Role of Static and Dynamic MR Imaging in Surgical Pelvic Floor Dysfunction
Abstract
Pelvic floor dysfunction (PFD) is a hidden women’s health epidemic in the United States, with over 10% of women having a lifetime risk for undergoing a surgical repair for this problem. Given the paucity of understanding of PFD pathophysiology and the high rate of recurrence and repeat surgery, imaging plays a major role in its clinical management, especially for the preoperative assessment of patients with multicompartment defects and failed surgical repairs. The recent development of fast magnetic resonance (MR) imaging sequences allows noninvasive, radiation-free, rapid, high-resolution evaluation of the entire pelvis in one examination. The H line, M line, organ prolapse (HMO) classification system, which is applied to dynamic MR images, allows consistent standardization and grading of various forms of PFD. In addition, the HMO system clearly defines and differentiates between the two main components of PFD: pelvic floor relaxation and pelvic organ prolapse. In addition to serving as an objective diagnostic tool in patients with surgical PFD, MR imaging has tremendous potential to be used as a research tool in trying to understand the pathophysiology of these complex disorders.
© RSNA, 2008
LEARNING OBJECTIVES FOR TEST 1
After reading this article and taking the test, the reader will be able to:
| •. | List the pathologic conditions that are collectively described as pelvic floor disorders. | ||||
| •. | Describe the role of MR imaging in management of pelvic floor disorders. | ||||
| •. | Identify various forms of pelvic floor dysfunction on MR images. | ||||
Introduction
Pelvic floor dysfunction is an umbrella term for a heterogeneous group of disorders affecting up to 50% of middle-aged and older parous women presenting with pelvic pain, pressure, dyspareunia, incontinence, incomplete emptying, and gross organ protrusion (,1,,2). Up to 10% of women in the United States develop pelvic floor dysfunction so severe that they require surgery (,3). However, only a small minority are likely to benefit exclusively from surgery. Pelvic floor dysfunction represents an important public health concern, with urinary incontinence alone affecting 10 million U.S. women with the resulting annual care cost at $10 billion (,4). Whereas exact mechanisms are subject to debate, risk factors include age, multiparity, complicated vaginal deliveries, obesity, collagen-related disorders, hysterectomy, and menopause (,5,,6). Possible causes include injury to the pelvic floor from surgery or childbirth, denervation of the musculature, fascial defects, and abnormal synthesis or degradation of collagen (,7).
Pelvic Organ Prolapse versus Pelvic Floor Relaxation
In pelvic floor relaxation, the pelvic sling , which is an aggregate term for active and passive pelvic floor support structures including muscles and connective tissue, becomes weakened and unresponsive. This results in pelvic floor dysfunction at rest and during activities that increase intraabdominal pressure such as coughing, sneezing, urination, or defecation.
Diagnosis and Treatment
Correspondence among symptoms, physical examination results, and a given anatomic derangement may be equivocal because pelvic anatomy is often distorted due to prior surgery such as hysterectomy or failure of multicompartment surgical repairs. Therefore, selecting the most appropriate surgical candidates on the basis of clinical findings alone is suboptimal. Clinical scoring systems such as the Pelvic Organ Prolapse Quantification system help standardize diagnosis but are less useful for surgical triage and planning, since they do not involve direct assessment of surgical anatomy (,5,,8).
Successful treatment involves understanding individual anatomic, medical, and psychosocial components of pelvic floor dysfunction. In the United States, 300,000–400,000 women undergo surgery with variable symptom relief, with about one-third of those being repeat operations (,5,,9–,11). Clearly, a robust and reproducible system for preoperative triage is necessary to minimize treatment failures.
At our institution, physical examination, fluoroscopy, and MR imaging are used in a complementary fashion in evaluation of bladder, vaginal, and rectal disorders. In fact, urodynamics and anorectal manometry are useful in functional testing and help localize problems to the anterior, middle, and posterior compartments. However, unlike MR imaging, conventional fluoroscopic imaging techniques do not allow simultaneous assessment of all of the pelvic compartments in one examination, do not allow assessment of the pelvic organs and the pelvic floor itself, and require multiple examinations and significant radiation exposure to the patient. Therefore, these examinations are used in a complementary fashion when assessing patients with pelvic floor dysfunction.
In this article, we use our extensive experience as a major tertiary referral center for pelvic floor dysfunction to describe current concepts in surgical pelvic floor dysfunction, including pelvic organ prolapse and pelvic floor relaxation, and to illustrate how MR imaging assists urogynecologists in localizing and grading pelvic floor dysfunction for surgical triage and planning.
Surgically Relevant Anatomy of the Pelvic Floor
Surgeons view the female pelvis as three functional compartments supported by three components of the pelvic sling: anterior compartment (bladder and urethra), middle compartment (vagina, cervix, uterus, and adnexa), and posterior compartment (anus and rectum) (,Fig 1,). The pelvic floor is a multilayered anatomic and functional unit, which consists of muscles and connective tissue in three contiguous supporting layers from cranial to caudal: the fascial layer (endopelvic fascia), the intermediate pelvic diaphragm, and the urogenital diaphragm.
Figure 1a. Normal female pelvic anatomy. (a) Midsagittal rapid half-Fourier T2-weighted MR image, obtained with a 1.5-T imaging unit, shows the bladder, urethra, and vagina. The base of the bladder is at the upper margin of the symphysis pubis. The urethra appears as a linear layered structure (*), with a hyperintense central mucosal band surrounded by a hypointense wall. The urethra forms an acute angle with the bladder base and extends 2–3 cm caudally through the muscular pelvic sling. The vagina appears as a linear tubular structure with a hyperintense mucosa and hypointense wall located just behind the bladder. The pelvic sling extends obliquely from the coccyx toward the posterior levator plate just behind the anorectal junction. (b) Corresponding axial image shows the urethra, vagina, and rectum. The pelvic sling appears as a symmetric hypointense concentric sling. Note the asymmetry of the levator muscles (*) in this asymptomatic patient. This finding can also be caused by uneven positioning of the patient in the imaging unit or by chemical shift artifact. Figure 1b. Normal female pelvic anatomy. (a) Midsagittal rapid half-Fourier T2-weighted MR image, obtained with a 1.5-T imaging unit, shows the bladder, urethra, and vagina. The base of the bladder is at the upper margin of the symphysis pubis. The urethra appears as a linear layered structure (*), with a hyperintense central mucosal band surrounded by a hypointense wall. The urethra forms an acute angle with the bladder base and extends 2–3 cm caudally through the muscular pelvic sling. The vagina appears as a linear tubular structure with a hyperintense mucosa and hypointense wall located just behind the bladder. The pelvic sling extends obliquely from the coccyx toward the posterior levator plate just behind the anorectal junction. (b) Corresponding axial image shows the urethra, vagina, and rectum. The pelvic sling appears as a symmetric hypointense concentric sling. Note the asymmetry of the levator muscles (*) in this asymptomatic patient. This finding can also be caused by uneven positioning of the patient in the imaging unit or by chemical shift artifact.

The endopelvic fascia consists of peritoneal reflections fused with the underlying connective tissue layer and provides important anterior and lateral passive support to the vagina, bladder, and urethra, helping impede abnormal descent during strain (,12,,13). The fascia around the vagina and the uterus forms the sacrouterine–cardinal ligament complex, which is superior to the pubocervical fascia between the bladder and the vagina (,13,,14). The sacrouterine–cardinal ligament complex suspends and supports the vaginal cuff as it pulls the cervix and the upper vagina superiorly and posteriorly toward the sacrum. Posteriorly, the sacrouterine–cardinal ligament fuses with the fascia around the rectum (prerectal fascia) and prevents pelvic floor descent (,15). The arcus tendineus fascia pelvis and the arcus tendineus levator ani are the connective tissue condensations of the obturator and levator ani fascia, which provide lateral support by anchoring organs to the pelvic side wall. Even though the fascia is not directly imaged at MR imaging, its failures in organ support can be inferred from pathologic movement of pelvic organs during dynamic MR imaging.
Caudal to the fascia is the muscular pelvic diaphragm, consisting mainly of the levator ani complex, which is composed of five muscle groups (pubovisceralis or pubococcygeus muscle, made up of puboperineal, pubovaginal, and puboanal components; iliococcygeus muscle; and puborectalis muscle). The pelvic diaphragm is tonically contracted at rest and helps maintain continence, especially during increased intraabdominal pressure. At rest, the muscles act as an anterior sling, drawing the urethra anteriorly and superiorly relative to the bladder base, resulting in an acute angle between the bladder neck and the urethra. Similar acute angulation is maintained between the anus and the rectum as the anus is drawn anterosuperiorly (,16,,17). The muscular pelvic diaphragm is usually assessed with axial and sagittal T2-weighted MR imaging.
