Long term Follow-up of Transvaginal Anatomical Implant of Mesh in Pelvic organ prolapse

Transvaginal mesh (TVM) is a minimally invasive but effective treatment for pelvic organ prolapse (POP). However, mesh exposure is a common and problematic complication after TVM. This study assessed the safety and long-term outcomes of TVM. A retrospective review was performed on the medical records of 175 consecutive patients who underwent TVM with the anatomical implant technique for pelvic organ prolapse at our center from April 2007 to December 2012. All operations were performed using TVM with the anatomical implant technique. Intraoperative variables, postoperative complications, and TVM outcomes were assessed. In average of 8 years (ranging from 4 to 10 years), the objective cure ratio reached 99.4%; and the subjective success rate of the TVM operation was 91.4%. Only 2 cases (1.1%) were identified as having mesh exposure. The reoperation rate was 4.0% (95% CI, 1.1–6.9%). No patients abstained from sex due to the operation or postoperative discomfort. Our anatomical implant technique for correcting POP is feasible in TVM procedures, which lead to favourable subjective and objective outcomes with the lowest rates of mesh exposure (1.1%) in published data. Therefore, performing TVM operations with the appropriate technique could consider to be permitted.

Patient positioning. A urethral catheter was inserted under general anaesthesia. Patients were placed in a dorsal lithotomy position, with the legs were supported by Allen stirrups and the arms were tucked beside the body, ensuring that pressure points were padded. Patients were then appropriately prepped and set in an 80° Trendelenburg position.

Anatomical implant technique. Well-performed hydrodissection beneath vaginal adventitia is crucial
to preserve the capillary network within the vaginal adventitia and for placing the mesh into the correct layer. During TVM procedures, the mesh was placed under the adventitial layer (beneath the full thickness of the vaginal wall), in the monolayer and continuous space consisting of visceral endopelvic fascia and vesicovaginal or rectovaginal adventitia. In details, 30-50 ml of normal saline was injected forward, left and right through one needle inserted to the middle line of the vaginal wall (Fig. 1a). The hydrodissection target layer was the potential septum beneath the vaginal adventitia. The vaginal wall was characterized by wide range swelling with normal surface after injection into the right layer, indicating that a water sac formed beneath the full thickness of the vaginal wall and the blood supply was appropriate (Fig. 2). If the injection failed to reach the correct layer, then only a focal water vacuole would be formed in the lamina propria or muscularis; and this area of the vaginal wall would become thin and white.
After hydrodissection, a vertical incision was made at vaginal wall to the water sac (Fig. 1b). Haemorrhaging could be seen until the incision penetrated the vaginal adventitia into the water sac layer; then clear, normal saline would leak from the incision without bleeding. Next, the water sac was cut to the left and right beside the vaginal adventitia to make a space for the mesh, clearing the operation field so the bleeding could be controlled. Notably, the capillary network can be seen in the transparent, thin layer of fascia on the back of the vaginal adventitia after a well-performed hydrodissection.  Four skin incisions were made on the genitocrural crease: two in the anteromedial edge of the obturator foramen at the level of the urethra and the other two at 2 cm below and 1 cm lateral to the first ones. Bilateral passage of the two upper cannula-equipped guides at 1-2 cm of the prepubic portion of the arcus tendinous fascia pelvis (ATFP) and bilateral passage of the two lower cannula-equipped guides at 1-2 cm of the distal portion of the ATFP (1 cm from the ischial spine) allowed each prosthetic arm to be passed through the obturator foramen (Fig. 1c). Next, the mesh was positioned tension-free under the bladder (Fig. 1d) and was fixed to the bladder neck and uterine isthmus with a Prolene suture to support the uterus. Posterior implant placement was performed similarly to the anterior implant per the manufacturer's instructions.
Electrocauterization was not performed during the operation, as it would damage the capillary network. Clipping the haemorrhagic area using tissue forceps is appropriate when incising the vaginal wall. A well-performed hydrodissection ensures a clear procedural field. A standardized procedure minimizes the surgical time,and adequately controls blood loss. The estimated blood loss was calculated by the suction volume in post-operation plus the estimated Laps blood loss (1 lap rinse equal to 10 ml).   Subjective success was defined when the patient reported the absence of a bulge and the absence of unacceptable symptoms, such as pain or incontinence 6 . The Chinese versions of the Pelvic Floor Distress Inventory (PFDI-20) [9][10][11] and the Patient Global Impression of Improvement (PGI-I) 12 were used to quantify the outcomes, while the Visual Analogue Scale (VAS) 13 was used to quantify pain, if any. Data analysis. Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) 19.0 for Windows (IBM, New York, NY). Unless specified, normally distributed continuous parameters were presented as the mean ± SD (standard deviation) and analysed using paired Student's t-test in the pre-and postoperative comparison; non-normal distribution parameters were presented as median (range); Categorical parameters were presented as number (percentage) and analysed using chi-square test. A p-value of less than 0.05 was considered statistically significant.

