Introduction

Currently, there are widespread variations in practice in the conservative and surgical management of prolapse in the UK. Five years ago, a national survey on the management of prolapse [1] in the UK was conducted which highlighted these variations. The objective of the national prolapse survey was to gain insight into the changing trends in the surgical management of various types of prolapse in different clinical settings, and to compare these changes to the first national UK prolapse survey conducted 5 years ago. Practice amongst the urogynaecologists working in the tertiary centres, generalists with a special interest in urogynaecology and the general gynaecologists within the UK were also assessed.

We hypothesised that there would be a significant change in the surgical trends for the management of prolapse particularly with the increasing popularity of grafts.

Methods

This was a postal questionnaire survey. The initial questionnaire, used in the survey of 2005, was developed following a pilot study which was carried out on the seven consultant gynaecologists at Worcestershire Royal Hospital. The method of questionnaire development is given in the initial article detailing the survey [1]. Case scenarios formulated for the first survey were modified to incorporate a further range of options, taking into account the current practice trends in surgical correction of prolapse. The questions constituting the final questionnaire incorporated the management of anterior vaginal wall prolapse (question 1), uterine prolapse in conjunction with vaginal wall prolapse (question 2), posterior vaginal wall prolapse (question 3), and vaginal vault prolapse (question 4) as in the previous survey. We also asked respondents on how they classified pelvic organ prolapse and the degree of follow-up patients received following prolapse surgery, and if they were using the BSUG database for auditing the results of their surgery.

The addresses were obtained from the Gynecare database of practising UK obstetricians and gynaecologists. This database is updated to include new consultants joining a trust; however, those leaving or retiring were not always accounted for. This would explain why a proportion of the questionnaires were returned without being completed. The database used for this survey was considerably smaller than the previous one because fewer consultants were registered as undertaking prolapse surgery. A covering letter describing the objectives of the study accompanied the questionnaire. After 4 weeks, the questionnaire was resent to non-respondents. This was again repeated after a further 4 weeks.

As this was a review of clinician's practice, we were advised by the Local Research and Ethics Committee that formal ethical approval was not required for the study. In addition, consent to use the information provided was obtained in the questionnaire from the respondents. Respondents who did not give consent to use the information provided were excluded from the analysis.

The analysis was performed by looking at the overall percentage response to each individual question. This was compared to the results from 5 years ago using the chi-square test and p values calculated to determine if the difference in the response was statistically significant. The responses between groups A (urogynaecologists) versus B (gynaecologist with a special interest in urogynaecology) and A versus C (generalists) were also compared as in the previous survey.

A p < 0.05 was set as statistically significant.

Results

Five hundred and forty-nine questionnaires were sent out of which 218 responses were returned (40%). Of these, 190 were completed giving a useable response rate of 35%. The 28 incomplete responses were from UK consultants who had left the trust, were now retired, were not performing prolapse surgery or were practising only obstetrics. Sixty-nine of 190 (36%) respondents from the previous survey also responded in this survey. Both the response rate and useable response rate were better than in the previous survey in which they were 33% and 28%, respectively, though overall numbers of questionnaires sent were far fewer and so too were the overall responses received.

Of the completed responses received, 20% (38/190) were from urogynaecologists (group A) working in tertiary centres, 52% (99/190) were from gynaecologists with a special interest in urogynaecology (group B) and 21% (39/190) were from general gynaecologists (group C). Seven percent (14/190) did not specify their designation. The 7% responses in which designation was not specified were analysed in the overall assessment, but were excluded from the analysis when comparing the three target groups. Compared to the previous survey, there was a greater response from the urogynaecologists and fewer responses from the generalists.

Anterior wall prolapse

For anterior vaginal wall prolapse, anterior colporrhaphy was the procedure of choice in 71% of respondents. This was a non-significant change from 5 years ago when it was 77%. With concomitant urodynamic stress incontinence, a Burch was the procedure of choice in only 1% compared to 11% 5 years ago. This change was significant (p < 0.003). Eighty-six percent of respondents would perform a midurethral tape combined with repair, whereas 5 years ago, this was 71% representing a significant rise (p < 0.015).

Eleven percent of respondents used a graft for primary prolapse compared to 10% in the previous survey, whereas 56% would do so for a recurrent anterior wall prolapse either alone or in combination with fascial plication. The use of graft for recurrent prolapse was a significant rise from the previous survey (p < 0.002). For both primary and secondary repairs, there was a significant increase in the use of synthetic grafts with a corresponding decrease in the use of biological grafts. Results are shown in Table 1.