The most caudal layer of the pelvic floor, the urogenital diaphragm, is composed mainly of connective tissue and the deep transverse perineus muscle, oriented transversely just deep to the pelvic diaphragm and anterior to the anorectal junction (,18,,19). It is a diamond-shaped combination of two contiguous triangles (anterior and posterior), whose common base is the transverse perineus muscle. The apex of the anterior triangle is the symphysis pubis and the sides are the pubic bones. The apex of the posterior triangle is the tip of the coccyx. The anterior triangle is penetrated by the urethra and vagina, whereas the posterior triangle is penetrated by the anus. The perineal body is directly anterior to the anal sphincter and anchors critical surrounding structures: the external anal sphincter, muscles of the urogenital diaphragm, and the puborectalis muscle (,20). The sacrouterine–cardinal ligament complex and the prerectal fascia also attach to the perineal body and function as a single support unit posteriorly, preventing descent of the pelvic floor (,15). Although the urogenital diaphragm is variably visualized in routine clinical MR imaging applications, it has been imaged with high resolution with the use of endoluminal coils (,19,,20).
Pelvic organ prolapse results when these support mechanisms to individual organs are disrupted. Defects in support structures leading to various degrees and combinations of pelvic organ prolapse and pelvic floor relaxation are described later. Surgical repair varies accordingly.
MR Imaging Provides Important Preoperative Information
MR imaging may be supplemented by fluoroscopically based functional testing for a thorough evaluation of pelvic floor disorders (,21). These tests include voiding urodynamics for the assessment of bladder and urethral function as well as evacuation proctography and anorectal manometry for the assessment of anorectal function. Although conventional fluoroscopic techniques may be used for anatomic assessment, they entail radiation exposure and are invasive as well as cumbersome, since they require single or often multiple organ opacification. In addition, these studies provide limited anatomic information because they do not directly image the pelvic floor or internal pelvic contents, including incidental lesions.
MR imaging has the following advantages: It lacks ionizing radiation, is noninvasive, and provides rapid, comprehensive, high-resolution, high-contrast evaluation of the entire pelvis, including support structures and organs (,22–,26). Moreover, MR imaging allows assessment of the behavior and relationship of pelvic organs and the pelvic floor statically (at rest) and dynamically (during active contraction or the Valsalva maneuver), providing an objective grading of prolapse and pelvic floor relaxation. This is especially useful for evaluation of multicompartment disorders (,14,,20,,25,,27–,29). Detection of incidental pathologic conditions, such as urethral and bladder diverticula, endometrial polyps, malignant lesions, fibroids, and adnexal lesions, with MR imaging is useful for treatment planning. Urogynecologists find MR imaging especially important in the preoperative planning of complex repairs in patients with failed surgical repairs.
One possible disadvantage of the MR imaging protocol used at our institution is that no gel is used to opacify the vagina or rectum, leading to difficulty in visualization of small rectoceles (,21,,25).
MR Imaging Techniques
Protocols vary by institution, with MR imaging studies performed with the patient in both supine and upright positions. At our center, as in a majority of others, patients are imaged in the supine position with pelvic or torso phased-array coils in a 1.5-T imaging unit. It is important to center the coil low on the pelvis to ensure visualization of prolapsed organs. For this examination to be reliable, these often elderly and nervous patients must be put at ease and be comfortable in the intimidating environment of an MR imaging unit. Thus, a relationship similar to that established between patients and staff in mammography must be established. Usually, a small core group of technologists must be trained in placing the patient at ease and giving her clear and simple instructions. Patient cooperation with instructions to rest and strain is critical for a useful examination (,Fig 2,). Reviewing the concept of the Valsalva maneuver with the technologists so as to coach the patient on how to perform it properly without lifting the pelvis is paramount.
Figure 2a. Example of squeezing and not straining during the dynamic phase of MR imaging performed in a 54-year-old woman with a complex history of urinary symptoms. (a) Resting midsagittal rapid half-Fourier T2-weighted image obtained at 1.5 T shows an incidentally detected 10.77-mm hypointense focus at the uterine fundus. This finding most likely represents an intramural fibroid. (b) Corresponding dynamic image shows contraction of the pelvic sling musculature instead of straining, as the hiatus is markedly shortened and the muscle is clearly contracted. Owing to the poor technique, no significant organ prolapse or pelvic floor relaxation was demonstrated on images from this study. Figure 2b. Example of squeezing and not straining during the dynamic phase of MR imaging performed in a 54-year-old woman with a complex history of urinary symptoms. (a) Resting midsagittal rapid half-Fourier T2-weighted image obtained at 1.5 T shows an incidentally detected 10.77-mm hypointense focus at the uterine fundus. This finding most likely represents an intramural fibroid. (b) Corresponding dynamic image shows contraction of the pelvic sling musculature instead of straining, as the hiatus is markedly shortened and the muscle is clearly contracted. Owing to the poor technique, no significant organ prolapse or pelvic floor relaxation was demonstrated on images from this study.

As part of the straining process, incontinence-related accidents are unavoidable and patients should be instructed appropriately. They should be dressed in a light cotton gown and remove their undergarments, continence garments, pads, and vaginal pessaries to maximize relative movement of pelvic organs. The bladder should be emptied prior to the study. Patients are placed in a supine position with a pelvic or torso phased-array coil. A towel is placed underneath the buttocks to absorb any urine or fecal accidents with the best alignment of the symphysis pubis and coccyx for optimal imaging in the midsagittal plane. Knees should be bent and supported lightly by a pillow with the legs somewhat apart so as not to interfere with organ prolapse, especially during the straining phase of the study. The entire rest-strain sequence can be repeated two or three times until the best images are obtained.
MR imaging sequences include the following: axial, sagittal, and coronal localizing images of the entire pelvis (repetition time msec/echo time msec = 15/5, field of view = 350–400 mm, matrix = 160 × 256, section thickness = 10 mm, intersection gap = 2 mm). With the patient at rest, the following static sequences of the entire pelvis are performed: axial, sagittal, and coronal rapid half-Fourier T2-weighted imaging sequence such as single-shot fast spin-echo (SE) or HASTE (half-Fourier acquisition single-shot turbo SE) (4.4/90, field of view = 350 mm, matrix = 128 × 256, section thickness = 6 mm, intersection gap = 2 mm) and axial dual-echo gradient-echo imaging. After the midline between the symphysis pubis and the coccyx is localized, the patient is instructed to progressively push forcefully and strain the pelvic floor inferiorly (Valsalva maneuver) for at least 10 seconds. In the midsagittal plane, six to eight rapid half-Fourier T2-weighted images are acquired at the same level while the patient progressively strains to maximal Valsalva maneuver. This procedure is also repeated in the axial plane with rapid half-Fourier T2-weighted sequences.
Interpretation and Grading of Pelvic Floor Disorders with MR Imaging: The HMO System
HMO Grading of Pelvic Floor Relaxation
To help standardize interpretation and grading of pelvic floor dysfunction with MR imaging, the HMO (H line, M line, organ prolapse) system was developed, which is applied to a midsagittal rapid half-Fourier T2-weighted image obtained during maximal patient strain (,14). On the midssagittal image obtained during maximal strain, three points of reference are first defined: A, the inferior margin of the symphysis pubis; B, the convex posterior margin of the puborectalis muscle sling (the posterior levator plate); and C, the junction between the first and second coccygeal segments (,Fig 3). Two anatomic fixed references in the HMO system are (a) the pubococcygeal line (PCL), which is drawn between points A and C, and (b) point B. The degree of pelvic floor relaxation is then graded as a measure of two components: hiatal widening (enlargement) and hiatal descent (,Fig 4,,,).
