Archives of Gynecology and Obstetrics

, Volume 287, Issue 5, pp 919–936

Meta-analysis of the efficacy and safety of the application of adjuvant material in the repair of anterior vaginal wall prolapsed

Authors

  • Hu Min
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Li Bingshu
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Cheng Yanxiang
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Chen Lu
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Sun Qing
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Zhu Xuejiao
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Wu Wenying
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Wu Debin
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Hong Shasha
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Ding Wenjuan
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Min Jie
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Zhang Xiaohong
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Guo Wenjun
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Chen Jianhua
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Liu Qian
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
  • Liu Yuling
    • Department of Gynecology and ObstetricsRenmin Hospital of Wuhan University
General Gynecology

DOI: 10.1007/s00404-012-2626-6

Cite this article as:
Min, H., Li, H., Bingshu, L. et al. Arch Gynecol Obstet (2013) 287: 919. doi:10.1007/s00404-012-2626-6

Abstract

Introduction

This study is a meta-analysis of the efficacy and safety of the application of adjuvant material in the repair of anterior vaginal wall prolapse and a sub-category analysis of the use of nonabsorbable synthetic mesh, biological graft and absorbable synthetic mesh.

Method

Pubmed, Embase and Ovid databases were searched for published randomized controlled trials from 1980 to February 2012 on the treatment of anterior vaginal wall prolapse with adjuvant materials. A comprehensive meta-analysis applying Revman5.1 analysis software was performed.

Results

A total of 20 randomized controlled trials including 2,313 participants were recognized. The result showed that repair with adjuvant materials was better and more effective; nevertheless, use of adjuvant materials resulted in longer duration of surgery and more peri-operative bleeding when compared with the control group, but no significant differences were observed between the two groups regarding visceral injury, postoperative pain, urinary tract infection rate, new stress incontinence and new dyspareunia.

Conclusion

Adjuvant material is worthy of clinical popularization, especially the biological graft type because of its lower anatomy failure rate and no difference in safety compared with the control group. However, exposure to adjuvant materials and erosion rate are high, which are the most important aspects to be improved.

Keywords

Anterior vaginal wall prolapseRepairAdjuvant materialsEfficacySafety

Introduction

Pelvic floor dysfunction leads to surgery in 11 % of women in their lifetime [1, 2]. Approximately 29.2 % of these women will require a second surgery, and the annual cost of surgery for prolapse alone is over $1 billion [3].

Following injury to pelvic floor, muscle and fascia tissue become weak or broken; if repair is undertaken on the basis of this original tissue, it is difficult to build a strong and stable pelvic floor; therefore, in an attempt to improve surgical outcomes and reduce prolapse recurrence, a variety of graft materials and meshes have been used. Novara [4] summarized a total of 815 articles about meshes published in 2004 and suggested that the application of adjuvant materials to anterior vaginal wall repair appeared to be a better choice than traditional repair.

Now, implanted adjuvant materials are being used with increasing frequency by pelvic reconstructive surgeons [5]. However, it still remains to be seen which of the available augmentation materials are best suited in prolapse repairs with respect to issues of long-term efficacy, complications and safety record.

An ideal adjuvant material is one that is biocompatible, chemically and physically inert, non-carcinogenic, mechanically strong, non-allergenic, nonmodifiable by body tissue, resistant to infection and inexpensive [6].

None of the currently available adjuvant materials fulfils this ‘ideal’. Adjuvant materials can be classified by material type, pore size and weave and also as biological or synthetic. Biological grafts are in the form of allograft (grafts from the host, e.g. fascia lata from the thigh), autografts (human tissues from another host) and xenografts (tissues from another species, e.g. porcine dermis). Synthetic adjuvant materials may be classified as absorbable and non-absorbable, and the pore size classified as macro-porous (>75 μm) and micro-porous (<10 μm), and the weave classified as multifilament and monofilament. The pore size and weave of mesh are the most important factors determining whether the body can mount an immune response against bacteria [7].

In vaginal surgery, the most common recurrence site is the anterior vaginal wall, and the reported failure rates ranged from 20 to 40 % [8].

We can assume that the factors that cause a prolapse in the first instance are also the cause of the recurrence [9]. Therefore, the assessment of the application of different augmentation materials in anterior vaginal wall repair could guide us in choosing the right materials in pelvic reconstructive surgery.

Through a meta-analysis on the collective literature, the current study attempts to address two principal questions regarding the application of augmentation materials in the repair surgery for anterior vaginal wall prolapse patients. First, what is the overall efficacy and safety outcome of patients treated with this approach? Second, which type of the augmentation materials is the most suitable for this patient population?

Methods

Study selection and data extraction

Using the relative keywords “anterior,” “vaginal prolapse,” “repair,” “mesh,” “adjuvant materials,” “complications,” “cystocele,” “efficacy,” and “safety,” searches in PubMed, Embase, and Ovid for English language articles published between January 1st, 1980 and February 30th, 2012 were conducted. The manual search was performed using the references cited in the acquired articles. We also searched previously published review articles on the topic to identify eligible studies.

