| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Faculty of Medicine, Cairo University, Kaser El Aini
St., Cairo 11511, Egypt.
2 Department of Anatomy, Faculty of Medicine, Cairo University, Cairo,
Egypt.
3 Department of Urology, Faculty of Medicine, Cairo University, Cairo,
Egypt.
Received February 21, 2007;
accepted after revision June 22, 2007.
Awarded first prize for outstanding presentation at the 12th Symposium on
Urogenital Radiology, ESUR 2005, Ljubljana, Slovenia.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Dissection of seven formalin-preserved cadavers (age at death, 25–50 years; no parity history available) was performed by a professor of anatomy to explore the anatomy of the urethral supporting ligaments and was followed by MRI of the cadaveric specimens with ligamentous markers in place and then by histologic analysis of the dissected ligaments. MRI of 17 healthy nulliparous women (age range, 20–35 years; mean age, 25.5 years) was then performed using T2-weighted, dual turbo spin-echo, balanced fast-field echo, and STIR sequences. A standardized grid system that allowed us to record structural observations on sequentially numbered axial MR images was used by a radiologist who then applied a 4-point grading scale to assess ligament visibility. Three authors—one radiologist, one anatomist, and one urologist—then compared the appearance of each ligament seen in a cadaveric specimen with its appearance on MR images of the same cadaver and on MR images of volunteers.
RESULTS. At cadaveric dissection we identified ventral and dorsal urethral ligaments. The ventral urethral ligaments included the pubourethral ligaments, which were found to consist of three separate components coursing anteroposterior from the bladder neck to the pubic bone; the periurethral ligament; and the paraurethral ligaments. Dorsal to the urethra, a slinglike ligament, which we believe should be named the "suburethral ligament," was identified. This ligament had a distinct plane of cleavage from the anterior vaginal wall. The MRI findings in the volunteers correlated with the MRI and gross anatomic findings in the cadavers. The proximal pubourethral, periurethral, paraurethral, and suburethral ligaments had visibility scores of 3 (moderately visible) or 4 (easily visible) on MRI in 47%, 65%, 47%, and 53% of volunteers, respectively.
CONCLUSION. Our results present evidence that may help resolve previous controversies regarding the MR appearance of the ventral urethral ligaments and that better define the course of the ligament dorsal to the urethra, the suburethral ligament. We hope that this detailed anatomic information about the structures involved in continence may lead eventually to improvements in the treatments for women with stress urinary incontinence.
Keywords: anatomy • cadaveric dissection • MRI • paraurethral ligaments • periurethral ligaments • pubourethral ligaments • stress urinary incontinence • suburethral ligaments • urethral supporting ligaments • urethropelvic ligaments
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The precise anatomy—including the number, orientation, sites of attachment, and names of the urethral supporting ligaments—has not been elucidated fully, and to our knowledge there is currently no uniform opinion concerning these structures [6–21]. Specifically, whether the ligaments ventral to the urethra have a transverse [22, 23] or anteroposterior [6, 9, 24] orientation and whether the ligaments dorsal to the urethra (whose names are uncertain) originate from the urethra itself [2, 5] or from the anterior vaginal wall [3, 4] are areas of disagreement.
Although more insight about stress urinary incontinence has been gained in recent decades [17–20], the underlying abnormalities that cause this condition are still not fully understood. As a result, the surgical techniques used to treat stress urinary incontinence have not been standardized [25]. This opinion was substantiated by a comprehensive review that documented the poor quality of published literature on surgical treatment of urinary incontinence and led the authors to conclude that "recommendations as to the best of clinical practice cannot be based on scientific evidence" [26].
This lack of a thorough understanding of the ligaments that support the urethra in women is largely due to the facts that previous investigations have focused on only the limited areas of this multifaceted apparatus that can be studied in women during surgery [23] and that this area is known by both anatomists and surgeons to be difficult to dissect and examine [24, 27]. As a result, in prior published MRI studies, investigators attempted to integrate the status of all the functional elements with the anatomic elements involved in the urinary continence mechanism and to consider the urethra and its supporting structures to work as a consolidated unit, not as separate organs [2, 20]. This is difficult to do without a preexisting clear understanding of the anatomy because it is only through precise anatomic definition of the structures involved in continence that the proper diagnosis can be made and successful surgical treatments rendered [5].