Results
Among the 175 patients, the mean age was 64.5 ± 11.4, the average parturition was 4.1 ± 1.8, the median duration of disease was 3 years (range from 1 month to 50 years, and a quartile of 1 year and 10 years) and the median follow-up was 8 years (range from 4 to 10 years). 12 patients (6.9%) had a surgical history of total or subtotal hysterectomy due to benign causes. Through In correlation analysis, patient age distribution was positively correlated with each point of POP-Q (p < 0.001), but duration of disease, BMI and parturition times were not (p > 0.10). In addition, pelvic prolapse symptoms were positively correlated with each POP-Q point (p < 0.001).
The anatomical implant technique was applied in all patient operations. 36 cases of Prolift A, 114 cases of Prolift T, 4 cases of Prolift P, 3 cases of Prosima A and 18 cases of Prosima C were performed; 25 cases of Tension-free Vaginal Tape Obturator (TVT-O) and 4 cases of Tension-free Vaginal Tape (TVT) were performed at the same period of surgery. The operation duration was 62.1 ± 17.1 minutes, and estimated intra-operative blood loss was 64.5 ± 46.1 ml. The reoperation rate was 4.0% (95% CI, 1.1-6.9%). Two bladder perforation cases were discovered during the operation. No intra-operative complications occurred in the remaining patients.
The postoperative POP-Q point distribution was improved (p < 0.001) from the pre-operation (Table 2). At a median of 8-year follow-up, only 1 patient had a relapsed prolapse (after Prosima C), and the objective cure ratio reached 99.4%. The average postoperative hospital stays were 4.0 ± 1.0 days, and urethral catheters were usually removed 2 days after the operation. Two cases (1.1%) were verified to have mesh exposure on the vaginal wall.
After TVM, patients' subjective symptoms were improved (Table 1), and the subjective success rate of the TVM operation performed in 175 patients was 91.4% (95% CI, 87.3-95.6%). 17 patients reported chronic pain and discomfort on the perineum, operative incision or puncture area (9.7%; 95% CI, 5.3-14.1%) ( Table 3). 24 patients with an average age of 50.0 ± 7.2 years were sexually active, in which 22 patients reported an improved quality of postoperative sex, and no patients abstained from sex due to postoperative discomfort (Table 3).
Ninety-six PFDI-20 questionnaires (54.9%) were effectively completed with a total of 97.0 ± 32.1 and 34.6 ± 19.0 points for pre-operation and post-operation, respectively (p < 0.01) (Fig. 3). The average postoperative PGI-I score was 1.43 ± 0.71 with statistical improvement (95% CI: 1.32-1.54) in comparison with preoperative PGI-I. When the patients were asked whether they would have the operation again with the clear knowledge of their postoperative life state, the responses were all positive; and they would recommend this operation to other patients.