Studies were selected if (1) they were randomized controlled trials; (2) participants were women diagnosed with anterior vaginal wall prolapse; (3) the interventions considered were anterior vaginal wall prolapse repair with mesh/graft with no restrictions on type of mesh/graft or technique used; (4) the comparator was another surgical technique using mesh/graft or a type of surgery that did not involve mesh/graft, and (5) report on at least one of the outcome measures mentioned below. And studies were excluded if (1) studies of women with prolapse caused by pelvic trauma, congenital disease, or prolapse after creation of a neovagina were excluded; (2) the outcomes were not clearly reported; (3) it is impossible to extract data from the study; (4) there were considerable overlap contents between studies; (5) non-randomized controlled studies, and (6) review articles, retrospective studies, and abstracts were not documented.

The full text articles were reviewed in a blinded fashion by two reviewers; studies were excluded if they did not pertain to anterior vaginal wall prolapse, or did not report treatment outcomes or original research (e.g., review articles). When the judgments of two reviewers were bifurcated, a further discussion would be presented to solve the problem. If no consensus was reached, a third reviewer intervened to make a final decision. Data were abstracted using a standardized form. The abstracts were reviewed by one reviewer who was blinded to the authors, in order to ensure that all the relevant studies would be included.

To assess the quality of the included studies, we considered the following domains according to the Cochrane’s Handbook: sequence generation, allocation concealment, blinding of study personnel and participants, incomplete outcome data, and selective outcome reporting. For each domain, we examined each paper’s description of methods and made a judgment regarding potential bias, according to three categories: low risk, high risk, or risk unclear. The result showed that the quality of included studies was high in quality.

In those studies that met the inclusion criteria, data were abstracted regarding subjective outcomes. The failure rate of the vaginal wall anatomy, operative time, intraoperative bleeding and postoperative visceral injury, pelvic pain, urinary infection, material exposure, material erosion, new urinary incontinence and new dyspareunia in the adjuvant materials repair group and in the without adjuvant materials repair group were collected and compared. In addition, the material types used in each literature were noted (Table 1).
Table 1

Information of included studies in meta-analysis comparing repair with or without adjuvant material in treating anterior vaginal wall prolapse

References

Experimental failure rate (n/N)

Control group failure rate (n/N)

Criteria of the study

Material

Anatomic recurrence definition

Follow-up period (m)

Surgery procedures

Type of the research

Altman [11]

69/176

114/174

18 years of age or older and presented with primary or recurrent prolapse of the anterior vaginal wall that was stage 2 or higher [(POP-Q) system] and with symptoms of vaginal bulging or pelvic heaviness

Polypropylene-mesh

POP-Q stage ≥II

12

E: prolapse repair with the transvaginal mesh kit;

C: traditional colporrhaphy

Multicenter, RCT

Del Roy [25]

3/39

13/39

Women ≥50 years old with greater anterior vaginal prolapse, with stress incontinence or not, requiring surgical correction

Polypropylene-mesh (NAZCA TC™ POP REPAIR SYSTEM)

9

E: polypropylene mesh repair;

C: colporrhaphy

RCT

Duggan [26]

9/19

7/16

Women presenting with anterior compartment prolapse ≥POP-Q stage 2

Prolift (Johnson & Johnson)

POPQ stage ≥II for the anterior compartment

6

E: anterior Mesh repair;

C: traditional anterior repair by midline plication

RCT

Nguyen [18]

5/38

17/38

Women ≥21 years old with anterior vaginal prolapse ≥POP-Q system stage II requiring surgical correction

Polypropylene mesh

Either point Aa or Ba was at stage II or beyond (−1 cm or lower)

12

E: polypropylene mesh repair by a transobturator approach;

C: anterior colporrhaphy

RCT

Nieminen [19]

14/105

40/97

Women with symptomatic anterior vaginal wall prolapse to the hymen or beyond

Low-weight polypropylene mesh (Parietene® light, Sofradim Co., Trévoux, France)

Anatomic recurrence of anterior vaginal wall prolapse

36

E: colporrhaphy reinforced with polypropylene mesh;

C: traditional anterior repair

RCT

Vollebregt [23]

6/56

34/58

Predominant cystocele ≥stage II (POP-Q system) and an indication for surgical correction

Avault anterior system (Bard, Covington, LA, USA).

≥Stage II cystocele (POP-Q system)

12

E: cystocele repair with trocar-guided transobturator mesh;

C: anterior colporrhaphy.

RCT

Sivaslioglu [22]

4/43

12/42

Patients diagnosed as having cystocele

Polypropylene mesh (Sofradim®, Parietene)

POP-Q stage ≥II

12

E: polypropylene mesh surgery; C: site-specific cystocele repair

RCT

Al-Nazer [30]

1/20

7/20

Stage 2 POPQ prolapse

Armless soft polypropylene (Gynemesh) mesh

12

E: mesh repair without anterior colporrhaphy; C: anterior colporrhaphy

RCT

Ali [29]

3/46

5/43

Women with grade 3 or 4 cystourethrocele (BW halfway system)

Tension-free polypropylene (Gynemesh PS)

Grade 2 or worse anterior wall prolapse (BW halfway system)

6

E: anterior colporrhaphy with mesh overlay;

C: anterior colporrhaphy alone

RCT

Rudnicki [28]

Women ≥55 years old with anterior vaginal prolapse ≥POP-Q system stage II

Polypropylene mesh (Avaulta Plus® Anterior Biosynthetic System)