Figure 3. Midsagittal T2-weighted single-shot fast SE relaxed image, obtained in a female patient who had undergone hysterectomy, shows anatomic landmarks used in the HMO classification system. Point landmarks are A (the inferior margin of the symphysis pubis), B (the posterior aspect of the puborectalis muscle sling), and C (the junction between the first and second coccygeal elements). Reference lines are the PCL, which is drawn from A to C and is a fixed anatomic reference line; H (the puborectal line), which represents the anteroposterior hiatal dimension and is drawn from A to B; and M, which is the shortest distance between B and the PCL and is a measure of pelvic floor descent. Figure 4a. Two components of pelvic floor relaxation. (a) Midsagittal T2-weighted single-shot fast SE image, obtained in a female patient in a relaxed position, shows the two vectors (components) of pelvic floor relaxation or prolapse: widening and descent. On radiologic images, these vectors are defined as an enlarging H and an enlarging M, respectively. H is the anteroposterior dimension of the levator hiatus, whereas M represents the descent of the levator from the PCL. (b) Diagram shows changes associated with pelvic floor descent. (c, d) Diagrams show a normal (c) and a pathologically widened (d) pelvic hiatus. Figure 4b. Two components of pelvic floor relaxation. (a) Midsagittal T2-weighted single-shot fast SE image, obtained in a female patient in a relaxed position, shows the two vectors (components) of pelvic floor relaxation or prolapse: widening and descent. On radiologic images, these vectors are defined as an enlarging H and an enlarging M, respectively. H is the anteroposterior dimension of the levator hiatus, whereas M represents the descent of the levator from the PCL. (b) Diagram shows changes associated with pelvic floor descent. (c, d) Diagrams show a normal (c) and a pathologically widened (d) pelvic hiatus. Figure 4c. Two components of pelvic floor relaxation. (a) Midsagittal T2-weighted single-shot fast SE image, obtained in a female patient in a relaxed position, shows the two vectors (components) of pelvic floor relaxation or prolapse: widening and descent. On radiologic images, these vectors are defined as an enlarging H and an enlarging M, respectively. H is the anteroposterior dimension of the levator hiatus, whereas M represents the descent of the levator from the PCL. (b) Diagram shows changes associated with pelvic floor descent. (c, d) Diagrams show a normal (c) and a pathologically widened (d) pelvic hiatus. Figure 4d. Two components of pelvic floor relaxation. (a) Midsagittal T2-weighted single-shot fast SE image, obtained in a female patient in a relaxed position, shows the two vectors (components) of pelvic floor relaxation or prolapse: widening and descent. On radiologic images, these vectors are defined as an enlarging H and an enlarging M, respectively. H is the anteroposterior dimension of the levator hiatus, whereas M represents the descent of the levator from the PCL. (b) Diagram shows changes associated with pelvic floor descent. (c, d) Diagrams show a normal (c) and a pathologically widened (d) pelvic hiatus.




The puborectal hiatus line (H line) allows grading of the maximal widening of the pelvic sling in the anteroposterior dimension during straining and is the linear distance between points A and B. The H line represents the most caudal part of the levator ani group (the puborectalis muscle). The abnormal widening of the pelvic floor, measured with the H line, is graded progressively when it exceeds 6 cm in length (,30). The second component of pelvic floor relaxation is the M line, which is the measure of muscular pelvic floor descent. The M line extends perpendicularly from the PCL to the posterior end of the H line (point B). Abnormal descent (M line) is progressively graded when its length exceeds 2 cm (,Table 1). With significant pelvic floor relaxation, variable degrees of pelvic floor widening (H line) and descent (M line) are present at dynamic MR imaging.
HMO Grading of Pelvic Organ Prolapse
MR Imaging Findings
General Considerations
On static images, the positions of the bladder, urethra, vagina, uterus, and rectum are noted (,Fig 1,). The uterus is often absent or atrophic. The rectum lies posteriorly and cranially to the transverse perineus muscle, making an acute angle with the anus. On axial images, the urethral cuff is located anteriorly and has a bull’seye configuration at T2-weighted imaging, with central hyperintense mucosa. The vagina has an H-shaped mucosal rim that is hyperintense on T2-weighted images. The perineum is seen as a diamond-shaped structure containing two triangles. The anorectum lies behind the transverse perineus. The anterior pelvic sling musculature is best evaluated on axial images because it appears relatively symmetric as a concentric sling, which is hypointense on T2-weighted images. The levator ani is assessed for asymmetry or gross disruptions on static and straining axial images.
At dynamic midsagittal T2-weighted MR imaging, the pelvic floor sling may normally descend and widen by up to 2 cm. Pelvic organs follow the pelvic floor inferiorly without protrusion through their respective hiatuses (no prolapse) or bulging into surrounding organs.
Pelvic Floor Relaxation
Pelvic floor relaxation has two components that can be graded during maximal strain: hiatal enlargement (H line) and pelvic floor descent (M line) (,Table 1). In asymptomatic patients, the H line is less than 6 cm and the M line is no more than 2 cm in length (,25,,30). With hiatal widening and relaxation of the supporting structures, the levator plate descends over time. As the pelvic floor is pathologically descending, so do the organs above it that it is supposed to support. Consequently, pelvic floor relaxation and organ prolapse are not synonymous and should be differentiated. In fact, many combinations of pelvic floor relaxation and pelvic organ prolapse can occur in a given patient. Although rare, gross focal tears in the levator ani complex may be appreciated at axial MR imaging during straining.
Pelvic floor relaxation develops over time as the fascia and the muscular levator sling become weakened (,Fig 4,,,). In patients without pelvic floor dysfunction, the fascia and the levator musculature create the so-called banana-shaped vaginal axis by elevating the distal vagina (,Fig 5). This normal configuration is often lost in cases of prolapse, as the vagina assumes a more vertical orientation. This pathologic state allows increases in intraabdominal pressure, which push the rectal wall forward, causing rectocele formation. Therefore, it is imperative to evaluate all of the pelvic floor compartments for concomitant organ prolapse in cases of pelvic floor relaxation (,Figs 6,, ,7,).
Figure 5. Diagram shows the normal banana-shaped configuration of the vaginal axis. This configuration is created by elevation of the distal one-third of the vaginal axis by the pelvic musculature and fascia. Figure 6a. Pelvic floor relaxation without accompanying organ prolapse in a 74-year-old woman with a history of three vaginal deliveries and hysterectomy who presented with constipation. (a) Midsagittal single-shot fast SE image obtained at 1.5 T during relaxation shows that the patient has undergone hysterectomy. (b) Corresponding straining image shows that the anteroposterior hiatal dimension (H line) is 6.6 cm and the pelvic floor descent (M line) is 3.5 cm, findings consistent with grade 1 pelvic floor relaxation and descent. There is no accompanying pelvic organ prolapse, with the urethra, bladder, vaginal apex, and bowel all above the H line. Figure 6b. Pelvic floor relaxation without accompanying organ prolapse in a 74-year-old woman with a history of three vaginal deliveries and hysterectomy who presented with constipation. (a) Midsagittal single-shot fast SE image obtained at 1.5 T during relaxation shows that the patient has undergone hysterectomy. (b) Corresponding straining image shows that the anteroposterior hiatal dimension (H line) is 6.6 cm and the pelvic floor descent (M line) is 3.5 cm, findings consistent with grade 1 pelvic floor relaxation and descent. There is no accompanying pelvic organ prolapse, with the urethra, bladder, vaginal apex, and bowel all above the H line. Figure 7a. Grade 1 pelvic floor widening and descent, grade 2 cystocele, and urethrocele in a 46-year-old multiparous woman with urinary frequency, urgency, constipation, a sensation of incomplete emptying, pelvic pressure, and pain. (a) Resting midsagittal single-shot fast SE image obtained at 1.5 T shows no pelvic floor prolapse or organ prolapse. (b) Corresponding dynamic image shows grade 1 pelvic floor widening with an anteroposterior hiatal length of 6.1 cm (H line) and pelvic floor descent of 2.0 cm (M line), a grade 2 cystocele with the bladder 3.1 cm below the H line, and a urethrocele with the urethra below the H line. Note the hypermobility of the urethra, which assumes a horizontal position during strain. The air in the bladder was secondary to catheterization. Figure 7b. Grade 1 pelvic floor widening and descent, grade 2 cystocele, and urethrocele in a 46-year-old multiparous woman with urinary frequency, urgency, constipation, a sensation of incomplete emptying, pelvic pressure, and pain. (a) Resting midsagittal single-shot fast SE image obtained at 1.5 T shows no pelvic floor prolapse or organ prolapse. (b) Corresponding dynamic image shows grade 1 pelvic floor widening with an anteroposterior hiatal length of 6.1 cm (H line) and pelvic floor descent of 2.0 cm (M line), a grade 2 cystocele with the bladder 3.1 cm below the H line, and a urethrocele with the urethra below the H line. Note the hypermobility of the urethra, which assumes a horizontal position during strain. The air in the bladder was secondary to catheterization.




