A recent study reports that the lifetime risk of any primary surgery for pelvic organ prolapse (POP) is 12.6%1 and up to 17%2 of women will undergo reoperation. The anterior compartment is the most common site of recurrence after support of the apical compartment.3–5 Data in the literature to guide a surgeon's decision on whether to perform a concomitant anterior repair (AR) at the time of vaginal apical correction of prolapse are sparse. Currently, this decision could be based on the extent of preoperative prolapse and/or the position of the anterior compartment after the apical correction intraoperatively, or the surgeon's personal belief that the additional procedure is warranted. An informal survey of surgeons in the Pelvic Floor Disorder Network (PFDN) suggested wide variation in the practice of adding a concomitant AR at the time of a vaginal apical suspension procedure, suggesting that this is a practice needing preliminary outcome data.
The National Institutes of Health-sponsored PFDN has conducted surgical trials that have generated evidence regarding vaginal apical suspension operations. In two of these studies, concomitant AR at the time of vaginal apical suspension was performed at the surgeon's discretion. With an enrollment total of more than 700 well-characterized participants, these studies provide a unique opportunity to compare prolapse outcomes in women undergoing native tissue vaginal apical suspension procedures with or without a concomitant AR.
The primary aim of this analysis was to compare the prevalence of anterior compartment prolapse 1 year after native tissue vaginal apical suspension in participants of surgical trials who underwent concurrent AR versus those who did not. Secondary aims included the following: (1) to compare the prevalence of overall prolapse 1 year after native tissue vaginal apical suspension in participants of surgical trials who underwent concurrent AR versus those who did not, (2) to measure the progression of the anterior compartment prolapse from baseline to 1 year postoperatively in each group, and (3) to determine whether preoperative anterior prolapse severity was a predictor of anatomic prolapse outcomes. We hypothesized that concomitant AR would be associated with a decreased prevalence of anterior vaginal prolapse at 1 year.
This study is a secondary analysis of the OPUS (Outcomes following vaginal Prolapse repair and mid Urethral Sling)6 and OPTIMAL7 (The Operations and Pelvic Muscle Training in the Management of Apical Support Loss) trials. Details of the trial designs and methods for OPUS8 and OPTIMAL9 as well as their primary outcomes have been previously described.6,7 All sites obtained local institutional review board approval and participants provided written research informed consent.
OPUS participants had stage 2–4 POP, did not have stress urinary incontinence, and were randomized to a tension free vaginal tape (TVT) or sham incisions at the time of vaginal prolapse surgery. To study the effects of AR during native tissue vaginal apical suspension, only participants in the OPUS trial with a sacrospinous ligament fixation (SSLF) or uterosacral ligament suspension (ULS) and either an AR or no procedure in the anterior compartment were included in this study (ie, patients with an AR only or a vaginal prolapse mesh procedure were not included). OPTIMAL participants had stage 2–4 POP with apical prolapse and stress urinary incontinence and were randomized to one of two native tissue vaginal apical suspensions (SSLF or ULS). All OPTIMAL participants had stress urinary incontinence symptoms, and TVT was routinely performed as part of the study surgery.
This anterior compartment analysis was planned to include data from both trials before outcomes were available for either study. In both studies, surgeons reported preoperatively whether they planned to do an AR at the time of the apical suspension and then reported at the time of surgery whether they actually performed an AR. The decision to perform an AR was left entirely to the discretion of the surgeon. Surgery was not concluded until the immediate surgical outcome was an anatomic success in all compartments. Postoperative POP quantification (POP-Q) measures were obtained by nonsurgeon study personnel.
Demographics, anthropomorphic measures, clinical characteristics, use of estrogen replacement therapy, previous urinary incontinence or POP treatment, and POP-Q10 were recorded at baseline. We report the individual surgeon rates of concomitant AR for all surgeons who participated in these surgeries. We compared the proportion of participants in each of the OPUS randomization groups (TVT vs no TVT) who underwent concomitant AR. Preoperative and postoperative POP-Q point Ba and the change in point Ba between baseline and 12 months after surgery were compared between groups who underwent SSLF or ULS with and without AR. Participants were considered an anterior anatomic success if (a) POP-Q point Ba was less than or equal to 0 cm, (b) the participant did not undergo subsequent surgery for anterior vaginal compartment prolapse, and (c) the participant did not use a pessary for prolapse at any point during the 1-year follow-up period. Participants were considered an overall anatomic success if (a) POP-Q points Ba, Bp, and C were less than or equal to 0 cm, (b) the participant did not undergo subsequent surgery for prolapse, and (c) the participant did not use a pessary for prolapse at any point during the 1-year follow-up period. Results were assessed for each specific vaginal vault procedure (SSLF or ULS) and for the total group combined (combined group). Data were also analyzed based on preoperative anterior compartment prolapse severity with a cut point that was selected a priori, dichotomizing point Ba ≤ +1 for “lower-stage” prolapse (stage ≤2) and point Ba > +1 for “higher-stage” prolapse (stage 3 or 4). The “all-stages” group included both lower- and higher-stage prolapse groups. Intraoperative complication rates were compared between the groups with and without concomitant AR.
