Int Urogynecol J (2008) 19:1593–1601 DOI 10.1007/s00192-008-0702-z

ORIGINAL ARTICLE

A 5-year prospective follow-up study of vaginal surgery for pelvic organ prolapse Ann Miedel & Gunilla Tegerstedt & Birgitta Mörlin & Margareta Hammarström

Received: 9 April 2008 / Revised: 3 June 2008 / Accepted: 19 July 2008 / Published online: 12 August 2008 # The International Urogynecological Association 2008

Abstract The objective of this study was to evaluate anatomic, functional, short- and long-term outcome of vaginal surgery for pelvic organ prolapse. This was a prospective observational study of 185 consecutive women planned for vaginal prolapse reconstructive surgery. Stage of prolapse, urinary incontinence (UI), bowel and mechanical symptoms were assessed preoperatively and at 1, 3 and 5 years postoperatively. The mean follow-up time was 53 months. The anatomic recurrence rate was 41.1% but less than half of them were symptomatic. Anterior compartment was most prone for recurrence and the majority of the recurrences took place within the first year. UI remained at the same level at 1-year follow-up. De novo urge occurred in 22.6% and de novo stress incontinence in 6.0%. An improvement was seen in difficulty in emptying bowel 1 year after surgery (54%). Patients were primarily cured from mechanical symptoms. Re-operation rate was 9.7%; if additional operation for incontinence was included, it was13.5%. Keywords Pelvic organ prolapse . Prospective study . Prolapse reconstructive surgery . Urinary incontinence A. Miedel (*) : G. Tegerstedt : M. Hammarström Department of Clinical Science and Education, Södersjukhuset, Section of Obstetrics and Gynaecology, Karolinska Institutet, 118 83 Stockholm, Sweden e-mail: [email protected] B. Mörlin Division of Obstetrics and Gynaecology, Department of Clinical Science, Karolinska Institutet Danderyds Hospital, Stockholm, Sweden

Introduction Pelvic organ prolapse is a common condition and a frequent indication for surgery [1, 2]. The preferred route for most prolapse surgery is vaginal [3, 4]and anterior and posterior colporrhaphy, dating back to the early nineteenth century, are among the most frequently, and still primarily, performed operations in gynaecological surgery [3] but there are surprisingly few prospective long-term data in the literature. The high rate of anatomical recurrence in prolapse surgery is well known and especially recurrent anterior wall prolapse [5, 6].This has during the last decades led to an alteration of surgical approach such as site-specific repair, paravaginal repair and introduction of mesh [4, 7]. The wide variety of surgical treatments available for prolapse indicates the lack of consensus to the optimal surgical treatment [7]. The surgical treatment of prolapse should aim at restoration of normal pelvic anatomy, reestablishment or maintenance of normal urinary, rectal and sexual functions and if possible not causing the woman any adverse effects. A woman with pelvic organ prolapse often presents with additional pelvic floor symptoms such as symptoms from bladder and bowel. Whether they are directly related to the prolapse can be difficult to evaluate. Studies addressing functional symptoms prior and after prolapse surgery are scarce. The purpose of this study was to prospectively study a group of women who underwent vaginal prolapse reconstructive surgery, in terms of anatomic and functional outcome, recurrence rate and long-term side effects

1594

Materials and methods Between Jan 1998 and Jan 2001, all patients planned for surgical treatment of symptomatic pelvic organ prolapse at Söder Hospital received information about the study together with planned schedule for operation. All patients were included who agreed to participate in the study. The women gave their approval to a nurse before the preoperative examination. Exclusion criteria included inability to answer questionnaire, dementia or other severe illness. A predefined protocol constructed for the study was used prior to surgery and included questions about obstetric and medical history, stress urinary incontinence (SUI), urge urinary incontinence, difficulty in bladder emptying, bowel function (constipation–evacuatory dysfunction, faecal incontinence and gas incontinence), urinary infections, smoking and estrogens. The questionnaire recorded whether the symptom was present (‘yes’ or ‘no’). No clinical examination was made for type of incontinence. The severity of the prolapse was graded according to Beecham system [8] where the classification was done in three degrees in relation to introitus at rest. Other variables examined were age, height, weight and chronic diseases such as diabetes mellitus, chronic lung disease, neurological disease and chronic lower back pain. Women with recurrent prolapse were not excluded. Depending on the location of the prolapse, surgery involved different types of repairs, all with a vaginal approach. Anterior repair was done by midline plication technique in all cases with Kelly plication for bladder neck stabilisation. For the posterior colporrhaphy, a midline incision in combination with levator muscle plication was performed The responsible surgeon decided whether antiincontinence procedures (tension-free vaginal tape (TVT)) were to be performed in combination with prolapse repair. Mesh (synthetic or biologic) was used primarily in one single case due to recurrent prolapse and in nine re-operations during follow-up at the discretion of the gynaecological surgeon. Perioperative morbidity was defined as haemorrhage requiring blood transfusion or re-operation, deep venous thrombosis or febrile morbidity. Postoperative evaluations were conducted 6–8 weeks after surgery and after 1, 3 and 5 years, respectively. All visits followed the same protocol with the same questions concerning urinary incontinence and bowel symptoms. Anatomic outcome was determined by vaginal examination. Once the subjects were diagnosed as failure, they did not re-enter as a new case but symptoms were recorded at the remaining follow-ups. The implementation of the Pelvic Organ Prolapse Quantification (POPQ) system as the standard for describing prolapse resulted in a graduate adaptation and at the 5-year follow-up objective failure was defined according to POPQ stage ≥2 at any site. Symptomatic prolapse was defined as mechanical symptoms such as feeling of vaginal bulge [4].

