Transischioanal trans-sacrospinous ligament rectocele repair with polypropylene mesh: a prospective study with assessment of rectoanal function
- First Online:
- Cite this article as:
- Mourtialon, P., Letouzey, V., Eglin, G. et al. Int Urogynecol J (2013) 24: 81. doi:10.1007/s00192-012-1813-0
- 232 Views
Introduction and hypothesis
Despite good anatomical outcomes of pelvic organ prolapse (POP) repair by the vaginal route using synthetic mesh, complications limit their use. Clinical data are needed to generalize prolapse mesh repair by the vaginal route. The current study aims to evaluate midterm rectoanal function and clinical outcomes after transischioanal rectocele repair using a medium weight polypropylene mesh.
Between March 2003 and June 2004, 230 patients with stage II–IV anterior and/or posterior POP were included in a prospective multicenter study. The current study is based on the analysis of the 116 patients who underwent a rectocele repair via the infracoccygeal route through the sacrospinous ligament. Anatomical cure was defined when rectocele was at stage <II in the Pelvic Organ Prolapse Quantification (POP-Q) system. Postoperative functional results were evaluated using the self-administered Pelvic Floor Distress Inventory (PFDI) and the Pelvic Floor Impact Questionnaire (PFIQ).
Of the 116 patients who received a posterior mesh with two arms via the infracoccygeal route through the sacrospinous ligament, midterm anatomical results were available for 78 women representing 67 % (78/116) of the operated patients. The mean follow-up was 36 (± 8.1) months. No rectal injury occurred during surgery. The objective success rate was 94.8 % and subjective (by patient satisfaction) was 93.23 %. Colorectal-Anal Impact (CRAI) and Colorectal-Anal Distress Inventory (CRADI) scores were both significantly decreased at midterm follow-up in comparison with baseline (42.7 at baseline vs 11.4 at 24- or 36-month follow-up, p = 0.001 for CRAI, and 81.1 vs 34.4, p < 0.001 for CRADI) highlighting the benefits of rectocele repair on colorectal-anal function.
Polypropylene mesh with two arms via the infracoccygeal route through the sacrospinous ligament has good anatomical results at midterm follow-up with significant improvement in symptoms and quality of life and is associated with few complications. Obstructive symptoms reported in cases of rectocele can be improved by transvaginal mesh repair.
KeywordsGenital prolapseVaginal surgeryMedium weight polypropylene meshQuality of lifeRectoanal function
Pelvic organ prolapse
Pelvic Organ Prolapse Quantification system
International Continence Society
Pelvic Floor Impact Questionnaire
Pelvic Floor Distress Inventory
Colorectal-Anal Impact Questionnaire
Colorectal-Anal Distress Inventory
Pelvic organ prolapse (POP) repair is one of the most common operations in postmenopausal women. Pelvic reconstructive surgery can be performed by the abdominal or vaginal route. The abdominal route has been considered the gold standard since the middle of the last century because anatomical success was higher than with the vaginal route . The use of a nonabsorbable prosthetic mesh in vaginal surgery was first introduced by Julian in 1996 , who showed in a randomized study a significant reduction of recurrence rate when a Marlex mesh was used as tissue support, in comparison with colporrhaphy.
Since then, several prospective randomized studies have shown that use of prosthetic reinforcement by the vaginal route is more effective than plication and/or colporrhaphy for anterior repair [3–8]. These papers report outcomes of cystocele repair but we have few data on long-term follow-up concerning rectocele repair with mesh. In a recent Cochrane review on surgical management of POP in women, Maher et al. concluded that the limited evidence suggests that posterior vaginal wall repair (traditional colporrhaphy) may have a better anatomical success rate than transanal repair in the management of the posterior vaginal wall prolapse but the clinical effects are uncertain . There was not enough evidence about whether to use graft materials in the posterior compartment.
Different techniques have been performed to treat rectocele. Abdominal surgery can be done by laparotomy or laparoscopy with fixation of the rectum to the promontory (rectopexy). Vaginal surgery can be done with classic plication of the fascia on the middle line or with mesh interposition. Some surgeons used a transperineal approach. More recently the stapled transanal rectal resection (STARR) was developed.
van den Esschert et al.  reported results of laparoscopic ventral rectopexy for obstructed defecation syndrome; 15 of 17 patients had an improvement of their defecation problem. The median operating time was 199 min; one conversion laparotomy was required, six patients had postoperative complications, and one required removal of rejected mesh.
