturator internus in exstrophy was 2.82 cm., not significantly different from the 2.76 cm. in controls (p ⫽ 0.862). The maximum width of the obturator internus was similar in exstrophy and control groups (0.72 and 0.83 cm., respectively, p ⫽ 0.21). We measured the axial divergence of the obturator internus, which in children with exstrophy was rotated an average of 8.1 degrees toward the midline. In controls the average angle of this rotation was 23.2 degrees, indicating that the obturator internus is 15.1 degrees more outwardly rotated in exstrophy than in normal cases (p ⬍0.001). The obturator internus in some classic bladder exstrophy cases pointed in an even more outward direction on the pelvis, whereas it would normally be directed inward (fig. 1, C and D). Obturator externus. In patients with exstrophy the mean length of the obturator externus was 3.37 cm. compared with 3.62 cm. in controls. This slight difference was not significant (p ⫽ 0.57). There was no statistically appreciable difference in the maximum width of the obturator internus in the exstrophy and control groups (p ⫽ 0.58). In patients and controls mean muscle width was 1.07 and 1.17 cm., respectively. We determined the axial divergence of the obturator externus. In exstrophy there was an average of 16.9 degrees more outward rotation of the obturator externus from the midline than in controls (p ⬍0.001). The obturator internus in the exstrophy group made an angle of only 28.4 degrees toward the midline, whereas in controls the average inward angle was 45.3 degrees. DISCUSSION
Formerly little anatomical information on the true nature of the pelvic floor in bladder exstrophy has been available. There has been a great deal of interest in better defining the pelvic bony anatomy in bladder and cloacal exstrophy in the last several years. In 1995 Sponseller et al first described in detail the bony anatomy of the pelvis using 2-D CT, which shows the passive support mechanisms of the pelvic floor, including the sacrum, coccyx, pubic ramus and ischium.4 Later Yazici et al confirmed and expanded these observations using 3-D CT.12 However, to our knowledge little is known to date of the normal active support of the pelvic floor in classic bladder exstrophy. This anatomical investigation has become especially important in the exstrophy population as the role of osteotomy is now more widespread. Also, in some small select groups of patients, mainly females, some degree of continence has been reported after bladder, posterior urethral,
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
1447
FIG. 3. A, displacement of levator (Lev.) ani to more posterior (Post.) position in patient with exstrophy, that is 68% posterior to anus versus normal controls. Also note shortened anterior (Ant.) segment of levator ani in exstrophy 32% anterior to anus versus 48% in controls. Obt. int., obturator internus. B, greater outward rotation of 15.1 degrees of obturator internus in exstrophy group versus controls. Also note that area encompassed by puborectalis is 2-fold that of controls and more flattened.
pelvic and abdominal wall closure alone without the need for a continence procedure of the bladder neck. Why do some patients achieve continence after closure alone? Is the pelvic anatomy in these cases different than in other exstrophy cases and how does it compare to normal age matched controls? The anatomy of the pelvic floor is complex. The pelvic floor consists of 3 supporting layers, namely the endopelvic fascia, pelvic diaphragm and urogenital diaphragm. The urogenital diaphragm, prostate, bulbocavernous muscle and corporeal bodies have been well described in classic bladder exstrophy on MRI scanning by Gearhart13 and Silver14 et al. However, almost nothing is known about the anatomy of the pelvic diaphragm in children with classic bladder exstrophy. In the past CT provided direct visualization of the pelvic musculature but it was limited to the transaxial plane. However, using the CT machine and software in our series each image was analyzed by multiplanar reformatting, providing anatomical views from the sagittal, coronal and axial planes. The pelvic diaphragm separates the pelvis above from the
perineum below. The obturator internus muscle serves as a frame for the attachment of the pelvic diaphragm to the pelvic bones. The pelvic diaphragm comprises striated muscles that close the pelvic outlet with their upper and lower fascial layer.6, 15 The major component of this layer is the levator ani muscles, which are composed of the lateral thin sheet of muscle called the iliococcygeus and a bulkier medial muscular sling called the pubococcygeus. The iliococcygeus attaches anteriorly to the pubic bone, arcus tendoneus of the levator ani, posterior attachment of the levator plate or anococcygeal raphe and coccyx. The more substantial part of the levator ani or pubococcygeus inserts bilaterally at the pubic rami and wraps around the midline structures of the bladder, urethra, vagina and rectum. The pubococcygeus muscle is funnel-shaped with a transverse portion called the levator plate and a vertical portion called the suspensory sling.15 At the level of the levator hiatus the plate bends downward sharply to form the suspensory sling. This most medial portion of the pubococcygeus, referred to
1448
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
FIG. 4. CT. A, coronal view shows angle (a) of right and left halves of levator ani in normal 1-year-old girl. B, coronal view demonstrates significant flattening in child with exstrophy of angle of right and left halves of levator ani. C, sagittal view reveals pelvic floor angle in child with normal pelvic floor. D, sagittal view shows more flattened pelvic floor angle in child with classic bladder exstrophy.