Table 1 Question 1: anterior vaginal wall prolapse
Full size table

When comparing the groups A, B and C, group A performed anterior colporrhaphy in 89% compared to 64% amongst group B and 74% in group C. Group A were therefore significantly more likely to perform anterior repairs compared to group C (89% vs 74%; p = 0.01) and compared to group B (89% vs 64%; p < 0.001). The use of PDS in the different groups was statistically more in group A when compared to group C (34% vs 5%; p < 0.0001) but was not statistically different for group B (34% vs 26%; p = 0.21). Three percent of group A, 16% of group B and 10% of group C used a graft for primary anterior wall repair either alone or in combination with fascial plication. The use of a graft for a primary anterior vaginal wall was not statistically different when comparing groups A and C (3% vs 10%; p = 0.82) but was significant less when comparing group A to group B (3% vs 16%; p = 0.002). For secondary redo anterior wall repairs, group A would use a graft either alone or in combination with fascial plication in 50%, group B in 58% and group C in 43%. The difference in the use of graft for secondary repairs was not statistically significant when comparing groups A and C or group A and B. The choice of procedure for treating concomitant anterior repair and USI was not statistically different in the three groups, and the midurethral tape in conjunction with an anterior midline plication repair was the procedure of choice. The Burch colposuspension was being performed by 5% in group A and 1% of group B, but none of the clinicians in Group C was performing the Burch. All three groups had a similar response when asked if they would operate on women who had not completed their family. This was not statistically different for the three groups (group A = 47%, group B = 45% and group C = 56%). The change in surgical approach if the patient was not sexually active was also not different amongst the three groups (group A = 18%; group B = 24%; group C = 23%).

Compared to 5 years ago, the greatest change in practice was in the uptake of graft in secondary repairs in all three groups and the associated increased use of synthetic graft with a corresponding decrease in the use of biological graft. The other significant difference was in the number of Burch procedures which had fallen dramatically since the previous survey in all three groups. In addition, this was being performed only by the specialists, i.e. group A and B, whereas 5 years ago, generalists were undertaking this procedure as well.

Uterovaginal wall prolapse

The second question assessed trends in the surgical management of second-degree uterine prolapse in conjunction with anterior vaginal wall prolapse. In women with uterovaginal prolapse, the procedure of choice was still a vaginal hysterectomy combined with a repair (82%) and had not changed from 5 years ago. Thirty-five percent of respondents would operate in women whose family was incomplete compared to 26% in the previous survey, and the procedure of choice was still a sacrohysteropexy (type I polypropylene mesh used to anchor the uterus to the sacrum). Results are given in Table 2.

Table 2 Question 2: uterine + vaginal wall prolapse (stage II)
Full size table

When comparing surgical practice in the management of uterovaginal wall prolapse in the three groups, the rates of performing preoperative urodynamics (UDS) were similar in the three groups (group A = 56%, group B = 62% or group C = 55%). The procedure of choice in the three groups was also similar with a vaginal hysterectomy and repair being the procedure of choice (group A = 88%; group B = 78%; group C = 85%). The procedure of choice for supporting the vault was also similar in all three groups, i.e. suturing the uterosacral ligaments to the vault (group A = 49%; group B = 54%; group C = 71%). None of the groups felt their procedure would change significantly in this scenario if the patient was not sexually active (group A = 5%; group B = 10%; group C = 8%). Offering a ring pessary till the patient's family was complete would be adopted by all three groups (group A = 63%; group B = 54%; group C = 62%). Uterine preservation surgery was offered by group A in 35%, group B in 39% and group C in 20% of cases.

Comparing the results to those from the previous survey, all responses were similar in the groups A, B and C except performance of uterine preservation surgery which was being done significantly less in the generalists, i.e. group C compared to group A or B.

Posterior vaginal wall prolapse

Question three assessed the surgical trends in the management of posterior vaginal wall prolapse. In women with posterior vaginal wall prolapse, the procedure of choice was posterior colporrhaphy with midline fascial plication in 66% of respondents. Twelve percent of respondents would use a graft for a primary posterior wall prolapse, and 49% would use a graft for a recurrent posterior wall prolapse. None of the results was significantly different to the results from 5 years ago. Though there was a marginal increase in the use of synthetic graft compared to biological graft, this change was not significantly different. Results are given in Table 3.

Table 3 Question 3: posterior vaginal wall prolapse
Full size table

The referral for anorectal studies was not statistically different in the three groups (A = 16%; B = 16%; C = 18%). The procedure of choice was similar in all three groups, i.e. posterior colporrhaphy with midline fascial plication (group A = 63%; group B = 62%%; group C = 75%). For primary repairs, 20% of group A, 13% of group B and 2% of group C would use a graft alone or in combination with fascial plication. More significantly, group A would use a graft for primary repair compared to group C (20% vs 2%, p < 0.001). However, there was no difference between groups A and B (20 vs 13%, p = 0.25). For recurrent posterior vaginal wall prolapse, the procedure of choice was a graft-reinforced repair with or without fascial placation in group A (52%), group B (44%) and group C (45%) with no significant difference. The patient's sexual status did not alter the approach of surgery in the three groups (A = 82%, B = 85%; C = 89%).