The aim of our study was to define the normal appearance of the ligaments supporting the urethra in women through careful anatomic dissection of cadavers. The dissections were followed by MRI of the cadaveric specimens so that the anatomic findings could be correlated precisely with the imaging findings. We then performed MR examinations in a series of healthy nulliparous women who had no symptoms of stress urinary incontinence to determine the detectability and appearance of these ligaments on MRI in healthy women.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Seven formalin-preserved female cadavers (age at death, 25–50 years) were made available for dissection by the department of anatomy. No data on cadaver parity or cause of death were available; however, rigorous criteria were required for cadavers to be included in the study. Cadavers were used only if they failed to show any evidence of an abnormally sized organ, an abnormal configuration of organs, an organ positioned away from its normal site, enlarged lymph nodes, an abnormal soft-tissue mass, a pelvic bone fracture, or intrapelvic hematoma or bruising.
Dissection was performed by a professor of anatomy with more than 10 years of experience in anatomic dissection. Dissection of each female cadaver began with a midline incision made in the anterior abdominal wall. The urinary bladder was identified and retracted backward to visualize any ligament connecting the pelvic wall to the bladder neck in the retropubic space. The pelvic region was then separated from the cadaver by making a transverse section at the level of the fourth lumbar vertebra. Subsequently, a midline sagittal incision was used to divide the pelvis into right and left sides. Each hemipelvis was then examined in detail for ligaments connecting the urethra to the pubic bone or to the pelvic sidewall. The sites of attachment of all detected ligaments to the urethra and the pelvic wall were determined. Digital images were taken of each of the visualized ligaments.
A ferromagnetic marker (nifedipine capsule) was fixed to each of the dissected ligaments; the nifedipine capsule was placed carefully to avoid any resulting distortion of the pelvic anatomy. MRI was then performed on each of the cadaveric sections with a phased-array pelvic coil using a 1.5-T unit (Gyroscan Powertrak 6000, NT, release 6.2.1, Philips Medical Systems). Axial, coronal, and sagittal T2-weighted turbo spin-echo (TSE) images were acquired with a slice thickness of 5 mm and slice gap of 0.7 mm. Other MRI parameters were as follows: TR/TE, 5,000/132; flip angle, 90°; matrix, 512 x 512; number of excitations, 2; and field of view, 240–260 mm. Dual TSE sequences in the three relevant orthogonal planes were also acquired. For those sequences, both proton density and T2-weighted MR images were obtained in the same series using the following parameters: slice thickness, 4 mm; slice gap, 0.5 mm; TR/first-echo TE, second-echo TE, 4,000/18, 120; flip angle, 90°; matrix, 256 x 256; number of excitations, 2; and field of view, 170 mm.
Histologic evaluation was then performed to determine whether the dissected structures were, in fact, true ligaments. This evaluation could be performed for all of the ligaments except the paraurethral ligaments. Because of their delicate nature, the paraurethral ligaments were destroyed as a result of the deep dissection needed to visualize the other urethral ligaments and were visualized in the cadavers only before division of the pelvis into right and left sides. All of the other dissected ligamentous-like and ligamentous structures could be removed from the cadaver. This procedure was performed by the same anatomist.
The dissected ligaments were fixed in 10% formalin, and all the specimens were sent to the department of histology. Paraffin-embedded sections, created in longitudinal and transverse sections of approximately 5 µ in thickness, were prepared, stained with H and E and Masson trichrome stains, and examined using light microscopy. The microscopic analyses were performed by a pathologist with more than 8 years of experience.