by others as the puborectalis sling, attaches to the midline viscera during passage through the pelvic floor. This muscle provides direct and indirect support for the vagina, bladder and urethra by drawing these structures ventrally toward the pubic bone.15, 16 In extensive new human anatomy dissections with microscopic confirmation Shafik et al reported that the puborectalis sling extension of the levator ani connects to the urethra at the bladder neck.16 Furthermore, since their dissection showed the urethra in an infralevator position, it is protected from the effects of intra-abdominal pressure. These new findings differ from those in previous investigations. The levator hiatus lies obliquely with a downward and forward slope toward the symphysis pubis. In infancy the anteroposterior diameter of the hiatus in infancy is 1.5 to 2 cm. and the side-to-side diameter is 1 to 1.5 cm.16 We observed no surprises in regard to the anatomical relationship of the obturator internus and externus in the exstrophy group. Length and width were the same as in normal controls and the degree of axial divergence or external rotation was the value expected based on previous studies. In fact, the degree of external rotation was almost the same as the amount of external rotation of the anterior pelvic ring as described by Sponseller et al.4 Our study had several important findings concerning the levator ani muscle group, which for the most part forms most of the musculature of the pelvic diaphragm. A major finding was that the levator in exstrophy cases is not only wider than in normal controls, as measured by the area encompassed by the puborectal sling, but also clearly flatter with an angle between the levator ani halves of 31 degrees less than in controls. This finding was further emphasized by the pelvic floor angle of the puborectal sling, which was 15 degrees more anterosuperior in rotation, causing the sling to have a more flattened and less conical shape than in normal controls. Another finding in this study was that in the exstrophy population 68% of the puborectalis sling is posterior to the rectum, whereas in normal children about half of the muscle is anterior and half is posterior to the rectum. Thus, in exstrophy the muscle is centered posterior away from the posterior urethra and bladder neck, where the effect of contraction of the puborectalis, which aids in maintaining a closed bladder neck, would be lessened. It has been observed to be especially true by our pediatric surgery colleagues, who
have noted that proper placement of the rectum within the puborectalis sling is essential for achieving rectal continence.10, 11 Since the width of the levator hiatus is 2-fold and the length is 1.3-fold, normal and the puborectalis sling is misplaced, it is easy to understand how pelvic organ prolapse can occur. As expected, the external rotation of the levator ani is similar to that of the obturator internus and externus, and the angle of the overlap of the levator ani and obturator internus was not different. Thus, 3-D CT of the pelvis in classic exstrophy revealed many interesting new facts. External rotation of the obturator group in exstrophy follows the diastasis and should be corrected by osteotomy. However, to our knowledge the result of closure on the levator group and size of the hiatus is currently unclear. One may hypothesize that in patients with some degree of continence after simple bladder and posterior urethral closure the levator group is repositioned and any continence achieved is a function of the pelvic floor alteration that occurs even without the aid of a normal urinary sphincter. However, one of us (J. P. G.) has noted that some degree of continence may be achieved in newborns without osteotomy, so that appropriate movement of the pelvic floor with or without osteotomy may aid in eventual continence. In addition, possibly some patients who attain some degree of continence after closure alone may have a lesser abnormality of the levator group with more of the muscle located anterior to the rectum instead of posterior, where it may act on the bladder neck and posterior urethra to increase outlet resistance. Further studies are currently under way to image the pelvic floor after primary closure for evaluating the effect of closure on the pelvic diaphragm and levator hiatus. CONCLUSIONS
To our knowledge this study is the first precise description of the pelvic floor muscular anatomy of classic bladder exstrophy in children before corrective surgery. Because of the multifactorial nature of urinary incontinence, imaging can be valuable not only preoperatively, but also postoperatively. Hopefully these studies may enable us to understand better the role of exstrophy closure and osteotomy on the configuration of the pelvic floor as well as their effect on eventual continence. REFERENCES