3

E: anterior repair with trocar-guided mesh;

C: conventional anterior repair

Multicenter RCT

Hiltunen [14]

7/104

9/86

Postmenopausal women with symptomatic anterior vaginal wall prolapse to the hymen or beyond

Low-weight polypropylene mesh

Above stage II during 12 months follow-up

12

E: anterior colporrhaphy reinforced with a self-tailored low-weight polypropylene mesh;

C: traditional anterior colporrhaphy

RCT

 

Patients with ages ranging between 50 and 75 years and diagnosed with AVP stage III or IV, or recurrent anterior vaginal prolapse

Synthetic monofilament polypropylene mesh (Nazca TC, Promedon, Córdoba, Argentina)

9

E: anterior vaginal prolapse repair with mesh

C: site-specific surgical repair of anterior vaginal prolapse repair without the use of mesh

RCT

Dyer [27]

3/24

3/23

Anterior prolapsed (POP-Q point Ba ≥0)

Porcine dermis graft (Pelvicol™)

POP-Q stage ≥II

24

E: vaginal paravaginal repairs using xenograft;

C: anterior colporrhaphy

RCT

 

1/27

  

Polypropylene mesh

  

E: vaginal paravaginal repairs using synthetic mesh

RCT

Menefee [16]

11/26

14/24

Women ≥18 years old with ≥stage II anterior vaginal wall prolapsed who were symptomatic, desired surgical correction

Porcine dermis

Anterior POP-Q at stage II or greater (Ba measurement of 1 or greater)

24

E: paravaginal repair with porcine dermis; C: anterior colporrhaphy

RCT

 

5/28

  

Polypropylene mesh

  

E: paravaginal repair with polypropylene mesh; C: –

RCT

Feldner [20]

4/29

11/27

Pre- and post-menopausal women referred for vaginal surgery who had anterior vaginal wall prolapse of stage II or more with point Ba ≥+1

Small intestine submucosa

POP-Q stage II or higher prolapse at 12 months follow-up for point Ba

12

E: –; C: traditional anterior colporrhaphy

RCT

Gandhi [12]

16/76

23/78

Women ≥18 years old with anterior vaginal wall prolapse to the hymen or beyond

Solvent dehydrated cadaveric fascia lata graft

Recurrent stage II anterior vaginal wall prolapsed (POP-Q system

12

E: anterior colporrhaphy with fascia lata graft;

C: anterior colporrhaphy

RCT

Guerette [13]

5/35

8/37

Women ≥18 years old with ≥stage II cystocele (POP-Q System, point Ba more than −1 cm)

Bovine pericardium graft (Veritas, Synovis Life Technologies, St. Paul, MN, USA)

Ba was or lower than −1 cm (POP-Q system)

12

E: bovine pericardium graft reinforced colporrhaphy; C: traditional anterior colporrhaphy

RCT

 

4/17

10/27

   

24

  

Hviid [15]

2/30

4/31

Women with ≥stage II (POP-Q, point Ba >−1.0) defects in the anterior vaginal compartment

Porcine skin collagen implant (Pelvicol® graft)

Ba was or lower than −1 cm (POP-Q system)

12

E: anterior repair with graft;

C: conventional anterior repair

RCT

Meschia [17]

7/98

20/103

Women with ≥stage II anterior vaginal wall prolapse (POP-Q System, point Ba 1 or greater)

Porcine collagen implant (Pelvicol™)

Ba was or lower than +1 cm (POP-Q system)

12

E: anterior vaginal repair with Pelvicol implant reinforcement; C: anterior vaginal repair

RCT

Sand [21]

18/73

30/70

Women planning to undergo reconstructive pelvic surgery with or without an anti-incontinence operation with a cystocele protruding to or beyond the hymenal ring

Polyglactin 910 mesh (Vicryl mesh; Ethicon, Somerville, NJ, USA)

Cystoceles beyond the hymenal ring

12

E: anterior colporrhaphy with mesh reinforcement; C: anterior colporrhaphy without mesh reinforcement

RCT

Weber [24]

15/26

36/57

All patients who underwent operation for anterior vaginal prolapsed anti-incontinence procedure other than suburethral plication was planned as part of their surgical procedures

Polyglactin 910 mesh (Vicryl mesh; Ethicon, Somerville, NJ, USA)

POP-Q stage ≥II at points Aa and Ba

(−1 cm or lower)

23.3

E: standard anterior colporrhaphy plus polyglactin 910 mesh;

C: standard anterior colporrhaphy

RCT

Primary outcomes for efficacy was vaginal wall rate, and all anatomy failure definition were all combined, for example, recurrent stage II anterior vaginal wall prolapse by the POP-Q system (Gandhi S), or POP-Q Ba ≥−1.0 (Hviid U), etc. Safety outcomes included operative time, intraoperative bleeding and postoperative visceral injury, pelvic pain, urinary infection, material exposure, material erosion, new urinary incontinence and new dyspareunia. For new urinary incontinence and dyspareunia, only women who were free of these symptoms at baseline were considered for these outcomes. All the comparisons of the above items will be made according to classification of the materials, which are absorbable synthetic mesh, nonabsorbable synthetic mesh and biological graft. In the case of a study with multiple treatment arms, each arm was classified into one of the corresponding groups outlined above. Wherever there were two or more publications by the same author on same topic, we assessed the quality of data from both and used the most up-to-date or longer follow-up for meta-analysis.