Volunteer Study
MR examinations were then performed of 17 healthy nulliparous volunteers
(age range, 20–35 years; mean 
SD, 25.5 3.8 years) who
did not have lower genitourinary tract symptoms. Institutional review board
approval was granted, and informed consent was obtained from all of the
volunteers. Each volunteer completed a validated questionnaire
[28], was interviewed, and was
examined by an experienced urogynecologist with 15 years of experience in
clinical evaluation and management of patients with genitourinary and pelvic
floor dysfunction. Clinical assessment established that none of the included
volunteers had lower genitourinary tract symptoms or other evidence of pelvic
floor dysfunction.
MRI was performed with patients in the supine position using a pelvic phased-array coil. All patients had comfortably full bladders. Neither oral nor IV contrast agent was administered. Axial, coronal, and sagittal T2-weighted images of the pelvic region were acquired, followed by axial dual TSE sequences using the same parameters as those used to image the cadaveric specimens. Additional images through the urethra and bladder neck region in the axial plane were obtained using two other techniques: axial oblique T2-weighted balanced fast-field echo (FFE) and STIR with fat suppression. The MRI parameters of the balanced FFE sequence were as follows: slice thickness, 4 mm; gap, 0.5 mm; TR/TE, 5.0/1.92; flip angle, 60°; matrix, 256 x 256; number of excitations, 3; and field of view, 345 mm. The parameters of the STIR sequences included a slice thickness of 4 mm, a slice gap of 0.5 mm, 2,752/15, a flip angle of 90°, a 256 x 256 matrix, 3 excitations, and a field of view of 300 mm.
Image Analysis
Each ligament found on cadaveric dissection was assessed on MR images with
a marker in place so its visibility and MRI characteristics could be
evaluated. The radiologist, with the anatomist and the urologists, then
compared each ligament seen in the cadaveric specimen with MR images of the
same cadaver and with MR images of volunteers. For the volunteers, a grid
system similar to that described by Chou and DeLancey
[29] was used to
systematically collect and record structural observations of different
ligaments on serial axial T2-weighted MR images. Cumulative data from the 17
volunteers were compiled so that a "normal" standard appearance
and location for each ligament could be defined. All MR examinations were
interpreted by the same radiologist, with 7 years of experience interpreting
pelvic floor MR examinations.
The grid system was centered at the arcuate pubic ligament. This ligament is a thick arch fiber connecting the lower borders of the symphyseal pubic surface bounding the pubic arch. Superiorly it blends with the interpubic disk and extends laterally attached to the inferior pubic rami; its inferior edge is separated from the anterior border of the urogenital diaphragm by an opening for the deep dorsal vein of the clitoris [30]. The arcuate pubic ligament was chosen as the reference point for the grid system because it could be localized when MR observations were compared among different women. The most cranial image on which the arcuate pubic ligament could be visualized was chosen as the reference level and was defined as "image A."
Sequential axial images cephalad to image A were denoted with positive numbers and those caudad, with negative numbers. The presence or absence of visualized ligaments supporting the urethra for each scanning level cephalad or caudad to image A was reported, together with the best sequence or sequences on which the ligament could be seen and on which side (left or right side or both sides) each ligament was most frequently detected. A 4-point grading scale was then used to indicate ligament visibility as follows: 1, not visible; 2, poorly visible; 3, moderately visible; and 4, easily visible. The mean and median visibility scores were calculated for each ligament. The mode was also determined.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Dissection and MR Findings of the Ventral Urethral Supporting Ligaments
Pubourethral ligaments (PULs)—Two broad linear structures
were seen in the Retzius space on either side of the midline lateral to the
symphysis pubis to extend from the bladder neck to the posterior aspect of the
pubic bone. When the pelvic region was divided at the midline into right and
left sagittal sections, these structures could be seen to consist of a group
of three distinct but related ligaments: All have a similar anteroposterior
orientation, but each connects different portions of the ventral urethral
surface to the pubic bone (on either side of the symphysis pubis)
(Fig. 1A). We refer to the most
cranial of these three ligaments as the "proximal PUL." This
ligament extends from the anterolateral aspect of the proximal urethra to the
inferior part of the posterior aspect of the pubic bone. The second ligament,
which we refer to as the "intermediate PUL," extends from the
middle third of the urethra to the inferior border of the pubic bone. The
third of these ligaments, which we refer to as the "distal PUL,"
originates from the distal third of the urethra and extends to the inferior
border of the pubic bone on its ventral surface.
|
|
|
|
|
|
|
|
|
On axial proton density MR images of the cadaveric specimens, this ligament was depicted as an intermediate-signal-intensity slinglike structure coursing ventral to the urethra (Figs. 2B and 2C). The visibility score of the periurethral ligament was 4 in the four cadavers.
