The definition of enterocele is controversial. Nichols and Randall1 described it as a sac of peritoneum that separates the rectum from the vagina, usually containing small intestine or omentum. It has also been postulated that vaginal enterocele is a condition in which there is peritoneum in contact with vaginal epithelium with no intervening fascia.2 Enteroceles have been reported in as few as 0.1% and as many as 16% of women who had gynecologic surgical procedures.3–5 This wide range may be a result of the difficulty of diagnosing enteroceles preoperatively. Few studies address the association between symptoms and enterocele.6,7 There are no diagnostic signs, symptoms, physical examination maneuvers, or laboratory tests.
The objective of this study was to characterize preoperative signs and symptoms of bowel function and vaginal prolapse in women with enteroceles, compared with women without them. We also wanted to determine whether enterocele was correlated with the extent of prolapse and degree of patient discomfort.
Materials and Methods
This study was approved by the Institutional Review Board at the Cleveland Clinic Foundation, and written informed consent was obtained from all subjects. Women who had surgery for pelvic organ prolapse, urinary incontinence, or both between June 1996 and February 1999 were asked to participate. Women were recruited from practices of four urogynecologists at our institution and completed standardized questionnaires on bowel function and local symptoms caused by vaginal prolapse. We have reported posterior vaginal prolapse and preoperative bowel function in this population8 but the current study focuses on women with enterocele defined by surgical findings and associated symptoms. All preoperative pelvic examinations used for analysis were done by the same research nurse using the prolapse grading system of the International Continence Society.9 The Baden classification of vaginal prolapse was used to examine the first 39 women recruited, before acceptance of International Continence Society terminology and standards. Data collected with the Baden system were not included in the analysis on vaginal sites. Vaginal examinations were done with Sims speculums with patients supine in the lithotomy position. Descent of prolapse was determined by Valsalva or cough maneuvers and confirmed as maximal by subjects. Methods, definitions, and descriptions conformed to standards recommended by the International Continence Society.9 Point Aa refers to the position of the midline anterior vagina 3 cm proximal to the hymen. Point Ba refers to the most distal extent of prolapse affecting the anterior wall. Point C refers to the most distal edge of the cervix or leading edge of the vaginal cuff after hysterectomy. Point D refers to the posterior fornix in the presence of a cervix. Point Ap refers to the midline posterior vagina 3 cm proximal to the hymen. Point Bp refers to the most distal extent of prolapse affecting the posterior wall. Positive measurements reflected protrusion beyond the hymen whereas negative measurements reflected protrusion above the hymen.
Each woman had surgery for uterovaginal prolapse, urinary incontinence, or both, as scheduled. Women with intraoperative enteroceles that required repair formed the study group and were compared with those who did not require repair. The presence of an enterocele that needed repair was judged by the four gynecologic surgeons who contributed subjects. Enteroceles of any size identified at surgery were repaired. Abdominal enterocele repair involved either a Moschcowitz or Halban obliteration of a deep cul de sac. From the vaginal approach, a herniated enterocele sac between the anterior and posterior vaginal walls, which admitted entry into the peritoneal cavity, constituted an enterocele in posthysterectomy women. Transvaginal enterocele repair involved recognition, entry, and excision of the enterocele sac with high ligation of the peritoneum using a double pursestring closure at its neck. A deep, redundant cul de sac that required wedge resection of the posterior vaginal wall during concomitant hysterectomy was also considered an enterocele repair in the presence of uterine prolapse.
Variables measured on a continuous scale are presented as medians and interquartile ranges and were analyzed using Wilcoxon rank-sum tests. Categoric or dichotomous variables were analyzed using χ2 or Fisher exact tests, as appropriate. Severity of bowel and vaginal prolapse symptoms was recorded on an ordinal scale from 1 to 10 and analyzed using Wilcoxon rank-sum tests. Controlling for stage of prolapse, the scale for vaginal prolapse symptoms was analyzed using a Cochran-Mantel-Haenszel test for general association. If frequency of bowel movements was less often than every other day (ie, two or fewer bowel movements per week), it was considered abnormal. Other symptoms of bowel function (straining, use of manual pressure for defecation, or fecal incontinence) were considered absent if the questionnaire response was “never or rarely.” Extent of prolapse was described with centimeter measurements of vaginal sites at points Aa, Ba, Ap, Bp, C, and D and analyzed using Wilcoxon rank-sum tests. Rectocele was defined by vaginal sites Bp or Ap at Stage II or worse. Those measurements also were converted to prolapse stages and analyzed using Mantel-Haenszel χ2 tests. The association of enterocele with previous surgery, menopausal status, age, and hormone replacement therapy (HRT) was assessed in a multiple logistic regression model with surgery group (enterocele versus others) as the outcome and those factors as covariates. The number of subjects in subgroups of each analysis did not always equal the total sample because of missing data. Statistical tests were two-sided and P values of .05 or lower were considered statistically significant. This was a descriptive study, so it was not designed to detect specific differences between groups, and lack of statistical significance should not be assumed to indicate lack of clinical significance.