1. Symington, J.: Further observations on the rectum and anal canal. J Anat, 46: 289, 1911 2. Berglas, B. and Rubin, I. C.: Study of the support structures of the uterus by levator myography. Surg Gynecol Obstet, 97: 677, 1953 3. Ruskin, M. J. and Pearl, R. K.: A three-dimensional model of the pelvic floor and anorectal anatomy. J Biocommun, 17: 22, 1990 4. Sponseller, P. D., Bisson, L. J., Gearhart, J. P. et al: The anatomy of the pelvis in the exstrophy complex. J Bone Joint Surg Am, 77: 177, 1995 5. Wall, L. L.: The muscles of the pelvic floor. Clin Obstet Gynecol, 36: 910, 1993 6. Strohbehn, K.: Normal pelvic floor anatomy. Obstet Gynecol Clin North Am, 25: 683, 1998 7. Klutke, C. G. and Siegel, C. L.: Functional female pelvic anatomy. Urol Clin North Am, 22: 487, 1995 8. Dietrich, R. B. and Kangarloo, H.: Pelvic abnormalities in children: assessment with MR imaging. Radiology, 163: 367, 1987 9. Mezzacappa, P. M., Price, A. P., Haller, J. O. et al: MR and CT demonstration of levator sling in congenital anorectal abnormalities. J Comp Assist Tomogr, 11: 273, 1987 10. Vade, A., Reyes, H., Wilbur, A. et al: The anorectal sphincter after rectal pull-through surgery for anorectal anomalies: MRI evaluation. Pediatr Radiol, 19: 179, 1989 11. Sato, Y., Pringle, K. C., Bergman, R. A. et al: Congenital anorectal anomalies: MR imaging. Radiology, 168: 157, 1988 12. Yazici, M., Sozubir, S., Kilicoglu, G. et al: Three-dimensional anatomy of the pelvis in bladder exstrophy: description of bone
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY pathology using 3 dimensional CT and its clinical relevance. J Pediatr Orthop, 18: 132, 1998 13. Gearhart, J. P., Yang, A., Leonard, M. P. et al: Prostate size and configuration in adults with bladder exstrophy. J Urol, 149: 308, 1993 14. Silver, R. I., Yang, A., Ben-Chaim, J. et al: Penile length in adulthood after exstrophy reconstruction. J Urol, 157: 999, 1997
1449
15. Shafik, A.: Levator ani muscle: new physioanatomical aspects and role in the micturation mechanism. World J Urol, 17: 266, 1999 16. Shafik, A.: A new concept of the anatomy of the anal sphincter mechanism and the physiology of deffication. VIII. Levator hiatus and tunnel: anatomy and function. Dis Colon Rectum, 22: 539, 1979
Vol. 166, 1444 –1449, October 2001 Printed in U.S.A.
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY USING 3-DIMENSIONAL COMPUTERIZED TOMOGRAPHY: INITIAL INSIGHTS ANDREW A. STEC,* HARPREET K. PANNU, YOUSEF E. TADROS, PAUL D. SPONSELLER, ELLIOT K. FISHMAN AND JOHN P. GEARHART From the Division of Pediatric Urology, Departments of Urology, Radiology and Orthopaedics, The Johns Hopkins School of Medicine, Baltimore, Maryland
ABSTRACT
Purpose: We present the pelvic floor anatomy of the major pelvic floor musculature in classic bladder exstrophy, including the levator ani, obturator internus and obturator externus. By improving our knowledge of pelvic floor anatomy we hope to understand better the relationship of the pelvic floor to the bony anatomy as well as the role of osteotomy in changing pelvic floor anatomy to enhance urinary control after surgery. Materials and Methods: 3-Dimensional computerized tomography was done in 6 boys and 1 girl, including 5 patients 2 days to 5 months old (mean age 7 months) undergoing primary closure and 2 who were 4 and 8 years old undergoing repeat closure. The pelvic floor musculature, including the levator ani, obturator internus and obturator externus, in these cases was compared to that in 26 age and sex matched controls. Results: The levator ani musculature encompasses a significantly wider area of 9.5 cm.2 in patients with classic bladder exstrophy than in controls. The anterior segment of the levator ani was shorter (1.2 cm.) and the posterior segment of the levator ani was longer (2.5 cm.) than in controls. The degree of divergence of the levator ani in classic exstrophy was significantly more outwardly rotated (38.8 degrees) than controls. In addition, the transverse diameter of the levator hiatus was 2-fold that in our control group and in that of published controls, while the length of the hiatus was 1.3-fold that in normal controls. There was also significant flattening, involving a 31.7 degree decrease in steepness between the right and left halves of the levator ani, of the puborectal sling in classic bladder exstrophy versus controls. Because of these findings, there is more anterior superior rotation in the pelvic floor in exstrophy cases. The obturator internus was more outwardly rotated (15.1 degrees) in exstrophy and the obturator externus also showed more outward rotation (16.9 degrees) than in controls. Conclusions: This study provides better understanding of the pelvic floor anatomy in classic bladder exstrophy. Significant differences have been documented in the pelvic floor in classic bladder exstrophy cases and controls. Hopefully these differences may have a pivotal role in providing new insight into long-term issues, such as urinary and fecal incontinence, and pelvic organ prolapse, in classic bladder exstrophy. KEY WORDS: osteotomy, urinary incontinence, bladder exstrophy, abnormalities, bladder
