- Email: [email protected]
Total colpocleisis with high levator plication for the treatment of advanced pelvic organ prolapse
Total colpocleisis with high levator plication for the treatment of advanced pelvic organ prolapse Walter S. von Pechmann, MD, Martina Mutone, MD, Joanne Fyffe, MS, and Douglass S. Hale, MD Indianapolis, Ind OBJECTIVES: Our purpose was (1) to report long-term objective and subjective outcome data after total colpocleisis with high levator plication and (2) to compare operative morbidity of total colpocleisis in patients with prior versus concurrent hysterectomy. STUDY DESIGN: The medical records of patients who underwent total colpocleisis with high levator plication between August 1, 1988, and December 31, 2000, were retrospectively reviewed. Patient characteristics, operative data, and objective outcome measures including pelvic organ prolapse staging measurements were obtained from subject records. Subjective outcome measures were obtained by a standardized telephone survey. A t test was used to compare continuous variables between patients who underwent prior versus concurrent hysterectomy. The Fisher exact test was used to test for association between patient group and each categorical variable. RESULTS: During the 12-year study period, 92 subjects underwent total colpocleisis with high levator plication. Of the 92 subjects, 90 (97.8%) underwent concurrent operations and 37 (40.2%) underwent simultaneous hysterectomy. With objective cure defined as absence of prolapse to the hymen, 90 subjects (97.8%) were objectively cured after a median follow-up of 12 months (range, 0-64 months). Of 62 subjects available for telephone follow-up, 56 (90.3%) reported being satisfied or very satisfied with how the surgery cured their prolapse after a median follow-up of 24 months (range, 13-161 months). Concurrent hysterectomy was associated with statistically significant increases in absolute change in hematocrit (9.5% without vs 11.9% with hysterectomy) and transfusion requirement (12.7% without vs 35.1% with hysterectomy). There was no significant difference in surgical complications between groups. CONCLUSION: Total colpocleisis is an effective operation for the treatment of advanced pelvic organ prolapse. Concurrent hysterectomy is associated with higher blood loss and transfusion requirements. (Am J Obstet Gynecol 2003;189:121-6.)
Key words: Colpocleisis, colpectomy, vaginectomy, vaginal hysterectomy, pelvic organ prolapse
Total colpocleisis with levator plication was first described in the English language literature by Edebohls1 in 1901. Colpocleisis, which requires removal of the vaginal epithelium, results in the loss of coital ability and therefore is appropriate for only a minority of patients. For selected patients with advanced and often recurrent prolapse, vaginal obliteration may be preferable to pelvic reconstruction. Although several series have described variations of the colpocleisis procedure resulting in high success rates, outcome reporting has been limited by a lack of reproducible objective and subjective data
From the Female Pelvic Medicine and Reconstructive Surgery Fellowship Program, Department of Obstetrics and Gynecology, Methodist Hospital of Indianapolis/Indiana University. Presented at the Twenty-Third Annual Meeting of the American UroGynecologic Society, San Francisco, Calif, October 6-18, 2002. Received for publication March 23, 2003; accepted March 23, 2003. Reprints not available from the authors. © 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.546
points.2-9 In addition, the effect of concurrent hysterectomy on operative morbidity is unknown. The objectives of this study were 2-fold: (1) to report, in accordance with recently established surgical outcome reporting guidelines,10 the long-term objective and subjective outcomes after total colpocleisis with high levator plication and (2) to compare the operative morbidity of total colpocleisis with and without concurrent hysterectomy. Material and methods This project was approved by the Institutional Review Board of Methodist Hospital of Indianapolis. All total colpocleisis operations performed at one institution between August 1988 and December 2000 comprised the study population. The surgical cases were identified by manual review of operative reports contained in all existing patient records of a tertiary care referralbased urogynecology practice. Operations described as partial colpocleisis or LeForte colpocleisis were excluded. Preoperative examination was performed with subjects 121
122 von Pechmann et al
seated in the semiupright position in a birthing chair, and measurements were taken during maximum Valsalva maneuver. Routine evaluation included preoperative urodynamic testing on all patients with advanced pelvic organ prolapse, but this was waived in select cases because of mitigating medical conditions or excessive patient travel time. Preoperative counseling included discussion regarding the loss of coital ability after total colpocleisis and potential effects on body self-image. All operations were performed under the direction of faculty in the Division of Female Pelvic Medicine and Reconstructive Surgery. Surgical technique. Vaginal hysterectomy with or without bilateral adnexectomy was performed in patients with intact uteri. After purse-string closure of the cul-de-sac peritoneum, the complete colpocleisis operation followed. In patients who had previously undergone hysterectomy, there was no attempt to enter and ligate or remove the enterocele sac. Colpocleisis was performed in the following manner: (1) the entire vaginal epithelium was mobilized by sharp dissection, (2) the prolapse was serially reduced with absorbable sutures placed into the vaginal muscularis beginning at the leading edge of the prolapse and continuing sequentially in interrupted purse-string fashion until the prolapsed tissues were superior to the level of the levator plate, (3) a support procedure was performed at the urethrovesical junction (pubourethral ligament plication, needle suspension, or suburethral sling), (4) the puborectalis and pubococcygeus muscles were approximated in the midline with either absorbable or delayed absorbable sutures in interrupted fashion to create a shelf from above the rectum to within one fingerbreadth of the urethrovesical junction, (5) indigo carmine was administered intravenously to cystoscopically confirm ureteral patency, (6) perineal body reconstruction was performed, and (7) the vaginal epithelial edges were trimmed and reapproximated, leaving a foreshortened, narrowed vaginal vault. Outcomes measurement. Data were obtained from three sources: office medical records, hospital records, and standardized telephone contacts. Office medical records were reviewed for demographic data, medical and surgical history, preopoerative and postoperative examination findings, and length of last follow-up. Before 1996, prolapse was graded according to the Baden and Walker system.11 Subsequent to 1996, prolapse was staged by the Pelvic Organ Prolapse-Quantification System of the International Continence Society.12 Objective cure was defined as the absence of prolapse of the anterior or posterior vaginal wall or vaginal apex to the hymen. Hospital charts were reviewed for type of anesthesia, types of concurrent operations performed, operative time, estimated blood loss, change in hematocrit, transfusion occurrence, incidence and types of complications, and length of stay. A designated research nurse who was not involved in direct patient care performed telephone fol-
July 2003 Am J Obstet Gynecol