Int Urogynecol J (2008) 19:1593–1601

At the 5-year follow-up, the women also completed a general Quality of Life questionnaire, the 36-item Medical Outcomes Study Short-Form Health Survey (SF36) [9].The SF36 yields an eight-scale profile of scores and is a generic measure of health status in opposition to others who targets specific diseases, ages or treatments. It thus can be useful to compare general and specific populations. An age-matched reference group (n=62) was randomly selected from the Swedish SF36 database. The first author (AM), who had not performed any of the prolapse surgery, completed all 5-year follow-up examinations and interviews. Statistical analyses were performed with SPSS 14.0 and odds ratio with 95% confidence interval were used to test differences in recurrences between groups. The study was approved by the Ethics Committee South at Karolinska Institutet. All subjects received written information before enrolment.

Results During the study period, a total of 248 women underwent prolapse surgery at the department and 185 women chose to participate. The women who chose not to participate were on average the same age (66.2 vs. 65.4). Baseline data Table 1 lists the baseline characteristics of the study population. All of the participating women had at least one vaginal birth. Of the 27 women who had undergone hysterectomy, only one case had previously been performed on prolapse indication; all the others were abdominal procedures due to fibroids and/or bleedings. In total, 22.2% (41/185) women had previous surgery for urinary incontinence (UI) or pelvic organ prolapse. The women presented with anatomic defects in different compartments, isolated or more often in combination. A protrusion of the prolapse beyond the introitus was present in 37.3%. Prolapse involving the anterior wall was the predominant finding (76.7%). Only 23.8% (n=44) were classified as having a defect in one compartment which mostly was the posterior wall. Table 2 includes preoperative symptoms among all women. The majority of the women complained of mechanical symptoms and pessaries were used by 40.6% (75/185). Urinary incontinence was present in 49.2%, equally presenting as urge, stress or mixed incontinence. Bowel symptoms were found among 42.7% of the women. The procedures performed are shown in Table 3. Seventeen gynaecologists performed the operations, where 57% of the operations were performed by three gynaecologists with a special interest in urogynaecology. A posterior vaginal

Int Urogynecol J (2008) 19:1593–1601

1595

Table 1 Baseline characteristics of the 185 women who underwent vaginal surgery for pelvic organ prolapse Characteristics

Number

Age, mean (SD) [range] Parity, mean (range) BMI, mean (range) Years after menopause mean (range) Estrogen replacement therapy, n/total (%) Local estrogen, n/total (%) Previous operations, n/total (%) Hysterectomy Incontinence Prolapse History, n/total (%) Chronic lung disease Diabetes mellitus Neurological disease Vaginal examination, n/total (%) Cystocele First degree Second degree Not staged Rectocele First degree Second degree Uterusdescens–uterusprolaps No. of sites involved (%) 1 2 3 Surgical procedures (%) Manchester procedure Vaginal hysterectomy With anterior and posterior colporrhaphy Anterior or posterior colporrhaphy Anterior colporrhaphy Posterior colporrhaphy Anterior + posterior colporrhaphy Cervix amputation Colpocleisis TVT