Three hundred ten women completed questionnaires, had standardized physical examinations, and proceeded with surgery. Seventy-seven (25%) had enterocele repairs. Demographic and clinical characteristics are shown in Table 1. Pertinent medical problems (diabetes, chronic cough, and neurologic deficit) and specific medication were approximately the same in each group; however, women with enteroceles were older, menopausal, had prior pelvic surgery, and had HRT compared with women without enteroceles when those factors were examined separately. After analyzing them in a multiple logistic regression model, the only factor that was independently associated with enterocele was age (P = .01). Thus, the association between enterocele and menopause, HRT, and prior surgery appears to be explained by the fact that those women were older. Concurrent surgical procedures are presented in Table 2.
Women with enteroceles had more advanced prolapse by stage of prolapse at most advanced site than women without enteroceles (Table 3). Women with enteroceles had more advanced prolapse at points Ap, Bp, and C but not at point D, and less advanced prolapse at points Aa and Ba.
To examine the possible overlap of symptoms in women with enteroceles and rectoceles, we separated subjects according to enterocele only (n = 20), rectocele only (n = 64), and enterocele and rectocele (n = 46). We were unable to group all women with enteroceles accordingly because of missing data. There were no significant differences in bowel symptoms among the groups (Table 4).
The extent to which women were bothered by symptoms of bowel function was reported on a scale of 1 to 10, with 1 being “not at all” and 10 being “extremely.” Median scores were 4.0 (interquartile range 1–8) for the enterocele group and 3.0 (interquartile range 1–7) for the nonenterocele group (P = .33). We also examined that score for the enterocele only, rectocele only, and enterocele and rectocele groups. Women with rectoceles (5.0, interquartile range 2–9) or enteroceles and rectoceles (6.0, interquartile range 2–9) were more bothered by bowel symptoms than women with enteroceles alone (2.5, interquartile range 1–5), although that difference did not reach statistical significance (P = .16).
Women rated the extent to which they were bothered by symptoms of vaginal prolapse on a similar scale of 1 to 10. Women with enteroceles were more bothered by symptoms of prolapse (8.0, interquartile range 5–10) than those without enteroceles (7.0, interquartile range 3–9) (P = .009). However, controlling for stage, there was no significant difference between the enterocele and nonenterocele groups on the scale for vaginal prolapse symptoms (P = .75).
We found no evidence that enteroceles were associated with specific symptoms of bowel function. Women with enteroceles were more bothered by symptoms of vaginal prolapse, but that was caused by more advanced prolapse rather than specific association with enteroceles. Symptoms of bowel function were not associated with enteroceles, although most enteroceles tend to be posterior (16 times more common than other enteroceles).10 It has been suggested that the clinical manifestations of enteroceles include rectal pressure, constipation, and incomplete emptying of the rectum,11 although our findings did not support that association. Although our study was not designed specifically to test hypotheses on differences between women with and without enteroceles with respect to bowel symptoms, our sample of 310 women provided adequate power to detect large differences. The small differences might not be clinically significant. A prospective study of thousands of women would be necessary to determine conclusively that women with enteroceles do not differ from women without enteroceles in bowel symptoms. Based on this study and our previous study on posterior vaginal prolapse and bowel function,8 we believe that bowel complaints are nonspecific symptoms. The only symptom significantly associated with enterocele was vaginal prolapse itself. Because symptoms other than prolapse are not associated with enterocele, the ability of an enterocele repair to relieve those symptoms (other than those related to vaginal protrusion) is uncertain.
Our study group was based on findings at surgery and requirements for enterocele repair. The comparison group was defined by the assumption that women who had vaginal reconstruction in which enteroceles were not repaired did not have enteroceles. Misclassification bias could occur if small enteroceles were not identified at surgery; however, it would not likely change our results. If clinically identified enteroceles were not associated with specific symptoms, then small enteroceles of unknown clinical significance would be less likely to be associated with symptoms. Therefore, what is a clinically relevant enterocele? There is no definition or standard for the diagnosis of enterocele; there is no clear distinction between clinically significant and insignificant enteroceles; and these issues require further research.