low-up on available patients during the month of January 2002. Telephone contacts included questions regarding satisfaction with surgical results, recurrence of prolapse, and regret over the loss of coital function. Subjective cure was defined as a response of “very satisfied” or “satisfied” when the patients were asked whether they were pleased with how well the operation cured the prolapse. Statistical analysis. Summary statistics were calculated for each patient group defined by whether the patients underwent previous versus concurrent hysterectomy. A t test was used to compare continuous variables between patient groups. The Fisher exact test was used to test for association between patient group and each categorical variable. Throughout the statistical comparisons, P < .05 was taken as significant. The statistical software SAS V8.02 (SAS Institute, Cary, NC) was used to perform the analysis. Results Patient characteristics. Age, parity, stage of prolapse, and number of previous prolapse surgeries are summarized in Table I. There was no significant difference between patient groups with regard to patient age, parity, marital status, medical history, type of urinary incontinence, stage of prolapse, length of follow-up, and objective or subjective cure rates. Medical comorbidities were common in this elderly population, with 75% having cardiovascular disease, 14% having pulmonary disease, 16% having diabetes, 10% having neurologic disorders, 50% having other medical comorbidities, and only 7% of patients having none. Previous surgery for prolapse was significantly more common in patients who had undergone prior hysterectomy. Of 75 patients who underwent preoperative urodynamic testing, 36 (48.0%) had stress urinary incontinence, 13 (17.3%) had detrusor overactivity, and 16 (21.3%) had mixed urinary incontinence. Operative data. Adjunctive procedures, blood loss and transfusion requirements, and surgical complications are summarized in Table II. Eighty-seven (94.6%) operations were performed under general anesthesia and five (5.4%) under regional anesthesia. Ninety of the 92 patients (97.8%) underwent adjunctive procedures at the time of colpocleisis. Two patients had no urethral support procedures and two others had pubourethral ligament plication performed to enhance distal anterior wall support in addition to another incontinence procedure. The operative time for the colpocleisis procedure could not be determined because of the frequency of adjunctive procedures. Mean total operating room time was 52 minutes longer when hysterectomy was performed concurrently (P < .001). The transfusion rate of both groups together was 21.7%. Complications. Reversible ureteral occlusion was suspected on the basis of intraoperative cystoscopy in one case in the nonhysterectomy group (1.8%) and three
von Pechmann et al 123
Volume 189, Number 1 Am J Obstet Gynecol
cases in the hysterectomy group (8.1%). The single case in the nonhysterectomy group and one case in the hysterectomy group resolved with removal of the most superior levator plication stitches. The second case in the hysterectomy group resolved after removal of the superior levator plication stitches and several purse-string and hemostatic sutures. The third case in the hysterectomy group occurred after removal of ureteral stents and removal of the levator stitches had no effect. A ureteral stent was passed again and achieved efflux of urine after being advanced 1 to 2 cm indicating edema of the intramural ureter as the source of obstruction. An indwelling stent was left in place for the first 2 postoperative weeks, after which resolution was confirmed by intravenous pyelography. Two cases (5.4%) in the hysterectomy group were converted to laparotomy. The first was converted to secure hemostasis because of bleeding from the ovarian vessels at the level of the infundibulopelvic ligament after adnexectomy. The second case was converted because of rupture of a diverticular abscess on cul-de-sac entry. One patient in the nonhysterectomy group sustained a small proctotomy that was recognized and repaired without incident. One patient in the hysterectomy group who had a preoperative diagnosis of metastatic lung cancer died of pulmonary complications 28 days postoperatively. Objective outcomes. Objective outcome data are summarized in Table III. Objective cure, defined as the absence of prolapse to the hymen, was achieved in 90 of 92 patients (97.8%). Median office follow-up time was 12 months (range 0-64 months). Data on postoperative vaginal length was available for 78 subjects; mean length was 2.4 cm (±1.0 cm). Recurrent vaginal prolapse occurred in 2 patients (2.2%). One patient (1.1%) had a recurrent enterocele 7 months postoperatively that was cured with vaginally approached reoperation. The second patient had asymptomatic enlargement of the genital hiatus and anterior wall descent to the hymen 64 months postoperatively, which required no further therapy. Postoperative rectal prolapse occurred in 2 patients (2.2%) within 6 months of surgery. The first patient underwent transanal excision and the second was managed conservatively. Of 46 patients (50%) for whom postoperative standing stress test data was available, 6 patients (13.0%) demonstrated recurrent stress urinary incontinence. Four had suburethral slings with cadaveric fascia, and two had pubourethral ligament plications performed at the time of colpocleisis. Three patients (3.3%) underwent subsequent procedures for treatment of stress urinary incontinence; two patients received periurethral collagen injection and one underwent tension-free vaginal tape suburethral sling placement without difficulty. There were no cases of de novo stress urinary incontinence. Subjective outcomes. Subjective outcome data are presented in Table III. Of the 92 subjects, 8 (8.7%) were deceased and another 13 (14.1%) were unavailable for
Table I. Patient characteristics of 92 women undergoing total colpocleisis Variable
Prior hysterectomy
Subjects (No. [%]) 55 (59.8%) Age (y) 77 (±6) Parity 3 (±2) Stage of prolapse I 0 (0%) II 1 (1.9%) III 28 (51.9%) IV 25 (46.3%) Previous prolapse operations 0 29 (52.7%) 1 17 (30.9%) 2 6 (10.9%) 3 1 (1.8%) 4 2 (3.6%)
Concurrent hysterectomy 37 (40.2%) 79 (±5) 4 (±2)
P value*
.07 .22 .23
0 (0% 0 (0%) 14 (37.8%) 23 (62.2%) <.001 35 (94.6%) 2 (5.4%) 0 (0%) 0 (0%) 0 (0%)
*t test was used to assess differences between groups for age and parity. Fisher exact test was used to determine the association between state of prolapse and hysterectomy and the association between number of previous prolapse operations and hysterectomy.
contact. Sixty-two subjects (67.4%) could be directly contacted by telephone, and information regarding reoperation was available for an additional 9 subjects (9.8%) through indirect contact. Median telephone follow-up time was 24 months (range 13-161 months). Telephone follow-up did not reveal any additional operations for vaginal or rectal prolapse or incontinence. Subjective cure, defined as a response of satisfied or very satisfied with how their surgery cured their prolapse, was achieved in 56 of 62 patients (90.3%). Seventy-one percent of those directly contacted by telephone reported being very satisfied or satisfied with their bladder function postoperatively. Regret over loss of coital ability occurred to some extent in 8 subjects (12.9%); 4 of the 8 stated that they would make the decision to have the colpocleisis procedure again, 3 were uncertain, and 1 would not. There was no significant difference in age at the time of surgery or number of previous prolapse surgeries between subjects who expressed regret and those who did not. Comment Review of the English language literature reveals several studies describing total “colpocleisis,” “colpectomy,” and “vaginectomy” for the treatment of advanced vaginal or uterovaginal prolapse. Reports from the first half of the 20th century were primarily descriptions of technique with little emphasis on outcomes. In 1901, Edebohls1 reported 4 cases of “panhysterokolpectomy” in which he described vaginal hysterectomy, vaginectomy, and purse-string reduction of the remaining viscera. Phaneuf,2 in 1935, reported another 5 cases, adding levator plication to the procedure. In 1938, Masson and Knepper3 reported 23 cases of vaginectomy, some with hysterectomy, followed by purse-string reduction and levator
124 von Pechmann et al
July 2003 Am J Obstet Gynecol
Table II. Operative data in 92 patients undergoing colpocleisis
Variable
Prior hysterectomy (n = 55)
Adjunctive procedures Incontinence operations Pubourethral 23 (41.8%) ligament plication Needle suspension 3 (5.5%) Suburethral sling 26 (47.3%) Tension-free 3 (5.5%) vaginal tape Other procedures 8 (14.5%) Estimated blood 396 (±221) loss (mL) Change in 9.5 hematocrit (%)† Transfusion 7 (12.7%) Ureteral occlusion 1 (1.8%) Laparotomy 0 (0%)
Concurrent hysterectomy (n = 37) P value*
10 (27.0%)
.19
4 (10.8%) 21 (56.8%) 2 (5.4%)
.43 .40 1.00
5 (13.5%) 578 (±494)
1.00 .04
11.9
.01
13 (35.1%) 3 (8.1%) 2 (5.4%)
.02 .30 .16
*t test used for variables represented as mean (±SD). Fisher exact test used for variables in italic. †Change in hematocrit refers to actual hematocrit values, not change expressed as a percentage.