The differences between the total repair with and without adjuvant material group were expressed as the risk ratio (RR) or mean difference with its 95 % confidence interval (CI); P < 0.05 was considered statistically significant. Statistical heterogeneity among RCTs was assessed by a χ2 test and expressed with the P value and I2 statistics [10]. I2 is the proportion of total variation contributed by between-study variability. In the presence of statistical heterogeneity, a random-effect model was used. In the absence of statistical heterogeneity, the fixed-effect model was used.

Finally, regarding possible publication bias, we performed a funnel plot and found no deviation from a symmetrical shape. Publication bias seems therefore unlikely. The statistical software used for this analysis was RevMan 5.1.

Results

As shown in Fig. 1, a total of 20 randomized controlled trials including 2,313 participants were recognized, 14 of which were full text [1124], 4 were from International Urogynecological Association (IUGA) conference reports [2528], and the other 2 were from review articles [29, 30]. In these studies, 11 articles were about nonabsorbable synthetic mesh researches, 2 articles were about nonabsorbable synthetic mesh and absorbable biological graft, 5 articles were about absorbable biological graft, and 2 articles were about absorbable synthetic mesh. The average follow-up time was: 3 months for 1 study [30], 6 months for 2 studies [26, 29], 9 months for 1 study [25], 12 months for 11 studies [11, 12, 14, 15, 17, 18, 2023, 30], 23.3 months for 1 study, 24 months for 2 studies, and 36 months for 1 study; the other one reported the follow-up result of both the 12- and 24-month studies.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig1_HTML.gif
Fig. 1

Search strategy for randomized controlled trials compared repair with or without adjuvant materials for treating anterior vaginal wall prolapse

The characteristics of included studies and the results of study quality description and assessment are shown in Table 2 and Fig. 2.
Table 2

Baseline of included studies in meta-analysis comparing repair with or without adjuvant material in treating anterior vaginal wall prolapse

References

Study Size

Age (mean ± SD)

Parity (number of deliveries; median, range)

Previous hysterectomy (n/N)

BMI (mean ± SD)

With estrogen treatment or Postmenopausal

Current smokers

E

C

E

C

E

C

E

C

E

C

E

C

Ali [29]

89

Al-Nazer [30]

40

Altman [11]

350

64.3 ± 9.8

65.1 ± 9.8

2 (0–7)

2 (0–6)

23.0 %

19.0 %

26.2 ± 3.4

25.0 ± 3.0

12.5 %

11.6 %

Del Roy [25]

78

Duggan [26]

35

Dyer [27]

73

Gandhi [12]

154

65.5 (11.6)

64.9 (11.7)

3 (1–7)

3 (1–10)

54 %

50 %

Guerette [13]

72

60.9 (34–80)

61.4 (36–80)

2.7 (1–7)

2.8 (0–5)

4/47

5/47

Hiltunen [14]

177

66 ± 9

65 ± 9

26.5 ± 3.5

27.2 ± 4.1

22 %

28 %

26.5 ± 3.5

27.2 ± 4.1

86

84

Hviid [15]

61

66 ± 9

65 ± 9

3 (0–11)

2 (1–10)

22 %

28 %

26.5 ± 3.5

27.2 ± 4.1

86

84

Menefee [16]

78

60 ± 10

61 ± 11

3 (1–8)

3 (1–8)

35 %

54 %

30 ± 5

31 ± 10

  

65 ± 7.0

 

3 (1–7)

 

50 %

 

28 ± 4

     

Meschia [17]

201

65 ± 8

65 ± 9

2 (0–6)

2 (0–5)

25.8 ± 4

25.1 ± 3

Nguyen [18]

76

61 ± 10.5

59 ± 9.5

3 (0–5)

3 (0–6)

43 %

31 %

28 ± 3

27 ± 4

27

18

Nieminen [19]

202

65 ± 9

66 ± 9

3 (0–11)

2 (1–10)

27 ± 4

27 ± 4

13

14

Feldner [20]

56

53.8 ± 9.7

56.3 ± 13.0

4.3 ± 1.8

4.0 ± 2.1

3/29

1/27

27.3 ± 4.9

27.5 ± 4.5

Rudnicki [28]

75

Sand [21]

143

65.0

62.5

2.8

2.8

58

46

3/70

2/73

Sivaslioglu [22]

85

57.7 ± 9.4

50.1 ± 9.9

3.1 ± 1.4

3.7 ± 1.9

29.4 ± 4.1

30.3 ± 5.6

Vollebregt [23]

114

60 ± 9.1

59 ± 8.6

2.4 ± 0.9

2.7 ± 1.9

24 ± 2.9

24 ± 3.6

Weber [24]

83

66.0 ± 11.2

65.6 ± 11.2

41 %

50 %

56 %

40 %

E mesh or graft reinforced repair, C repair without mesh or graft; –, not stated

https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig2_HTML.gif
Fig. 2

Judgments about each risk of bias item presented as percentages across all included studies