Compared to the survey 5 years ago, more generalists (group C) were performing graft reinforced surgery for recurrent prolapse, but other parameters remained unchanged.

Vaginal vault prolapse

Question four assessed the management of vaginal vault prolapse (VVP). Seventy-three percent of respondents would operate on a vault prolapse. Forty-three percent would perform UDS prior to surgery. The procedure of choice was an abdominal sacrocolpopexy (44%) where a type I polypropylene mesh is used to anchor the vaginal vault to the sacrum. When there was associated occult incontinence, 35% of respondents who would operate would perform an additional incontinence procedure at the time of surgery. Results are shown in Table 4.

Table 4 Question 4: vault prolapse
Full size table

The individual operative rates in the three groups were variable (group A = 95%; group B = 90%; group C = 69%). The number of group A respondents who would operate was significantly greater than the numbers in group C who would operate (95% vs 69%; p < 0.001), but no different to the number of group B respondents who would operate (95% vs 90%; p = 0.3). Preoperative UDS in the three groups varied (group A = 50%; group B = 47%; group C = 23%). This was statistically similar for group A and B but significantly greater in group A compared to group C (50% vs 23%; p < 0.001). The procedure of choice in both group A (68%) and group B (44%) was an abdominal sacrocolpopexy (SCP). In group C, however, a SCP procedure accounted for 30% of the surgery performed by this group for VVP. This was significantly less than the respondents in group A performing these procedures (68% vs 30%; p < 0.001). The preferred procedure for a VVP in group C was a sacrospinous/ileococcygeal fixation. Five years ago, the preferred procedure in this group was a standard repair. There were minimal variations in the three groups when comparing who would perform an anti-incontinence procedure at the time of prolapse surgery (group A = 39%; group B = 35%; group C = 30%). This was not statistically different for group A and B or group A and C. The choice of procedure was not influenced by the patient's sexual status (group A = 13%; group B = 13%; group C = 3%). This was not statistically different in the three groups.

Compared to 5 years ago, practice trends amongst the three groups remained unchanged for the management of vault prolapse, though more generalists were performing sacrospinous fixations than 5 years ago.

The method of classification used for prolapse varied. Most group A respondents (58%) and group B (41%) used the pelvic organ prolapse-quantitative assessment (POP-Q) [2] as the preferred criteria for classifying prolapse, but only 21% of group C were using the POP-Q. Use of the POP-Q was significantly more in group A when compared to group B (58% vs 41%; p < 0.01) or group C (58% vs 21%; p < 0.0001). There was an overall rise in the use of the POP-Q from the previous survey in both groups B and C but a fall in group A. Group C preferred to classify prolapse in degrees (first, second and third or procidentia).

Eighty-two percent of all respondents saw their patients back in gynaecology outpatients. This is a marginal fall from the previous survey (91%). This varied from 6 weeks to 6 months and occasionally longer. Group A followed up patients in 92% of cases, Group B in 83% and Group C in 72%.

Thirty-five percent of all respondents were using the BSUG database, but 65% were not using it. Use was significantly higher amongst group A (58%) when compared to group B (35%) or C (10%).

Discussion and conclusions

There are wide variations in the management of different types of prolapse. Overall, surgical practice in the management of pelvic organ prolapse has not altered dramatically in the past 5 years. The uptake of grafts has increased significantly in anterior compartment defects, and the rise was seen predominantly in patients with recurrent prolapse. There was also a significant reversal in the use of synthetic grafts with a corresponding decrease of biological grafts for the anterior compartment both in primary repairs and recurrent cases. Basic trends in prolapse surgery remain unchanged in the different groups of clinicians practising urogynaecology in the UK, i.e. the generalists, gynaecologists with a special interest, and the urogynaecologists. Unlike the previous survey, there was no obvious comparable trend when comparing the different management options amongst urogynaecologists, gynaecologists with a designated special interest in urogynaecology, and the general gynaecologists. In the UK, changes in the management of prolapse conform to the emerging evidence which is both reassuring and suggestive of conscientious practice.