Demographic and baseline characteristics were compared between participants with and without a concurrent AR within each vaginal vault procedure group using t tests for continuous measures, Kruskal-Wallis tests for scale measures, mean score tests with modified ridit scores for ordinal categorical measures, and χ2 tests for categorical measures. Change from baseline in Ba was analyzed within and across vaginal vault procedure groups using analysis of covariance models. For subgroup analyses based on preoperative POP-Q stage of the anterior vaginal compartment, the models included a corresponding categorical baseline stage variable and its interaction with concurrent AR status. Analyses of the success outcome were conducted in a similar fashion except using logistic models in place of analysis of covariance models. Analyses of outcome data were adjusted for baseline POP-Q points Ba and C as well as age and previous hysterectomy. Logistic regression models were also used to evaluate probability of success as a function of baseline POP-Q point Ba. Data were analyzed with SAS Version 9.4 (SAS Institute, Cary, NC) software.
A total of 701 women (ULS 476 [68%], SSLF 225 [32%]) met inclusion criteria for this study from the previously described PFDN surgical trials.6,7 Overall, concomitant AR was performed in 63% (441/701) of the participants: 59% (133/225) of SSLF procedures and 65% (308/476) of ULS procedures (N.S.). Changes to the preoperative plan occurred infrequently: AR planned but not performed in 9% (61/701) and performed although not planned in 1% (10/701).
In the OPTIMAL trial, all participants except two received a TVT. In the randomized population of OPUS in which one half of the participants received a TVT, a similar proportion in the TVT group (66%) and in the sham (no TVT) group (57%) underwent concomitant AR.
Table 1 displays baseline demographics and clinical characteristics of the study participants. Women who underwent concomitant AR were older, more often postmenopausal, on estrogen, not Hispanic/Latina, had previously undergone hysterectomy, and had more advanced anterior prolapse. The women with concomitant AR had on average 0.7 cm greater POP-Q point Ba (ie, greater anterior prolapse) in the SSLF group (P = 0.05) and 0.9 cm greater in the ULS group (P < 0.001) than women who did not undergo AR. However, women who underwent an AR did not have significantly more overall (POP-Q stage) or apical (POP-Q point C) prolapse at baseline.
Figure 1 displays a scatter plot of individual surgeon characteristics demonstrating how many surgeries each surgeon performed and what percent of those surgeries included a concomitant AR. Thirty-six surgeons at 10 clinical sites (range of 1–5 surgeons per site) performed the 701 surgeries. Thirty-five of the 36 surgeons are board certified in the Female Pelvic Medicine and Reconstructive Surgery subspecialty. The mean number of surgeries per surgeon in these trials was 20 (median = 15) and ranged from 1 to 103; the surgeon-specific proportion of surgeries with concomitant AR ranged from 0% to 100%; six (17%) of the surgeons performed AR less than 15% of the time and 12 (33%) performed an AR more than 85% of the time. There was no apparent relationship between the number of surgeries done and rate of concomitant AR. No significant differences in rates of intraoperative complications were observed between the groups with and without AR.
The primary outcome of our study is found in Table 2. Overall, anterior anatomic outcomes were marginally better with a concomitant AR (82% vs 80%, P = 0.03). This difference was largely the result of improvements in anterior success observed in the higher-stage prolapse subgroup (76% vs 69%, P = 0.03) and from the SSLF group (81% vs 73%, P = 0.02). The largest percentage group difference in anatomic success rates (17%) was observed among patients with higher-stage prolapse (preoperative Ba > +1) undergoing an SSLF (74% vs 57%, P = 0.02). Concomitant AR did not significantly improve anterior success rates for women undergoing ULS regardless of the preoperative stage of anterior prolapse. Similarly, there was no evidence that anterior success rates improved with an AR with either surgical procedure in any lower-stage prolapse group.
Overall anatomic success rates (which required success for points C and Bp, in addition to Ba) trended closely behind anterior anatomic success rates (Table 3) confirming the well-known observation that the anterior wall is the most common site for anatomic failures after apical vaginal native tissue surgeries. For the entire study group, overall anatomic outcomes were better with a concomitant AR (76% vs 72%, P = 0.002).