The cohort of the French Ugytex Study Group including 230 patients with symptomatic POP from 13 centers provides data on rectocele repair using a polypropylene mesh with two arms via the infracoccygeal route through the sacrospinous ligament. This cohort study aims to evaluate the use of a medium weight polypropylene mesh protected by an absorbable hydrophilic coating in prolapse repair by the vaginal route. Promising short-term results have already been published. We provide now the final results on midterm (>24 months follow-up) anatomical success, rectoanal function, and complications following rectocele mesh repair.
Material and methods
Institutional Review Board or Ethics Committee approval is not required in France for a cohort study using a medical device with a European or French approval for sale on the market.
The preoperative evaluation consisted of a medical history and urogynecological examination. All departments used a standardized case report form. Clinical evaluations were performed using a simplified International Continence Society (ICS) Pelvic Organ Prolapse Quantification staging system (POP-Q), assessing anterior vaginal wall prolapse, uterine or vaginal vault prolapse, and posterior vaginal wall prolapse on maximum Valsalva effort in the seated semi-lithotomy position. A bivalve speculum was used to appreciate POP-Q with a full bladder. In order to evaluate symptoms and quality of life, each patient was asked to answer the French version of the validated Pelvic Floor Distress Inventory (PFDI) and the Pelvic Floor Impact Questionnaire (PFIQ) . The Urinary Dysfunction Measurement (MHU) scale and the presence and severity of dyspareunia were also investigated. Each patient recruited had at least a symptomatic vaginal wall prolapse at stage II–IV in the ICS POP-Q staging system (Ba or Bp ≥ −1) and an impairment of her quality of life.
All patients were operated by the vaginal route using a specially designed prosthetic mesh (Ugytex™, Sofradim-Covidien, Trévoux, France).
It is a medium weight (38 g/m2) and macroporous (pore size 1.5×1.7 mm) polypropylene monofilament mesh offering tissue ingrowth and connective differentiation. Furthermore, the mesh is coated by a hydrophilic film composed of atelocollagen, polyethylene glycol, and glycerol. The absorbable coating is supposed to protect delicate pelvic viscera from the risk of acute inflammatory reaction during the healing inflammatory peak.
Concomitant procedures such as vaginal hysterectomy or sacrospinous suspension were performed according to each surgeon. When the patient had associated preoperative stress urinary incontinence, a midurethral sling was placed after the anterior vaginal skin closure, by a separate incision. Intra- and postoperative complications were recorded on a standardized case report form. Immediate postoperative complications were defined as those occurring during the first 30 days.
Postoperatively, patients were instructed to rest for 2 weeks; they were allowed to return to work after 4 weeks and to participates in sports or to have sexual intercourse after 6 weeks.
The postoperative follow-up was scheduled at 6 weeks, 6 months, 1 year, 2 years, and then 3 years after surgery. At each visit, urogynecological examination was performed using the POP-Q system on maximum Valsalva effort in the seated semi-lithotomy position. Objective anatomical cure was defined when the anterior vaginal wall for the cystocele and the posterior vaginal wall for the rectocele were at stage 0 (optimal outcome) or I (satisfactory outcome).
Baseline data, medical and surgical histories, surgery, and then pre- and postoperative clinical assessment data were collected using a standardized case report form. Pre- and postoperative functional results of symptoms, quality of life, and sexuality were evaluated using the self-administered PFDI and PFIQ, the MHU scale, and questions related to presence and severity of dyspareunia.
We report the results of patients who underwent a rectocele repair via the infracoccygeal route through the sacrospinous ligament with at least 24 months follow-up.
Statistical analysis was performed using Minitab v15.0 for Windows® (Minitab Inc., State College, PA, USA).
Qualitative data were compared using a Pearson chi-square or a Fisher exact test if the assumptions of the chi-square (theoretical frequency less than 5) were not met, while continuous data were compared using a t test (parametric) or nonparametric Mann–Whitney–Wilcoxon test if t test assumptions were not met. A p value < 0.05 was considered statistically significant.
Baseline demographics and previous surgery
Median (min.; max.)