In 1912 Symington proposed that the musculature of the pelvic floor should be studied by examining series of gross pathology sections.1 It was not until 1953 that the levator ani muscles were first imaged by Berglas and Rubin, who directly injected contrast medium into the musculature and then performed radiography.2 In 1990 Ruskin and Pearl discussed the need for and development of an anatomically accurate 3-dimensional (D) model of the pelvic floor, deeming it necessary for academic study and learning.3 They proposed that a 3-D model of the pelvic floor was the next logical step after 2-D scans and cadaver dissection for learning about the pelvic floor. In 1995 in the initial preliminary study of the bony pelvic anatomy in the exstrophy complex Sponseller et al reported 30% shorter length of the pubic bones in the exstrophy pelvis,4 generating interest in the true anatomical nature of this entity. The pelvic floor musculature is extremely important in the question of why some children with exstrophy achieve some degree of continence even before bladder neck repair, while others cannot. Surprisingly to our
knowledge there have been no previous studies of the pelvic floor musculature in children with classic bladder exstrophy and no published descriptions of this specific anatomy. Thus, while operative treatment of bladder exstrophy has markedly improved in the last 2 decades, investigation of the pelvic floor is a potential avenue for continued surgical improvement. In the past even when osteotomy was performed and the pelvic ring was closed, the levator muscle group did not receive attention since it did not originate from or insert into adjacent bony structures connected by a joint.5 A major issue arises, in that the levator ani has an important role in providing urinary and fecal continence, supporting the bladder and uterus, and preventing pelvic organ prolapse.6 These issues are broad enough to include other pelvic floor muscles, such as the obturator internus, which acts as an intermediary for the attachment of the pelvic floor muscles to the pelvis and as a framework for the pelvic diaphragm attachment to bone.7 In this initial study of children with classic bladder exstrophy we performed computerized tomography (CT) with 3-D modeling to investigate the pelvic floor musculature. While magnetic resonance imaging (MRI) is acknowledged to be the best means of evaluating soft tissue, CT with 3-D
Accepted for publication May 25, 2001. Supported by a grant from the American Foundation for Urologic Disease/American Urological Association Research Scholar Program. * American Foundation for Urologic Disease Research Scholar. 1444
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
reconstruction has been shown to be an adequate diagnostic tools for soft tissue abnormalities and it has been performed for years for anorectal imaging.8 –11 MATERIALS AND METHODS
We studied 6 boys and 1 girl with classic bladder exstrophy, including 5 who were 2 days to 15 months old (mean age 7 months) undergoing primary closure, and a 4 and an 8-year-old child undergoing repeat closure without previous osteotomy. The primary and repeat closure groups were compared with 16 and age and sex matched controls, respectively. In all cases spiral CT of the abdomen and pelvis was done before primary closure of the exstrophy bladder and osteotomy. Spiral CT involved 3 mm. collimation, 6 mm. table speed and 3 mm. reconstruction in the newborns, and 5 mm. collimation, 8 mm. table speed and 5 mm. reconstruction in the other pediatric patients. For comparison with the study group pelvic CT images of 26 age and sex matched children were obtained from the department of radiology. These patients were screened and the pelves were considered normal. The major reason for CT in the controls were abdominal trauma, Wilm’s tumor and neuroblastoma. The CT series of each patient was loaded onto an SGI O2 Workstation (Silicon Graphics, Mountain View, California) running 3-D Virtuoso CT/MR, R2.5.2, VA30, GMA (Siemens, Erlangen, Germany). This software allowed each series of images to be analyzed by multiplanar reformatting, providing anatomical views from the sagittal, coronal and axial planes. The software was used to render interactive 3-D models of the pelvic anatomy. It provided the advantage of establishing a horizontal axial plane as well as a vertical coronal plane. These established planes were configured to compensate for any rotation of the child on the CT table or any other delineating factors present at the time of CT, eliminating variability across all patient images. Thus, we obtained data using the established planes as a consistent reference for each patient. All measurements, angles, lengths, widths, relationships among muscles and overall