Effect of anterior vaginal wall prolapse repair by modified transvaginal mesh surgery: a retrospective cohort study

To explore clinical outcomes and complications of modified Transvaginal mesh (M-TVM) for advanced anterior vaginal wall prolapse in 1 year follow-up. 574 patients underwent TVM surgeries from 2019 to 2020 were collected and divided into TVM group and M-TVM group, all preoperative and postoperative data was obtained and compared between the two groups. 285 women were involved eventually, including 181 in TVM group and 104 in M-TVM group. No significant difference of general conditions was found between these two groups. After long-term follow-up, patients in TVM group were more likely to suffer from pelvic pain than M-TVM group (P = 0.046). Meshes seemed much wider in M-TVM group (4.5 ± 0.69 cm) than in TVM group (3.0 ± 0.91 cm). No matter TVM or M-TVM, surgeries can significantly change point Aa and Ba when compared to preoperative data. Compared to TVM group, point C and D were significant changed in patients in M-TVM group after surgery (P < 0.001) M-TVM is a commendable procedure that can significant correct anterior prolapse with mesh extended wider, and also supply stable apical support at the same time.


Introduction
Anterior vaginal wall prolapse is the most common type of pelvic organ prolapse (POP), which is regarded as a benign condition that occurs in 40-60% of women [1]. The etiology of anterior vaginal wall prolapse was different. One early theory focused on attenuation and weakening of the underlying vaginal layers [2]. Another theory suggests that anterior wall defects generate due to a break in the connective tissue of the pelvic floor musculature to the fibrous thickening of the levator ani fascia or the white line [3]. More recently, it has been proposed that most anterior wall defects result from impairment of the pubovisceral portion of the levators and the apical vaginal suspension complex (namely the cardinal and uterosacral ligaments) [4]. These new theories may remind the surgeons that enhanced apical support at the same time of repairing anterior vaginal wall may be a new strategy of improving efficiency of anterior vaginal wall prolapse repair surgeries.
Treatments of anterior vaginal wall prolapse are challenging. Anterior colporrhaphy used to be the most frequent approach for repairing anterior vaginal wall prolapse, but with the high risk of recurrence rate [5]. Transvaginal mesh (TVM) surgery showed a great result of high cure rate and low recurrence [6]. However, the most controversial issue of TVM is complication [7]. Therefore, how to avoid or at least get maximum reducing of postoperative complications is the burning question for urogynecology surgeons. Previous literature shows that postoperative mesh shrinkage is one common cause of pelvic pain, which is the most common complication after TVM [8]. VELEMIR reported recurrence after TVM repair is associated with severe mesh retraction [9]. Some believed mesh shrinkage is related to insufficient intraoperative spreading [10]. Improving the protocol of TVM and enhancing the efficiency of surgery may be the key points.
A modified TVM procedure was introduced to our center in 2019 for anterior vaginal wall prolapse repair. It showed good outcomes on avoiding postoperative mesh shrinkage and getting advanced apical support. In this study, we retrospectively recruited the patients undergoing classical TVM and modified TVM (M-TVM). Complications and outcomes were collected and compared to gain truly understanding of these two different surgeries for anterior prolapse.

Patient population
This was a retrospective, single-center based cohort study with 574 women who were underwent anterior pelvic floor repair surgeries between January 1, 2019 and October 30, 2020. The inclusion criteria were as follows: (1) primary diagnosis of III or IV grade anterior vaginal wall prolapse; (2) complete data acquisition; (3) no other systemic diseases. Patients who met the following were excluded: (1) had other procedures during surgery, such as tension-free vaginal tape (TVT) surgery, etc.; (2) company with stage III or IV apical or posterior vaginal wall prolapse; (3) had pelvic surgeries or hysterectomy before. 299 patients who were uterus preserved met the criteria and signed informed consent, after data checking, 285 participants were recruited finally. They were divided into 2 groups according to the different procedure taken. Those are traditional TVM group (n = 181), and modified TVM group (n = 104) (see Fig. 1).