65.4 (13.3) [32–89] 2.4 (0–15) 25.5 (19–38) 18 (0–50) 57 (31.0) 80 (43.7) 27 (14.6) 21 (11.4) 24 (13.0) 18 (9.7) 13 (7.0) 12 (6.5)

postoperatively due to difficulty in emptying bladder; both of them had undergone concomitant incontinence surgery. Urinary infection occurred in 6.1% (n=12), febrile morbidity treated with antibiotics in 2.7% (n=5), vaginal haematoma in 2.2% (n=4) and there were two re-operations (1.1%) for surgical complications (bleeding and abscess). None of the patients required blood transfusion. Postoperative follow-ups One hundred seventy-two women (93.0%) were evaluated 1 year after surgery and 151 (81.6%) at the 3-year and 143 (77.3%) at the 5-year follow-up. The mean follow-up time in the whole group was 53.2 months and 123 women (66.5%) attended all four follow-ups. Anatomical recurrences

142 (76.7) 68 (36.7) 53 (28.6) 21 (11.3) 69 (37.3) 22 (11.9) 84 (45.4) 44 (23.8) 72 (38.9) 69 (37.3) 74 (40.0) 36 (19.5) 30 5 7 (3.8) 36 (19.5) 25 (8.7) 2 (1) 4 (2.2) 32 (17.3)

Vaginal support defects are described according to Beecham preoperatively.

wall prolapse was diagnosed in less than half of the women but almost all of the women (91.9%) underwent a posterior colporrhaphy as part of the prolapse surgery according to clinical routine at the time. Over half of the operations (56.0%; 104/185) addressed all three compartments. There were 32 (17.3%) concomitant TVT procedures. Three women had vault prolapse; two underwent colpocleisis and one woman underwent anterior and posterior colporrhaphy with a synthetic mesh (Prolene). Complications There were no serious perioperative or postoperative complications. Two patients required a suprapubic catheter

For the anatomical recurrences, for patients whose 5-year follow-up data were not available, we inputted the data at their last observation (Fig. 1). Anatomical recurrences in operated compartment were 74/185 (40.0%) and prolapse in a different compartment 16/ 185 (8.6%). Among all the anatomical recurrences, 44/185 (23.8%) were symptomatic (Table 3). More than half of the anatomical recurrences (57/90; 63.3%) were diagnosed within the 1-year follow-up for all different procedures. Failure, including the anterior compartment, was the most common regardless of procedure. The recurrence rate was highest for the single procedure in the anterior wall but the numbers of this procedure were small. The lowest recurrence rate was found for single procedure in the posterior wall although a considerable proportion was diagnosed with prolapse in another site at follow-up (Table 3). Due to symptomatic anatomic recurrence, 18 women (9.7%) proceeded to further surgery during follow-up. Additional operations for SUI were performed in seven patients. In an attempt to find predictive factors for recurrence, the women were divided into two groups according to outcome at follow-up—no anatomic recurrence and anatomical recurrence. No differences were found concerning parity, body mass index or previous gynaecological operations. Younger age and involvement of more than two sites preoperatively turned out to be significantly associated with having an anatomical recurrence (Table 4). Functional symptoms at follow-up Functional symptoms at all follow-ups are shown in Table 2. Self-reported urinary incontinence according to protocol remained at almost the same percentage level before and after surgery but for the individual women it changed. The

1596

Int Urogynecol J (2008) 19:1593–1601

Table 2 Functional symptoms preoperatively and at all follow-up Time Months; mean (SD) [range] Mechanical symptoms Vaginal bulging–discomfort, n (%) Urinary symptoms No incontinence, n (%) Urge incontinence, n (%) Stress incontinence, n (%) Mixed incontinence, n (%) Bowel symptoms No bowel symptoms, n (%) Diff emptying–constipation, n (%) Gas incontinence, n (%) Faecal incontinence, n (%)

Preop (n=185)

1 year (n=173) 13.1 (2.8) [5–29]

3 year (n=152) 38.0(3.6) [27–51]

5 year (n=143) 62.5 (5.8) [50–84]

152 (82.2)

24 (14)

16 (10)

28 (20)

91 28 32 34

(49) (15) (17) (18)

83 51 20 18

(48) (29) (12) (10)

68 49 16 18

(45) (32) (10) (12)

70 30 13 13

(49) (21) (9) (9)

106 61 27 21

(57) (33) (15) (11)

118 25 27 21

(68) (14) (16) (12)

95 34 23 9

(62) (22) (15) (6)

74 41 36 16

(52) (29) (25) (11)

Women with urge UI–mixed UI preoperatively were relieved of urge UI in 33.7% (11/31) after anterior repair (Table 5).

overall prevalence of faecal incontinence and incontinence of flatus remained unchanged except for gas incontinence at 5-year follow-up. Since UI and bowel symptoms may develop over time, we chose to comment on the changes in UI and bowel symptoms 1 year postoperatively.