Most study participants (62/7011[89%]) contributed 12-month outcome data. In participants who underwent concomitant AR, Ba improved on average −1.8 (1.3) cm in the (Ba ≤ +1) lower-stage prolapse group and −4.4 (2.2) cm in the (Ba > +1) higher-stage prolapse group (Table 3). Similarly, in the participants not undergoing AR, Ba improved −1.4 (1.4) cm in the (Ba ≤ +1) group and −3.8 (2.1) cm in the (Ba > +1) group. Overall, improvements in point Ba from baseline to 12 months were observed in both vaginal vault procedure groups, and there was no significant difference in improvement in point Ba between women who underwent concomitant AR and those who did not (P = 0.27).
Overall, the anterior anatomic success rate was 81% (508/627) and the overall anatomic success was 75% (469/627) at 12 months (Table 4). Participants with lower-stage anterior vaginal compartment prolapse at baseline demonstrated a higher successful anterior prolapse outcome than those with higher-stage baseline anterior prolapse (93% vs 74%, P < 0.001) and had a significantly higher overall anatomic success rate (88% vs 61%, P < 0.001).
To determine whether there is a possible POP-Q inflection point that is predictive of overall anatomic success rates, we plotted the predicted probability of overall anatomic success against baseline anterior vaginal compartment prolapse as measured by preoperative Ba (Fig. 2). The associated receiver operator curve is also shown (Fig. 3). A point Ba of +2.5 had predicted probability of success of 0.76 with a sensitivity of 0.64 and a specificity of 0.62. Overall anatomic success rates were predicted at less than 50% if baseline Ba was greater than +6 cm. We also plotted the predicted probability of overall anatomic success rates for maximum preoperative prolapse from either Ba, C, or Bp (instead of just Ba), and the results and receiver operator curve curve were very similar (results not shown). Subgroup analysis for the SSLF and the ULS groups was similar to the overall group (data not shown).
The clinical question we wanted to address in this study is whether a concomitant AR should be performed at the time of native tissue vaginal apical surgery. We could not find high-quality evidence in the literature to guide this decision and indeed the variability related to surgeon bias on this topic was confirmed in our study. Without a randomized trial, our study provides some information from a large, well-characterized data set. Our hypothesis was confirmed, and our study population showed a modest general benefit of concomitant AR completed at the time of native tissue vaginal apical suspension for 1-year anterior anatomic outcomes. Although the difference in the overall group for both procedures was small, this difference was significant when controlling for preoperative baseline prolapse severity. There may be a larger benefit for those women with higher-stage baseline anterior prolapse and for women undergoing SSLF as we found a large difference favoring AR in women with higher-stage baseline anterior prolapse undergoing SSLF. These results and conclusions did not change when the outcome included postoperative prolapse in any compartment. We did not find any difference in intraoperative complications related to the AR, and we think that it is unlikely that there are long-term complications related to the procedure. Our results suggest that the current practice of surgeon discretion on whether to perform an AR or not in women with lower-stage prolapse or in women undergoing ULS remains viable. In addition, there is evidence in our analysis that an AR should be considered for women with higher-stage prolapse undergoing an SSLF.
Several investigators have reported that anterior compartment support defects are largely secondary to apical support loss.11,12 Our study supports that assertion, given that many women with anterior support loss had favorable surgical outcomes although they did not undergo specific AR. It is likely that the postoperative axis of the vagina is directed more posteriorly with the SSLF than the ULS, increasing the mechanical stress on the vulnerable anterior compartment.
Given the standardized inclusion criteria for these two clinical trials, these results suggest that the individual surgeon's practice pattern on whether concomitant AR should be performed with vaginal apical surgery was a major factor determining whether an AR was performed or not; the performance of a concomitant AR was very surgeon dependent with 17% of the surgeons performing an AR less than 15% of the time and 33% of surgeons performing an AR more than 85% of the time. Overall, however, our concomitant AR rate of 63% in this study is consistent with a 58% rate in results from a large single site doing ULS.13 Therefore, the findings of this study seem to be generalizable, given the participation of multiple centers and surgeons. The anterior wall is the most common site of recurrence after a vaginal apical surgery, consistent with our findings. The 81% anatomic success rate for the anterior compartment decreased to 75% when apical and posterior measures were added.