63 (33; 91)
24.8 (19.5; 37.1)
2 (0; 5)
Previous prolapse repair
Previous surgery for incontinence
Cystocele repair (all patients have a Ugytex mesh)
Sacrospinous ligament fixation
ICS POP-Q stages and Bp measurement
Baseline (n = 78)
6 months (n = 70)
1 year (n = 62)
2 years (n = 64)
3 years (n = 57)
Stage 0a, c
Stage Ia, c
Median (min.; max.)
1 (−5; 10)
−4 (−8; 1)
−4 (−7; 1)
−4 (−6; 1)
−2.9 (−6; 1)
Comparison of anatomical outcomes of patients with or without anterior repair
Patients without anterior repair (n = 13)
Patients with anterior repair (n = 65)
Mean (SD) median (min.; max.)
Mean (SD) median (min.; max.)
−3.2 (2.1) −3 (−7; 1)
3.2 (3.1) 2 (−2; 9)
−4.3 (1.4) −5 (−6; −2)
−2.1 (1.5) −2 (−6; 2)
0.8 (1.9) 1.5 (−3; 3)
0.8 (3.8) 0 (−5; 10)
−3.2 (2.3) −3.5 (−6; 1)
−2.8 (2.8) −3 (−6; 0)
When considering the posterior compartment at 3 years, the erosion rate was 11.5 % (9/78). Three of nine patients needed surgery subsequent to an erosion. The surgery rate for erosion on the posterior compartment was 3.8 % (3/78). All the other erosions were managed medically with topical estrogen treatment and vaginal antiseptic. No patient required removal of the posterior mesh.
n = 78
Perioperative adverse events
Rupture of lateral arm of the mesh
Immediate postoperative adverse events
Hematoma of pararectal fossa requiring surgery
Late adverse events
Erosion without surgery
Surgery for erosion
Surgery for recurrence
Surgery for de novo prolapse
De novo dyspareunia
CRADI (PFDI) and CRAIQ (PFIQ) scores
p (t test)
Mean (SD) median (min.; max.)
Mean (SD) median (min.; max.)
42.7 (53.2) 10.3 (0; 197.2)
11.4 (33.3) 0 (0; 214.4)
81.1 (61.5) 69 (0; 258.6)
34.4 (44.2) 13.4 (0; 202.9)
The subscale of the Colorectal-Anal Distress Inventory (CRADI) was more specifically analyzed showing that 70.5 % (48/68) reported obstructive symptoms, 53 % (35/66) incontinence symptoms, 70 % (47/67) irritative/pain symptoms, and 23.8 % (16/67) symptoms of rectal prolapse.
Obstructive scale score
No. of patients with item > 1
% of patients with item > 1
p (t test)
We did not find a correlation between Bp measurement and obstructive scale of symptoms.
Concerning sexual activity, 64.5 % of the patients (49/76) were sexually active before surgery versus 61.8 % (47/76) after surgery. One patient reported de novo dyspareunia after surgery but had sexual activities. One patient reported dyspareunia before surgery that disappeared after the surgery.
This study is the first with the longest follow-up about rectocele repair with polypropylene mesh by the transischioanal route (MEDLINE; 1966–January 2012; English and French languages; search terms: “rectocele,” “posterior repair,” “mesh,” “graft”). Of the patients, 93.2 % are satisfied with the surgery and the failure rate is low at about 5 % (4/78). For the four patients with anatomical failure using the POP-Q staging system, only one was dissatisfied with the results. All these patients had a stage II on the POP-Q. The one who was dissatisfied had POP-Q point Bp at +1, whereas the three who were satisfied had POP-Q point Bp at 0. In accordance with other authors we think that the POP-Q staging system is deficient in accounting for the reality between symptoms and anatomy especially for stage II [15, 16] because stage II ranges from −1 to +1 and symptoms in patients are very different depending on whether the bulge is inside or outside of the vagina. Some patients report very good satisfaction, the objective anatomical cure is optimal (stage 0), but the score for symptoms still remains high. This is consistent with Cundiff and Fenner’s finding that patient satisfaction correlates with alleviation of defecatory symptoms but not anatomical cure .
Comparing our cure rate based on the POP-Q with other studies of transvaginal repair with polypropylene mesh, the anatomical success rate with transischioanal repair ranges from 91 % (Altman et al. ) to 100 % (Flam , Gauruder-Burmester et al. ).