Assessment of study quality

The 20 studies included are all English language randomized controlled studies. In summary, risks of bias in random sequence generation (30 %, 6/20), incomplete outcome data addressed (35 %, 7/20) and selective reporting (35 %, 7/20) were low, whereas allocation concealment (55 %, 11/20), blinding of participants and personnel (55 %, 11/20) and blinding of outcome assessment (80 %, 16/20) were low. Risk of bias in free of other bias (80 %, 16/20) were unclear. Baseline information of these studies is shown in Table 2. Overall, these 20 studies were considered to be of high methodological quality (Fig. 3)
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig3_HTML.gif
Fig. 3

Funnel plot for report bias for all included studies

Meta-analysis results of operative outcomes

Anatomy failure rate

Compared with repair without adjuvant materials, the application of the adjuvant materials to anterior vaginal wall repair totally reduced vaginal front wall prolapse anatomy failure rate, and the difference between the two groups was obvious (P < 0.00001) (RR 0.51, 95 % CI [0.41, 0.64]); according to the material classification calculation, biological graft, nonabsorbable synthetic mesh all appeared to have a lower anatomy failure rate comparing with the simple repair; the difference between the two groups was obvious, and the effects were (P = 0.02, RR 0.41, 95 % CI [0.29, 057]), (P = 0.001, RR 0.60, 95 % CI [0.45, 0.82]), respectively, and the comparison of repair with or without absorbable synthetic mesh showed no obvious difference (P > 0.05). The forest plot is shown in Fig. 4.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig4_HTML.gif
Fig. 4

Forest plot of the comparison of anatomy failure rate between repair with or without adjuvant material in treating anterior vaginal wall prolapse

Blood loss

Comprehensive analysis of intraoperative bleeding demonstrated that application of adjuvant material in repair of the anterior vaginal wall prolapse augmented the blood loss compared with the simple repair, and the difference between the two groups was obvious (P = 0.01) (MD 35.00, 95 % [CI 6.90, 63.11])., Comprehensive analysis of nonabsorbable synthetic mesh met similar results (P = 0.002, MD 41.64, 95 % [CI 15.21, 70.06]): absorbable biological graft showed no significant difference when compared with the control group, P value was (P = 0.66).Meta analysis was not done in absorbable synthetic because of a lack of literature data. Forest plot is shown in Fig. 5.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig5_HTML.gif
Fig. 5

Forest plot of the comparison of intraoperative blood loose between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

The operation time

Adjuvant material repair totally took longer operation time than simple repair, the difference between the two groups was obvious (P < 0.00001) (MD 16.33, 95 % CI [8.12, 24.54]), nonabsorbable synthetic mesh and biological graft appeared similar result, the effect value were (P = 0.002, MD 19.10, 95 % CI [7.28, 30.93]) and (P = 0.001, MD 14.90, 95 % CI [6.03, 23.76]). Meta-analysis was not done in absorbable synthetic because of a lack of literature data. The forest plot is shown in Fig. 6.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig6_HTML.gif
Fig. 6

Forest plot of the comparison of operation time between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

Visceral injury incidence

Overall material repair group visceral injury incidence was 7/463 (1.51 %) compared with 1/457 (0.22 %) of the control group with no obvious difference (P > 0.05, RR 7.12, 95 % CI [0.89, 57.27]), nonabsorbable synthetic mesh appeared similar result (P = 0.07), absorbable biological graft and absorbable the synthetic of mesh did not do meta analysis because too little literature reported positive case. Forest plot is shown in Fig. 7.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig7_HTML.gif
Fig. 7

Forest plot of the comparison of visceral injury incidence between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

Material exposure rate

Because material exposure and erosion were impossible in the control group that not using the adjuvant material, so these two items were conducted with expressive analysis. The rate of nonabsorbable synthetic mesh was 3.23–17.31, 5.32 % on average, absorbable biological graft exposure rate was 0.00–1.00, 0.78 % on average, no research of absorbable synthetic mesh reported this data (Table 3).
Table 3

Descriptive analysis of the material exposure rate in the application of materials in surgery of treating anterior vaginal wall prolapsed

Material types

References

Exposure number/total number

Incidence (%)

Summary

Nonabsorbable synthetic mesh

Altman [11]

6/186

3.23

26/489 (5.32 %)

Hiltunen [14]

18/104

17.31

Nguyen [18]

2/37

5.41

Nieminen [19]

8/104

7.69

Vollebregt [23]

2/58

3.45

Meschia [17]

1/100

1.00

1/129 (0.78 %)

Feldner [20]

0/29

0.00

Biologic graft

Summary

 

27/618

4.37 %

Material erosion rate

Integrated erosion incidence was 7.69 %, that of nonabsorbable synthetic mesh was 6.52–14.81 %, 10.61 % on average; absorbable biologic graft was 0.00–4.17 %, 2.53 % on average, and that of absorbable synthetic mesh was 2.86 %. The forest plot is shown in Fig. 8 (Table 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig8_HTML.gif
Fig. 8

Forest plot of the comparison of postoperative pain rate between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

Table 4

Descriptive analysis of the material erosion rate in the application of materials in surgery of treating anterior vaginal wall prolapsed

Material types

References

Exposure number/total number

Incidence (%)

Summary

Nonabsorbable synthetic mesh

Ali [29]