muscle orientations were obtained directly from these anatomical models using the software package tools. Levator ani. The area encompassed by the puborectal sling was determined by reference to the medial cortices of the pubis and defined as the area between the right and left halves of the levator ani from their origin on the coccyx to their insertion points on the pubis. Insertion points are the sites of muscle attachment and we use the words origin and insertion as in anatomical texts. The anterior length of the levator ani was measured at the approximate level of the medial cortices of the pubis as the length from the anterior edge of the rectum to the levator insertion on the pubis. The posterior length of the levator ani was determined at the approximate level of the medial cortices of the pubis as the length along the levator from the anterior edge of the rectum to its origin on the coccyx. The width of the levator ani was considered its the maximum width in the coronal plane. The angle created between the right half of the levator ani and the midline of the body is an indicator of the outward rotation or divergence of the halves of the levator ani. This angle was also measured with reference to the medial cortices of the pubis, which standardized the level at which measurements were made in all patients. In the coronal plane the angle between the right and left levator ani halves was used to measure the steepness of the sides of the puborectal sling. The width and length of the levator hiatus was measured at the level of the pelvic floor. In the sagittal plane the pelvic floor angle was defined as the angle from vertical as the muscle extended from the coccyx. The angle between the obturator internus and levator ani was measured
1445
at the level of the triradiate cartilage, which is the medial growth center of the acetabulum, and the site at which the pubis, ischium and ilium unite and usually fuse at puberty. The angle is where the levator ani and obturator internus cross, allowing us to examine the spatial relationship of the location of these 2 muscles (fig. 1, E). Obturator internus. The length of the obturator internus was determined just below the level of the triradiate cartilage as the length along the muscle from its origin on the pubis to its insertion on the greater trochanter. The width of the obturator internus was considered its maximum width in the coronal plane. The axial divergence of the obturator internus was defined as the angle between a line drawn parallel to the body midline and one down the length of the obturator internus, allowing us to measure the outward rotation of the muscle (fig. 1, C and D). Obturator externus. At the approximate level of the medial cortices of the pubis, the length of the obturator externus was measured down the muscle from its origin on the pubis to its insertion in the trochanteric fossa. The width of the obturator externus was considered its maximum width in the axial plane. The axial divergence of the obturator externus was considered the outward rotation of the muscle, as measured by the angle between a line drawn parallel to the midline of the body and one from the insertion to origin point along the length of the muscle (fig. 2, D and E). RESULTS
We calculated p values by the 2-tailed t test assuming unequal variance with p ⬍0.05 considered significant (see table). Levator ani. We determined the area encompassed by the puborectal sling. The levator ani encompassed a mean area of 9.5 cm.2 in patients with classic bladder exstrophy and 4.9 cm.2 in controls. Thus, the levator ani in children with exstrophy supports an almost 2-fold greater body cavity area than the levator in normal children (fig. 1). No difference was noted in the overall length of the levator ani in exstrophy
FIG. 1. CT. A, body cavity area supported by levator ani (a) in normal 1-year-old girl. B, body cavity area supported by levator ani in 8-month-old girl with classic bladder exstrophy. C, axial view of lateral divergence of obturator internus in pelvic floor of child with classic bladder exstrophy. D, axial view of obturator externus (b) in pelvic floor of child with classic bladder exstrophy. E, our method of measuring angle between levator ani and obturator internus where 2 muscles overlap.
1446
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
FIG. 2. CT. A, angle (a) of axial divergence of levator ani muscle in control. B, angle of axial divergence of levator ani muscle in patient with exstrophy. C, our method of measuring levator ani length. Line at anterior edge of rectum separates anterior (a) and posterior (b) portion of levator ani in 10-month-old boy with classic bladder exstrophy. Note significant posterior location of levator ani. D, axial view of lateral divergence of obturator internus (a) in classic bladder exstrophy. E, axial view of lateral divergence of obturator externus (b) in classic bladder exstrophy.