After assessing patients with MR imaging, our urogynecologists perform a novel transvaginal mesh repair to limit levator descent and hiatal widening (,31). This surgical mesh reconstruction recreates the normal pelvic floor and vaginal support unit lost by prior disruption of sacrouterine-cardinal ligaments and the prerectal fascia. This repair also involves suspension of the vaginal vault from the lower sacrum to recreate the normal banana-shaped vaginal axis for vaginal vault support. Patients may also undergo cystocele repair with or without a sling procedure for incontinence; enterocele repair; or rectocele repair, depending on the results of preoperative imaging triage. With this approach, symptom improvement has been reported in 98% of patients and recurrences have decreased to 4% (,15). However, these are preliminary findings, and longer patient follow-up is needed to assess the long-term efficacy of this novel surgical approach.
Pelvic Organ Prolapse
Anterior Compartment: Cystoceles.—
A cystocele is prolapse of the bladder through its respective hiatus in the anterior compartment. In most cases, women present with some degree of simple stress urinary incontinence, which is usually diagnosed clinically and may not require MR imaging. In more severe cases, the posterior wall of the bladder descends disproportionally more than the anterior wall, resulting in a downward and clockwise bladder rotation as well as urethral prolapse, as the urethra rotates into a horizontal configuration known as urethral hypermobility (,Figs 7,–,,,,,,,,11).
Figure 8a. Hydronephrosis, endometrial hyperplasia or malignancy, and severe cystourethrocele in a 77-year-old woman with mixed incontinence. (a) Sagittal rapid half-Fourier T2-weighted image shows severe hydronephrosis. (b) Sagittal MR image obtained in a different plane shows a 29.24-mm mass with heterogeneous high signal intensity, a finding suspicious for endometrial malignancy. There is also a 0.55-mm hypointense lesion, which is most consistent with a benign fibroid. (c) Straining midsagittal MR image shows an obvious severe cystourethrocele. Figure 8b. Hydronephrosis, endometrial hyperplasia or malignancy, and severe cystourethrocele in a 77-year-old woman with mixed incontinence. (a) Sagittal rapid half-Fourier T2-weighted image shows severe hydronephrosis. (b) Sagittal MR image obtained in a different plane shows a 29.24-mm mass with heterogeneous high signal intensity, a finding suspicious for endometrial malignancy. There is also a 0.55-mm hypointense lesion, which is most consistent with a benign fibroid. (c) Straining midsagittal MR image shows an obvious severe cystourethrocele. Figure 8c. Hydronephrosis, endometrial hyperplasia or malignancy, and severe cystourethrocele in a 77-year-old woman with mixed incontinence. (a) Sagittal rapid half-Fourier T2-weighted image shows severe hydronephrosis. (b) Sagittal MR image obtained in a different plane shows a 29.24-mm mass with heterogeneous high signal intensity, a finding suspicious for endometrial malignancy. There is also a 0.55-mm hypointense lesion, which is most consistent with a benign fibroid. (c) Straining midsagittal MR image shows an obvious severe cystourethrocele. Figure 9a. Cystourethrocele with urethral hypermobility, vaginal vault prolapse, and minimal pelvic floor relaxation in a 73-year-old woman whose chief symptom was vaginal vault prolapse. (a) Resting midsagittal MR image shows a mild cystocele and some urethral hypermobility even without strain. The vaginal apex (*) is above the H line. (b) Corresponding straining image shows a severe cystourethrocele with urethral hypermobility and vaginal vault prolapse, with the bladder and the vaginal vault (*) below the hiatus. There is also minimal pelvic floor relaxation, with the hiatus (H line) and the descent (M line) measuring 7.1 cm and 3.0 cm, respectively. Figure 9b. Cystourethrocele with urethral hypermobility, vaginal vault prolapse, and minimal pelvic floor relaxation in a 73-year-old woman whose chief symptom was vaginal vault prolapse. (a) Resting midsagittal MR image shows a mild cystocele and some urethral hypermobility even without strain. The vaginal apex (*) is above the H line. (b) Corresponding straining image shows a severe cystourethrocele with urethral hypermobility and vaginal vault prolapse, with the bladder and the vaginal vault (*) below the hiatus. There is also minimal pelvic floor relaxation, with the hiatus (H line) and the descent (M line) measuring 7.1 cm and 3.0 cm, respectively. Figure 10a. Mild pelvic floor prolapse and widening, uterine prolapse, cystourethrocele, minimal rectocele, and Bartholin cyst in a 59-year-old woman with a vaginal bulge and incontinence. (a) Straining midsagittal single-shot fast SE image shows an anteroposterior hiatal dimension of 7.9 cm and pelvic floor descent of 2.1 cm, findings consistent with grade 1 or mild pelvic floor descent and widening. The urethra is 1.6 cm below the H line and the bladder is 4.3 cm below the H line, findings consistent with a grade 3 cystourethrocele. The vaginal apex is below the hiatus, a finding consistent with uterine prolapse. A minimal rectocele is also present. (b) Axial MR image shows a 1.9 × 1.6-cm structure in the distal left paravaginal region, a finding that represents a Bartholin cyst. The imaging findings were confirmed at surgery; the patient underwent vaginal hysterectomy, transvaginal paravaginal repair of the grade 4 cystocele with soft polypropylene mesh, sling bladder neck suspension, and rectocele repair with soft polypropylene mesh. Figure 10b. Mild pelvic floor prolapse and widening, uterine prolapse, cystourethrocele, minimal rectocele, and Bartholin cyst in a 59-year-old woman with a vaginal bulge and incontinence. (a) Straining midsagittal single-shot fast SE image shows an anteroposterior hiatal dimension of 7.9 cm and pelvic floor descent of 2.1 cm, findings consistent with grade 1 or mild pelvic floor descent and widening. The urethra is 1.6 cm below the H line and the bladder is 4.3 cm below the H line, findings consistent with a grade 3 cystourethrocele. The vaginal apex is below the hiatus, a finding consistent with uterine prolapse. A minimal rectocele is also present. (b) Axial MR image shows a 1.9 × 1.6-cm structure in the distal left paravaginal region, a finding that represents a Bartholin cyst. The imaging findings were confirmed at surgery; the patient underwent vaginal hysterectomy, transvaginal paravaginal repair of the grade 4 cystocele with soft polypropylene mesh, sling bladder neck suspension, and rectocele repair with soft polypropylene mesh. Figure 11. Pelvic floor widening and descent with multicompartment organ prolapse in a 68-year-old multiparous woman with a vaginal bulge, urgency, and incontinence who had undergone vaginal hysterectomy and anterior repair of vaginal vault prolapse. Straining midsagittal MR image shows severe grade 3 hiatal enlargement, with a 10.6-cm H line, and moderate pelvic floor descent, with a 4.6-cm M line. There is a grade 4 severe cystourethrocele or procidentia, with the cystourethral junction 5 cm below the hiatal line. Although the vaginal apex is not clearly seen, it is below the hiatal line. An enterocele and a rectocele are also present, with small bowel contents and the rectum within the peritoneal sac well below the H line and the anorectal junction 3.9 cm below the H line. The patient underwent transvaginal paravaginal repair of the grade 4 cystocele with soft polypropylene mesh, as well as repair of the rectocele and the perineum.







In these high-grade cystourethroceles, symptoms of stress incontinence may paradoxically be masked by kinking at the bladder neck resulting from the transverse orientation of the urethra and may be unmasked only if the bladder prolapse is repaired (,14). High-grade cystourethroceles are otherwise symptomatic, since prolapse of more than one anatomic compartment and higher degrees of pelvic floor relaxation are usually present. If the bladder prolapse is severe enough, the muscular pelvic floor may entrap the ureters and create ureteral obstruction or hydronephrosis, which can be easily assessed at MR imaging (,Fig 8,,).
In the preoperative work-up of cystoceles, we perform a video urodynamic pressure evaluation in addition to MR imaging. The voiding video urodynamic study uncovers urge incontinence from detrusor instability, which is usually treated medically, and stress urinary incontinence masked by urethral hypermobility. MR imaging findings facilitate surgical planning by assessing for the presence of multicompartment organ prolapse, pelvic floor relaxation, and incidental pathologic conditions (,Figs 8,,–,,,,,,,,,,14). This approach has excellent clinical correlation and supplants the need for a variety of invasive fluoroscopic examinations such as cystocolpoproctography (,32–,39).