3/46

6.52

21/198 (10.61 %)

Duggan [26]

2/19

10.53

Dyer [27]

4/27

14.81

Menefee [16]

4/28

14.29

Sivaslioglu [22]

3/43

6.98

Rudnicki [28]

5/35

14.29

Dyer [27]

4/27

14.81

Biologic graft

Feldner [20]

0/29

0.00

2/79 (2.53 %)

Menefee [16]

1/26

3.85

Dyer [27]

1/24

4.17

Absorbable synthetic mesh

Weber [24]

1/35

2.86

1/35 (2.86 %)

Summary

 

24/312

7.69 

 

Postoperative pain rate

Adjuvant material repair totally appeared no obvious difference in postoperative pain when compared with simple repair group (P = 0.61), absorbable biological graft appeared similar result (P = 0.07), and that of nonabsorbable synthetic mesh repair was significantly higher than simple repair, the difference between the two groups was obvious (P = 0.02) (RR 2.19, 95 % CI [1.11, 4.32]), absorbable synthetic mesh was not reported in literature. The forest plot is shown in Fig. 8.

Postoperative urinary tract infection rate

Incidence of postoperative urinary tract infections of the overall adjuvant material repair was not higher than that of control group, the difference between the two groups was not significant (P = 0.06), nonabsorbable synthetic mesh and absorbable biological graft showed similar result, P value was 0.06 and 0.68, respectively; absorbable synthetic mesh did not carry out meta analysis because of lack of reported data. The forest plot is shown in Fig. 9.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig9_HTML.gif
Fig. 9

Forest plot of the comparison of postoperative urinary tract infection rate between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

De novo stress urinary incontinence

Three articles from literature had reported data of this complication, all were from trials on nonabsorbable synthetic mesh, a total of 641 patients were enrolled, meta-analysis revealed that the comparison of de novo stress urinary incontinence rate in adjuvant materials repair simple repair group had no obvious difference (P = 0.54). The forest plot is shown in Fig. 10.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig10_HTML.gif
Fig. 10

Forest plot of the comparison of de novo stress urinary incontinence rate between repair with or without adjuvant material in treating anterior vaginal wall prolapsed

De novo dyspareunia

The comparison of de novo dyspareunia rate in adjuvant materials repair simple repair group had no obvious difference, the same result was calculated in nonabsorbable synthetic mesh subclassification and biological graft subclassification, P value was 0.89 and 0.41, respectively, absorbable synthetic mesh did not carry out meta analysis because of lack of reported data. The forest plot is shown in Fig. 11.
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-012-2626-6/MediaObjects/404_2012_2626_Fig11_HTML.gif
Fig. 11

Forest plot of the comparison of de novo dyspareunia rate between repair with or without adjuvant material in treating anterior vaginal wall prolapse

Discussion

We searched published, randomized controlled trials relevant to the treatment of anterior vaginal wall prolapse with adjuvant materials, and compared the efficacy and safety of anterior vaginal wall prolapse repair with and without adjuvant materials, and the comparison was made according to classification of the materials, which are absorbable synthetic mesh, nonabsorbable synthetic mesh and biological graft.

Efficacy is mainly about anatomy failure rate, but the definitions of anatomy failure were different in the included studies, Sivaslioglu [22] defined it as the leading edge of cystocele was <−1 cm in relation to hymen, Ali [29] defined it according to Baden-Walker system, while Nguyen [18] defined it as either point Aa or Ba was at stage II or beyond (−1 cm or lower) according to POP-Q system, and most studies defined it as stage II or greater anterior or vaginal prolapse after the surgery. In 2001, the National Institutes of Health Terminology Workshop for Researchers in Female Pelvic Floor Disorders recommended that vaginal support was considered “optimal” when both points Aa and Ba were at stage 0 (−3 cm), and “satisfactory” when both points Aa and Ba were at stage I (−2 cm) and improved from preoperative staging according.

The 2001 National Institutes of Health (NIH) Workshop on Standardization of Terminology for Researchers in Pelvic Floor Disorders recommended arbitrary definitions for treatment success [31, 32]. The definition of “optimal anatomic outcome,” designated as “cure,” requires perfect anatomic support [pelvic organ prolapse quantification system (POP-Q) stage 0] and “satisfactory anatomic outcome” requires support higher than 1 cm proximal to the hymen.

However, data suggest that 75 % of women presenting for annual gynecologic exams without symptoms of POP would not meet the definition of “optimal anatomic outcome” and almost 40 % would not meet the definition of “satisfactory anatomic outcome” [33].

Study from Barber [34] suggested that the absence of vaginal bulge symptoms postoperatively has a significant relationship with a patient’s assessment of overall improvement, while anatomic success alone does not.

In our analysis, 17 of the 20 included studies clearly defined the anatomic failure or success and displayed detailed data, but most studies did not specify the condition of the vaginal bulge symptomatic relief, or described in the form of different kinds of questionnaires which were difficult to pool the data, so more studies including investigation of vaginal bulge symptomatic relief after the repair of anterior vaginal wall prolapse with or without adjuvant materials are expected to be carried out in clinic.