Bowel symptoms at 1-year follow-up Among all women who preoperatively reported difficulty in emptying bowel (n=61), all but one underwent a posterior repair. Fifty-seven of them came to the 1-year follow-up visit and 75% (95%CI 64–87) were relieved of their symptom. At 5-year follow-up, 49/61 were evaluated and 55.1% (27/49) were still asymptomatic. For gas and faecal incontinence, 54% (95% CI 34–74) and 53% (95%CI 30– 75), respectively, were relieved of their symptoms at the 1-year follow-up (Table 6).

Urinary incontinence at 1-year follow-up Women with no UI preoperatively were diagnosed with de novo urge UI in 22.6%, de novo SUI in 6.0% and de novo mixed in 4.8% at 1-year follow-up. After anterior repair without concomitant incontinence surgery, women with SUI–mixed UI preoperatively had a resolution of SUI of 42.3% (11/26).

Table 3 Different surgical procedures and type of recurrence during follow-up Manchester procedure

Vaginal hysterectomy with posterior colporrhaphy n=4

Anterior + Anterior Posterior Colpocleisis posterior colporrhaphy colporrhaphy colporrhaphy n=25 n=7 n=36 n=4

67.7 (33–86) 67.6 (43–76) 52.3 48.3 33 (44.6) 15 (50.0)

55.2 (34–81) 56.8 2 (50.0)

67.7 (32–84) 61.1 (38–79) 58.3 (35–89) 78.2 (73–86) 54.1 57.0 57.3 50.5 10a (40.0) 4a (57.1) 7a (19.4) 2 (50) 1 (50)

20 (60) 5 (15)

1 1

7 (63) 2 (18) 1 (9) 1 (9) 5/11 (45.5) 1 (4)

n=74 Age, mean (range) Follow-up time, months; mean Anatomic recurrences, n/tot (%)a Site of recurrence–new compartment prolapseb, n Ant Post Uterus–vault Combinations Symptomatic last visit Proceeded to further surgery, n/tot (%)

Vaginal hysterectomy with anterior and posterior colporrhaphy n=30

11 (73)

1 (7) 8 (24) 3 (20) 19/33 (57.6) 6/15 (40) 7 (9.5) 4 (13.3)

1 0/3 0

4 () 1b (20)

4/5 (80) 2 (28.6) 2a

9 b(45) 7 () 3b(15) 1b () 7/20 (35) 3 (8.3) 1a

1 (50)

2 (50) 1 (25)

Not shown one enterocele operation with no recurrence at follow-up, one vaginal hysterectomy with no recurrence at follow-up, two cervix amputation only with no recurrence at follow-up, one vag hyst with anterior colporrhaphy with an asymptomatic recurrence a Refers to operated compartment b For single procedures new compartment prolapse are also shown

Int Urogynecol J (2008) 19:1593–1601

1597

248 women asked to participate

two re-operations. Six received synthetic mesh (Prolene) and three received biological mesh (Pelvichol) in the anterior wall. Of the synthetic mesh, 2/6 had a recurrent prolapse at 5-year follow-up and 3/3 with biological grafts. The only subject who received a synthetic mesh primarily experienced mesh erosion which was treated with excision 14 months after. At the 5-year follow-up, she had an asymptomatic recurrence

63 women not included refused/excluded 185 underwent vaginal surgery for pelvic organ prolapse Lost to follow up n=42 Postop 5 no recurrences 1 year follow up 14 no recurrences 7 recurrences 3 year follow up 9 no recurrences 6 recurrences

1 no follow up data

Lost to follow-up Of the 42 women lost to 5-year follow-up, their mean follow-up time was 20.4 months (SD 13.9). Thirteen had been diagnosed as recurrences and five required reoperation. Thirteen out of the 185 died during the followup, with no relation to operation. Six women suffered from dementia, two had moved from the region and the rest did not want to participate. An estimation of an overall adverse result of the 29 women (lost to follow-up but at their last observation had no recurrence) and the objective cure rate would only have been 37.5% (Fig. 1).

