Introduction

It is estimated that approximately 50% of women who are alive today at the age of 50 will live into their 90s [1]. In this large group of older women more patients with vaginal or uterine prolapse will need surgery in the future as vaginal and uterine prolapse render the patients unnecessarily immobile and bothered. It is of utmost importance to help this group of women even though many of these patients because of concomitant diseases are not suited for more extensive surgery. For some decades colpectomy and colpocleisis have been considered obsolete and considered associated with a high incidence of de novo stress urinary incontinence. However, in recent years these operations have been reevaluated and are again being performed by gynecological surgeons. The purpose of this study was to report objective and subjective outcome data with special emphasis on urinary incontinence and patient satisfaction.

Patients and methods

Forty-two patients with vaginal or uterovaginal prolapse were prospectively evaluated during the period from October 1997 to September 2002. All patients were examined, operated, and followed up by only two surgeons.

Prior to operation a patient history was obtained and gynecological examination performed. All patients were asked about their present sexual life and the possibility of sexual life in the future. If the patients were in any way in doubt about the possibility of sexual intercourse in the future, a colpectomy or colpocleisis was not performed. The main complaint of all women was prolapse but all patients were asked about their present state of continence. Further investigations with frequency-volume charts and urodynamics were not performed because many of the women were very old and could not comply with the investigations. Patients with a uterus were treated with colpocleisis and those with prior hysterectomy with colpectomy. The reason for performing a colpocleisis on patients with a uterus was to be able to detect any bleeding from the uterus at a later stage. Postoperative prolapse in the lateral channels has been described, which is why these channels should be avoided when unnecessary. Furthermore, purse-string sutures are somewhat faster to perform than rows of interrupted sutures.

The patients were followed up 3 months after surgery and a standardized telephone interview was performed in September 2003. We asked the patients three questions: “Are you satisfied with your operation and if not, why not?” “Do you ever experience incontinence and if yes, when?” “Do you think there is a recurrence of your prolapse?”

A total colpectomy is defined as an excision of the vaginal lining whereas colpocleisis refers to closure of the vaginal vault. In order to perform a colpectomy or colpocleisis, a protrusion of the anterior vaginal wall, vaginal top, and posterior vaginal wall must be present.

Surgical procedure in total colpectomy

The operation can be performed under general, regional, or local anesthesia. A circumscribing incision is made through the vaginal mucosa. Anteriorly, the incision is made at a minimum of 3 cm from the urethral meatus in such a way that there is no interference with the urethra. Posteriorly, the incision is made 1–0.5 cm from the introitus and laterally to the level of the hymenal ring. Two midline incisions and two lateral incisions can be made, and total excision of the vaginal mucosa is performed using sharp and blunt dissection as close to the vaginal mucosa as possible. Two marking sutures are placed at the top of the vault.

The raw surface of the denuded vaginal prolapse is reduced into the pelvis using purse-string sutures with absorbable material with approximately 1 cm distance starting at the top of the vault where the marking sutures are placed. Bit by bit the protruding vagina is pushed up into the pelvis before the purse-string suture is tied. The sutures are put through the endopelvic fascia, but in many cases this fascia is very thin and might be almost impossible to identify. If an enterocele is present, no effort is made to open the enterocele sac. Finally, the introital flaps are approximated with a row of horizontal sutures using absorbable sutures.

Surgical procedure in colpocleisis

The operation can be performed under general, regional, or local anesthesia. A fractionated abrasio is performed in connection with the operation. We perform the colpocleisis ad modum Le Fort as described in 1825. The procedure is used in patients with their uterus intact and is preferred to colpectomy because discharge from the uterus can be expelled. Rectangular strips of vaginal mucosa of approximately equal size are removed from the anterior and posterior surfaces of the protruding vagina leaving a canal of approximately 3 cm in each side. Again care is taken not to remove vaginal mucosa from the area beneath the urethra. The anterior and posterior surfaces are approximated to each other with rows of interrupted stitches resulting in a lateral tunnel in each side through which any discharge from the uterus is able to pass.