plication with the added modification of perineorrhaphy. In 1950, Williams4 reported “about sixty cases” of vaginal hysterectomy with vaginectomy and levator plication but no suture reduction. Although no recurrences were reported in these early series, outcomes were typically described in one or two sentences as “satisfactory.” Studies from the latter half of the century placed increasing emphasis on outcomes reporting but with no specific criteria regarding what constituted cure or recurrence. In 1951, Adams5 reported a series of 30 cases, 11 with concurrent hysterectomy, in which he performed purse-string suture reduction followed by urethral and levator plication and perineorrhaphy. He reported no recurrences with a median follow-up of approximately 5 years. In 1960, Symmonds and Pratt6 reported 10 cases of vaginectomy with levator plication with a greater than 5year median follow-up. In the 9 patients available for follow-up, 2 patients required reoperation: 1 for recurrent vaginal prolapse and another for treatment of bilateral hernias into the ischiorectal fossa. A third patient expressed regret over the loss of coital ability. In 1997, DeLancey and Morley 8 reported on a series of 33 patients with posthysterectomy prolapse who underwent vaginectomy, suture reduction, and levator plication. There was one case of recurrence identified after a mean objective follow-up time of 7.7 months and mean subjective followup of 34.6 months. These authors were the first to report the use of a standardized questionnaire regarding recurrence and subjective outcomes. All women but one denied remorse over loss of coital ability; one woman said she had “accepted” her sexual inactivity. In 2001, Grody et al9 identified no recurrences on follow-up survey in a
series of 17 cases of vaginectomy, suture reduction, “vaginal paravaginal cystourethropexy,” and perineorrhaphy with 5 cases of concurrent hysterectomy. Using defined outcome criteria, this study found a 97.8% objective cure in 92 cases, confirming the high success rates reported by previous investigators. With objective cure defined as absence of prolapse to the hymen, two patients had recurrence. One of these was an asymptomatic prolapse of the anterior wall to the hymen and the patient remained “very satisfied.” The second subject had undergone previous abdominally approached paravaginal defect repair and had persistent anterior wall support at the time of her colpocleisis. She subsequently had a symptomatic recurrent enterocele that was cured with vaginally approached ligation of the enterocele sac. Some investigators have noted that enterocele is thought to be the most likely form of recurrence and have therefore recommended routine entry into the enterocele sac to achieve excision and ligation for this reason.13 Others believe that the theoretical benefits do not warrant the increased risk of ileus and other complications.8 It is noteworthy that one of the two cases in this series converted to laparotomy involved rupture of a diverticular abscess after peritoneal entry for hysterectomy. The risk of such a complication may well supercede the risk of recurrence from leaving the enterocele sac intact. Several previous studies have reported postoperative stress urinary incontinence after both total4,6,7,9 and partial colpocleisis.14 Because of these reports as well as previous faculty experience, 90 patients (97.8%) received some form of urethral support procedure. Suburethral slings were typically performed only in patients with urodynamically confirmed stress urinary incontinence. Patients with marked preoperative voiding dysfunction typically underwent less obstructive procedures such as pubourethral ligament plication and two patients had no urethral support procedures performed to minimize postoperative voiding difficulties. Determining the true incidence of postoperative stress urinary incontinence would require routine postoperative urodynamic evaluation and/or standardized subjective assessments, which were not performed in this study. On the basis of the 46 (50%) cases in whom postoperative standing stress test data were available, there was a 13.0% incidence of recurrent stress urinary incontinence but no cases of de novo stress urinary incontinence. Standing stress testing was performed at variable bladder volumes, which may have resulted in underreporting. When necessary, reoperation for urinary incontinence is not precluded by the total colpocleisis procedure. Further investigation of the problem of postcolpocleisis stress urinary incontinence is clearly warranted. The development of postoperative rectal prolapse, as occurred in two subjects in this study, has not been reported previously. It is unclear to what extent colpocleisis may play a contributing role, but several factors should be
von Pechmann et al 125
Volume 189, Number 1 Am J Obstet Gynecol
considered. This study population has several risk factors for the development of rectal prolapse, including female sex, advanced age, and, in some cases, chronic constipation.15,16 Moreover, the frequent coexistence of rectal and genital prolapse and their association with pelvic floor neuropathy is well established.17-19 Colpocleisis with levator plication uses the dysfunctional pelvic floor musculature as a physical barrier to visceral descent by narrowing the levator hiatus. Consequently, intra-abdominal pressure finds the path of least resistance in the remaining anorectal hiatus. Further studies are needed to determine the strength of the association between colpocleisis and rectal prolapse, but screening for rectal prolapse should be considered in patients undergoing this procedure. Among the 62 subjects available for phone follow-up, satisfaction with how the surgery cured the prolapse was greater than 90%, consistent with the high objective cure rate. Because this was a retrospective study, validated psychometric instruments designed to assess the impact of surgery on subjective parameters were not used. In addition, the structure of the telephone questionnaire did not allow for control of confounders such as bowel or bladder dysfunction when patients were questioned specifically regarding cure of the prolapse. Nonetheless, several measures were taken to minimize bias in obtaining subjective outcome data. A nurse not involved in patient care administered all questionnaires, which were structured to provide subjects with a range of responses. Among the 8 patients who expressed regret, there was variability in both the degree of regret expressed over loss of coital ability and over the decision to have the surgery. This illustrates a spectrum of regret that may not have been captured in previous studies that reported lower regret rates. In addition, subjects who were available to complete the survey may have been healthier as a group than their counterparts and therefore more prone to regret, resulting in overreporting due to selection bias. Ridley7 has stated that colpocleisis should be “reserved for those few well-individualized patients in whom sexual function is distinctly not a factor and the problems of vaginal prolapse are primary.” Indeed, all subjects in this study fell into this category. However, because one third of patients over age 78 years remain sexually active,20 the importance of adequate preoperative screening and counseling cannot be overemphasized. It may be the case that women suffering the effects of massive vaginal prolapse obtain a different perspective regarding sexual activity after they have obtained resolution of their symptoms. Our experience has been that many women will resume the social activities they had avoided preoperatively and may unexpectedly develop new relationships. All these factors need to be explored during preoperative counseling. In addition, it is important that the patient understand that loss of coital ability does not preclude other forms of sexual activity, stimulation, or orgasm for
Table III. Postoperative outcomes of 92 patients undergoing total colpocleisis Objective cure Subjective cure* Very satisfied Satisfied Somewhat satisfied Not satisfied Regret over loss of coital ability† Never Somewhat Yes Unable to answer
90/92 (97.8%) 38/62 (61.3%) 18/62 (29.0%) 3/62 (4.8%) 3/62 (4.8%) 52/62 (83.9%) 6/62 (9.7%) 2/62 (3.2%) 2/62 (3.2%)
*In response to the question, “Are you satisfied with how well the surgery repaired the dropped organs?” †In response to the question, “Have you ever regretted losing the ability to have intercourse?”