Clinical procedures
All enrolled patients had no absolute contraindications to these two surgical methods. After being fully informed of the relevant advantages and disadvantages of these two surgical methods by surgeons, patients chose surgical procedures by themselves according to their own considerations.
The surgery was performed with the patient in the lithotomy position under general anesthesia. The entire thickness of the anterior vaginal wall was incised and dissected to medial margin of descending pubic ramus. Fingered palpation was made between the pubic arch and the ischial spine to ensure that the mesh straps could be passed through at this level. 4 arms polypropylene mesh (TiLOOP total 4, pfm, Nṻrnberg, Germany) was used for anterior vaginal repair. For TVM, the first 0.5 cm skin incision was made at the inner folds of thighs at the level of the urethral meatus. The second skin incision was made 1 cm laterally and 2 cm below the first incision. Then a puncture guide device was inserted in each of these skin incisions and passed though the obturator externus muscle, the obturator membrane and the obturator internus muscle, the first one passing to the pubic ramus before descending and emerging in the vagina, the other one following a downward at the beginning of arcus tendinous fasciae pelvis. The distal ends of the mesh implant were then captured and pulled through the device until they appeared through the skin incisions. After the Fig. 1 The process of capturing patients process was repeated on the other side the anterior mesh straps were positioned tension-free. (Fig. 2A).
M-TVM procedure the protocol was almost the same at the beginning until cutting the 4 skin incisions. The first skin incision was almost the same position as in TVM procedure. The second skin incision was made approximately 4 cm laterally and 4 cm down from the anus. A puncture guide device was pushed through the incision, through the buttocks, and below the plane of the lavator ani muscle, with fingers behind the dissected vaginal wall constantly controlling the progress. The guide was advanced until it was in contact with the middle part of the sacrospinous ligament. The process was repeated on the other side and the posterior mesh straps were positioned tension-free. (Fig. 2B and figure supplemental Fig. 1).
The POP-Q classification was used to evaluate level of prolapse according to the International Continence Society [11]. Examination was performed with women in dorsal lithotomy position. Following inspection of the vulva and perineum, the labia were separated and prolapse was noted.
Women were asked to perform Valsalva maneuver (women were asked to pinch the nose, close the mouth, and exhale like as inflating the balloon, bear down for 10-15 s) for measurement of Aa, Ba, C, Ap, Bp, D, gh, pb, tvl.
3/4 dimensional transperineal ultrasound was performed 6 months after operation with 3 senior sonographers. Width and length of meshes were measured in vesicovaginal gap follow Dietz methods [12] (supplemental Fig. 2).
All patients came to out-patient department for postoperative examination at 3, 6, and 12 months after surgeries. Pelvic examinations were taken including symptom checklist, a stress test, POP-Q test and transperineal ultrasound follow-up.
Follow-up examinations were performed by surgeons themselves. Complications directly related to the surgeries were acquired by analyzing questionnaires that women taken at each visit when we follow-up, which include pelvic floor function (PFDI20, PFIQ7) [13], sexual function (PISQ-12) [14,15], quality of life (EQ5D-3L main score [16] and visual analogic scale (VAS) [17], and improvement (PGII) [18]. Vaginal examinations were conducted to check for mesh erosion and assess vaginal support according to the POP-Q evaluation. To evaluate recurrence, we included a composite outcome defined by either anatomical recurrence (any point Ba, C, or Bp > 0 cm beyond the hymen), and/or retreatment (reintervention for prolapse recurrence).

Data collection
All patient data were complete and collected from hospital medical records, which include general condition, detailed medical history, surgical characteristics, and all information concerning postoperative complications.

Statistical analysis
Sample size was calculated according to the formula as follows: (n: sample size : admissible error σ:variation index) According to the literatures [5,6], the expected population rate was 0.0288, we estimated σ = 5, = 0.03, α = 0.05, then the sample size was at least 120. In consideration of 10% of lost to follow-up, the sample size would be 132.
Statistical analysis was performed after normality testing (Kolmogorov-Smirnov and Shapiro-Wilk testing). Continuous variables are expressed as means with SDs. For categorical variables, counts and percentages are presented. One-way ANOVA was used to compare normally distributed continuous, Kruskal-Wallis test was used in the event of abnormal distributions. Contingency tables with Chisquared test were used to assess the relationship between quantitative variables. Statistical analyses were performed using SPSS 20.0 software for windows.