A secondary finding of this study is that it provides evidence for the common clinical belief that participants with higher-stage prolapse are more likely to have worse anatomic outcomes after native tissue vaginal apical repairs. In this study, those participants with higher-stage baseline anterior prolapse had significantly lower anterior and overall anatomic success rates than those with lower-stage prolapse. The predicted probability for anatomic success after a native tissue vaginal suspension decreases markedly for preoperative prolapse greater than 2.5 cm and is less than 50% if point Ba is greater than 6 cm. These data provide some evidence for doing other (eg, nonnative tissue) vaginal suspensions for patients with high-degree prolapse. This information is likely to be helpful in counseling patients considering specific prolapse surgeries or in designing future clinical trials for varying degrees of prolapse.
The strengths of our study are the large number of participants (>700), the 10 sites, and the 35 surgeons that make our study generalizable. Our observations are certainly limited by the lack of randomization for concomitant AR; however, evaluation of the study results in light of those potential biases supports our conclusions. We hypothesized that concomitant AR would be associated with a decreased risk of anterior vaginal prolapse at 1 year. The study showed a modest overall benefit in the unadjusted comparisons (82% success with AR to 80% without AR), but after adjusting for the observed confounding between AR and prolapse severity, the benefit was significant (P = 0.03). Furthermore, the benefit was clearly demonstrated among the most severe prolapse patients, allowing us to be fairly confident of these conclusions about the benefit of AR.
As with all studies, we recognize some limitations, including the point that our outcomes were only anatomic. We accepted this limitation as we wanted to address whether anatomy correction surgery produced an anatomic result, and we wanted a clean, easily understood anatomic outcome. Ideally, primary outcomes for most prolapse trials should also have subjective and symptomatic outcomes, and these composite outcomes probably better reflect patients' desired outcomes.
Additional surgery that is not associated with clinically relevant improvements in surgical outcome increases health care cost, surgical time, and potential surgical morbidity although we found no evidence of surgical morbidity. We hope that surgeons will find this information useful for surgical planning and counseling. Future surgical investigations evaluating anatomic outcomes may benefit from our treatment and effect size estimates to plan appropriately powered randomized surgical trials to eliminate surgeon bias as a possible confounder. Such studies are needed to provide the high-quality evidence that patients, colleagues, and payers deserve. For these native tissue vaginal apical surgeries, there is a strong relationship between preoperative prolapse severity and worse postoperative anatomic outcomes; this is important for patient counseling and possibly important for surgical procedure selection.
1. Wu JM, Matthews CA, Conover MM, et al. Lifetime risk of stress urinary incontinence or pelvic organ prolapse surgery
. Obstet Gynecol
2. Denman MA, Gregory WT, Boyles SH, et al. Reoperation 10 years after surgically managed pelvic organ prolapse
and urinary incontinence. Am J Obstet Gynecol
2008;198(5):555. e1–555. e5.
3. Holley RL, Varner RE, Gleason BP, et al. Recurrent pelvic support defects after sacrospinous ligament fixation
for vaginal vault prolapse. J Am Coll Surg
4. Fialkow MF, Newton KM, Weiss NS. Incidence of recurrent pelvic organ prolapse
10 years following primary surgical management: a retrospective cohort study. Int Urogynecol J Pelvic Floor Dysfunct
5. Morgan DM, Rogers MA, Huebner M, et al. Heterogeneity in anatomic outcome of sacrospinous ligament fixation
for prolapse: a systematic review. Obstet Gynecol
6. Wei JT, Nygaard I, Richter HE, et al. A midurethral sling to reduce incontinence after vaginal prolapse repair. N Engl J Med
7. Barber MD, Brubaker L, Burgio KL, et al. Comparison of 2 transvaginal surgical approaches and perioperative behavioral therapy for apical vaginal prolapse: the OPTIMAL randomized trial. JAMA
8. Wei J, Nygaard I, Richter H, et al. Outcomes following vaginal prolapse repair and mid urethral sling (OPUS) trial—design and methods. Clin Trials
9. Barber MD, Brubaker L, Menefee S, et al. Operations and pelvic muscle training in the management of apical support loss (OPTIMAL) trial: design and methods. Contemp Clin Trials
10. Bump RC, Mattiasson A, Bo K, et al. The standardization of terminology of female pelvic organ prolapse
and pelvic floor dysfunction. Am J Obstet Gynecol
11. Rooney K, Kenton K, Mueller ER, et al. Advanced anterior vaginal wall prolapse is highly correlated with apical prolapse. Am J Obstet Gynecol
12. Crosby EC, Sharp KM, Gasperut A, et al. Apical descent in the office and the operating room: the effect of prolapse size. Female Pelvic Med Reconstr Surg
13. Silva WA, Pauls RN, Segal JL, et al. Uterosacral ligament vault suspension: five-year outcomes. Obstet Gynecol