The endopelvic fascia is a connective tissue structure which comprises the most superior layer of support in pelvic floor. It attaches to the lateral pelvic side wall bilaterally at the arcus tendinous levator ani (ATLA), providing lateral support for pelvic organs (Fig. 13.1). Different components of the endopelvic fascia are named based on their anatomic locations (e.g., pubovaginal/pubocervical, rectovaginal, parametrium, and paracorpium fascia). A few identifiable components of the endopelvic fascia on MRI include the urethral ligaments, perineal body, cardinal, and uterosacral ligaments.³ Because the majority of endopelvic fascia is not directly visualized on imaging, the intact status of the fascia is inferred by normal appearance of the pelvic organs. Defects in various levels of the endopelvic fascia manifest as deformity of the vaginal and/or paravaginal structures on imaging.

The pelvic diaphragm is formed by levator ani muscles and coccygeus muscle. The levator ani is a striated muscle group, constantly in contraction, providing resting tone to the pelvic floor. Further contraction and relaxation of the levator ani muscles plays an important role in pelvic floor function. The levator ani comprises three components: puborectalis, pubococcygeus, and iliococcygeus muscles. Anteriorly, these muscles attach to the posterior aspect of the pubic symphysis and inferior pubic ramus, laterally extend along the tendinous attachment of obturator internus muscle (arcus tendinous), and posteriorly attach to the coccyx. The levator plate (anococcygeal ligament) is the tendinous insertion of the levator ani muscle to the coccyx and
is a near-horizontally oriented structure parallel to the posterior wall of the rectum (Fig. 13.2A). The levator ani muscles demarcate the urogenital hiatus, where the urogenital apparatus resides. The puborectalis muscle is the inferiormost muscle in the levator ani group and is a U-shaped sling that extends from the posterior aspect of the pubic symphysis wrapping around the anorectum and forms the margins of the pelvic floor hiatus (Fig. 13.2B). The levator hiatus has a mean area of 11 cm² at rest and 14 cm² on maximal strain in nulliparous women.⁶ The effect of hiatal enlargement on pelvic organ prolapse (POP) is discussed in the following sections. The puborectalis is the main muscle that maintains the anorectal angle (ARA) and therefore fecal continence. Puborectalis muscle relaxation results in widening of the ARA and pelvic floor hiatus. Furthermore, the puborectalis provides additional support to the periurethral ligaments. The pubococcygeus arises from the superior pubic ramus and attaches to the levator plate posteriorly. The iliococcygeus functionally supports the vagina and appears as a sheet-like muscle arising from the arcus tendineus laterally and inserts upon the levator plate posteriorly. It appears as a slightly convex structure superiorly on coronal images (Fig. 13.2C). The coccygeus muscle is the most posterior aspect of the pelvic diaphragm and extends from the ischial spine to the midline coccyx.

The urogenital diaphragm is a fibromuscular layer of connective tissue located below the pelvic diaphragm (see Fig. 13.2C). It has a triangular shape and extends horizontally from the ischium on either side to the perineal body in the midline with attachment anteriorly to the pubic symphysis. The perineal body comprises the thick connective tissue in the perineum centrally between the anal verge and anterior urogenital triangle and is an attachment site for multiple pelvic floor structures such as perineal musculature, external anal sphincter, and rectovaginal fascia. In the female pelvis, the urethra and vagina traverse the urogenital diaphragm.³,⁷,⁸