According to our meta-analysis, the total numbers of anatomic failure in material repair group and the non-material repair group were 18.60 % (208/1,118) and 37.84 % (420/1,110), the difference was obvious, demonstrating that repair with adjuvant materials could improve the success rate of the anterior vaginal wall prolapse surgery, supporting the application of adjuvant materials in clinic.

If we made the sub-category analysis according to absorbable synthetic mesh, biological graft and nonabsorbable synthetic mesh, there appeared the same result. In absorbable synthetic mesh studies, the average failure rates in the experimental and the control groups were 33.33 % (33/99) and 51.97 % (66/127), respectively, the difference was significant (P < 0.01). The average failure rates in biological graft studies and nonabsorbable synthetic mesh were 15.09 % (48/318) versus 25.30 % (83/323) (RR 0.60, 95 % CI [0.45, 0.82] and 18.12 % (127/701) versus 41.06 % (271/660) (RR 0.41, 95 % CI [0.29, 0.57]), respectively. Thus, biological graft appeared the lowest anatomy failure rate, then is the nonabsorbable synthetic mesh, and absorbable synthetic mesh has the highest anatomy failure rate.

We try to repair the position of the pelvic organs to their original anatomic positions. Ideally, we have four main goals: no anatomic prolapse, no functional symptoms, patient satisfaction, and the avoidance of complications [35].

Complications are very important factors that we should consider when choosing the adjuvant materials, while there was no detailed data on this. In our meta-analysis, we pooled the data in a total of nine safety items, they are operative time, intraoperative bleeding, postoperative visceral injury, pelvic pain, urinary infection, material exposure, material erosion, de novo urinary incontinence and de novo dyspareunia

Generally speaking, repair surgery adding adjuvant materials prolonged the surgery time (MD 16.33, 95 % CI [8.12, 24.54], P < 0.01), and thus increased the bleeding in the surgery (MD 35.00, 95 % CI [6.90, 63.11], P = 0.01) totally, so did the sub-category analysis, the effect value of surgery time for biological graft and nonabsorbable synthetic mesh studies were longer than the control group, the effect value was (MD 14.63, 95 % CI [5.76, 23.50], P = 0.001) and (P < 0.002, MD 19.10, 95 % CI [7.28, 30.93]), respectively. In general, the more delicate the surgery, the more operation time it takes, so more surgery time does not suggest that the surgical procedure is not good, it is actually a sign of professionalism. On the other hand, we can see that only the repair adding nonabsorbable synthetic mesh increase bleeding in the surgery (MD 42.64, 95 % CI [15.21, 70.06]), but the biological graft did not (P > 0.05).

The types of postoperative visceral injury only included bladder injury and ureter damage, rectum damage or urethral injury was not reported. In addition, the incidences in most of the studies were 0, including three studies assessing biological graft [14, 15, 18], one study assessing nonabsorbable synthetic mesh [12] and one article investigating both biological graft and nonabsorbable synthetic mesh [16]. All feasible data for meta analysis was from nonabsorbable synthetic mesh literature, and the result showed that, visceral injury incidence in nonabsorbable synthetic mesh repair group and control group appeared no obvious difference (P = 0.07).

Exposure is that vaginal mesh visualized through separated vaginal epithelium [36], while erosion ‘state of being worn away’, as by friction or pressure [37].

The presenting symptoms of these two complications depended on the material of the organ involved, which can be vagina, urethra, bladder or bowel. There were particular complications for the application of adjuvant materials.

In this meta-analysis, the pooled material exposure of the literatures including information of this item was 4.37 % totally, and the nonabsorbable synthetic mesh exposure rate was 3.23 % [11] to 17.31 % [14], and 5.32 % on average. Biological graft for exposure rate ranged from 0.00 to 1.00 %, and was 0.78 % on average, lower than previous studies. None of the literatures in absorbable synthetic mesh contains such description. From this meta-analysis, biological graft showed lower materials exposure rate than nonabsorbable synthetic materials.

There were many factors that would affect the exposure and erosion rate, concomitant hysterectomy during placement of synthetic mesh in pelvic reconstructive surgery could increase the possibility of the happen of these two complications, etc.

Data in the materials erosion showed that, nonabsorbable synthetic mesh erosion rate was 6.52–14.81 %, and was 10.61 %on average. Moreover, that of the biological graft was 0.00–4.17 %, and 2.53 % on average. Only one literature on absorbable synthetic mesh described erosion data, reported incidence was 2.86 %.

This suggests that nonabsorbable synthetic mesh erosion rate is obviously higher than that of absorbable synthetic mesh and biological graft. Because the erosion could chronically develop to form fistula in organs, it could significantly decrease the patients’ quality of life and prevent the clinical application of these adjuvant materials. Therefore, the improvement of this complication is very important for the long-term popularization of these materials.

In other studies, the risk of vaginal exposure and erosion is considered to be at 10.7–12.27 % [38, 39], higher than the value calculated in this analysis. Data above is all from observations in 1 year follow-up. There is variation in the timing of mesh erosion, it has been reported as early as 6 weeks and as late as 7 years after surgery [4042]. The difference may result from the short follow-up period, what is more, variable methodology, techniques of placement of the materials, type of materials may also take part of responsibility.