Figure 12. Important incidental finding in the endometrium in a 68-year-old woman who presented with incontinence and frequency. Sagittal MR image shows an area of cystic endometrial thickening (arrow), which is highly suggestive of malignancy or endometrial hyperplasia. Further gynecologic studies were recommended. Figure 13a. Mild pelvic floor widening and descent, multicompartment prolapse, and incidental findings of a subserosal fibroid and an ovarian cyst in a 51-year-old multiparous woman who presented with lower pelvic pain and pressure, a vaginal bulge, urinary frequency, and dyspareunia. (a) Resting midsagittal MR image shows no organ prolapse. There is a 3.3-cm subserosal fibroid at the uterine fundus (*). (b) Straining midsagittal MR image shows a hiatal anteroposterior dimension of 6.8 cm (H line) and pelvic floor descent of 3.8 cm (M line), findings consistent with grade 1 pelvic floor widening and descent. In addition, there is a mild cystourethrocele, with the urethra and bladder located 0.9 cm and 1.2 cm below the hiatal line, respectively. A moderate peritoneocele is seen, with the peritoneum 2.9 cm below the H line. There is a mild rectocele, with the rectum 1 cm below the hiatus. The subserosal fibroid is also seen (*). (c) Axial T2-weighted image shows a hyperintense left adnexal lesion measuring 4.4 cm (arrow), which most likely represents an ovarian or paraovarian cyst. Figure 13b. Mild pelvic floor widening and descent, multicompartment prolapse, and incidental findings of a subserosal fibroid and an ovarian cyst in a 51-year-old multiparous woman who presented with lower pelvic pain and pressure, a vaginal bulge, urinary frequency, and dyspareunia. (a) Resting midsagittal MR image shows no organ prolapse. There is a 3.3-cm subserosal fibroid at the uterine fundus (*). (b) Straining midsagittal MR image shows a hiatal anteroposterior dimension of 6.8 cm (H line) and pelvic floor descent of 3.8 cm (M line), findings consistent with grade 1 pelvic floor widening and descent. In addition, there is a mild cystourethrocele, with the urethra and bladder located 0.9 cm and 1.2 cm below the hiatal line, respectively. A moderate peritoneocele is seen, with the peritoneum 2.9 cm below the H line. There is a mild rectocele, with the rectum 1 cm below the hiatus. The subserosal fibroid is also seen (*). (c) Axial T2-weighted image shows a hyperintense left adnexal lesion measuring 4.4 cm (arrow), which most likely represents an ovarian or paraovarian cyst. Figure 13c. Mild pelvic floor widening and descent, multicompartment prolapse, and incidental findings of a subserosal fibroid and an ovarian cyst in a 51-year-old multiparous woman who presented with lower pelvic pain and pressure, a vaginal bulge, urinary frequency, and dyspareunia. (a) Resting midsagittal MR image shows no organ prolapse. There is a 3.3-cm subserosal fibroid at the uterine fundus (*). (b) Straining midsagittal MR image shows a hiatal anteroposterior dimension of 6.8 cm (H line) and pelvic floor descent of 3.8 cm (M line), findings consistent with grade 1 pelvic floor widening and descent. In addition, there is a mild cystourethrocele, with the urethra and bladder located 0.9 cm and 1.2 cm below the hiatal line, respectively. A moderate peritoneocele is seen, with the peritoneum 2.9 cm below the H line. There is a mild rectocele, with the rectum 1 cm below the hiatus. The subserosal fibroid is also seen (*). (c) Axial T2-weighted image shows a hyperintense left adnexal lesion measuring 4.4 cm (arrow), which most likely represents an ovarian or paraovarian cyst. Figure 14. Moderate hiatal enlargement with mild pelvic floor descent accompanied by pancompartment failure including severe enterocele, mild cystocele, and minimal rectocele in a 74-year-old woman with a vaginal bulge, urinary urgency, and incontinence. Straining midsagittal MR image shows grade 2 pelvic hiatal enlargement, with an anteroposterior dimension of 8.2 cm (H line), and grade 1 pelvic floor descent (M line), with the puborectal line 2.9 cm below the PCL. There is a grade 3 enterocele, with the small bowel loops 5.7 cm below the H line, and a mild cystocele, with the bladder 0.5 cm below the hiatus. A minimal rectocele was also seen on dynamic straining images. These findings were confirmed at surgery; the patient underwent sacrospinous transvaginal vault suspension and enterocele and rectocele repair with soft polypropylene mesh.




On the midsagittal T2-weighted image, prolapses of the bladder, urethra, vagina, and rectum are each assessed and graded separately along with any associated pelvic floor relaxation to help determine their relative clinical relevance (,Table 2). The relationship of the bladder neck to the vertical urethra during rest and any abnormal clockwise rotation and kinking to a transverse position (urethral hypermobility) may be easily assessed. Separate assessments of urethral and bladder prolapse are important because high-grade cystoceles can hide underlying urethral hypermobility, masking stress urinary incontinence.
Surgeons characterize cystoceles by the anatomic defect involved: central, lateral (paravaginal), or their combination (,21). Loss of bladder support is thought to result from defects in the pubocervical fascia, which attaches to the arcus tendineus fascia pelvis anterolaterally and to the cervix posteriorly. The most commonly seen lateral defect results from fascial weakening in its attachment to the arcus tendineus fascia pelvis. Central defects result from fascial disruptions in the midline (,14).
Treatment ranges from nonsurgical interventions such as a pessary in mild cases to either abdominal or vaginal approach surgery, with sutures and mesh used to approximate the original fascial support with or without an incontinence surgery (sling procedure). Anterior colporrhaphy is performed for central defects with plication of the vesicopelvic fascia and bladder neck to close the central defect (,7). A paravaginal repair is advocated with either retropubic or vaginal access in lateral defects. In cases of urethral hypermobility, a suburethral sling procedure is performed to provide midurethral support and avoid subsequent stress incontinence (,40). Additional procedures may be necessary, depending on which concurrent abnormalities are present.
Uterine and Vaginal Vault Prolapse.—
Uterine and vaginal vault prolapses are defects of the middle compartment (,Figs 9,, ,10,). In severe cases of complete eversion, cervical and uterine prolapses are seen as a bulging mass outside the external genitalia as the vaginal walls form a sac containing the prolapsing organs. This makes diagnosis of any concominant pelvic organ prolapse very difficult (,25). Grade 4 uterine prolapse often manifests as insidious progressive ureteral obstruction (,21). Uterine procidentia is diagnosed when the uterus prolapses completely outside of the hiatus. The laxity of the uterosacral ligaments allows the cervix to move anteriorly, resulting in progressive uterine retroversion and subsequent prolapse (,41).
Because vaginal vault prolapse is usually associated with prolapse of other organs, comprehensive assessment of the entire pelvis with MR imaging is particularly important (,7). As explained earlier, the banana-shaped vaginal axis created by elevation of the distal vagina by the ligaments and muscles is often lost in cases of prolapse. The vagina is then pathologically displaced inferiorly on dynamic MR images, and its distal portion moves anteriorly. Patients often develop concurrent enteroceles as the small bowel descends in the potential space of the cul-de-sac, best demonstrated on straining dynamic midsagittal MR images (,Fig 11).
Uterine prolapse is usually treated with hysterectomy, whereas some milder cases of vaginal vault prolapse may be managed with pelvic floor muscle exercises and pessaries. The uterine size and concomitant uterine or ovarian pathologic conditions (especially malignancy) are very important in triaging patients to vaginal or abdominal hysterectomy. MR imaging easily demonstrates any concurrent urethral, uterine, or ovarian disorders (,Figs 8,,, ,10,, ,12, ,13,,). Vaginal repairs are usually accomplished by reattaching the vaginal apex to the sacrospinous ligament, iliococcygeus muscle, or uterosacral ligaments (,7). In posterior culdoplasty, the posterior vaginal fornix is suspended from the uterosacral ligaments, which are brought together in the midline to obliterate the cul-de-sac. In abdominal sacrocolpopexy, the vaginal apex is fixed to the sacrum with sutures or mesh (,31).
Enteroceles, Peritoneoceles, and Sigmoidoceles.—
In the posterior compartment, small bowel and peritoneal mesenteric structures can bulge and protrude through the rectovaginal space (cul-de-sac) and posterior perineum, resulting in enteroceles, sigmoidoceles, and peritoneoceles or mesenteroceles, depending on their contents (,30). Compared with all other forms of organ prolapse, these herniation defects present the biggest diagnostic challenge at physical examination, especially when multiple organs are involved. Dynamic MR imaging is ideally suited to preoperative characterization of these bulges (,Figs 11, ,13,,–,,15,).