Results

Patient characteristics

The mean age of the patients at operation was 79 years (range: 62–101). Colpocleisis was performed in 25 patients and colpectomy in 17.

In the colpectomy group 11 patients had a prior abdominal hysterectomy and 6 a prior vaginal hysterectomy. Hysterectomy was performed a mean of 20.4 years (range: 7–51) (median 15 years) before the colpectomy in the abdominal hysterectomy group and a mean of 14.7 years (range: 3–26) (median 17.5 years) in the vaginal hysterectomy group.

Twenty patients had a prior prolapse operation: 15 had one, 4 had two, and 1 five prior prolapse operations; 16 patients had no prior operation.

Operative data and complications

The patients were hospitalized for a mean of 2 days (range: 0–6). One patient in the colpocleisis group experienced postoperative bleeding and resuturation was performed a few hours after the operation. One patient in the colpocleisis group experienced severe vaginal discharge for 2 weeks after the operation. No other complications were observed.

Objective outcomes

One patient lived very far away and did not attend the control after 3 months and could not since be contacted by telephone. Forty-one patients attended the clinical follow-up after 3 months. No recurrence of the vaginal or uterovaginal prolapse was seen after 3 months, but three patients had a small asymptomatic low rectocele.

There are missing data on incontinence after the operation on 1 of the 41 patients who attended the follow-up visit after 3 months. Seventeen patients experienced incontinence before the operation and 23 no incontinence. Three months after the operation 12 patients still experienced incontinence and in 5 patients the incontinence had disappeared. There were no cases of de novo incontinence. Among the 12 patients with persisting incontinence after the operation, 7 found their incontinence unchanged, 4 improved, and 1 worsened. No attempt was made to discriminate between stress and urge incontinence. One patient had a permanent catheter because of retention for several years before the operation. After the operation she no longer needed a catheter. She could empty her bladder and was dry.

The telephone interview was performed a mean of 46 months (range: 12–91) after the operation. Six patients had died of causes unrelated to the operation. Seven patients could not be contacted. Twenty-nine patients answered the telephone interview. Twenty-six patients (90%) stated that they were satisfied with the operation. Three patients stated that they were not entirely satisfied with the operation. One patient said it was because of the small rectocele even though she had no symptoms from it. One patient still experienced some descensus of the vagina but not beyond the introitus, and one patient complained of pain in the lower abdomen but no recurrence. One of the three patients with a small rectocele 3 months after the operation at the time of the telephone interview stated that she was satisfied with no recurrence and the third patient with a small rectocele after the operation could not be contacted.

Eleven patients were incontinent at the time of the telephone interview and 18 patients were not. One patient was incontinent after 3 months and was now dry. Five patients had become incontinent since the 3-month control. One patient had become very dement and was now in a nursing home (the staff answered the questionnaire), one patient experienced urge incontinence now and then, and three patients experienced incontinence when coughing, but stated that it was no problem.

Discussion

The primary underlying causes of uterovaginal or vaginal prolapse are related to weakness in the supporting connective tissue of the pelvic structures aggravated by multiparity, obstetric trauma, prior surgery, and chronically increased abdominal pressure caused by obesity, chronic cough, or heavy lifting [1].

In the colpectomy group we observed that the prolapse occurred many years after the hysterectomy. There seemed to be a shorter span of years between the hysterectomy and the prolapse in the vaginal hysterectomy group (14.7 years) than in the abdominal hysterectomy group (20.4 years), but this might be due to the fact that many of the vaginal hysterectomy patients already had some vaginal prolapse at the time of the operation.