Av. Exstrophy Av. Controls Levator ani: Area (cm.2) Total length (cm.) Posterior segment length (cm.) Width (cm.) Axial divergence (degrees) Angle between halves (degrees) Pelvic floor angle (degrees) Angle between obturator internus and levator ani (degrees) Obturator internus: Length (cm.) Width (cm.) Axial divergence (degrees) Obturator externus: Length (cm.) Width (cm.) Axial divergence (degrees)
9.5 3.76, 1.20 2.56 0.44 38.8 75.4 111.8 53.7
p Value
4.9 3.61, 1.74 1.87 0.44 23.3 43.7 127.3 51.3
0.015 0.721, 0.047 0.049 0.997 0.008 0.008 0.027 0.501
2.82 0.72 8.1
2.76 0.83 23.2
0.862 0.207 0.001
3.37 1.07 28.4
3.62 1.17 45.3
0.572 0.575 0.001
cases and controls (3.76 and 3.61 cm., respectively, p ⬎0.7). The anterior segment of the levator ani was 1.20 cm. in exstrophy cases and 1.74 cm. in normal controls. This significant difference in the length of the anterior segment of the levator ani indicates that in normal children 48% of the levator ani is located anterior of the rectum, whereas in children with exstrophy only 32% of the muscle is anterior (p ⬍0.05, fig. 3). In exstrophy the mean length of the posterior segment of the levator ani was 2.56 cm., significantly greater than 1.87 cm. in controls (p ⬍0.05). In children with exstrophy 68% of the puborectal sling is centered posterior to the urethra, and away from the underside of the bladder and urethra, whereas in normal children the levator is more centered with only 52% of the muscle located posterior (fig. 3). There was no significant difference in the maximum width of the levator ani muscle in patients with classic bladder exstrophy (0.44 cm.) and controls (0.44 cm.). We also determined the axial divergence of the levator ani. In children with exstrophy each half of the levator ani was outwardly rotated a mean of 38.8 degrees from the midline (fig. 2, B). In controls there was only 23.3 degrees of outward rotation. Comparing the 2 groups
showed that each half of the levator ani in exstrophy cases was an average of 15.5 degrees more outwardly rotated than normal, which was significant (p ⫽ 0.008). Coronal images of the levator ani enabled us to measure the angle between the right and left halves of the puborectal sling. In exstrophy and control cases the average angle was 75.4 and only 43.7 degrees, respectively. The 31.7 degree decrease in steepness between the right and left halves of the levator ani indicates significant flattening of the puborectal sling in exstrophy (p ⫽ 0.008, fig. 4). The mean transverse width of the hiatus (45 mm.) was 2-fold that (20 mm.) in age matched controls 45 versus 20 mm. (p ⫽ ⬍0.05). The mean length of the hiatus was 1.3-fold that in controls (33.2 versus 25 mm., p ⫽ ⬍0.05). The average puborectal sling in exstrophy made an angle of 111.8 degrees from vertical at the point where it came off of the coccyx, which was significantly shallower than the 127.3 degrees in controls (p ⬍0.03). The 15.5 degree more anterosuperior rotation of the pelvic floor in exstrophy causes the puborectal sling to have a more flattened and less defined conic shape than normal. The angle between the obturator internus and levator ani as they overlap in the pelvic floor was not statistically different in the exstrophy and control groups (53.7 and 51.3 degrees, respectively, p ⫽ 0.05). Obturator internus. The mean length of the obturator internus in exstrophy was 2.82 cm., not significantly different from the 2.76 cm. in controls (p ⫽ 0.862). The maximum width of the obturator internus was similar in exstrophy and control groups (0.72 and 0.83 cm., respectively, p ⫽ 0.21). We measured the axial divergence of the obturator internus, which in children with exstrophy was rotated an average of 8.1 degrees toward the midline. In controls the average angle of this rotation was 23.2 degrees, indicating that the obturator internus is 15.1 degrees more outwardly rotated in exstrophy than in normal cases (p ⬍0.001). The obturator internus in some classic bladder exstrophy cases pointed in an even more outward direction on the pelvis, whereas it would normally be directed inward (fig. 1, C and D). Obturator externus. In patients with exstrophy the mean length of the obturator externus was 3.37 cm. compared with 3.62 cm. in controls. This slight difference was not significant (p ⫽ 0.57). There was no statistically appreciable difference in the maximum width of the obturator internus in the exstrophy and control groups (p ⫽ 0.58). In patients and controls mean muscle width was 1.07 and 1.17 cm., respectively. We determined the axial divergence of the obturator externus. In exstrophy there was an average of 16.9 degrees more outward rotation of the obturator externus from the midline than in controls (p ⬍0.001). The obturator internus in the exstrophy group made an angle of only 28.4 degrees toward the midline, whereas in controls the average inward angle was 45.3 degrees. DISCUSSION
Formerly little anatomical information on the true nature of the pelvic floor in bladder exstrophy has been available. There has been a great deal of interest in better defining the pelvic bony anatomy in bladder and cloacal exstrophy in the last several years. In 1995 Sponseller et al first described in detail the bony anatomy of the pelvis using 2-D CT, which shows the passive support mechanisms of the pelvic floor, including the sacrum, coccyx, pubic ramus and ischium.4 Later Yazici et al confirmed and expanded these observations using 3-D CT.12 However, to our knowledge little is known to date of the normal active support of the pelvic floor in classic bladder exstrophy. This anatomical investigation has become especially important in the exstrophy population as the role of osteotomy is now more widespread. Also, in some small select groups of patients, mainly females, some degree of continence has been reported after bladder, posterior urethral,
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
1447
FIG. 3. A, displacement of levator (Lev.) ani to more posterior (Post.) position in patient with exstrophy, that is 68% posterior to anus versus normal controls. Also note shortened anterior (Ant.) segment of levator ani in exstrophy 32% anterior to anus versus 48% in controls. Obt. int., obturator internus. B, greater outward rotation of 15.1 degrees of obturator internus in exstrophy group versus controls. Also note that area encompassed by puborectalis is 2-fold that of controls and more flattened.