Discussion
This study was conducted at a single high-volume center to describe the anatomical implant technique. The operation's objective success rate was 99.4%, and the subjective success rate was 91.4%. Only 2 cases (1.1%) were verified to have mesh exposure without symptoms. Our results demonstrate that the transvaginal anatomical implant technique is effective and safe. Therefore, performing the TVM operation using the appropriate technique should be considered to permit.
The unique advantages of this vaginal surgery include, having no incision in the abdominal wall; being performed via the abdominal cavity; having a short operating time and a fast postoperative recovery; and being relatively safe for older and chronic patients. Synthetic mesh application enhances the success rate of the vaginal pelvic reconstructive operation [14][15][16][17] , solving the postoperative problems of the traditional repair operation, such as diminished vaginal volume and dysuria, and achieves the goal of recovering pelvic functions 14 Transvaginal incision delayed healing 2 1 Cd/T2/S1 1Bd/T2/S1 Conservative treatment for one case, re-sutured for another in office.
POP recurrence 1 -ASC was performed 9 months after the initial surgery. Further surgery success.
One case with small amount of vaginal discharge.
Bladder wall mesh exposure 1 4 A/T4/S3 Holmium laser cystoscopy lithotripsy and remove of foreign body Symptoms relieved after further operation.
No severe frequency occurrence with limited impact on quality of life.
Urinary incontinence 26 4B/T3/S1 Further surgery for severe incontinence (4 cases during follow-up) and expectant or conservative treatment for others.
Symptoms controlled by re-interventions.
Unusual vaginal discharge 4 1 C/T3/S1 1 Cc/T3/S1  long-term retention of the synthetic mesh may result in complications 19 . Therefore, some experts believe that the TVM operation should not be recommended after weighing the advantages and disadvantages; because although it may result in a better recovery, but the related complications generate other problems 3,4 . A primary complication of the synthetic mesh is mesh exposure. Some research shows that the mesh exposure rate is between 4% and 19% 14,20 . Only 2 patients (2/175, 1.1%) experienced mesh exposure in our study, which is the lowest in the published data on TVM. This mesh exposure rate is even lower than that of ACS (3%) 21 or LSC (2.5%) 21 .
The anatomical implant technique aims to maintain the integrity of all vaginal wall layers; and their blood is supplied by placing the mesh in the correct anatomical space. The full thickness of the vaginal wall is composed of three layers: the lamina propria, muscularis, and adventitia. The adventitia is a variable, discrete layer of collagen, elastin, and adipose tissue containing blood vessels, lymphatic vessels, and nerves. The capillary network within the vaginal adventitia is crucial for the tissue activity and healing ability of the vaginal wall. Stiffness and mesh exposure in the vaginal wall are related to the damaged capillary network within the vaginal adventitia. Therefore, the TVM procedure should minimize injury to this area, which helps to decrease haemorrhaging and to preserve the intact blood supply to the vaginal wall, ensuring tissue vitality and preventing possible complications from the mesh.
In surgical terms, the layer exposed after dissecting between the lamina propria and muscularis in the anterior vaginal wall is the "pubocervical fascia" 22,23 , and the same layer in the posterior vaginal wall is the "rectovaginal fascia" 24,25 . Traditional prolapse repair surgery for POP is performed based on these fascias. TVM procedures place the mesh under the adventitial layer (beneath the full thickness of the vaginal wall), in the monolayer and continuous space consisting of the visceral endopelvic fascia and the vesicovaginal or rectovaginal adventitia, but different from the fascias in traditional prolapse repair. Otherwise, a portion of the mesh will remain under the lamina propria layer or in the muscularis of the vaginal wall, rendering the vaginal wall incomplete.
Mesh exposure is also related to the surgeon's experience 25 . We devote ourselves to standardizing and improving our operational skills, and we believe that positive technical exchange and training reduces the incidence of mesh exposure. Through follow-up visits with 175 patients, the lower mesh exposure rates should be attributed to proper operating skills. We cannot conclude that the main factor in reducing mesh exposure is the anatomical implant technique; however, this technique has better postoperative outcomes, and thus requires further study. We also expect that more clinical research will report the TVM experiences of the patients.