The etiology for POP is multifactorial. For years, it was speculated that vaginal deliveries contribute to development of POP. Levator ani avulsion which is defined as detachment of the muscle from the pubic insertion occurs in about 10% to 30% of vaginal deliveries. Several studies have demonstrated presence of macroscopic and microscopic abnormalities in the levator ani after childbirth. Replacement of muscle with fibrosis in women with stress urinary incontinence (SUI) and/or POP has been validated on postmortem studies.⁹,¹⁰,¹¹ More recent studies have shown that avulsion of the levator ani, resulting in muscle weakness and widening of the levator hiatus, predisposes to POP years later.¹²,¹³,¹⁴,¹⁵,¹⁶ In evaluation of 423 women with vaginal delivery, Handa et al.¹⁵ found 64 (15%) women experienced some degrees of levator ani avulsion. Compared to the control group, these women had larger dimensions of the levator hiatus; the odds of POP were increased by approximately 50% for every 5-cm² increase in the levator hiatus area. Furthermore, in this group, pelvic floor muscle strength (measured during voluntary pelvic muscle contraction with a perineometer) was inversely and significantly associated with the odds of prolapse. Imaging abnormalities of levator ani avulsion have been described on MRI and translabial three-dimensional ultrasound (US).¹³,¹⁷,¹⁸ The pertinent imaging findings on US are discussed in Chapter 14. Levator muscle defects on MRI may manifest as absence or detachment of the muscle or scarring (Fig. 13.3). DeLancey et al.¹³ showed injury to the levator ani frequently involves the pubovisceral portion of the muscle that arises from the inner surface of the pubic bone just lateral to the vagina but also involves the iliococcygeal muscle.¹⁹ In evaluation of a small cohort of primiparous women 6 to 7 weeks after normal vaginal delivery, Shi et al.²⁰ demonstrated that majority
of levator ani avulsions occurred at the origin. The tears of the pubococcygeus were located at the pubic origin. The tears of the iliococcygeus were located at or near the fascia of the obturator internus. All tears of the pubococcygeus were associated with focal pubic bone marrow edema. However, in a few cases, bone marrow edema was not accompanied by associated tears. Both MRI and US can provide valuable information about puborectalis muscle tears; however, the origin and insertion of the iliococcygeus are better evaluated on MRI. Therefore, in cases of suspected severe injury to the puborectalis, MRI may be preferred to evaluate any associated injury to the iliococcygeus muscle.²¹

Patient preparation for MR defecography starts with patient education. Prior to the examination, the referring physicians should explain the procedure, discuss the importance of adequate patient effort, and alleviate any concerns that patients might have about defecating during an imaging examination. Radiology personnel should ease patient anxiety by explaining clearly the instructions used during the image acquisitions and addressing any concerns patients might have upon arrival to their imaging appointments. MR defecography can be performed with the patient in sitting or supine positions. Imaging in sitting position most closely mimics physiologic positioning during defecation; however, it requires open magnets, which are not readily available at most institutions. Most centers perform MR defecography in the supine position. The patients typically lie on their back on the MRI table with a pillow or wedge under their knees for comfort and to help promote defecation.²⁵,²⁶ Patients are asked to evacuate the rectal gel on command into an MR compatible enema ring and/or with physical barriers such as absorbable pads covering the surface of the MRI table. Several studies have shown comparable results of MR defecography in supine versus sitting/upright position.²⁷,²⁸ Kumar and colleagues²⁹ showed that supine MR defecography demonstrated significantly higher prevalence and degree of cystocele and urethral hypermobility as compared to upright voiding cystourethrogram (VCUG). In preparation for the examination, contrast is instilled in the rectum. Although various institutions used different types and volume of contrast, approximately 120 to 180 mL of US gel is generally recommended.²⁶,³⁰ Some patients may need a larger volume to induce the sensation of rectal fullness for defecation (particularly in patients presenting with impaired defecation). Administration of vaginal contrast is optional. It may be helpful if there is history of prior surgery such as urethral sling or vaginal mesh, although routine use is not typically recommended.²⁶ Many institutions ask patients to drink a standardized volume (e.g., 16 oz) of water, 15 to 20 minutes before the examination to ensure some bladder filling. However, overdistention of the bladder should be avoided because it may prevent prolapse of other compartments.

Although specific sequences and parameters may differ between institutions, in general, static sequences include T2-weighted images (T2WI) in three planes (sagittal, axial, coronal) and, in many cases, a T1-weighted sequence in at least a single plane. These images are used for evaluation of pelvic anatomy including the pelvic organs, muscles, osseous structures, as well as pelvic floor ligaments and fascia. Dynamic imaging is typically performed while the patient evacuates the rectal gel or during other maneuvers (e.g., Kegel) using T2-weighted or balanced steady-state sequences acquired at the rate of approximately one image per second through a single midsagittal plane. This results in a series of images that can be viewed in cine mode to assess change in location of the organs during the maneuvers. Additional similar cine images may be acquired in axial or coronal planes to assess for paramedian pathologies (e.g., lateral rectocele). Each dynamic sequence should be acquired for approximately 30 seconds, and at minimum, three defecation acquisition should be acquired; ideally, the defecatory sequence should be repeated until complete rectal emptying to ensure adequate patient effort. If the patient is unable to fully evacuate on the defecatory phase, he or she may be asked to complete evacuation in a restroom prior to returning to the machine for postdefecation cine images with maximal Valsalva maneuver which may demonstrate more severe dysfunction.⁴