pelvic and abdominal wall closure alone without the need for a continence procedure of the bladder neck. Why do some patients achieve continence after closure alone? Is the pelvic anatomy in these cases different than in other exstrophy cases and how does it compare to normal age matched controls? The anatomy of the pelvic floor is complex. The pelvic floor consists of 3 supporting layers, namely the endopelvic fascia, pelvic diaphragm and urogenital diaphragm. The urogenital diaphragm, prostate, bulbocavernous muscle and corporeal bodies have been well described in classic bladder exstrophy on MRI scanning by Gearhart13 and Silver14 et al. However, almost nothing is known about the anatomy of the pelvic diaphragm in children with classic bladder exstrophy. In the past CT provided direct visualization of the pelvic musculature but it was limited to the transaxial plane. However, using the CT machine and software in our series each image was analyzed by multiplanar reformatting, providing anatomical views from the sagittal, coronal and axial planes. The pelvic diaphragm separates the pelvis above from the
perineum below. The obturator internus muscle serves as a frame for the attachment of the pelvic diaphragm to the pelvic bones. The pelvic diaphragm comprises striated muscles that close the pelvic outlet with their upper and lower fascial layer.6, 15 The major component of this layer is the levator ani muscles, which are composed of the lateral thin sheet of muscle called the iliococcygeus and a bulkier medial muscular sling called the pubococcygeus. The iliococcygeus attaches anteriorly to the pubic bone, arcus tendoneus of the levator ani, posterior attachment of the levator plate or anococcygeal raphe and coccyx. The more substantial part of the levator ani or pubococcygeus inserts bilaterally at the pubic rami and wraps around the midline structures of the bladder, urethra, vagina and rectum. The pubococcygeus muscle is funnel-shaped with a transverse portion called the levator plate and a vertical portion called the suspensory sling.15 At the level of the levator hiatus the plate bends downward sharply to form the suspensory sling. This most medial portion of the pubococcygeus, referred to
1448
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY
FIG. 4. CT. A, coronal view shows angle (a) of right and left halves of levator ani in normal 1-year-old girl. B, coronal view demonstrates significant flattening in child with exstrophy of angle of right and left halves of levator ani. C, sagittal view reveals pelvic floor angle in child with normal pelvic floor. D, sagittal view shows more flattened pelvic floor angle in child with classic bladder exstrophy.
by others as the puborectalis sling, attaches to the midline viscera during passage through the pelvic floor. This muscle provides direct and indirect support for the vagina, bladder and urethra by drawing these structures ventrally toward the pubic bone.15, 16 In extensive new human anatomy dissections with microscopic confirmation Shafik et al reported that the puborectalis sling extension of the levator ani connects to the urethra at the bladder neck.16 Furthermore, since their dissection showed the urethra in an infralevator position, it is protected from the effects of intra-abdominal pressure. These new findings differ from those in previous investigations. The levator hiatus lies obliquely with a downward and forward slope toward the symphysis pubis. In infancy the anteroposterior diameter of the hiatus in infancy is 1.5 to 2 cm. and the side-to-side diameter is 1 to 1.5 cm.16 We observed no surprises in regard to the anatomical relationship of the obturator internus and externus in the exstrophy group. Length and width were the same as in normal controls and the degree of axial divergence or external rotation was the value expected based on previous studies. In fact, the degree of external rotation was almost the same as the amount of external rotation of the anterior pelvic ring as described by Sponseller et al.4 Our study had several important findings concerning the levator ani muscle group, which for the most part forms most of the musculature of the pelvic diaphragm. A major finding was that the levator in exstrophy cases is not only wider than in normal controls, as measured by the area encompassed by the puborectal sling, but also clearly flatter with an angle between the levator ani halves of 31 degrees less than in controls. This finding was further emphasized by the pelvic floor angle of the puborectal sling, which was 15 degrees more anterosuperior in rotation, causing the sling to have a more flattened and less conical shape than in normal controls. Another finding in this study was that in the exstrophy population 68% of the puborectalis sling is posterior to the rectum, whereas in normal children about half of the muscle is anterior and half is posterior to the rectum. Thus, in exstrophy the muscle is centered posterior away from the posterior urethra and bladder neck, where the effect of contraction of the puborectalis, which aids in maintaining a closed bladder neck, would be lessened. It has been observed to be especially true by our pediatric surgery colleagues, who