Follow up n=185 94 no recurrence 90 recurrence

Fig. 1 Study profile and result

Dyspareunia At the 5-year follow-up, 73/143 had been sexually active after surgery and 19 (26%) had symptoms of dyspareunia. Five women (2.7%) had undergone a second operation due to dyspareunia–vaginal tightness. Re-operations Re-operations were performed 5–60 months (mean 27 months) after the primary operation Two women underwent

SF36 at 5-year follow-up At the 5-year follow-up, 85.3% (122/143) completed the SF36 health survey. The SF36 dimension scores were generally comparable to those of an age-matched reference group except for social functioning where the scores tended to be lower especially for women younger than 65 years old.

Table 4 In an attempt to find predictive factors for recurrence, the women were divided into three groups according to outcome at follow-up—no anatomic recurrence, non-symptomatic recurrence and symptomatic recurrences. Age and more than two sites involved preoperatively showed significant differences between groups Baseline characteristic

Age mean (SD) [range] Parity mean (range) No. of sites involved preop 1 2 3 BMI mean (range) Preoperative symptoms Urinary incontinence Any bowel symptoms Previous surgery Hysterectomy Incontinence Prolapse a b

Anatomical recurrence, n=90

Odds ratioa (95% confidence interval)

p value

68.2 (12.4) [35–86] 2.2 (0–5)

62.3 (13.7) [32–89] 2.5 (0–15)

0.97 (0.94; 0.99) 1.14 (0.92; 1.43)

0.003 0.232

28 35 31 25.3

15 37 38 25.9

1.97 (0.90; 4.30)b 2.29 (1.04; 5.02)b 1.05 (0.96; 1.14)

0.087 0.039 0.254

No anatomical recurrence, n=94

(30) (37) (33) (18.9–36.8)

(17) (41) (42) (19.4–38.3)

53 (56) 42 (45)

41 (46) 36 (40)

1.54 (0.86; 2.77) 1.26 (0.70; 2.28)

0.143 0.432

13 (14) 9 (10) 11 (12)

14 (16) 12 (13) 13 (14)

1.15 (0.50; 2.67) 1.45 (0.58; 3.74) 1.27 (0.54; 3.06)

0.741 0.425 0.581

Odds ratio describes the change in odds of having a recurrence when increasing the specific variable by one unit. Odds ratio compared to one site involved preoperatively.

1598

Int Urogynecol J (2008) 19:1593–1601

Table 5 Urinary incontinence preoperatively and 1 year after prolapse surgery that included anterior repair and no concomitant incontinence surgery (n=111) No incontinence preoperatively, n=66 Age mean(SD) [range] Postoperatively (%) No incontinence Urge urinary incontinence Stress urinary incontinence Mixed urinary incontinence Cured 1 year De novo 1 year (%)

Urge urinary incontinence preoperatively, n=19

65 (11.8)[33–83]

65 (11.8) [33–86]

42 16 5 2

5 (26) 13 (74) 0 1 (5) 5 (26) 18/111 (16)

(64) (25) (8) (3)

Stress urinary incontinence preoperatively, n=14 72 (9.8) [47–82] 4 2 4 4 6 5/111

(29) (14) (29) (29) (43) (4)

Mixed urinary incontinence preoperatively, n=12 70 (9.8) [55–86] 2 3 4 3 2 7/111

(17) (25) (33) (25) (17) (6)

Data are expressed as number/total (%).

Discussion This prospective observational study was designed to describe the anatomical and functional outcome of vaginal pelvic reconstructive surgery. At the 5-year follow-up, information concerning anatomy and symptoms covered over 77% of all patients. Our study confirms that prolapse surgery is associated with a high anatomical recurrence rate and the compartment most prone for recurrence is the anterior compartment. Posterior repair had the lowest recurrence in the same compartment but more new failures were found in other compartments. The outcome did not favour vaginal hysterectomy over Manchester operation which still could be an alternative for women who want to preserve their uterus. Despite a high anatomical recurrence rate, a considerable proportion of the women were asymptomatic corresponding to what was found in a randomised trial between three different surgical techniques of anterior colporrhaphy [5]. Their anatomical success rate ranged from 40% to 60% but a high rate of symptomatic improvement was noticed. Our follow-up time was longer but the majority of the recurrences took place within the first year after surgery. The failure rate in prolapse surgery depends on the definition of failure, whether failure is defined as subjective or objective failure. A corrective procedure in one compartment might predispose for prolapse in a different anatomic compartment—is this to be considered as recurrence? Or could it be interpreted as an initial failure to

