Objective
The objective of the study was to compare pelvic structure location on magnetic resonance imaging (MRI) during maximal Valsalva among women with posterior prolapse and those with normal support.
Study Design
Subjects (n = 37) had posterior vaginal wall (PVW) prolapse of +1 cm or greater. All underwent midsagittal, dynamic MRI. Structure locations (distal vagina, apex, perineal body, external anal sphincter) were determined. PVW length, levator and urogenital hiatus diameters, and prolapse diameter were measured.
Results
Subjects had more caudal structures ( P < .001) and larger hiatus diameters ( P < .005); the posterior wall was longer, whereas the straight-line distance between the apex and distal vagina was shorter. In enteroceles, the apex was more ventrally displaced compared with rectoceles ( P = .003). Unlike apical descent (r = –0.3; P = .1), PVW length and point Bp were correlated with MRI prolapse size (r = 0.5; P = .002; r = 0.7; P < .001, respectively).
Conclusion
At maximal Valsalva on MRI, structures are more caudal in women with posterior prolapse. The posterior vaginal wall is longer; this length strongly correlates with prolapse size.
Our understanding of the mechanisms of posterior vaginal wall prolapse is still evolving. Various imaging modalities, especially magnetic resonance imaging (MRI), have proven useful in delineating normal pelvic anatomy, studying changes in pelvic floor descent in the postpartum period, and diagnosing and staging of pelvic organ prolapse (POP). Changes in the motion of pelvic viscera with aging and POP on MRI have been documented and have provided many insights into the mechanisms behind pelvic floor failure.
Imaging studies that compare women with posterior vaginal wall prolapse with those with normal support make it possible to quantify the posterior vaginal wall structural deformations that are associated with rectocele and enterocele. To date, the use of MRI to study the posterior vaginal wall (PVW) has been limited to delineating normal anatomy and documenting the clinical occurrence of rectocele and enterocele.
Our objective was to use MRI to detect differences in vaginal wall position and perineal descent, as well as posterior vaginal wall length and apical descent between women with PVW prolapse and those with normal support. These investigations will allow us to evaluate parameters not necessarily easily captured on physical examination.
Materials and Methods
This was a secondary analysis of a case-control study at the University of Michigan (Ann Arbor, MI) looking at POP. The study was approved by the university’s institutional review board (institutional review board #1999-0395). Women were recruited from the urogynecology clinics, as well as community advertisements, and all underwent the pelvic organ prolapse quantification (POP-Q). To be considered a subject, women had to have prolapse of the PVW that extended at least 1 cm below the hymenal ring. This criterion was met by 35 women in the original study. Women were accepted as controls (n = 37) if they were asymptomatic and had normal vaginal support on straining, which was defined as all portions of the vaginal compartment at least 1 cm above the hymen.
All women were matched for age, race, and number of vaginal deliveries. Women were excluded if they had a prior history of surgery for POP or other pelvic floor dysfunction; if they were pregnant within the last year; had a history of chronic steroid use; had previous radiation to the pelvis; or were immunocompromised, with an increased risk of infection from testing. Women with a prior hysterectomy were included as long as it had not been done for prolapse and if their prolapse occurred at least 2 years after surgery.
All women underwent a POP-Q examination with all points (except for genital hiatus, levator hiatus, and total vaginal length) recorded at maximum Valsalva in the lithotomy position at a 45° angle. Genital and levator hiatus measurements were done at rest to obtain accurate measurements of the anatomy, not influenced by the size of the subjects’ prolapse. Dynamic MRI was performed in the supine position as well. Images were taken in the axial, sagittal, and coronal planes using a fast spin proton density technique. Scans were performed on a 1.5 T superconducting magnet (Signa; General Electric Medical Systems, Milwaukee, WI). Slice thickness was 4 mm, with a gap of 1 mm, yielding 5 mm image spacing. Twenty cubic centimeters of ultrasound gel was added to the vagina to better delimit its location and boundaries.
Based on MRI appearance, PVW prolapse among subjects was further classified as either rectocele predominant or enterocele predominant. In this study, an enterocele was judged to be present when the potential space in the cul-de-sac of Douglas was abnormally dilated by the small intestine or mesenteric fat, similar to the definition suggested by Gousse et al. Two experienced researchers independently agreed that 9 women had enterocele-predominant posterior prolapse (n = 9). Those with rectocele-predominant prolapse (n = 28) did not have associated enteroceles. One of the 9 enterocele-predominant subjects had a rectocele.