Results
A total of 285 patients who signed inform consents were recruited eventually. All the surgeries were performed by 3 experienced surgeons, difference comparison of postoperative complications and recurrence rate among these surgeons were taken and showed no significant difference among them (see supplemental Table 1). Table 1 shows the general conditions of patients in these two groups. No significant differences were found. It revealed that the objectives of the present study were undifferentiated. About 98% patients in this study were postmenopausal, and almost 80% of them were suffered stage III anterior vaginal wall prolapse, 55% were companied with stage I or II apical prolapse. Table 2 shows surgical characteristics and complications during or after operation. Data during operation seemed no difference between these two groups except operative time. It seems M-TVM (43.6 ± 4.220 min) cost little much time than TVM group (42.1 ± 4.136 min). Complications related to surgery in the follow-up periods in Table 2 showed no significant difference, which included mesh exposure (P = 0.601), infection (P = 0.464), de novo urinary stress incontinence (P = 0.457), bladder outlet obstruction (P = 0.365) and Dyspareunia (P = 0.610). In TVM group, patients were more likely to suffer pelvic pain than M-TVM group after surgery (P = 0.046)0.14 patients in TVM group and 3 in M-TVM group were found recurrence of POP after surgery (P = 0.076). Significant difference was found in width of mesh 6 months after surgeries between these two groups (3.0 ± 0.91 versus 4.5 ± 0.69) (P < 0.001). Meshes implanted in patients in M-TVM group (4.5 ± 0.69 cm) seemed much wider than these in TVM group (3.0 ± 0.91 cm). No difference found in length of meshes (P = 0.721).
In this study, scores of pelvic floor function questionaries (PFDI20, PFIQ7) getting lower in most patients except the ones recrudesced after surgery during the follow-up period(n = 17). Scores of sexual functions (PISQ-12) seemed no significant change before and after surgery. Quality of life (EQ5D-3L main score and visual analogic scale (VAS)) and Improvement (PGI-I) are postoperative questionnaires, and it also seems no significant difference between these two groups of these two evaluated methods. Table 3 revealed the preoperative and postoperative anatomical characteristics according to POP-Q classification. Surgeries, no matter TVM or M-TVM, can significantly change point Aa and Ba on POP-Q (P < 0.05). No difference was found between these two groups with point Aa and Ba preoperatively or postoperatively. M-TVM procedures can significantly change point C and D after surgery compared to TVM surgery. Point C (− 6.1 ± 0.60 cm) and point D (− 8.1 ± 0.56) were much higher in M-TVM group than these two points in TVM group (− 4.5 ± 1.8 cm, − 6.5 ± 1.79 cm) 1 month after operation. So as they are in 3 months, 6 months and 1 year after surgeries. Surgeries were not changing Ap, Bp and TVL and there was no difference between these two group about these 3 data.
To reveal the differences of point C and D changing between these two groups, we compared patients companied with I or II grade apical prolapse in these two groups, which were mild apical prolapse in TVM group(n = 96) and mild apical prolapse in M-TVM group(n = 55). General conditions were no difference. POP-Q data revealed that both procedures could significant change point C and point D after surgery when patients companied with apical prolapse (P < 0.05), and numbers of point C and D were much higher in M-TVM group(P < 0.001) (see Table 4).