both the patient and her partner. This study demonstrates that preoperative counseling is not universally successful in preventing postoperative regret. Although this finding may lead to the temptation to omit colpocleisis as an option even for patients who meet Ridley’s selection criteria, concerns about patient regret need to be balanced with respect for the patient autonomy implicit in informed consent. Information regarding operative morbidity is scarce in previous studies. Only two studies have reported data regarding blood loss, and neither of these provided preoperative and postoperative laboratory values.8,9 The mean estimated blood loss in this study was higher than what has been previously reported8 and is likely accounted for by the frequency of concurrent procedures. Ninety of 92 subjects underwent bladder neck support procedures, more than half of which were suburethral slings, and 13 subjects underwent additional adjunctive procedures. The 21.7% overall transfusion rate in this study is consistent with the 29.4% rate reported in the one previous study that reported transfusion data.9 We think that these transfusion rates may simply reflect a low threshold for transfusion in elderly patients with frequent medical comorbidities. Because the difference in change in hematocrit was only 2.4% greater in the hysterectomy group, the nearly 3-fold difference in transfusion requirements is difficult to explain. This may have occurred partly as a result of the added duration of the procedure with hysterectomy, which often led to the decision to transfuse intraoperatively. Although significant, the blood loss incurred during colpocleisis is typically gradual and easily controlled, producing less stress on a weakened myocardium than the type of acute hemorrhage that can occur during reconstructive procedures such as sacral colpopexy or sacrospinous ligament fixation.8 Reversible ureteral obstruction occurred more frequently and conversion to laparotomy occurred exclusively in the concurrent hysterectomy group. Ureteral obstruction during colpocleisis has not been described
126 von Pechmann et al
previously. The prevalence of hydronephrosis has been shown to increase with advancing severity of pelvic organ prolapse.21 Upper urinary tract imaging studies were not routinely obtained, but the same pathophysiologic mechanisms that lead to hydronephrosis may render patients with advanced pelvic organ prolapse more vulnerable to ureteral compromise at the time of surgery. All cases were reversed without major interventions, confirming the importance of intraoperative cystoscopy. The only case in which ureteral stents were placed preoperatively was complicated by loss of ureteral flow after stent removal. This occurred early in the series, after which preoperative stent placement was abandoned. Of the two cases converted to laparotomy, both were directly related to hysterectomy. Vaginal hysterectomy and total colpocleisis were chosen in the patients with intact uteri because of the higher reported success rates with total as opposed to LeForte colpocleisis.14 Nonetheless, LeForte colpocleisis should be considered as a potentially less morbid alternative in patients who have not previously undergone hysterectomy. This case series has several limitations as a result of its retrospective nature. Postoperative prolapse staging was generally recorded by the primary surgeons and was, therefore, subject to observer bias. The clinical outcomes with respect to incontinence were not uniformly or quantitatively measured, preventing conclusions regarding the incidence of postoperative stress urinary incontinence. Baseline preoperative data that would allow evaluation of the effect of surgery on quality of life using a validated psychometric instrument were not available. Consistently recorded racial data on this cohort of patients was not available, although our patient population is predominantly white. It may be that candidacy rates and outcomes would be different in populations of different racial compositions. This study also has several strengths. This is the largest series to date describing outcomes from total colpocleisis for advanced prolapse. The data are reported according to recently established guidelines for surgical outcome studies, including seven of the eight domains of information recommended by the American Urogynecologic Society.10 The definitions of objective and subjective cure were explicitly stated and subjective outcomes were determined by the use of a standardized questionnaire including questions regarding regret over the loss of coital ability. This is also the first study to assess differences in operative morbidity from colpocleisis with or without concurrent hysterectomy. The finding that concurrent hysterectomy is associated with increased blood loss and transfusion requirements may affect how patients are
July 2003 Am J Obstet Gynecol
counseled regarding the risks and benefits of the various surgical options for treatment of pelvic organ prolapse. In summary, total colpocleisis with high levator plication is an effective operation for the treatment of advanced pelvic organ prolapse in patients who do not anticipate future sexual activity. This study could not have been completed without the gracious assistance of Marty Morris, RN, and Elizabeth Fuller, MSN, RN, CNS. REFERENCES
1. Edebohls GM. Panhysterokolpectomy: a new prolapsus operation. Med Rec N Y 1901;60:561-4. 2. Phaneuf TE. The place of colpectomy in the treatment of uterine and vaginal prolapse. Am J Obstet Gynecol 1935;30:544-53. 3. Masson JC, Knepper PA. Vaginectomy. Am J Obstet Gynecol 1938;36:94-9. 4. Williams JT. Vaginal hysterectomy and colpectomy for prolapse of the uterus and bladder. Am J Obstet Gynecol 1950;59:365-70. 5. Adams HD. Total colpocleisis for pelvic eventration. Surg Gynecol Obstet 1951;2:321-4. 6. Symmonds RE, Pratt JH. Vaginal prolapse following hysterectomy. Am J Obstet Gynecol 1960;79:899-909. 7. Ridley JH. Evaluation of the colpocleisis operation: a report of fifty-eight cases. Am J Obstet Gynecol 1972;113:1114-8. 8. DeLancey JOL, Morley GW. Total colpocleisis for vaginal eversion. Am J Obstet Gynecol 1997;176:1228-32. 9. Grody MHT, Merchia V, Nyirjesy P, Kaplan E, Chatwani J. Total colpocleisis: a prospective study. J Pelvic Surg 2001;7:72-8. 10. Wall LL, Versi E, Norton P, Bump R. Evaluating the outcome of surgery for pelvic organ prolapse. Am J Obstet Gynecol 1998;178:877-9. 11. Baden WF, Walker T, Lindsey JH. The vaginal profile. Tex Med 1968;64:56-8. 12. Bump RC, Mattiason 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. 13. Nichols D, Randall C. Massive eversion of the vagina. In: Mitchell C, editor. Vaginal surgery. Baltimore: Williams & Wilkins; 1996. p. 351-83. 14. Karram M, Sze E, Walters M. Surgical treatment of vaginal vault prolapse. In: Karram M, Walters M, editors. Urogynecology and reconstructive pelvic surgery. 2nd ed. St. Louis: Mosby; 1999. p. 235-56. 15. Kupfer C, Goligher J. One hundred consecutive cases of complete prolapse of the rectum treated by operation. Br J Surg 1970;57:481-7. 16. Ihre T, Seligson U. Intussusception of the rectum—internal procidentia: treatment and results in 90 patients. Dis Colon Rectum 1975;18:391-6. 17. Peters W, Smith M, Drescher C. Rectal prolapse in women with other defects of pelvic floor support. Am J Obstet Gynecol 2001;184:1488-95. 18. Parks A, Swash M, Urich H. Sphincter denervation in anorectal incontinence and rectal prolapse. Gut 1977;18:656-65. 19. Swash M, Snooks S, Henry M. Unifying concept of pelvic floor disorders and incontinence. J R Soc Med 1985;78:906-11. 20. Rogers R. Sexual function in women with pelvic floor disorders. Am Urogynecol Soc Q Rep 2002;21:1-3. 21. Beverly C, Walters MD, Weber AM, Piedmonte MR, Ballard LA. Prevalence of hydronephrosis in patients undergoing surgery for pelvic organ prolapse. Obstet Gynecol 1997;90:37-41.