To date, rectocele repair with polypropylene mesh is still a subject of discussion. Indeed, data concerning cystocele repair show a low rate of recurrence after use of polypropylene mesh interposition compared to plication even if the rate of symptomatic recurrence did not differ between the two groups. For posterior compartment repair by the transvaginal route, we did not find any randomized controlled trial comparing polypropylene mesh interposition versus plication of the rectovaginal fascia. Sand et al. in 2001  and Paraiso et al. in 2006  examined rectocele repair with and without mesh reinforcement.
In the randomized controlled trial (RCT) of Paraiso et al. , patients were randomly assigned to either posterior colporrhaphy, site-specific rectocele repair, or site-specific rectocele repair augmented with a porcine small intestinal submucosa graft (Fortagen, Organogenesis Inc., Canton, MA, USA). The primary outcome was anatomical cure of posterior vaginal prolapse with a POP-Q point Bp ≤ to −2 at 1 year after surgery. The results were significantly poorer in the group who underwent rectocele repair with graft augmentation. In this study graft augmentation does not seem to improve results, but the technique of mesh placement is very different from the transischioanal route. Failure rate was 46 % for mesh augmented repair, 22 % for site-specific repair, and 14 % for posterior colporrhaphy. In this study 37 % of the patients received a hysterectomy, 62 % an anterior colporrhaphy, and 78 % a perineorrhaphy.
In our study, it does not seem that rectocele repair with mesh leads to de novo POP in the untreated compartment. It is true that the Ba point was very different at baseline for our study.
In a recent paper, Withagen et al.  found that using the Prolift (Ethicon Women’s Health and Urology, Somerville, NJ, USA) system 23 % of all patients developed a de novo POP stage ≥ II in the untreated compartment. This occurred, respectively, in 46 and 25 % of patients after an isolated anterior and isolated posterior Prolift™. These results suggest that overcorrection of the treated compartment leads to de novo POP in the untreated compartment. On the other hand, Hiltunen et al.  did not find any difference in the posterior compartment between the groups with anterior mesh versus the group without mesh, but in this study patients of both groups had a posterior colporrhaphy in case of posterior wall prolapse. Furthermore, the anterior mesh was not fixed by the transobturator route. In a paper comparing “Perigee and/or Apogee” (American Medical Systems, Minnetonka, MN, USA) versus “Prolift anterior and/or posterior,” Long et al.  found no difference between the two groups except that the POP-Q points Aa and Ba of the Prolift group were significantly higher than in the other group (p < 0.001). This could explain the high rate of de novo POP in the untreated compartment. Finally, the need for posterior repair with mesh in case of anterior mesh to avoid de novo rectocele is still unclear. Probably many factors play a role: stage of prolapse at first repair, strength of the repair with mesh, techniques, mesh types, and patient factors (high body mass index, poor collagen, important physical activities).
In their review on efficacy and safety of using mesh or grafts in surgery for POP, Jia et al.  reviewed posterior vaginal wall repair including 9 studies involving 417 women. They concluded that there were too few data reported for any of the outcomes to draw reliable conclusions or to carry out further statistical analyses.
In our study, peri- and postoperative complications are rare. Any rectal injury occurs during surgery.
Hemorrhage requiring blood transfusion is the only one major complication of the study. In our opinion, surgery for mesh exposure is acceptable. No infection of mesh occurred.
In this current study, we demonstrated a significant improvement of the CRADI score after rectocele repair with mesh by the transischioanal route.
Obstructive symptoms were the most frequently reported symptoms related to rectocele in our study. Surgical repair with mesh can significantly improve obstructive symptoms.
This result is consistent with the paper of Gustilo-Ashby et al.  reporting bowel symptoms 1 year after surgery of rectocele from patients of the randomized trial of Paraiso et al. . They demonstrated resolution and improvement in all bowel symptoms evaluated in a randomized prospective trial of three different surgical techniques of rectocele repair by the vaginal route . They also found that anatomical cure is associated with a reduced risk of postoperative staining and a feeling of incomplete emptying. Restoration of normal posterior vaginal wall support was only associated with a reduced risk of bothersome incomplete emptying, but not with other bowel symptoms.
Improvement of defecatory symptoms after rectocele repair remains unclear; several studies reported good outcomes while others showed impairment of bowel symptoms.
Kahn and Stanton  demonstrated an increase in constipation, incomplete bowel emptying, and incontinence of feces, but this study was retrospective and the technique was a traditional colporrhaphy.