Discussion
In this study, we present a retrospective study on 285 patients undergoing TVM or M-TVM surgeries and demonstrated the modified procedure turned out good outcome in 1 year follow-up. In accordance with previous study [19,20], most patients suffered from III or IV prolapse were postmenopausal women aged above 60 years old. Pelvic floor prolapses always affect two or more compartment [21,22], even we excluded patients with severe apical prolapse, there still almost half of the recruited ones accompany with grade I or II apical prolapse. Requiring of surgeries for POP repairing is increasing recently [19]. Although TVM surgery showed good outcomes and low recurrence rate in long-term follow-up, the US FDA banned the use of transvaginal mesh for prolapse with long-term follow-up outcome in FDA research in April 2019 [23]. However, TVM surgery still has been approved in many countries in the world. Recently, many prospective and retrospective studies with a significant amount of follow-up confirm the TVM technique's efficacy and low morbidity [24][25][26][27]. In present research, TVM procedures, no matter TVM or M-TVM, are demonstrated being reliable with high cure rate (96.6%, 275/285) and few complications in long-term follow-up.
Nowadays, improving procedure to avoid complications makes surgery more reliable. Studies reported some complications such as recurrence, mesh exposure and pelvic pain, may associated with mesh shrinkage, which is defined as the mesh did not remain flat but folded up on itself, either during implantation or immediately after closure [28]. In H.P. Dietz's opinion, surgical technique seems to play a role as fixation of mesh to underlying tissue results in a flatter, more even appearance [19].Sunil et al. reported a cadaver model study stimulated apical support after sacrocolpopexy, and found out increasing mesh width is associated with better vaginal apical support, less mesh elongation, and higher failure loads [29]. In traditional TVM surgery, recurrence of upper part of anterior wall prolapse may related to the mesh shrinkage caused by the shift of deep branch band when the deep branch puncture is not in right place or fixed insecurely. To avoid mesh shrinkage and also considering the theory of apical support in anterior wall defect, we modified the anchoring points with same size and tailor of meshes, which turned out different widths and good apical support floor repair, POP-Q Pelvic Organ Prolapse quantification  after surgeries. TVM and M-TVM both significantly change the point Aa and Ba after surgery, and remain stable in the follow-up period. In addition, implants in M-TVM group are wider and flatter to address paravaginal defects than those in TVM. This improvement is also confirmed by following up results that decreasing in the proportion of vaginal and pelvic pain, which is consider to be one result of mesh shrinkage in literatures [8,10,19]. Predictably, M-TVM can also improve the point C and D. The latter regroup about patients with mild apical prolapse also proved it. These results can be explained by De Lancey theory that the TVM procedure offered only the level II attachment support, but M-TVM procedure offered additional force for level I suspension. In the present study, comparing with traditional TVM, M-TVM can reconstruct anterior compartment validly and significantly reinforce apical support by improving the cervix to normal position (upper level of ischial spine), which may result in more stable outcome of surgery and decrease the recurrence rate. Though there was no difference in the two groups in our study in 1 year follow-up, a bit longer time of following up may get different results.
There are some similarities between M-TVM and The Surelift system, which reported as a surgical mesh kit and comprised a monofilament polypropylene mesh with 6 adjustable fixation points to correct anterior and apical prolapse [30]. The existing literature on the outcomes of Surelift remains extremely limited. But it obviously gets more damages and increase risks and time of surgery because 6 arms need 6 cut incisions. So we modified the procedure to the M-TVM with 4 arms, which remain the latter two arms suspended to sacrospinous ligament with tension-free and the former two arms cross through obturator.
Limitations of the present study were as follows: first, our study were that of a single-center, retrospective design, and duration of 1 year follow-up, which may not be sufficiently long to draw substantial conclusions. Second, the population in this study was Asian women, which may not present the other populations. Third, there still remain high prevalence of de novo USI after anterior pelvic floor repair, this is in accordance with literatures [31] and may be explained by the theory that occult USI always masked by severe POP. Fourth, to service patients better, following up period were done by patients' own surgeons, this may cause some bias with non-double blind.

Conclusion
In a word, M-TVM can significant correct anterior prolapse with mesh extended wider, and also supply stable apical support at the same time.
Author contributions PH Data collection, Data analysis, Manuscript writing, LL Project development, Data collection, YW Project development, JT Data collection, LL Project development, Data analysis, Manuscript revising.
Funding This work was supported by Grant number 81671441 (Lubin LIU) and 82171622(Lubin LIU) from national natural science foundation of China.

Data availability
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.