Figure 15a. Peritoneocele and pelvic floor enlargement with descent in a 60-year-old woman with a history of hysterectomy and cystocele repair who presented with a worsening vaginal bulge, frequency, and nocturia. (a) Resting midsagittal MR image shows no organ prolapse or pelvic floor prolapse. There is no uterus, a finding consistent with the surgical history. (b) Straining MR image shows mild pelvic hiatal enlargement of 7.4 cm (H line) and moderate pelvic floor descent of 5 cm (M line). There is a mild cystocele, with the bladder 0.5 cm below the puborectal line. A severe peritoneocele containing fat is clearly demonstrated. These findings were confirmed at surgery; the patient underwent mesh repair of the vaginal vault prolapse with modified sacrospinal fixation, rectocele repair, repair of the large enterocele (omentocele), and levator plate repair. The air in the bladder was secondary to catheterization. Figure 15b. Peritoneocele and pelvic floor enlargement with descent in a 60-year-old woman with a history of hysterectomy and cystocele repair who presented with a worsening vaginal bulge, frequency, and nocturia. (a) Resting midsagittal MR image shows no organ prolapse or pelvic floor prolapse. There is no uterus, a finding consistent with the surgical history. (b) Straining MR image shows mild pelvic hiatal enlargement of 7.4 cm (H line) and moderate pelvic floor descent of 5 cm (M line). There is a mild cystocele, with the bladder 0.5 cm below the puborectal line. A severe peritoneocele containing fat is clearly demonstrated. These findings were confirmed at surgery; the patient underwent mesh repair of the vaginal vault prolapse with modified sacrospinal fixation, rectocele repair, repair of the large enterocele (omentocele), and levator plate repair. The air in the bladder was secondary to catheterization.

Enteroceles may be simple or complex (,21), differentiation of which can be achieved with MR imaging. Simple enteroceles have no associated vaginal vault prolapse. In contrast, complex enteroceles occur with other forms of anterior or posterior vaginal vault prolapse. Accordingly, surgical repair may involve an enterocele repair with or without vaginal vault prolapse repair. MR imaging is useful in differentiating enteroceles and high rectoceles, enabling more efficient surgical planning with safer planes for intraoperative dissection (,21,,35,,42). MR imaging has been shown to be superior to dynamic cystocolpoproctography, which fails to demonstrate up to 20% of enteroceles (,35,,42). Surgical repairs of enteroceles aim to obliterate the cul-de-sac (,14).
Rectoceles.—
Rectoceles are also located in the posterior compartment and involve either prolapse of the rectum through the hiatus or its bulging into the posterior vaginal wall (,Figs 16,–,,,,,,,,,,,21). They result from defects in the prerectal and pararectal fascia as well as the rectovaginal septum (,41). It is important to correlate imaging findings with symptoms because mild rectoceles can be asymptomatic. Rectoceles are usually accompanied by other pelvic floor disorders, resulting in organ competition for space in the vagina. Therefore, rectoceles are often missed at physical examination (,43).
Figure 16a. Grade 2 pelvic floor widening and descent with a moderate rectocele in a 78-year-old multiparous woman with difficulty urinating. (a) Resting rapid half-Fourier T2-weighted image obtained at 1.5 T shows no uterus, a finding consistent with the patient history. (b) Straining image shows grade 2 enlargement of the puborectal hiatus, with an H line of 8 cm, and grade 2 pelvic floor descent, with an M line of 4.5 cm; a moderate rectocele, with the rectum 3.9 cm below the puborectal hiatus; and a minimal cystocele. Figure 16b. Grade 2 pelvic floor widening and descent with a moderate rectocele in a 78-year-old multiparous woman with difficulty urinating. (a) Resting rapid half-Fourier T2-weighted image obtained at 1.5 T shows no uterus, a finding consistent with the patient history. (b) Straining image shows grade 2 enlargement of the puborectal hiatus, with an H line of 8 cm, and grade 2 pelvic floor descent, with an M line of 4.5 cm; a moderate rectocele, with the rectum 3.9 cm below the puborectal hiatus; and a minimal cystocele. Figure 17a. Anterior rectoceles. (a, b) Resting (a) and straining (b) midsagittal MR images of a 41-year-old woman who had undergone cystectomy show an anterior rectocele, with the rectal wall pushing on the posterior vaginal wall (arrow in b). (c, d) Resting (c) and straining (d) midsagittal MR images of a 55-year-old woman show a grade 2 anterior rectocele (arrow in d) and a grade 1 cystourethrocele with urethral hypermobility. Figure 17b. Anterior rectoceles. (a, b) Resting (a) and straining (b) midsagittal MR images of a 41-year-old woman who had undergone cystectomy show an anterior rectocele, with the rectal wall pushing on the posterior vaginal wall (arrow in b). (c, d) Resting (c) and straining (d) midsagittal MR images of a 55-year-old woman show a grade 2 anterior rectocele (arrow in d) and a grade 1 cystourethrocele with urethral hypermobility. Figure 17c. Anterior rectoceles. (a, b) Resting (a) and straining (b) midsagittal MR images of a 41-year-old woman who had undergone cystectomy show an anterior rectocele, with the rectal wall pushing on the posterior vaginal wall (arrow in b). (c, d) Resting (c) and straining (d) midsagittal MR images of a 55-year-old woman show a grade 2 anterior rectocele (arrow in d) and a grade 1 cystourethrocele with urethral hypermobility. Figure 17d. Anterior rectoceles. (a, b) Resting (a) and straining (b) midsagittal MR images of a 41-year-old woman who had undergone cystectomy show an anterior rectocele, with the rectal wall pushing on the posterior vaginal wall (arrow in b). (c, d) Resting (c) and straining (d) midsagittal MR images of a 55-year-old woman show a grade 2 anterior rectocele (arrow in d) and a grade 1 cystourethrocele with urethral hypermobility. Figure 18a. Large cystourethrocele and rectocele without anterior bulging in a 53-year-old woman with vaginal prolapse. (a) Resting MR image shows a baseline rectocele (arrow), with the rectum below the hiatus (white line). (b) Straining midsagittal MR image shows a severe cystourethrocele and the rectocele (arrow), which does not demonstrate anterior bulging. Contrast this appearance with those of the anterior rectoceles in ,Figure 17,,,. Figure 18b. Large cystourethrocele and rectocele without anterior bulging in a 53-year-old woman with vaginal prolapse. (a) Resting MR image shows a baseline rectocele (arrow), with the rectum below the hiatus (white line). (b) Straining midsagittal MR image shows a severe cystourethrocele and the rectocele (arrow), which does not demonstrate anterior bulging. Contrast this appearance with those of the anterior rectoceles in ,Figure 17,,,. Figure 19a. Small rectocele and grade 1 pelvic floor descent in a 79-year-old woman with a long-standing history of dark brown vaginal discharge likely secondary to adenomyosis and fibroids who presented with urinary incontinence. (a) Resting midsagittal T2-weighted single-shot fast SE image obtained at 1.5 T shows a hypointense mass measuring 4.3 × 5.2 cm in the uterus, a finding highly suggestive of a fibroid. There are also numerous small cystic structures in the myometrium with a poorly defined junctional zone, findings most consistent with adenomyosis. There is no pelvic organ or pelvic floor prolapse. (b) Corresponding straining image shows a normal anteroposterior hiatal size of 5.8 cm (H line) and grade 1 pelvic floor descent of 3 cm (M line). The only organ prolapsing through the hiatus is the rectum, which extends 1.5 cm below the H line, a finding consistent with a grade 1 rectocele. Figure 19b. Small rectocele and grade 1 pelvic floor descent in a 79-year-old woman with a long-standing history of dark brown vaginal discharge likely secondary to adenomyosis and fibroids who presented with urinary incontinence. (a) Resting midsagittal T2-weighted single-shot fast SE image obtained at 1.5 T shows a hypointense mass measuring 4.3 × 5.2 cm in the uterus, a finding highly suggestive of a fibroid. There are also numerous small cystic structures in the myometrium with a poorly defined junctional zone, findings most consistent with adenomyosis. There is no pelvic organ or pelvic floor prolapse. (b) Corresponding straining image shows a normal anteroposterior hiatal size of 5.8 cm (H line) and grade 1 pelvic floor descent of 3 cm (M line). The only organ prolapsing through the hiatus is the rectum, which extends 1.5 cm below the H line, a finding consistent with a grade 1 rectocele. Figure 20a. Pelvic floor laxity with prolapse, grade 2 enterocele, and grade 1 rectocele in a 74-year-old multiparous woman with a history of hysterectomy, two pelvic prolapse surgeries, and cadaveric sling urethropexy who presented with worsening incontinence. (a) Resting midsagittal single-shot fast SE image obtained at 1.5 T shows no prolapse. (b) Dynamic midsagittal MR image shows an anteroposterior hiatal dimension of 8.8 cm (H line) and pelvic floor descent of 8 cm (M line), findings consistent with grade 2 hiatal enlargement and grade 3 pelvic floor prolapse. There is a grade 2 enterocele, which extends 2.1 cm below the H line, and a grade 1 rectocele. These findings were confirmed at surgery; the patient underwent a sling procedure with soft polypropylene mesh, as well as rectocele and perineal repair. Figure 20b. Pelvic floor laxity with prolapse, grade 2 enterocele, and grade 1 rectocele in a 74-year-old multiparous woman with a history of hysterectomy, two pelvic prolapse surgeries, and cadaveric sling urethropexy who presented with worsening incontinence. (a) Resting midsagittal single-shot fast SE image obtained at 1.5 T shows no prolapse. (b) Dynamic midsagittal MR image shows an anteroposterior hiatal dimension of 8.8 cm (H line) and pelvic floor descent of 8 cm (M line), findings consistent with grade 2 hiatal enlargement and grade 3 pelvic floor prolapse. There is a grade 2 enterocele, which extends 2.1 cm below the H line, and a grade 1 rectocele. These findings were confirmed at surgery; the patient underwent a sling procedure with soft polypropylene mesh, as well as rectocele and perineal repair. Figure 21. Grade 1 hiatal enlargement and pelvic floor descent with grade 1 cystourethrocele and rectocele in a 35-year-old multiparous woman with vaginal prolapse and urinary frequency. Straining midsagittal rapid half-Fourier T2-weighted image obtained at 1.5 T shows mild pelvic hiatal enlargement and pelvic floor descent, with an H line of 8.0 cm and an M line of 2.7 cm. There is a grade 1 cystourethrocele, with the bladder and urethra 0.9 cm below the H line, and a grade 1 rectocele, with the rectum 1.3 cm below the hiatus.