Key words: Colpocleisis, colpectomy, vaginectomy, vaginal hysterectomy, pelvic organ prolapse
Total colpocleisis with levator plication was first described in the English language literature by Edebohls1 in 1901. Colpocleisis, which requires removal of the vaginal epithelium, results in the loss of coital ability and therefore is appropriate for only a minority of patients. For selected patients with advanced and often recurrent prolapse, vaginal obliteration may be preferable to pelvic reconstruction. Although several series have described variations of the colpocleisis procedure resulting in high success rates, outcome reporting has been limited by a lack of reproducible objective and subjective data
From the Female Pelvic Medicine and Reconstructive Surgery Fellowship Program, Department of Obstetrics and Gynecology, Methodist Hospital of Indianapolis/Indiana University. Presented at the Twenty-Third Annual Meeting of the American UroGynecologic Society, San Francisco, Calif, October 6-18, 2002. Received for publication March 23, 2003; accepted March 23, 2003. Reprints not available from the authors. © 2003, Mosby, Inc. All rights reserved. 0002-9378/2003 $30.00 + 0 doi:10.1067/mob.2003.546
points.2-9 In addition, the effect of concurrent hysterectomy on operative morbidity is unknown. The objectives of this study were 2-fold: (1) to report, in accordance with recently established surgical outcome reporting guidelines,10 the long-term objective and subjective outcomes after total colpocleisis with high levator plication and (2) to compare the operative morbidity of total colpocleisis with and without concurrent hysterectomy. Material and methods This project was approved by the Institutional Review Board of Methodist Hospital of Indianapolis. All total colpocleisis operations performed at one institution between August 1988 and December 2000 comprised the study population. The surgical cases were identified by manual review of operative reports contained in all existing patient records of a tertiary care referralbased urogynecology practice. Operations described as partial colpocleisis or LeForte colpocleisis were excluded. Preoperative examination was performed with subjects 121
122 von Pechmann et al
seated in the semiupright position in a birthing chair, and measurements were taken during maximum Valsalva maneuver. Routine evaluation included preoperative urodynamic testing on all patients with advanced pelvic organ prolapse, but this was waived in select cases because of mitigating medical conditions or excessive patient travel time. Preoperative counseling included discussion regarding the loss of coital ability after total colpocleisis and potential effects on body self-image. All operations were performed under the direction of faculty in the Division of Female Pelvic Medicine and Reconstructive Surgery. Surgical technique. Vaginal hysterectomy with or without bilateral adnexectomy was performed in patients with intact uteri. After purse-string closure of the cul-de-sac peritoneum, the complete colpocleisis operation followed. In patients who had previously undergone hysterectomy, there was no attempt to enter and ligate or remove the enterocele sac. Colpocleisis was performed in the following manner: (1) the entire vaginal epithelium was mobilized by sharp dissection, (2) the prolapse was serially reduced with absorbable sutures placed into the vaginal muscularis beginning at the leading edge of the prolapse and continuing sequentially in interrupted purse-string fashion until the prolapsed tissues were superior to the level of the levator plate, (3) a support procedure was performed at the urethrovesical junction (pubourethral ligament plication, needle suspension, or suburethral sling), (4) the puborectalis and pubococcygeus muscles were approximated in the midline with either absorbable or delayed absorbable sutures in interrupted fashion to create a shelf from above the rectum to within one fingerbreadth of the urethrovesical junction, (5) indigo carmine was administered intravenously to cystoscopically confirm ureteral patency, (6) perineal body reconstruction was performed, and (7) the vaginal epithelial edges were trimmed and reapproximated, leaving a foreshortened, narrowed vaginal vault. Outcomes measurement. Data were obtained from three sources: office medical records, hospital records, and standardized telephone contacts. Office medical records were reviewed for demographic data, medical and surgical history, preopoerative and postoperative examination findings, and length of last follow-up. Before 1996, prolapse was graded according to the Baden and Walker system.11 Subsequent to 1996, prolapse was staged by the Pelvic Organ Prolapse-Quantification System of the International Continence Society.12 Objective cure was defined as the absence of prolapse of the anterior or posterior vaginal wall or vaginal apex to the hymen. Hospital charts were reviewed for type of anesthesia, types of concurrent operations performed, operative time, estimated blood loss, change in hematocrit, transfusion occurrence, incidence and types of complications, and length of stay. A designated research nurse who was not involved in direct patient care performed telephone fol-
July 2003 Am J Obstet Gynecol
low-up on available patients during the month of January 2002. Telephone contacts included questions regarding satisfaction with surgical results, recurrence of prolapse, and regret over the loss of coital function. Subjective cure was defined as a response of “very satisfied” or “satisfied” when the patients were asked whether they were pleased with how well the operation cured the prolapse. Statistical analysis. Summary statistics were calculated for each patient group defined by whether the patients underwent previous versus concurrent hysterectomy. A t test was used to compare continuous variables between patient groups. The Fisher exact test was used to test for association between patient group and each categorical variable. Throughout the statistical comparisons, P < .05 was taken as significant. The statistical software SAS V8.02 (SAS Institute, Cary, NC) was used to perform the analysis. Results Patient characteristics. Age, parity, stage of prolapse, and number of previous prolapse surgeries are summarized in Table I. There was no significant difference between patient groups with regard to patient age, parity, marital status, medical history, type of urinary incontinence, stage of prolapse, length of follow-up, and objective or subjective cure rates. Medical comorbidities were common in this elderly population, with 75% having cardiovascular disease, 14% having pulmonary disease, 16% having diabetes, 10% having neurologic disorders, 50% having other medical comorbidities, and only 7% of patients having none. Previous surgery for prolapse was significantly more common in patients who had undergone prior hysterectomy. Of 75 patients who underwent preoperative urodynamic testing, 36 (48.0%) had stress urinary incontinence, 13 (17.3%) had detrusor overactivity, and 16 (21.3%) had mixed urinary incontinence. Operative data. Adjunctive procedures, blood loss and transfusion requirements, and surgical complications are summarized in Table II. Eighty-seven (94.6%) operations were performed under general anesthesia and five (5.4%) under regional anesthesia. Ninety of the 92 patients (97.8%) underwent adjunctive procedures at the time of colpocleisis. Two patients had no urethral support procedures and two others had pubourethral ligament plication performed to enhance distal anterior wall support in addition to another incontinence procedure. The operative time for the colpocleisis procedure could not be determined because of the frequency of adjunctive procedures. Mean total operating room time was 52 minutes longer when hysterectomy was performed concurrently (P < .001). The transfusion rate of both groups together was 21.7%. Complications. Reversible ureteral occlusion was suspected on the basis of intraoperative cystoscopy in one case in the nonhysterectomy group (1.8%) and three