identify all defects during the first procedure? It might also reflect a further progression of lack of pelvic support. Women with pelvic support defects often present with an involvement of more than one compartment together with additional urinary, bowel and mechanical symptoms which can make it difficult to randomise into one standard prolapse operation. Thus, in randomised trials between single-site procedures, other prolapse procedures have concurrently been performed [5, 10]. The re-operation rate in our study was 9.7%, and if we included operation for UI postoperatively the rate was 13.5% which is consistent with the literature [1]. The re-operation rate depends on the surgeons attitude to re-operation or the fact that many women considered as surgical failures–recurrences are satisfied with their surgical result and asymptomatic because the prolapse have been elevated above the hymen [5, 11]. In our study, there were no restrictions for re-operations and women were offered a second operation if desired. Like Whiteside et al. [12], we found women with recurrent prolapse to be younger and to have more sites affected by prolapse preoperatively but there were no differences between groups concerning urinary incontinence or bowel symptoms. Operating younger women with respect to maintaining sexual function could have influenced the outcome but women at younger age are possibly more physically active which also could affect prolapse recurrence except for genetic or connective tissue disorders which is not addressed in this study.

Table 6 Bowel symptoms preoperatively and 1 year after prolapse surgery that included posterior repair (n=158). More than one symptom can exist in some individuals

Age (range; SD) Cured 1 year (%; 95% CI) De novo 1 year Data are expressed as number/total (%).

Diff emptying–constipation preoperatively, n=57

Gas incontinence preoperatively, n=24

Anal incontinence preoperatively, n=19

62 (12.7) [33–82] 43/57 (75; 64–87) 10/158 (6)

67 (11.6) [45–81] 13 /24 (54; 34–74) 16/158 (10)

69 (11.2) [47–89] 10/19 (53; 30–75) 11/158 (7)

Int Urogynecol J (2008) 19:1593–1601

The prevalence of UI in our patient population remained at an almost constant level preoperatively and postoperatively although for the individual women there were changes in type of UI. This was also an older population where many of the women have high rates of concurrent UI [13]. Examining a younger population of women undergoing prolapse surgery, the results may have turned out differently. Meta-analysis in the Cochrane review of surgical management of prolapse [7] found the impact of pelvic organ prolapse surgery on continence issues to be limited and inconclusive. Comparing our results to what is described in this review, we had a lower rate of de novo SUI (10% vs 21%) which could possibly be explained by the large number of pessaries used preoperatively, unmasking occult SUI [14]. The cure rate of SUI after anterior repair without concomitant incontinence procedure in our study was 42% which corresponds to former studies with subjective cure rates of 46–52% [15] but even higher cure rates have been reported [16]. Few studies address the effect of prolapse surgery on urge UI and results differ. Digesu et al. [17] showed a significant improvement in overactive bladder symptoms after surgical repair of symptomatic anterior vaginal wall prolapse (stage ≥2) and a resolution of urge UI in 45% while Nguyen and Bhatia [18] reported a cure rate of 63% after surgical repair of uterine and/or vaginal vault prolapse compared to our cure rate of 33%. Milani et al. on the other hand showed, in a study of vaginal repair with Prolene mesh, a high anatomical success rate of 93% with no change in urge or stress UI preoperatively and postoperatively [19] Although the relationship between stage of prolapse and bowel symptoms has been found to be weak [20, 21], women with prolapse have been shown in a recent study to be more likely to suffer from difficult defaecation and faecal incontinence compared to matched controls [22].We found an improvement of difficulty in emptying bowels after traditional posterior colporrhaphy. Approximately one third of the patients in our study experienced difficulties with emptying their bowels preoperatively and half of them were relieved of their symptoms 1 year after surgery. After 5 years, two thirds were still asymptomatic. This is in concordance with what several other authors have found after discrete posterior repair [19, 23–25]. The standard posterior colporrhaphy without identification of fascial defects has been criticised for rather predisposing patients to bowel symptoms through inadequate anatomic restoration and Kahn and Stanton [26] reported an increase of evacuation disorders and constipation after posterior repair. Our results differ from theirs except for dyspareunia. Mellgren et al. [27] on the other hand showed in a prospective study of posterior colporrhaphy with levator plication a resolution of constipation in half of their patients.