On the axial T2-weighted MR images of the healthy volunteers, this ligament also appeared as a slinglike structure anterior to the urethra. The proximal PUL was attached to the ventral aspect of the periurethral ligament (Fig. 2D). On axial oblique balanced FFE MR images, the periurethral ligament also was seen to connect the medial aspect of the puborectalis muscle on both sides (Fig. 2E). The visibility score was 3 or 4 in 11 volunteers and 2 in six volunteers.
Paraurethral ligaments—An additional set of ventral ligaments was identified. These tiny thin paraurethral ligaments could be visualized on axial proton density MR images of the cadaveric specimens as obliquely oriented linear structures of intermediate signal intensity connecting the lateral wall of the urethra to the periurethral ligaments. The same configuration was seen on the MR images of the volunteers (Fig. 3A, 3B, 3C). The visibility score was 3 in three cadavers and 2 in one cadaver; this score was 3 or 4 in eight volunteers, 2 in three volunteers, and 1 in six volunteers.
|
|
|
|
|
|
|
|
|
In volunteers, these structures were identified on axial T2-weighted MR images obtained at the level of the proximal and middle urethra. At these levels, the urethra was seen to lie on a supporting shelflike layer located ventral to the anterior vaginal wall (Fig. 2D). On T2-weighted TSE MR sequences, this supporting layer consisted of a low-signal-intensity ligamentous structure overlying another high-signal-intensity layer that was intimately related to the anterior vaginal wall. The ligamentous part of the supporting layer could be traced, more reliably than the high-signal layer part, to extend anterolaterally toward both lateral pelvic sidewalls at the site of origin of the levator ani from the obturator internus muscles.
On axial fat-suppression STIR MR images, the ligamentous component was depicted as a predominantly low-signal-intensity structure dorsal to the urethra that runs posterior to the urethra and forms a suburethral sling (Fig. 5C). On a more cephalic image, a plane of cleavage was clearly depicted between the anterior vaginal wall and this ligament (Fig. 5D). To our knowledge, this structure, which we refer to as the "suburethral ligament," has not been described previously.
|
|
Histologic analysis of the suburethral ligament—In specimens obtained at the lateral attachment of the suburethral ligament to the pelvic sidewalls for histologic evaluation, this ligament was noted to contain skeletal muscle fibers interspersed with collagen fibers (Fig. 5E).
|
|
|
|
|
|
|
|
The proximal PUL, periurethral, paraurethral, and suburethral ligaments were assigned visibility scores of 3 or 4 in 47%, 65%, 47%, and 53% of volunteers, respectively. Of these ligaments, the periurethral and suburethral ligaments were most frequently and most easily visualized. The best way to see the whole length of a proximal PUL was to locate it first on a sagittal sequence and then to obtain axial oblique balanced FFE thin-slice images through the regions where the ligament was present (Fig. 2E).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There has been some confusion about the anatomy of the female ventral urethral supporting ligaments. Although DeLancey [22] has stated that the pubovesical muscle or ligament originates as an extension of the detrusor muscle at the level of the bladder neck and runs as a transverse band across the anterior portion of the urethrovesical junction and then runs laterally to the arcus tendineus fascia, Vazzoler and colleagues [24] reported that this structure actually consists of three paired PULs that run in an anteroposterior direction from the bladder neck to the symphysis pubis. Our study has confirmed the findings of Vazzoler et al. by showing that the pubourethral structure consists of three closely associated paired ligaments: the proximal, intermediate, and distal PULs.
|
Unfortunately, the intermediate and distal PULs are not as consistently or as easily visualized on MR images. The difference between cadaveric specimens and volunteers regarding the inability to visualize the same ligaments in both groups also was encountered by Tan et al. [1]. This difference was suspected to be due to one of three possibilities: the difference in age between cadavers and volunteers, anatomic variations, or use of cadaver sectioning.