To evaluate the defecatory dysfunction, we used the obstructive subscale of the CRADI. The CRADI is a scale of the PFDI. The PFDI was validated with the PFIQ by Barber et al. in 2001 . The PFDI has 46 items and contains 3 scales, the Urinary Distress Inventory (UDI), the Pelvic Organ Prolapse Distress Inventory (POPDI), and the CRADI. The CRADI consists of 17 items related to lower gastrointestinal tract dysfunction and has 4 subscales (obstructive, incontinence, pain/irritation, and rectal prolapse). The following three items are used to assess the obstructive subscale: Question 8: Do you usually have to push on the vagina or around the rectum to have or complete a bowel movement?; Question 9: Do you feel you need to strain too hard to have a bowel movement?; Question 10: Do you feel you have not completely emptied your bowels at the end of a bowel movement? The obstructive subscale demonstrated the strongest association with vaginal or perineal splinting and a diagnosis of defecatory dysfunction and also correlated with a history of hard straining more than 25 % of the time. Altomare et al. had validated  a new scoring system to assess obstructed defecation syndrome (ODS). They concluded that the ODS score offers a validated severity of disease index in grading the severity of disease and monitoring the efficacy of therapy.
No study had compared the obstructive subscale of the PFDI to the ODS scoring system for patients with ODS. According to Barber et al. the PFDI serves the role of both a symptom inventory and a measure of the degree of bother and distress caused by the broad array of pelvic symptoms.
Our study shows a significant improvement for all the items of the obstructive subscales of the PFDI. This suggests that ODS was due to rectocele; indeed anatomical or functional abnormalities can be responsible for ODS. Functional abnormalities included anismus, dyssynergy defecation, and non-relaxation syndrome and anatomical abnormalities included rectocele, rectal intussusception (internal rectal prolapse), mucosal prolapse, enterocele, and perineal descent. Rectocele and rectal intussusception, however, are the two most frequent anatomical defects associated with ODS. Patients were not looking for all these pathologies. Furthermore, no more exploration such as defecography or rectal manometry had been performed to exclude the other etiologies.
We did not find a relation between obstructive bowel symptom and Bp measurement as in others studies , confirming that symptoms are independent of anatomical prolapse stage.
Some biases existed in this study. Many other surgical procedures were done during the rectocele repair with mesh. This affects the entire pelvic floor dynamics and can have a role in anatomical and symptom improvement. Different techniques of cystocele repair with mesh were performed (retropubic, four arms in the obturator foramen, free) and could alter the results. All pre- and postoperative clinical examinations were evaluated by the surgeons who performed the surgery and not by an independent observer. POP-Q measurements, PFIQ, and PFDI were not completed for all of the patients and up to 33 % were lost to follow-up. No questionnaire about sexual activities was used. The authors did not receive any grant regarding this study. The clinical research department of Sofradim-Covidien (Trévoux, France) realized the data collection (database) using the case report form processed by the surgeon himself. Sofradim-Covidien participated in the data analysis and has not been involved at all in the contents of the manuscript.
In conclusion, this study reports good anatomical and functional outcomes with a low rate of complications. Obstructive bowel symptoms are the most common with rectocele and can be improved with this repair. The relation between symptom improvement and posterior vaginal wall repair is unclear and probably included factors other than anatomical cure. More studies are needed to compare rectocele repair with or without mesh and with other approaches (abdominal, perinea, and transanal) and improvement of bowel symptoms.
We thank all the other investigators of the French Ugytex Study Group for their participation in the study: Jean-Louis Bénifla and Fabien Demaria (Rothschild Hospital, Paris), Jérome Blanchot and Pierre-Louis Broux (La Sagesse Private Hospital, Rennes), Michel Cosson and Jean-Philippe Lucot (Jeanne De Flandre University Hospital, Lille), Patrick Delporte (Public Hospital, Dunkerque), Hervé Fernandez (Antoine Béclere University Hospital, Clamart), Christine Frayret (Public Hospital, Chartres), François Hacquin (Santa Maria Private Hospital, Nice), Kazeem-Marc Maaliki (Franche Comté Private Hospital, Besançon), Loïc Marpeau and Fabrice Sergent (University Hospital, Rouen).
We thank Emilie Joyeux for her contribution to the picture.
Conflicts of interest
Renaud de Tayrac is consultant for Boston scientific. This study was funded by Sofradim-Covidien.