von Pechmann et al 123
Volume 189, Number 1 Am J Obstet Gynecol
cases in the hysterectomy group (8.1%). The single case in the nonhysterectomy group and one case in the hysterectomy group resolved with removal of the most superior levator plication stitches. The second case in the hysterectomy group resolved after removal of the superior levator plication stitches and several purse-string and hemostatic sutures. The third case in the hysterectomy group occurred after removal of ureteral stents and removal of the levator stitches had no effect. A ureteral stent was passed again and achieved efflux of urine after being advanced 1 to 2 cm indicating edema of the intramural ureter as the source of obstruction. An indwelling stent was left in place for the first 2 postoperative weeks, after which resolution was confirmed by intravenous pyelography. Two cases (5.4%) in the hysterectomy group were converted to laparotomy. The first was converted to secure hemostasis because of bleeding from the ovarian vessels at the level of the infundibulopelvic ligament after adnexectomy. The second case was converted because of rupture of a diverticular abscess on cul-de-sac entry. One patient in the nonhysterectomy group sustained a small proctotomy that was recognized and repaired without incident. One patient in the hysterectomy group who had a preoperative diagnosis of metastatic lung cancer died of pulmonary complications 28 days postoperatively. Objective outcomes. Objective outcome data are summarized in Table III. Objective cure, defined as the absence of prolapse to the hymen, was achieved in 90 of 92 patients (97.8%). Median office follow-up time was 12 months (range 0-64 months). Data on postoperative vaginal length was available for 78 subjects; mean length was 2.4 cm (±1.0 cm). Recurrent vaginal prolapse occurred in 2 patients (2.2%). One patient (1.1%) had a recurrent enterocele 7 months postoperatively that was cured with vaginally approached reoperation. The second patient had asymptomatic enlargement of the genital hiatus and anterior wall descent to the hymen 64 months postoperatively, which required no further therapy. Postoperative rectal prolapse occurred in 2 patients (2.2%) within 6 months of surgery. The first patient underwent transanal excision and the second was managed conservatively. Of 46 patients (50%) for whom postoperative standing stress test data was available, 6 patients (13.0%) demonstrated recurrent stress urinary incontinence. Four had suburethral slings with cadaveric fascia, and two had pubourethral ligament plications performed at the time of colpocleisis. Three patients (3.3%) underwent subsequent procedures for treatment of stress urinary incontinence; two patients received periurethral collagen injection and one underwent tension-free vaginal tape suburethral sling placement without difficulty. There were no cases of de novo stress urinary incontinence. Subjective outcomes. Subjective outcome data are presented in Table III. Of the 92 subjects, 8 (8.7%) were deceased and another 13 (14.1%) were unavailable for
Table I. Patient characteristics of 92 women undergoing total colpocleisis Variable
Prior hysterectomy
Subjects (No. [%]) 55 (59.8%) Age (y) 77 (±6) Parity 3 (±2) Stage of prolapse I 0 (0%) II 1 (1.9%) III 28 (51.9%) IV 25 (46.3%) Previous prolapse operations 0 29 (52.7%) 1 17 (30.9%) 2 6 (10.9%) 3 1 (1.8%) 4 2 (3.6%)
Concurrent hysterectomy 37 (40.2%) 79 (±5) 4 (±2)
P value*
.07 .22 .23
0 (0% 0 (0%) 14 (37.8%) 23 (62.2%) <.001 35 (94.6%) 2 (5.4%) 0 (0%) 0 (0%) 0 (0%)
*t test was used to assess differences between groups for age and parity. Fisher exact test was used to determine the association between state of prolapse and hysterectomy and the association between number of previous prolapse operations and hysterectomy.
contact. Sixty-two subjects (67.4%) could be directly contacted by telephone, and information regarding reoperation was available for an additional 9 subjects (9.8%) through indirect contact. Median telephone follow-up time was 24 months (range 13-161 months). Telephone follow-up did not reveal any additional operations for vaginal or rectal prolapse or incontinence. Subjective cure, defined as a response of satisfied or very satisfied with how their surgery cured their prolapse, was achieved in 56 of 62 patients (90.3%). Seventy-one percent of those directly contacted by telephone reported being very satisfied or satisfied with their bladder function postoperatively. Regret over loss of coital ability occurred to some extent in 8 subjects (12.9%); 4 of the 8 stated that they would make the decision to have the colpocleisis procedure again, 3 were uncertain, and 1 would not. There was no significant difference in age at the time of surgery or number of previous prolapse surgeries between subjects who expressed regret and those who did not. Comment Review of the English language literature reveals several studies describing total “colpocleisis,” “colpectomy,” and “vaginectomy” for the treatment of advanced vaginal or uterovaginal prolapse. Reports from the first half of the 20th century were primarily descriptions of technique with little emphasis on outcomes. In 1901, Edebohls1 reported 4 cases of “panhysterokolpectomy” in which he described vaginal hysterectomy, vaginectomy, and purse-string reduction of the remaining viscera. Phaneuf,2 in 1935, reported another 5 cases, adding levator plication to the procedure. In 1938, Masson and Knepper3 reported 23 cases of vaginectomy, some with hysterectomy, followed by purse-string reduction and levator
124 von Pechmann et al
July 2003 Am J Obstet Gynecol
Table II. Operative data in 92 patients undergoing colpocleisis
Variable
Prior hysterectomy (n = 55)
Adjunctive procedures Incontinence operations Pubourethral 23 (41.8%) ligament plication Needle suspension 3 (5.5%) Suburethral sling 26 (47.3%) Tension-free 3 (5.5%) vaginal tape Other procedures 8 (14.5%) Estimated blood 396 (±221) loss (mL) Change in 9.5 hematocrit (%)† Transfusion 7 (12.7%) Ureteral occlusion 1 (1.8%) Laparotomy 0 (0%)
Concurrent hysterectomy (n = 37) P value*
10 (27.0%)
.19
4 (10.8%) 21 (56.8%) 2 (5.4%)
.43 .40 1.00
5 (13.5%) 578 (±494)
1.00 .04
11.9
.01
13 (35.1%) 3 (8.1%) 2 (5.4%)
.02 .30 .16
*t test used for variables represented as mean (±SD). Fisher exact test used for variables in italic. †Change in hematocrit refers to actual hematocrit values, not change expressed as a percentage.