have noted that proper placement of the rectum within the puborectalis sling is essential for achieving rectal continence.10, 11 Since the width of the levator hiatus is 2-fold and the length is 1.3-fold, normal and the puborectalis sling is misplaced, it is easy to understand how pelvic organ prolapse can occur. As expected, the external rotation of the levator ani is similar to that of the obturator internus and externus, and the angle of the overlap of the levator ani and obturator internus was not different. Thus, 3-D CT of the pelvis in classic exstrophy revealed many interesting new facts. External rotation of the obturator group in exstrophy follows the diastasis and should be corrected by osteotomy. However, to our knowledge the result of closure on the levator group and size of the hiatus is currently unclear. One may hypothesize that in patients with some degree of continence after simple bladder and posterior urethral closure the levator group is repositioned and any continence achieved is a function of the pelvic floor alteration that occurs even without the aid of a normal urinary sphincter. However, one of us (J. P. G.) has noted that some degree of continence may be achieved in newborns without osteotomy, so that appropriate movement of the pelvic floor with or without osteotomy may aid in eventual continence. In addition, possibly some patients who attain some degree of continence after closure alone may have a lesser abnormality of the levator group with more of the muscle located anterior to the rectum instead of posterior, where it may act on the bladder neck and posterior urethra to increase outlet resistance. Further studies are currently under way to image the pelvic floor after primary closure for evaluating the effect of closure on the pelvic diaphragm and levator hiatus. CONCLUSIONS
To our knowledge this study is the first precise description of the pelvic floor muscular anatomy of classic bladder exstrophy in children before corrective surgery. Because of the multifactorial nature of urinary incontinence, imaging can be valuable not only preoperatively, but also postoperatively. Hopefully these studies may enable us to understand better the role of exstrophy closure and osteotomy on the configuration of the pelvic floor as well as their effect on eventual continence. REFERENCES
1. Symington, J.: Further observations on the rectum and anal canal. J Anat, 46: 289, 1911 2. Berglas, B. and Rubin, I. C.: Study of the support structures of the uterus by levator myography. Surg Gynecol Obstet, 97: 677, 1953 3. Ruskin, M. J. and Pearl, R. K.: A three-dimensional model of the pelvic floor and anorectal anatomy. J Biocommun, 17: 22, 1990 4. Sponseller, P. D., Bisson, L. J., Gearhart, J. P. et al: The anatomy of the pelvis in the exstrophy complex. J Bone Joint Surg Am, 77: 177, 1995 5. Wall, L. L.: The muscles of the pelvic floor. Clin Obstet Gynecol, 36: 910, 1993 6. Strohbehn, K.: Normal pelvic floor anatomy. Obstet Gynecol Clin North Am, 25: 683, 1998 7. Klutke, C. G. and Siegel, C. L.: Functional female pelvic anatomy. Urol Clin North Am, 22: 487, 1995 8. Dietrich, R. B. and Kangarloo, H.: Pelvic abnormalities in children: assessment with MR imaging. Radiology, 163: 367, 1987 9. Mezzacappa, P. M., Price, A. P., Haller, J. O. et al: MR and CT demonstration of levator sling in congenital anorectal abnormalities. J Comp Assist Tomogr, 11: 273, 1987 10. Vade, A., Reyes, H., Wilbur, A. et al: The anorectal sphincter after rectal pull-through surgery for anorectal anomalies: MRI evaluation. Pediatr Radiol, 19: 179, 1989 11. Sato, Y., Pringle, K. C., Bergman, R. A. et al: Congenital anorectal anomalies: MR imaging. Radiology, 168: 157, 1988 12. Yazici, M., Sozubir, S., Kilicoglu, G. et al: Three-dimensional anatomy of the pelvis in bladder exstrophy: description of bone
PELVIC FLOOR ANATOMY IN CLASSIC BLADDER EXSTROPHY pathology using 3 dimensional CT and its clinical relevance. J Pediatr Orthop, 18: 132, 1998 13. Gearhart, J. P., Yang, A., Leonard, M. P. et al: Prostate size and configuration in adults with bladder exstrophy. J Urol, 149: 308, 1993 14. Silver, R. I., Yang, A., Ben-Chaim, J. et al: Penile length in adulthood after exstrophy reconstruction. J Urol, 157: 999, 1997
1449
15. Shafik, A.: Levator ani muscle: new physioanatomical aspects and role in the micturation mechanism. World J Urol, 17: 266, 1999 16. Shafik, A.: A new concept of the anatomy of the anal sphincter mechanism and the physiology of deffication. VIII. Levator hiatus and tunnel: anatomy and function. Dis Colon Rectum, 22: 539, 1979