Pelvic pain is another common persistent complication after the prolapsed repair surgery. Meta-analysis suggested that the comparison of the pelvic pain incidence between repair with and without adjuvant materials showed no obvious difference (P > 0.05). Biological graft sub-category analysis appeared the same result (P > 0.05), while repair using nonabsorbable synthetic mesh significantly increased the rate value (P = 0.02), this result may be for the reason that absorbable synthetic mesh contracts in the long-term application.

The common types of postoperative infection are surgical site infection and urinary tract infection, but literatures enrolled in this analysis seldom make the description of surgical site or material site infection data. Interestingly, according to the meta-analysis, totally, the difference of urinary tract infection in repair using adjuvant materials group and repair without adjuvant materials group was not significant (RR 0.56, 95 % CI [0.31, 1.02], P = 0.06). In addition, sub-category analysis of nonabsorbable synthetic mesh and biological graft also did not show obvious difference between the two groups, P value is 0.06 and 0.69, respectively, suggesting that the application of each type of the adjuvant material did not increase the incidence of urinary tract infection. Vaginal meshes with mono-filament and large pore size (>75 μm) are known to be associated with reduced risk of infection, but in meta-analysis, the mesh group and the graft analysis both showed the characteristic of not adding the infection rate after application, support their usage in clinic.

Reports on stress urinary incontinence are different in the literatures, some recorded postoperative stress urinary incontinence proportion, some recorded new stress urinary incontinence happen ratio. This meta-analysis is mainly about the latter, and the literatures containing the urinary tract infection information are all studies on nonabsorbable synthetic materials. The result showed that the comparison of new stress urinary incontinence between nonabsorbable synthetic mesh repair group and the control group showed no obvious difference (P = 0.54), suggesting that repair adding nonabsorbable synthetic material did not increase the incidence of new stress urinary incontinence.

How are the patients satisfied with postoperative sexual life is a very important aspect of evaluating the prolapse repair surgery. This meta-analysis revealed that total analysis and sub-category analysis both implied that repair adding nonabsorbable synthetic material did not increase occurred the incidence of new stress urinary incontinence. The average rate of new dyspareunia rate were 7.63 % (11/175) for overall analysis, 7.89 % (9/114) for nonabsorbable synthetic mesh, 8.22 % (6/73) for biological graft, none absorbable synthetic mesh literatures described the results of this item.

Comment

Comprehensive analysis showed that, the application of the adjuvant materials in the repair surgery in treating anterior vaginal prolapsed had lower anatomy failure rate compared with the surgery without adjuvant materials, that is, repair with adjuvant materials could get a better effectiveness. On the other hand, adjuvant materials performed an effect of longer operating time, more peri-operative bleeding when compared with the control group, but the two groups around in visceral injury, postoperative pain, urinary tract infection rate, new stress incontinence and new dyspareunia had no significant differences, meanwhile, adjuvant materials exposure and erosion rate is too high, which are the most important aspects to be improved. Overall, the adjuvant materials are safe and very reliable.

According to the sub-category analysis, studies on absorbable synthetic mesh is very few, a number of complications failed to do Meta-analysis on this type of adjuvant material, but in clinic, nonabsorbable synthetic mesh and biological graft are actually used more often. If we compared these two types of materials in effectiveness and safety, we can see that, biological graft showed lower postoperative anatomy failure rate, namely the better effectiveness. At the same time, when compared with the control group, biological graft increased less operation time and the amount of blood loss, lower material exposure and erosion rate, it also did not increase postoperative pain and new dyspareunia and new stress urinary incontinence. Generally, the efficacy and safety of biologic graft are superior to nonabsorbable synthetic mesh, but there remains a small risk of prion or viral infection estimated to be approximately 1 in 2 million [43]. The other important issue with some biological grafts is their ability to provide long-term support to the grafted repair. Absorbable biological graft was developed in the condition of many complications occured after the application of nonabsorbable synthetic mesh. But nowadays, xenografts have been found to degrade after implantation with loss of graft integrity and strength, both in animal [44] and human studies and showed high short-term failure [45]. Nowadays, allogenic graft materials using fascia lata [45] and rectus sheath [46] have been used for abdominal vault suspension.

Clinic procedure is more complicated, the elevated costs suggest that biological mesh implantation should be evaluated in correlation to the needs of each clinical case; what is more, it may be of value in patients where synthetic mesh is inappropriate, such as presence of infection. Above all, adjuvant material is worthy of clinical popularization, especially the biological graft and nonabsorbable synthetic type.

Of course, there are some problems in this analysis, such as the incidence of some rare complications like material erosion, new stress incontinence, are different with the other cohort study or case report data, so we expect to carry out of more rigorous large randomized controlled study in clinical. On the other hand, due to the limited follow-up time of the studies included in this analysis (the shortest 3 months, the longest was 36 months), some long-term complications failed to be observed, therefore, good quality and adequately powered trials with much longer-term follow-up are also expected to be initiated in clinic. What is more, the treatment cost received more and more attention of the clinical doctors, when we choose a certain materials or tools to treat disease, the price/performance ratio should also be taken into consideration. However, the research on this aspect is quite rare; we look forward to evidence-based medicine’s research on this aspect, this can help the clinical workers to make the most cost-efficient treatment programme for the patients according to the patients’ economy state.

Conflict of interest

There is not any potential conflict of interest.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012