The limited visibility of the proximal PUL and poor visibility of the intermediate and distal PULs on routine MRI could, in part, be responsible for prior controversy about the presence and orientation of the ventral urethral ligaments, a debate that has even extended to confusion concerning the names given to these ligaments, which have included "pubourethral," "preurethral," "pubourethralis," and "pubovesical" ligaments [15].
We have also confirmed the presence of a previously described [1, 2] separate periurethral ligament: a thin T2 hypointense structure connecting the medial aspects of the puborectalis muscle that courses ventral to the proximal and mid urethra.
Another controversy has concerned the nature and attachment of the structures that support the dorsal aspect of the proximal and mid urethra. Specifically, there has been a lack of consensus as to whether the ligament dorsal to the urethra, named the "urethropelvic ligament" by Klutke et al. [5] and the "pubourethral ligament" by Kim et al. [2], arises primarily from the lateral aspect of the urethra or primarily from the vagina and periurethral tissue to attach laterally to the pelvic wall, as described by DeLancey [3, 4]. Our results from both the anatomic cadaveric dissections and the subsequent MRI evaluations of cadavers and volunteers prove that the ligament dorsal to the urethra runs posterior to it in the form of a suburethral sling and that there is a plane of cleavage between this ligament and the anterior vaginal wall. Due to the course of this ligament, we suggest that a more apt name for it is "suburethral ligament." To our knowledge, the fact that this ligament is separate from the anterior vaginal wall and runs beneath the urethra rather than being attached to the urethra has not been described previously.
The grid system used in our study allowed us to compare similar structures in different women using standardized landmarks. The importance of applying such a grid system to collect data about a subject as confusing and controversial as the ligaments supporting the female urethra cannot be overstated. This system also facilitates distinction of normal anatomy from occasionally encountered anatomic variants. For example, at a scanning level in which the periurethral ligament is visible only sometimes, such as on image 6 above the arcuate pubic ligament, the failure to visualize this ligament can be normal; however, in regions where it is always seen, such as on images 2 and 3 above the arcuate pubic ligament, the failure to visualize the same ligament can be considered abnormal when correlated with a patient's symptoms and with clinical findings. Such information eventually may serve to improve our understanding of the functional urethral anatomy and of the anatomic rationale for successful surgical repair.
We used a phased-array coil instead of an endovaginal coil, even though the local spatial resolution of the latter is superior, because recent studies have shown that the quality of MR images obtained using a phased-array coil suffices for evaluation of the urethral supporting system [23, 29]. Also, by using a phased-array system, we avoided disturbing the normal structural relationships that might have resulted from placing a coil in the vagina.
Knowledge of normal and variant anatomy obtained through cadaveric dissection is essential to gain an understanding of human anatomy, including the anatomy of the ligaments supporting the female urethra. Our study attempts to shed some light on the urethral supporting ligaments. Based on our findings, we are now able to define the normal anatomic locations and some of the anatomic variations of the urethral supporting ligaments. We hope that with this improved knowledge of female periurethral anatomy, other series, including outcome-related investigations, can be initiated to evaluate further the role of MRI in the management of patients with stress urinary incontinence, because previous reports have estimated that each year 300,000 women require surgery for the treatment of pelvic organ prolapse and stress urinary incontinence [31, 32]. In addition, hundreds of operations have been described for the management of stress urinary incontinence [33]. Up to 29% of those patients, in fact, were forced to undergo repeated operations [34]. These statistics show that stress urinary incontinence is a problem and that treatment has often not been optimal. Nevertheless, the common occurrence of repeated operations indicates the need for improved treatments [35].