On midsagittal MRI images at maximum Valsalva, a sacrococcygeal-inferior pubic point (SCIPP) line (x-axis) and perpendicular y-axis were drawn, as described in our previous work. Using these axes, locations of the distal vagina, vaginal apex (most superior point of the PVW), midperineal body, and midexternal anal sphincter were determined as x,y coordinates in centimeters. The length of the entire posterior vaginal wall from apex to distal vagina, levator and urogenital hiatus diameters, and levator plate angle were measured using these same midsagittal images at maximal Valsalva. These, as well as coordinate locations of structures at maximum Valsalva, were compared among all subjects and controls ( Figure 1 , A) and among women with rectocele and enterocele-predominant prolapse using Image J 1.4l software (National Institutes of Health, Bethesda, MD).
On MRI, the widest anterior-posterior prolapse diameter was also calculated as the distance between the most anterior point of the puborectalis and most ventral point of the PVW, similar to methods described on defecography ( Figure 1 , B). Using the straight-line distance from the vaginal apex to the distal vagina on imaging in cases and controls, we were able to compare the groups and determine a surrogate measure for apical downward displacement ( Figure 1 , B).
Means, SDs, and frequency distributions for discrete variables were used to characterize the groups. Comparisons between the groups were made using Student t tests. Effect sizes (Cohen’s d) were calculated for significant differences. An alpha of less than 0.05 was considered to be significant.
Results
Overall, there were 35 controls and 37 subjects. Women were similar with respect to age, body mass index (BMI), vaginal parity, menopausal status, and hysterectomy status ( Table 1 ). Similar percentages of women in the 2 cohorts recalled a history of at least 1 forceps-assisted vaginal delivery and were currently smokers. Women with PVW prolapse were more likely to be currently using hormone therapy.
| Characteristic | Subjects (n = 37) | All controls (n = 35) | P value |
|---|---|---|---|
| Age, y | 58.9 (11.0) | 58.9 (12.1) | 1.0 |
| BMI, kg/m 2 | 28.6 (6.5) | 27.1 (7.1) | 0.3 |
| Vaginal parity | 2.7 (1.6) | 2.8 (1.7) | 0.9 |
| Recalled forceps-assisted vaginal delivery, % | 51.9 | 28.0 | 0.1 |
| White, % | 59.5 | 52.9 | 0.6 |
| Postmenopausal, % | 84.4 | 75.8 | 0.5 |
| Current use of hormone therapy, % | 33.3 | 5.9 | 0.006 |
| Hysterectomy, % | 37.8 | 17.6 | 0.07 |
| Current smoker, % | 15.2 | 21.9 | 0.5 |
Subjects had a higher degree of both anterior (points Aa and Ba) and posterior prolapse (points Ap and Bp), as well as larger resting genital hiatuses during clinical POP-Q examination ( Table 2 ). Differences in rectoceles and enteroceles on POP-Q were seen only apically (points C and D) ( Table 2 ).
| POP-Q, cm | Controls (n = 35) | All subjects (n = 37) | Rectocele predominant (n = 28) | Enterocele predominant (n = 9) | P value controls vs all subjects | P value rectoceles vs enteroceles |
|---|---|---|---|---|---|---|
| Aa | –1.7 (0.9) | –0.5 (1.5) | –0.6 (1.2) | –0.1 (2.5) | < .001 | .4 |
| Ba | –1.7 (0.9) | +0.2 (2.4) | –0.1 (1.5) | 1.0 (4.2) | < .001 | .2 |
| C | –6.3 (1.5) | –3.7 (4.1) | –4.9 (3.0) | –0.3 (5.5) | < .001 | .003 |
| D | –8.4 (1.3) | –6.9 (3.1) | –7.8 (1.2) | –3.4 (5.9) | .02 | .002 |
| Ap | –1.8 (0.8) | +1.1 (1.4) | +1.3 (1.0) | 0.7 (2.1) | < .001 | .3 |
| Bp | –1.8 (0.8) | +1.8 (2.1) | +1.5 (1.1) | +2.7 (3.7) | < .001 | .1 |
| GHrest | 2.7 (0.8) | 4.0 (0.9) | 4.1 (0.9) | 3.7 (0.3) | < .001 | .2 |
| LHrest | 6.6 (1.5) | 7.3 (1.3) | 7.4 (1.3) | 6.7 (0.6) | .11 | .2 |
| TVL | 10.4 (1.3) | 10.4 (1.3) | 10.5 (1.2) | 9.3 (2.1) | .90 | .4 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