Owing to these shortcomings, imaging studies including MR imaging and evacuation proctography are an integral part of the preoperative patient evaluation and are routinely performed as complementary examinations. We have recognized that with dynamic MR imaging alone, subtle or low-grade rectoceles may be missed, whereas with fluoroscopic defecography the pelvic floor cannot be reliably evaluated (,42). Hence, both examinations are performed to evaluate the anatomy and function in a complementary fashion. A major disadvantage of fluoroscopic defecography is the high radiation exposure, which poses a problem in women of childbearing age. Some investigators perform MR defecography, which has not been widely studied. A major difficulty with this technique has been demonstrating complete emptying of the sac.
As with other forms of pelvic organ prolapse, the diagnosis of a rectocele at dynamic MR imaging is usually established on the midsagittal dynamic T2-weighted image acquired during maximal strain as either rectal protrusion below the H line or rectal bulging anterior to the transverse perineus muscle. At evacuation proctography, a variety of additional disorders such as rectal mucosal prolapse, solitary rectal ulcers, and incompletely emptying anterior rectoceles may also be more confidently evaluated. Surgical treatment of patients may involve rectovaginal fascia repair or posterior fixation of the rectum with possible sigmoid or rectal resection (,14).
Current Work and Future Directions
Most recent work in the area of female pelvic floor disorders has focused on use of dynamic and three-dimensional MR imaging techniques by various researchers. MR imaging is being used to define the problem radiologically in hopes of gaining a better understanding of mechanisms leading to pelvic floor and organ prolapse (,24,,44–,48). Hoyte et al (,49) showed statistically significant differences in levator muscle thickness and percentages of gaps among women with no symptoms, those with urodynamic stress incontinence, and those with pelvic organ prolapse. Recently, it was possible to identify and describe five subdivisions of the levator ani muscle group by using MR imaging in normal volunteers. This information might be used in making correlations between specific injuries and the resulting pathologic condition (,50).
Most current MR imaging protocols image patients in a supine position. Thus, it is possible that these images are unable to demonstrate the full extent of pelvic floor or organ prolapse, which is physiologically most pronounced when the patient is standing, prompting current efforts in imaging sitting patients with open-magnet MR units (,4,,51,,52). MR imaging will likely become increasingly used in the management of surgical pelvic floor disorders, especially in cases of failed surgical repairs.
Conclusions
Pelvic floor dysfunction is an important medical and social problem in the female population. A minority of patients with these disorders need surgery. Unfortunately, pelvic floor disorders are not completely understood in the medical community. MR imaging is an important diagnostic resource in triaging patients to surgery and in helping surgeons plan specific repairs. It provides clinicians with an objective assessment of the problem in a single examination. In addition, MR imaging has tremendous potential to be used as a research tool in trying to understand the pathophysiology of these complex disorders. It is hoped that better insight into the mechanisms of this disease will lead to more efficient treatment planning.
![]() |
![]() |
References
- 1
, Entman SS. Pelvic organ prolapse and stress urinary incontinence. Med Clin North Am1995;79:1473–1479. Crossref, Medline, Google ScholarRush CB - 2
, Bent AD. Genital prolapse with and without urinary incontinence. J Reprod Med1990; 35:792–798. Medline, Google ScholarHarris TA - 3
, Smith VJ, Bergstrom JO, Collins JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol1997;89:501–506. Crossref, Medline, Google ScholarOlsen AL - 4
, Versi E, Mulkern RV, Lerner MH, Griffiths DJ, Jolesz FA. MR imaging of the female pelvic floor in the supine and upright positions. J Magn Reson Imaging1996;6:961–963. Crossref, Medline, Google ScholarFielding JR - 5
. Classification and evaluation of prolapse. Best Pract Res Clin Obstet Gynaecol2005; 19:895–911. Crossref, Medline, Google ScholarMouritsen L - 6
. Practical MR imaging of female pelvic floor weakness. RadioGraphics2002;22:295–304. Link, Google ScholarFielding JR - 7
, Haab F. What’s new in prolapse surgery? Curr Opin Urol2003;13:315–323. Crossref, Medline, Google ScholarDeval B - 8
, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol1996;175:10–17. Crossref, Medline, Google ScholarBump RC - 9
. The hidden epidemic of pelvic floor dysfunction: achievable goals for improved prevention and treatment. Am J Obstet Gynecol2005; 192:1488–1495. Crossref, Medline, Google ScholarDeLancey JO - 10
, Weber AM, Meyn L. Procedures for pelvic organ prolapse in the United States, 1979–1997. Am J Obstet Gynecol2003;188:108–115. Crossref, Medline, Google ScholarBoyles SH - 11
, Weber AM, Meyn L. Procedures for urinary incontinence in the United States, 1979– 1997. Am J Obstet Gynecol2003;189:70–75. Crossref, Medline, Google ScholarBoyles SH - 12
. Pelvic floor disorders: the role of fascia and ligaments. Clin Obstet Gynecol1993;36: 926–938. Crossref, Medline, Google ScholarNorton PA - 13
. Normal pelvic floor anatomy. Obstet Gynecol Clin North Am1998;25:683–705. Crossref, Medline, Google ScholarStrohbehn K - 14
, Kaufman HS, Cundiff GW, Genadry R, Bluemke DA, Fishman EK. Dynamic MR imaging of pelvic organ prolapse: spectrum of abnormalities. RadioGraphics2000;20:1567–1582. Link, Google ScholarPannu HK - 15
, Deng DY, Rodriguez LV, Raz S. Repair of vaginal vault prolapse and pelvic floor relaxation using polypropylene mesh. Neurourol Urodyn2005;24:654–658. Crossref, Medline, Google ScholarRutman MP - 16
, Hotzinger H, Fritsch H. Tomographical anatomy of the pelvis, pelvic floor, and related structures. Clin Anat1997;10:223–230. Crossref, Medline, Google ScholarFrohlich B - 17
, Sawhney R, DeLancey JO. Levator ani muscle anatomy evaluated by origin-insertion pairs. Obstet Gynecol2004;104:168–173. Crossref, Medline, Google ScholarKearney R - 18
, Schiemann T, Taupitz M, Kooijman H, Hamm B, Nicolas V. Sectional depiction of the pelvic floor by CT, MR imaging and sheet plastination: computer-aided correlation and 3D model. Eur Radiol2001;11:659–664. Crossref, Medline, Google ScholarBeyersdorff D - 19
, Stoker J, Zwamborn AW, Entrius KA, Calame JJ, Lameris JS. Female pelvic floor: endovaginal MR imaging of normal anatomy. Radiology1998;206:777–783. Link, Google ScholarTan IL - 20