1599

In a randomised trial between posterior colporrhaphy without levator plication, site-specific repair and posterior repair augmented with porcine graft, significant improvement of defaecatory dysfunction after surgery was found with no differences between treatment groups [10]. The prevalence of faecal incontinence preoperatively among our women with concomitant prolapse was slightly lower compared to previous works [28] but Morgan et al. reported an even lower percentage. Posterior colporrhaphy in our study group did not aggravate bowel disorders already present such as gas or faecal incontinence postoperatively which was reported by Kahn and Stanton [26] rather we found a reduction consistent with that of Gustilo-Ashby et al. [23]. There are several limitations to this study. There is a lack of a consistent system of classifications due to policy changes during the last decade. When the study started in 1998, the POPQ had been introduced but not fully adopted; most of the articles published at that time did not even use a standardised system [29]. Since the POPQ system was formally adopted as the standard for describing prolapse and the fact that POPQ provided an objective system for describing pelvic support, we gradually changed the method during follow-up. Recurrence according to Beecham in the first 2 years of follow-up was a prolapse to the introitus which would correspond to a POPQ stage ≥2. Another weakness is that no validated pelvic floor questionnaires were present in the Swedish language at that time and no objective measures were made of SUI or urge urinary incontinence. The questions concerning bowel symptoms as difficulty in emptying bowel and constipation could have, for the patients, a broader meaning in terms of hard lumpy stool, incomplete evacuation and/or digital manipulation which was not thoroughly explored. Moreover, the SF36 was only used at the 5-year follow-up and whether quality of life was improved after pelvic reconstructive surgery in our patients cannot be concluded in this study. However, this has been shown in other studies [30]. Dyspareunia and sexual function was not assessed prior to the operation and is therefore difficult to evaluate in this study. The prevalence of dyspareunia is similar to other studies where posterior colporrhaphy has been shown to be associated with the development of dyspareunia at a prevalence of 21–27% [10, 26]. The principal strength of this study, however, is that it was a prospective study with a noteworthy high participation at long-term follow-up which can be a difficulty in performing studies in an elderly patient population. Furthermore, at 5-year follow-up, there was one independent observer. To our knowledge, there are few prospective studies to which we can compare our findings but our results are probably generalisable to women undergoing multiple reconstructive

1600

surgeries demonstrating a low perioperative and postoperative complication rate, few adverse effects and a higher proportion of subjective cure than anatomic restoration. However, more data are still needed to evaluate the effect of surgery on urinary incontinence and bowel symptoms [7].

Conclusion Our long-term follow-up of standard prolapse surgery showed a high rate of anatomic recurrence but only half of the women were symptomatic. Younger age and several sites involved preoperatively were associated with anatomic recurrence. The compartment most prone for recurrence was the anterior wall. Most of the failures occurred within the first year after surgery. Our data showed an alleviation of mechanical symptoms and an improvement of difficulty in emptying bowel after a posterior repair. No aggravations of gas or faecal incontinence were observed. This study supports a certain improvement of stress and urge urinary incontinence after an anterior repair but also demonstrates the difficulty in predicting outcome of urinary incontinence after prolapse surgery in any individual woman. Some will note improvement and some women will acquire another type of incontinence. The functional outcome of prolapse surgery is complex and not always foreseeable. Acknowledgment The study was supported by grants from the Vårdal foundation (www.vardal.se). Conflicts of interest None.

References 1. Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL (1997) Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 89:501–506 2. Luber KM, Boero S, Choe JY (2001) The demographics of pelvic floor disorders: current observations and future projections. Am J Obstet Gynecol 184:1496–1501 discussion 1501–1503 3. Boyles SH, Weber AM, Meyn L (2003) Procedures for pelvic organ prolapse in the United States, 1979–1997. Am J Obstet Gynecol 188:108–115 4. Jelovsek JE, Maher C, Barber MD (2007) Pelvic organ prolapse. Lancet 369:1027–1038 5. Weber AM, Walters MD, Piedmonte MR, Ballard LA (2001) Anterior colporrhaphy: a randomized trial of three surgical techniques. Am J Obstet Gynecol 185:1299–1304 discussion 1304–1306 6. Shull BL, Bachofen C, Coates KW, Kuehl TJ (2000) A transvaginal approach to repair of apical and other associated sites of pelvic organ prolapse with uterosacral ligaments. Am J Obstet Gynecol 183:1365–1373 discussion 1373–1374 7. Maher C, Baessler K, Glazener CM, Adams EJ, Hagen S (2007) Surgical management of pelvic organ prolapse in women. Cochrane Database Syst Rev CD004014