plication with the added modification of perineorrhaphy. In 1950, Williams4 reported “about sixty cases” of vaginal hysterectomy with vaginectomy and levator plication but no suture reduction. Although no recurrences were reported in these early series, outcomes were typically described in one or two sentences as “satisfactory.” Studies from the latter half of the century placed increasing emphasis on outcomes reporting but with no specific criteria regarding what constituted cure or recurrence. In 1951, Adams5 reported a series of 30 cases, 11 with concurrent hysterectomy, in which he performed purse-string suture reduction followed by urethral and levator plication and perineorrhaphy. He reported no recurrences with a median follow-up of approximately 5 years. In 1960, Symmonds and Pratt6 reported 10 cases of vaginectomy with levator plication with a greater than 5year median follow-up. In the 9 patients available for follow-up, 2 patients required reoperation: 1 for recurrent vaginal prolapse and another for treatment of bilateral hernias into the ischiorectal fossa. A third patient expressed regret over the loss of coital ability. In 1997, DeLancey and Morley 8 reported on a series of 33 patients with posthysterectomy prolapse who underwent vaginectomy, suture reduction, and levator plication. There was one case of recurrence identified after a mean objective follow-up time of 7.7 months and mean subjective followup of 34.6 months. These authors were the first to report the use of a standardized questionnaire regarding recurrence and subjective outcomes. All women but one denied remorse over loss of coital ability; one woman said she had “accepted” her sexual inactivity. In 2001, Grody et al9 identified no recurrences on follow-up survey in a
series of 17 cases of vaginectomy, suture reduction, “vaginal paravaginal cystourethropexy,” and perineorrhaphy with 5 cases of concurrent hysterectomy. Using defined outcome criteria, this study found a 97.8% objective cure in 92 cases, confirming the high success rates reported by previous investigators. With objective cure defined as absence of prolapse to the hymen, two patients had recurrence. One of these was an asymptomatic prolapse of the anterior wall to the hymen and the patient remained “very satisfied.” The second subject had undergone previous abdominally approached paravaginal defect repair and had persistent anterior wall support at the time of her colpocleisis. She subsequently had a symptomatic recurrent enterocele that was cured with vaginally approached ligation of the enterocele sac. Some investigators have noted that enterocele is thought to be the most likely form of recurrence and have therefore recommended routine entry into the enterocele sac to achieve excision and ligation for this reason.13 Others believe that the theoretical benefits do not warrant the increased risk of ileus and other complications.8 It is noteworthy that one of the two cases in this series converted to laparotomy involved rupture of a diverticular abscess after peritoneal entry for hysterectomy. The risk of such a complication may well supercede the risk of recurrence from leaving the enterocele sac intact. Several previous studies have reported postoperative stress urinary incontinence after both total4,6,7,9 and partial colpocleisis.14 Because of these reports as well as previous faculty experience, 90 patients (97.8%) received some form of urethral support procedure. Suburethral slings were typically performed only in patients with urodynamically confirmed stress urinary incontinence. Patients with marked preoperative voiding dysfunction typically underwent less obstructive procedures such as pubourethral ligament plication and two patients had no urethral support procedures performed to minimize postoperative voiding difficulties. Determining the true incidence of postoperative stress urinary incontinence would require routine postoperative urodynamic evaluation and/or standardized subjective assessments, which were not performed in this study. On the basis of the 46 (50%) cases in whom postoperative standing stress test data were available, there was a 13.0% incidence of recurrent stress urinary incontinence but no cases of de novo stress urinary incontinence. Standing stress testing was performed at variable bladder volumes, which may have resulted in underreporting. When necessary, reoperation for urinary incontinence is not precluded by the total colpocleisis procedure. Further investigation of the problem of postcolpocleisis stress urinary incontinence is clearly warranted. The development of postoperative rectal prolapse, as occurred in two subjects in this study, has not been reported previously. It is unclear to what extent colpocleisis may play a contributing role, but several factors should be
von Pechmann et al 125
Volume 189, Number 1 Am J Obstet Gynecol
considered. This study population has several risk factors for the development of rectal prolapse, including female sex, advanced age, and, in some cases, chronic constipation.15,16 Moreover, the frequent coexistence of rectal and genital prolapse and their association with pelvic floor neuropathy is well established.17-19 Colpocleisis with levator plication uses the dysfunctional pelvic floor musculature as a physical barrier to visceral descent by narrowing the levator hiatus. Consequently, intra-abdominal pressure finds the path of least resistance in the remaining anorectal hiatus. Further studies are needed to determine the strength of the association between colpocleisis and rectal prolapse, but screening for rectal prolapse should be considered in patients undergoing this procedure. Among the 62 subjects available for phone follow-up, satisfaction with how the surgery cured the prolapse was greater than 90%, consistent with the high objective cure rate. Because this was a retrospective study, validated psychometric instruments designed to assess the impact of surgery on subjective parameters were not used. In addition, the structure of the telephone questionnaire did not allow for control of confounders such as bowel or bladder dysfunction when patients were questioned specifically regarding cure of the prolapse. Nonetheless, several measures were taken to minimize bias in obtaining subjective outcome data. A nurse not involved in patient care administered all questionnaires, which were structured to provide subjects with a range of responses. Among the 8 patients who expressed regret, there was variability in both the degree of regret expressed over loss of coital ability and over the decision to have the surgery. This illustrates a spectrum of regret that may not have been captured in previous studies that reported lower regret rates. In addition, subjects who were available to complete the survey may have been healthier as a group than their counterparts and therefore more prone to regret, resulting in overreporting due to selection bias. Ridley7 has stated that colpocleisis should be “reserved for those few well-individualized patients in whom sexual function is distinctly not a factor and the problems of vaginal prolapse are primary.” Indeed, all subjects in this study fell into this category. However, because one third of patients over age 78 years remain sexually active,20 the importance of adequate preoperative screening and counseling cannot be overemphasized. It may be the case that women suffering the effects of massive vaginal prolapse obtain a different perspective regarding sexual activity after they have obtained resolution of their symptoms. Our experience has been that many women will resume the social activities they had avoided preoperatively and may unexpectedly develop new relationships. All these factors need to be explored during preoperative counseling. In addition, it is important that the patient understand that loss of coital ability does not preclude other forms of sexual activity, stimulation, or orgasm for
Table III. Postoperative outcomes of 92 patients undergoing total colpocleisis Objective cure Subjective cure* Very satisfied Satisfied Somewhat satisfied Not satisfied Regret over loss of coital ability† Never Somewhat Yes Unable to answer
90/92 (97.8%) 38/62 (61.3%) 18/62 (29.0%) 3/62 (4.8%) 3/62 (4.8%) 52/62 (83.9%) 6/62 (9.7%) 2/62 (3.2%) 2/62 (3.2%)
*In response to the question, “Are you satisfied with how well the surgery repaired the dropped organs?” †In response to the question, “Have you ever regretted losing the ability to have intercourse?”