DeLancey [36] stated in 1996 that stress urinary incontinence results from damage to specific muscles, fascial structures, and nerves of the pelvic floor. The author suggested that if we begin to define the damage occurring in each element of the continence mechanism we should be able to select precise treatment plans on the basis of the different abnormalities found in different patients. This may help clinicians switch from the current empiric approach of treatment, which is based on a patient's symptom complex, to treatment of an individual patient's specific neuromuscular and fascial defect that results in symptoms. In a more recent study [35], DeLancey stated that at present there are no validated tests that have been proven to make these distinctions and that this kind of research is desperately needed and is now within our reach given the dramatic changes in MRI, which has revolutionized our ability to see the pelvic floor in living patients.
We believe that our study, which has refined the understanding of the locations of the normal ligaments supporting the urethra and correlated anatomic findings with imaging findings, may be an important initial step in refining the treatment of stress urinary incontinence, because it is only through detailed understanding of the normal continence mechanism that abnormalities can be identified and corrected. Promising results have been reported with respect to the ability of MRI to depict abnormalities in women with stress urinary incontinence and that information has been used to change patient management. Recently, we have been able to delineate the precise underlying anatomic defect that is responsible for symptoms in individual patients presenting with stress urinary incontinence [37]. In that study, we found differences in the underlying structural defects in women with stress urinary incontinence. These defects involved one or more elements of the urethral supporting structures. Of course, the next step is to determine how patients with these defects will be treated best, but this research has yet to be performed. Indeed, MRI evaluations of women with pelvic floor dysfunction provided significant information that altered clinical management in 41.6% of patients with urinary incontinence in one study [38], and MRI and cystocolpoproctography findings led to changes in the initial operative plan in 41% of patients with pelvic floor dysfunction in another study [39].
Our study has a number of limitations. One of the most important limitations, which also affects previously performed studies regarding the urethral supporting ligaments, is the variation in number and appearance of the ligaments, even in dissected specimens [22, 24]. Of course, reliance on the anatomy of female cadavers of varying ages may also be a problem. Because the parity of the female cadavers used in our study is not known, we cannot determine whether any of them had abnormal urethral anatomy, especially the one cadaver in whom no urethral ligaments were detected. We are also uncertain about why there were differences in which MR sequence best showed the ligaments in living volunteers (TSE T2-weighted) and in cadavers (proton density). It might be due to postmortem fixation with formalin or to other unidentified factors.
|
In conclusion, cadaveric dissections, histologic examinations, and subsequent MRI of marked urethral supporting ligaments of cadavers and MRI of periurethral anatomy in nulliparous women yielded detailed information about the nature of ligaments supporting the urethra. We believe that our results have helped to resolve previous controversies regarding the ventral urethral ligaments and have also facilitated the definition of a new course and orientation of a ligament dorsal to the urethra, one that we believe should be referred to as the "suburethral ligament." We hope that this detailed anatomic information about the structures involved in continence may lead eventually to improvements in the treatments for women with stress urinary incontinence.
Acknowledgments
The first author owes special thanks and gratitude to Ahmed Samy, head of
the radiology department, and to all her professors and senior staff members
in the Department of Radiology, Faculty of Medicine, Cairo University, for
providing her the time and the facilities needed to conduct and publish this
research, as well as to all operators of the MRI unit who have worked with
her, especially Mona Ali and Mervate Mohamed, for their utmost cooperation
during the tedious process of imaging the cadaveric specimens.
We gratefully acknowledge Richard Cohan, Department of Radiology, University of Michigan Hospital, Ann Arbor, MI, for his important contributions, assistance, and sincere advice during manuscript preparation, as well as Jurgen Futterer and Jelle Barentsz, Department of Radiology, University Medical Centre Nijmegen, The Netherlands, for their thoughtful suggestions and valuable comments on the manuscript.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R. F. El Sayed, S. El Mashed, A. Farag, M. M. Morsy, and M. S. Abdel Azim Pelvic Floor Dysfunction: Assessment with Combined Analysis of Static and Dynamic MR Imaging Findings Radiology, August 1, 2008; 248(2): 518 - 530. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