, Halligan S, Bartram CI. Pelvic floor imaging. Radiology2001;218:621–641. Link, Google ScholarStoker J - 21
, Raz S. Diagnostic imaging of pelvic floor dysfunction. Curr Opin Urol2001;11: 423–428. Crossref, Medline, Google ScholarRodriguez LV - 22
, Kelvin FM, Fitzgerald K, Hale DS, Benson JT. Association of compartment defects in pelvic floor dysfunction. AJR Am J Roentgenol1999;172:439–444. Crossref, Medline, Google ScholarMaglinte DD - 23
, Aubard Y, Berkane V, Camezind-Vidal MA, Mares P, Rouanet JP. Magnetic resonance imaging of the pelvic floor. Abdom Imaging2003; 28:217–225. Crossref, Medline, Google ScholarMaubon A - 24
, Schierlitz L, Zou K, Flesh G, Fielding JR. Two and 3-dimensional MRI comparison of levator ani structure, volume, and integrity in women with stress incontinence and prolapse. Am J Obstet Gynecol2001;185:11–19. Crossref, Medline, Google ScholarHoyte L - 25
, Vasavada SP, Barbaric ZL, Gousse AE, Raz S. Grading pelvic floor prolapse and pelvic floor relaxation using dynamic magnetic resonance imaging. Urology1999;54:454–457. Crossref, Medline, Google ScholarComiter CV - 26
. MR imaging of pelvic floor relaxation. Radiol Clin North Am2003;41:747–756. Crossref, Medline, Google ScholarFielding JR - 27
, Anthuber C, Baron A, Kohz P, Reiser M. Dynamic MR colpocystorectography assessing pelvic floor descent. Eur Radiol1997;7: 1309–1317. Crossref, Medline, Google ScholarLienemann A - 28
, Norton PA. Urogynecology and pelvic floor dysfunction: epidemiology and natural history of pelvic floor dysfunction. Obstet Gynecol Clin North Am1998;25:723–746. Crossref, Medline, Google ScholarBump RC - 29
, Martel-Boncoeur MP, Juhan V, et al. Static and dynamic magnetic resonance imaging of the pelvic floor [in French]. J Radiol2000;81(12 suppl):1875–1886. Medline, Google ScholarMaubon A - 30
, Marumoto AK, Raz S. Magnetic resonance imaging of the perineum and pelvic floor. Top Magn Reson Imaging2001;12:83–92. Crossref, Medline, Google ScholarBarbaric ZL - 31
, Raz S. Repair of vaginal vault prolapse and pelvic floor relaxation using polypropylene mesh. Curr Opin Obstet Gynecol2006;18: 555–559. Crossref, Medline, Google ScholarMourtzinos A - 32
, DeGreforio G, Allman KH, Kundt G, Dohnicht S. Comparison of cystourethrography and dynamic MRI in bladder neck descent. J Comput Assist Tomogr2000;24:382–388. Crossref, Medline, Google ScholarGufler H - 33
, Sachlarides TJ, Dominguez JM, Brubaker LT, Smith CS. Four-contrast defecography: pelvic “flooroscopy.” Dis Colon Rectum1995;38:695–699. Crossref, Medline, Google ScholarAltringer WE - 34
, Lombard R, Jehaes C, et al. Colpocystodefecography. Dis Colon Rectum1993;36: 1015–1021. Crossref, Medline, Google ScholarHock D - 35
, Anthuber C, Baron A, Reuser M. Diagnosing enteroceles using dynamic magnetic resonance imaging. Dis Colon Rectum2000;43: 205–212. Crossref, Medline, Google ScholarLienemann A - 36
, Hale DS, Maglinte DD, Patten BJ, Benson JT. Female pelvic organ prolapse: diagnostic contribution of dynamic cystoproctography and comparison with physical examination. AJR Am J Roentgenol1999;173:31–37. Crossref, Medline, Google ScholarKelvin FM - 37
, Maglinte DD, Hale DS, Benson JT. Female pelvic organ prolapse: a comparison of triphasic dynamic MR imaging and triphasic fluoroscopic cystocolpoproctography. AJR Am J Roentgenol2000;174:81–84. Crossref, Medline, Google ScholarKelvin FM - 38
, Hamada A. Evaluation of pelvic descent disorders by dynamic contrast roentgenography. Dis Colon Rectum2000;43(10 suppl):S6– S11. Crossref, Medline, Google ScholarTakano M - 39
, Retzky S, Smith C, Saclarides T. Pelvic floor evaluation with dynamic fluoroscopy. Obstet Gynecol1993;82:863–868. Medline, Google ScholarBrubaker L - 40
, Maher CF. Mesh augmentation during pelvic-floor reconstructive surgery: risks and benefits. Curr Opin Obstet Gynecol2006;18:560–566. Crossref, Medline, Google ScholarBaessler K - 41
, Stothers L, Chopra A. Vaginal reconstructive surgery for incontinence and prolapse. In: Walsh PC, Retik AB, Vaughan ED, Wein AJ, eds. Campbell’s urology. 2nd ed. Philadelphia, Pa: Saunders, 1998; 1059–1094. Google ScholarRaz S - 42
, Barbaric ZL, Safir MH, Madjar S, Marumoto AK, Raz S. Dynamic half Fourier acquisition single shot turbo spin-echo magnetic resonance imaging for evaluating the female pelvis. J Urol2000;164:1606–1613. Crossref, Medline, Google ScholarGousse AE - 43
, Ropert A, Vilotte J, et al. How accurate is clinical examination in diagnosing and quantifying pelvirectal disorders? A prospective study in a group of 50 patients complaining of defecatory difficulties. Dis Colon Rectum1993;36:430–438. Crossref, Medline, Google ScholarSiproudhis L - 44
, Hsu Y, Ashton-Miller JA, DeLancey JO. Measurement of the pubic portion of the levator ani muscle in women with unilateral defects in 3-D models from MR images. Int J Gynaecol Obstet2006;92:234–241. Crossref, Medline, Google ScholarChen L - 45
, Summers A, Hussain HK, Guire KE, DeLancey JO. Levator plate angle in women with pelvic organ prolapse compared to women with normal support using dynamic MR imaging. Am J Obstet Gynecol2006;194:1427–1433. Crossref, Medline, Google ScholarHsu Y - 46
, Chen L, DeLancey JO, Ashton-Miller JA. Vaginal thickness, cross-sectional area, and perimeter in women with and those without prolapse. Obstet Gynecol2005;105:1012–1017. Crossref, Medline, Google ScholarHsu Y - 47
, Dumanli H, Schreyer AG, et al. MR-based three-dimensional modeling of the normal pelvic floor in women: quantification of muscle mass. AJR Am J Roentgenol2000;174:657–660. Crossref, Medline, Google ScholarFielding JR - 48
, Jakab M, Reid WM, Berger LA, Hoyte L. Three-dimensional magnetic resonance imaging assessment of levator ani morphologic features in different grades of prolapse. Am J Obstet Gynecol2003;188:910–915. Crossref, Medline, Google ScholarSingh K - 49
, Jakab M, Warfield SK, Shott S, Flesh G, Fielding JR. Levator ani thickness variations in symptomatic and asymptomatic women using magnetic resonance–based 3-dimensional color mapping. Am J Obstet Gynecol2004;191:856–861. Crossref, Medline, Google ScholarHoyte L - 50
, Hsu Y, Kearney R, Stein T, Umek WH, DeLancey JO. Appearance of the levator ani muscle subdivisions in magnetic resonance images. Obstet Gynecol2006;107:1064–1069. Crossref, Medline, Google ScholarMargulies RU - 51
, Griffiths DJ, Versi E, Mulkern RV, Lee ML, Jolesz RA. MR imaging of pelvic floor continence mechanisms in the supine and sitting positions. AJR Am J Roentgenol1998;171:1607–1610. Crossref, Medline, Google ScholarFielding JR - 52
, Hetzer FH, Roos JE, Treiber K, Marincek B, Hilfiker PR. Dynamic MR imaging of the pelvic floor performed with patient sitting in an open-magnet unit versus with patient supine in a. closed-magnet unit. Radiology2002;223:501–508. Link, Google ScholarBertschinger KM