Prolapse of the vaginal vault is a condition that greatly affects the quality of life of many elderly women. The patients have bloody, smelling discharge from the prolapse and often problems emptying the bladder and rectum. Many women have problems walking and are sometimes bedridden because of the prolapse. Because of associated medical problems, these patients are often poor candidates for operations such as sacrospinous vaginal vault suspension operations, sacrouterine site-specific suspension operations, or abdominal sacropexies.

Postoperative infection after colpectomy or colpocleisis is rarely seen. Kohli et al. [2] described a previously unreported complication in a patient undergoing an uncomplicated Le Fort colpocleisis. The patient returned 8 months after the operation with an enlarged uterus filled with purulent material that necessitated subsequent hysterectomy. No cases of infection occurred in the present study. A rare case of late development of prolapse of the small intestine via the vagina despite intact colpocleisis 3 years after has also been reported [3].

Denehy et al. [4] compared Le Fort colpocleisis (21 patients) with vaginal hysterectomy, anterior colporrhaphy, and posterior colpoperineoplasty (42 patients). Patients operated by colpocleisis were older and the operation time was halved. Kelly plication was performed in all women in the colpocleisis group. Any enterocele sac was excised. It is stated that urologic symptoms if present were ameliorated. It is not stated whether any de novo incontinence was observed. One recurrence in the colpocleisis group was observed.

In order to avoid de novo stress incontinence after the operation, Langmade et al. [5] performed a partial colpocleisis in which the posterior repair was done in such a way as to form a pencil-sized vagina. In 102 cases no recurrence was found and no patients had de novo incontinence. Two patients with preoperative stress incontinence were unchanged after the operation. Operation time was an average of 78 min and the operation was clearly more extensive than colpocleisis and colpectomy. Coitus was still not possible.

Barclay [6] reported on 85 patients who underwent colpectomy. He stressed the importance of avoiding removal of the vaginal mucosa under the urethrovesical angle. There was no control of the patients and therefore no comments on postoperative incontinence.

DeLancey et al. [7] reported on 33 patients with colpectomy. They found the operation simple, safe, and effective. The enterocele sac was not opened or repaired as it was recognized that the colpocleisis would close the potential space into which the enterocele might protrude. One patient had recurrence of the prolapse and a repeat operation was performed. Three women had incontinence that developed gradually over the years but none shortly after surgery. Five women had stress incontinence before the operation and 12 poor urethral support. In these 17 women, 14 patients had fascial plication under the urethra, 2 a needle suspension, and 1 a pubovaginal sling. We agree with DeLancey that the operation completely obliterates the vaginal canal and that there is no place for the enterocele to protrude below the pelvic floor. Therefore, we never opened and removed the enterocele. We did not perform any suspension operations on the urethra and did not find any de novo incontinence after the operation. We find it of importance to avoid operation in the area under the urethra in order not to compromise the nerves and tissue in this area and in order not to pull the urethra down. We realize, however, that this is still an unsolved question just like the question about unmasked incontinence in patients with anterior colporrhaphy. Further studies are required to solve this issue.

Some surgeons advocate a high perineorrhaphy in all patients to reduce recurrence. We only performed perineorrhaphy in cases with a very low or missing perineum. We did not find any cases of recurrence of the vaginal prolapse, but we did find three cases of low nonsymptomatic rectocele. A perineorrhaphy, however, does not prevent or cure a rectocele.

Pechmann et al. [8] in a study on 92 subjects performed hysterectomy before colpectomy on all patients with uterovaginal prolapse and found higher blood loss and transfusion requirements in the patients who had hysterectomy. They also performed a telephone interview with 62 patients and 90.3% were satisfied or very satisfied. We found the same results without having to perform a hysterectomy.

We performed colpectomy and colpocleisis in a very old age group. We found colpectomy or colpocleisis to be an effective operation in selected old patients and no risk of de novo incontinence when care was taken not to involve the area beneath the urethra and urethrovesical junction. The incidence of urinary incontinence rises with age especially after the age of 70, which might be the reason for more women being incontinent at the time of the telephone interview than at the 3-month examination.