Defecography was not used in this study. Although some reports support preoperative defecography to identify enteroceles and other defects, to modify surgery,12 it has limitations.13 In a previous study, abnormal radiologic features on defecography were found in 47 normal women. Asymptomatic enteroceles were found in two of 20 women and asymptomatic rectoceles were found in 17 of 21 women. There are no standard guidelines regarding the testing procedure or standard methods of interpretation.13
The only standard technique for assessing uterovaginal prolapse is based on recommendations of the International Continence Society. That system has proven interobserver and intraobserver reliability and clinical use.14 Those standards are validated definitions that allow comparisons of published series from different institutions as well as longitudinal evaluations of individual women. Intuitively, points Ap, Bp, C, and D would correspond to enterocele defects. Of those points, only point D was not associated with an enterocele, which was consistent with the finding that women who had hysterectomies might be more likely to develop enteroceles.
The results of this study suggest that certain aspects of histories and physical examinations of women with prolapse are associated with enteroceles, including advanced age, menopausal status, and pelvic surgery. Hysterectomy or vaginal surgery might predispose women to apical vaginal vault weakness, which might be aggravated by increasing propulsive effects of intra-abdominal pressure and the dragging effects of gravity.5 That suggests a possible iatrogenic disruption of vaginal supportive tissue, which causes a change in vaginal axis1 or loss of continuity of fibrous connective tissue,2 and contributes to enterocele formation. In our study, incontinence procedures were associated with enteroceles only when surgery was by the vaginal approach. We realize that vaginal prolapse repairs commonly accompany vaginal incontinence procedures; however, the Burch procedure was associated with enterocele formation in other studies. In a study of 131 women with histories of a Burch colposuspension, 22% needed rectocele repair and 10% needed enterocele repair at least 3 years after initial surgery.15 Burch postulated that this was caused by elevation of the anterior vagina, which led to changes in the slope of the posterior wall and predisposed those subjects to enterocele.16 The association of Burch procedure with enterocele requires further study.
1. Nichols DH, Randall CL. Vaginal surgery. 4th ed. Baltimore: Williams & Wilkins, 1996:336–50.
2. Richardson AC. The anatomic defects in rectocele and enterocele. J Pelvic Surg 1995;1:214–21.
3. Austin RC, Damstra EF. New fascia plastic repair of enterocele. Surg Gynecol Obstet 1955;101:297–304.
4. Weed JC, Tyrone C. Enterocele. An analysis of 52 cases. Am J Obstet Gynecol 1950;60:324–32.
5. Ranney B. Enterocele, vaginal prolapse, pelvic hernia: Recognition and treatment. Am J Obstet Gynecol 1981;140:53–60.
6. Ostrzenski A. A new laparoscopic retroperitoneal posterior culdoplasty technique. J Reprod Med 1999;44:504–10.
7. Spence-Jones C, Kamm MA, Henry MM, Hudson CN. Bowel dysfunction: A pathogenic factor in uterovaginal prolapse and urinary stress incontinence. Br J Obstet Gynaecol 1994;101:147–52.
8. Weber AM, Walters MD, Ballard LA, Booher DL, Piedmonte MR. Posterior vaginal prolapse and bowel function. Am J Obstet Gynecol 1998;179:1446–50.
9. Bump RC, Mattiasson A, Bo K, Brubaker LP, DeLancey JO, Klarskov P, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996;175:10–7.
10. Masson JC. Vaginal hernia. Surg Gynecol Obstet 1928;47:36–8.
11. Holley RL. Enterocele: A review. Obstet Gynecol Surv 1994;49:284–93.
12. Brubaker L, Retzky S, Smith C, Saclarides T. Pelvic floor evaluation with dynamic fluoroscopy. Obstet Gynecol 1993;82:863–8.
13. Shorvon PJ, McHugh S, Diamant NE, Somers S, Stevenson GW. Defecography in normal volunteers: Results and implications. Gut 1989;30:1737–49.
14. Hall AF, Theofrastous JP, Cundiff GC, Harris RL, Hamilton LF, Swift SE, et al. Interobserver and intraobserver reliability of the proposed International Continence Society, Society of Gynecologic Surgeons, and American Urogynecologic Society pelvic organ prolapse classification system. Am J Obstet Gynecol 1996;175:1467–70.
15. Wiskind AK, Creighton SM, Stanton SL. Incidence of genital prolapse after Burch colposuspension. Am J Obstet Gynecol 1992;167:399–404.
© 2000 The American College of Obstetricians and Gynecologists
16. Burch JC. Urethrovaginal fixation to Cooper's ligament for correction of stress incontinence, cystocele, and prolapse. Am J Obstet Gynecol 1961;81:281–4.