Int Urogynecol J (2008) 19:1593–1601 8. Beecham CT (1980) Classification of vaginal relaxation. Am J Obstet Gynecol 136:957–958 9. Ware JE Jr, Gandek B (1998) Overview of the SF-36 health survey and the International Quality of Life Assessment (IQOLA) Project. J Clin Epidemiol 51:903–912 10. Paraiso MF, Barber MD, Muir TW, Walters MD (2006) Rectocele repair: a randomized trial of three surgical techniques including graft augmentation. Am J Obstet Gynecol 195:1762–1771 11. Diez-Itza I, Aizpitarte I, Becerro A (2007) Risk factors for the recurrence of pelvic organ prolapse after vaginal surgery: a review at 5 years after surgery. Int Urogynecol J Pelvic Floor Dysfunct 18:1317–1324 12. Whiteside JL, Weber AM, Meyn LA, Walters MD (2004) Risk factors for prolapse recurrence after vaginal repair. Am J Obstet Gynecol 191:1533–1538 13. Bump RC, Mattiasson A, Bo K et al (1996) The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 175:10–17 14. Bergman A, Koonings PP, Ballard CA (1988) Predicting postoperative urinary incontinence development in women undergoing operation for genitourinary prolapse. Am J Obstet Gynecol 158:1171–1175 15. Glazener CM, Cooper K (2001) Anterior vaginal repair for urinary incontinence in women. Cochrane Database Syst Rev CD001755 16. Bergman A, Koonings PP, Ballard CA (1989) Primary stress urinary incontinence and pelvic relaxation: prospective randomized comparison of three different operations. Am J Obstet Gynecol 161:97–101 17. Digesu GA, Salvatore S, Chaliha C, Athanasiou S, Milani R, Khullar V (2007) Do overactive bladder symptoms improve after repair of anterior vaginal wall prolapse? Int Urogynecol J 18:1439–1443 18. Nguyen JK, Bhatia NN (2001) Resolution of motor urge incontinence after surgical repair of pelvic organ prolapse. J Urol 166:2263–2266 19. Milani R, Salvatore S, Soligo M, Pifarotti P, Meschia M, Cortese M (2005) Functional and anatomical outcome of anterior and posterior vaginal prolapse repair with Prolene mesh. Bjog 112:107–111 20. Weber AM, Walters MD, Ballard LA, Booher DL, Piedmonte MR (1998) Posterior vaginal prolapse and bowel function. Am J Obstet Gynecol 179:1446–1449 discussion 1449–1450 21. da Silva GM, Gurland B, Sleemi A, Levy G (2006) Posterior vaginal wall prolapse does not correlate with fecal symptoms or objective measures of anorectal function. Am J Obstet Gynecol 195:1742–1747 22. Morgan DM, DeLancey JO, Guire KE, Fenner DE (2007) Symptoms of anal incontinence and difficult defecation among women with prolapse and a matched control cohort. Am J Obstet Gynecol 197:509e1–509e6 23. Gustilo-Ashby AM, Paraiso MF, Jelovsek JE, Walters MD, Barber MD (2007) Bowel symptoms 1 year after surgery for prolapse: further analysis of a randomized trial of rectocele repair. Am J Obstet Gynecol 197:76e1–76e5 24. Kenton K, Shott S, Brubaker L (1999) Outcome after rectovaginal fascia reattachment for rectocele repair. Am J Obstet Gynecol 181:1360–1363 discussion 1363–1364 25. Cundiff GW, Weidner AC, Visco AG, Addison WA, Bump RC (1998) An anatomic and functional assessment of the discrete defect rectocele repair. Am J Obstet Gynecol 179:1451–1456 discussion 1456–1457 26. Kahn MA, Stanton SL (1997) Posterior colporrhaphy: its effects on bowel and sexual function. Br J Obstet Gynaecol 104:82–86

Int Urogynecol J (2008) 19:1593–1601 27. Mellgren A, Anzen B, Nilsson BY et al (1995) Results of rectocele repair. A prospective study. Dis Colon Rectum 38: 7–13 28. Meschia M, Buonaguidi A, Pifarotti P, Somigliana E, Spennacchio M, Amicarelli F (2002) Prevalence of anal incontinence in women with symptoms of urinary incontinence and genital prolapse. Obstet Gynecol 100:719–723

1601 29. Muir TW, Stepp KJ, Barber MD (2003) Adoption of the pelvic organ prolapse quantification system in peer-reviewed literature. Am J Obstet Gynecol 189:1632–1635 discussion 1635–1636 30. Barber MD, Amundsen CL, Paraiso MF, Weidner AC, Romero A, Walters MD (2007) Quality of life after surgery for genital prolapse in elderly women: obliterative and reconstructive surgery. Int Urogynecol J Pelvic Floor Dysfunct 18:799–806