both the patient and her partner. This study demonstrates that preoperative counseling is not universally successful in preventing postoperative regret. Although this finding may lead to the temptation to omit colpocleisis as an option even for patients who meet Ridley’s selection criteria, concerns about patient regret need to be balanced with respect for the patient autonomy implicit in informed consent. Information regarding operative morbidity is scarce in previous studies. Only two studies have reported data regarding blood loss, and neither of these provided preoperative and postoperative laboratory values.8,9 The mean estimated blood loss in this study was higher than what has been previously reported8 and is likely accounted for by the frequency of concurrent procedures. Ninety of 92 subjects underwent bladder neck support procedures, more than half of which were suburethral slings, and 13 subjects underwent additional adjunctive procedures. The 21.7% overall transfusion rate in this study is consistent with the 29.4% rate reported in the one previous study that reported transfusion data.9 We think that these transfusion rates may simply reflect a low threshold for transfusion in elderly patients with frequent medical comorbidities. Because the difference in change in hematocrit was only 2.4% greater in the hysterectomy group, the nearly 3-fold difference in transfusion requirements is difficult to explain. This may have occurred partly as a result of the added duration of the procedure with hysterectomy, which often led to the decision to transfuse intraoperatively. Although significant, the blood loss incurred during colpocleisis is typically gradual and easily controlled, producing less stress on a weakened myocardium than the type of acute hemorrhage that can occur during reconstructive procedures such as sacral colpopexy or sacrospinous ligament fixation.8 Reversible ureteral obstruction occurred more frequently and conversion to laparotomy occurred exclusively in the concurrent hysterectomy group. Ureteral obstruction during colpocleisis has not been described
126 von Pechmann et al
previously. The prevalence of hydronephrosis has been shown to increase with advancing severity of pelvic organ prolapse.21 Upper urinary tract imaging studies were not routinely obtained, but the same pathophysiologic mechanisms that lead to hydronephrosis may render patients with advanced pelvic organ prolapse more vulnerable to ureteral compromise at the time of surgery. All cases were reversed without major interventions, confirming the importance of intraoperative cystoscopy. The only case in which ureteral stents were placed preoperatively was complicated by loss of ureteral flow after stent removal. This occurred early in the series, after which preoperative stent placement was abandoned. Of the two cases converted to laparotomy, both were directly related to hysterectomy. Vaginal hysterectomy and total colpocleisis were chosen in the patients with intact uteri because of the higher reported success rates with total as opposed to LeForte colpocleisis.14 Nonetheless, LeForte colpocleisis should be considered as a potentially less morbid alternative in patients who have not previously undergone hysterectomy. This case series has several limitations as a result of its retrospective nature. Postoperative prolapse staging was generally recorded by the primary surgeons and was, therefore, subject to observer bias. The clinical outcomes with respect to incontinence were not uniformly or quantitatively measured, preventing conclusions regarding the incidence of postoperative stress urinary incontinence. Baseline preoperative data that would allow evaluation of the effect of surgery on quality of life using a validated psychometric instrument were not available. Consistently recorded racial data on this cohort of patients was not available, although our patient population is predominantly white. It may be that candidacy rates and outcomes would be different in populations of different racial compositions. This study also has several strengths. This is the largest series to date describing outcomes from total colpocleisis for advanced prolapse. The data are reported according to recently established guidelines for surgical outcome studies, including seven of the eight domains of information recommended by the American Urogynecologic Society.10 The definitions of objective and subjective cure were explicitly stated and subjective outcomes were determined by the use of a standardized questionnaire including questions regarding regret over the loss of coital ability. This is also the first study to assess differences in operative morbidity from colpocleisis with or without concurrent hysterectomy. The finding that concurrent hysterectomy is associated with increased blood loss and transfusion requirements may affect how patients are
July 2003 Am J Obstet Gynecol
counseled regarding the risks and benefits of the various surgical options for treatment of pelvic organ prolapse. In summary, total colpocleisis with high levator plication is an effective operation for the treatment of advanced pelvic organ prolapse in patients who do not anticipate future sexual activity. This study could not have been completed without the gracious assistance of Marty Morris, RN, and Elizabeth Fuller, MSN, RN, CNS. REFERENCES
1. Edebohls GM. Panhysterokolpectomy: a new prolapsus operation. Med Rec N Y 1901;60:561-4. 2. Phaneuf TE. The place of colpectomy in the treatment of uterine and vaginal prolapse. Am J Obstet Gynecol 1935;30:544-53. 3. Masson JC, Knepper PA. Vaginectomy. Am J Obstet Gynecol 1938;36:94-9. 4. Williams JT. Vaginal hysterectomy and colpectomy for prolapse of the uterus and bladder. Am J Obstet Gynecol 1950;59:365-70. 5. Adams HD. Total colpocleisis for pelvic eventration. Surg Gynecol Obstet 1951;2:321-4. 6. Symmonds RE, Pratt JH. Vaginal prolapse following hysterectomy. Am J Obstet Gynecol 1960;79:899-909. 7. Ridley JH. Evaluation of the colpocleisis operation: a report of fifty-eight cases. Am J Obstet Gynecol 1972;113:1114-8. 8. DeLancey JOL, Morley GW. Total colpocleisis for vaginal eversion. Am J Obstet Gynecol 1997;176:1228-32. 9. Grody MHT, Merchia V, Nyirjesy P, Kaplan E, Chatwani J. Total colpocleisis: a prospective study. J Pelvic Surg 2001;7:72-8. 10. Wall LL, Versi E, Norton P, Bump R. Evaluating the outcome of surgery for pelvic organ prolapse. Am J Obstet Gynecol 1998;178:877-9. 11. Baden WF, Walker T, Lindsey JH. The vaginal profile. Tex Med 1968;64:56-8. 12. Bump RC, Mattiason 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. 13. Nichols D, Randall C. Massive eversion of the vagina. In: Mitchell C, editor. Vaginal surgery. Baltimore: Williams & Wilkins; 1996. p. 351-83. 14. Karram M, Sze E, Walters M. Surgical treatment of vaginal vault prolapse. In: Karram M, Walters M, editors. Urogynecology and reconstructive pelvic surgery. 2nd ed. St. Louis: Mosby; 1999. p. 235-56. 15. Kupfer C, Goligher J. One hundred consecutive cases of complete prolapse of the rectum treated by operation. Br J Surg 1970;57:481-7. 16. Ihre T, Seligson U. Intussusception of the rectum—internal procidentia: treatment and results in 90 patients. Dis Colon Rectum 1975;18:391-6. 17. Peters W, Smith M, Drescher C. Rectal prolapse in women with other defects of pelvic floor support. Am J Obstet Gynecol 2001;184:1488-95. 18. Parks A, Swash M, Urich H. Sphincter denervation in anorectal incontinence and rectal prolapse. Gut 1977;18:656-65. 19. Swash M, Snooks S, Henry M. Unifying concept of pelvic floor disorders and incontinence. J R Soc Med 1985;78:906-11. 20. Rogers R. Sexual function in women with pelvic floor disorders. Am Urogynecol Soc Q Rep 2002;21:1-3. 21. Beverly C, Walters MD, Weber AM, Piedmonte MR, Ballard LA. Prevalence of hydronephrosis in patients undergoing surgery for pelvic organ prolapse. Obstet Gynecol 1997;90:37-41.