The proportion of responses and the usable response rate (35%) were both better than in the previous national survey [1], even though overall responses were lower. Fewer questionnaires were sent out in this survey compared to the previous one. This might be because prolapse surgery is becoming more specialised hence explaining the smaller size of the Gynecare database from which the mailing list was obtained. This may also explain why fewer generalists and more urogynaecologists responded compared to the previous survey. The number of registered consultant members of the BSUG database is 229; hence, the response rate albeit small is still likely to be valid.

Following the introduction of graft use in pelvic organ prolapse (POP), there were initial concerns regarding the sudden increase in graft repairs for primary procedures particularly with the limited evidence. This does not appear to have happened, and it is reassuring that the increased uptake has been with secondary procedures in anterior compartment defects and in accordance with currently available evidence. It was also reassuring that the use of biological graft was falling particularly with a lack of evidence in their use [3]. The uptake for the posterior vaginal wall defects has remained unchanged and may be due to the lack of evidence of its benefit in the posterior vaginal wall prolapse [3]. The caution in the uptake of grafts could be related to the lack of long-term evidence of benefit, associated complications, particularly with trocar devices, and recommendations from NICE as well as the financial implications of using these devices on the NHS.

Specialised surgeries such as the Burch, sacrohysteropexy and sacrocolpopexy are being performed less frequently by the generalists than 5 years ago. Even amongst the specialists, the number of these procedures being performed is significantly less than in the past. It could be argued that these specialised procedures should therefore be performed in specialist centres or by those performing an ‘adequate’ volume of such surgery in non-tertiary centres to maintain skills and improve outcomes. Collecting data on complications and outcomes, by using such tools as the BSUG database, is becoming increasingly important in this context.

The lower overall follow-up rate following prolapse surgery may be reflective of increasing pressures from the primary care trusts to keep follow-up patients in outpatients to a minimum. This is likely to fall further particularly with mounting pressures to maintain appropriate ratios of new to follow-up in clinics. The follow-up rates amongst the specialists (group A and B) compared to the previous survey may be stable because of the complex nature of procedure performed by them, hence the insistence by clinicians to see these patients for follow-up. A reduction in the follow-up has implications for monitoring of outcomes and problems with assessing success and satisfaction rates amongst patients particularly with this role being gradually delegated to primary care.

There is increasing use of the BSUG database which provides outcomes data and is a powerful audit tool for individual clinicians. In order for the database to provide denominator data however so that results can be meaningful in epidemiological studies, the uptake and use of the database need be much higher than the current rates of usage. This needs to be encouraged.

It was difficult to compare the results of this survey with practice in other countries as apart from a similar survey in Australia and New Zealand [4], there have not been similar surveys in other parts of the world to draw comparisons. There has been a survey in South Africa [5] comparing practice between the urogynaecologists and urologists; however, this was solely for anterior compartment defects. In the survey from South Africa, the preferred procedure for surgical correction of anterior vaginal wall prolapse was an anterior colporrhaphy, similar to the findings in our survey. In this survey, however, both paravaginal repair and graft repairs were more prevalent and being performed by 41.9% and 55.1%, respectively. This was significantly greater than in the UK survey. This survey did not look formally at the differences between primary and recurrent repairs. In addition, urologists were performing a significant proportion of the prolapse work for the anterior compartment which is not routine practice here in the UK.

Earlier this year, Vanspauwen et al. [4] published the results of their survey of Australian and New Zealand practitioners. This survey was based on the Initial National UK prolapse survey conducted 5 years ago using the same subset of questions and a similar analysis criteria. The commonest procedure for anterior repair in their survey was also an anterior colporrhaphy, but synthetic graft was being used by a significantly greater proportion of clinicians for secondary repair compared to the UK (75% vs 56%, p < 0.005). For primary repair of the anterior compartment defect, results were not statistically significant. The management of uterovaginal prolapse and posterior vaginal wall prolapse was similar with no significant difference. For apical defects, the procedure of choice for vault prolapse was an abdominal approach in the UK with the sacrocolpopexy being the preferred operation, whereas in the Australian survey, the vaginal approach was preferred, and the commonest procedures were a sacrospinous fixation followed by a vaginal graft.

The management of prolapse has always been associated with discrepancy and variation. In the absence of robust RCT data, this is likely to continue. With a lifetime risk of 50% [6] for POP, high risk of recurrence [7] and a rising annual incidence of surgery for this condition, there is an urgent need to establish standards in the management of this condition. Collection of prospective data should be considered in which the operative and clinical details of women undergoing prolapse surgery both standard and with mesh/graft can be recorded. This may be feasible through the BSUG database; however, this requires greater uptake by clinicians so that sufficient efficacy and safety data can be gathered to guide the management in future. In addition, a need for adequately powered studies with sufficient follow-up is required to validate the efficacy of procedures before they are used.