ORIGINAL ARTICLE

POSTERIOR VAGINAL COMPARTMENT REPAIRS:

WHERE ARE THE MAIN ANATOMICAL DEFECTS?

Bernard T. Haylen, Sushen Naidoo, Stephen J Kerr,

Chin H Yong, Warwick Birrell

St Vincent’s Clinic, Darlinghurst. N.S.W. Australia

Bernard T. Haylen MD FRANZCOG FRCOG CU Urogynaecologist

St Vincent’s Hospital, Darlinghurst. N.S.W. Australia

Sushen Naidoo, MB BS FRANZCOG Fellow in Urogynaecology

Chin H Yong, MB ChB FRANZCOG Fellow in Urogynaecology

Kirby Institute, University of New South Wales, Kensington.N.S.W. Australia

Stephen Kerr, MIPH, PhD

Mater Hospital, North Sydney. N.S.W. Australia

Warwick Birrell, MB BS, FRANZCOG, CREI Gynaecologist

Correspondence to:

1

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

Associate Professor B.T. Haylen,

Suite 904, St Vincent’s Clinic,

438 Victoria Street,

Darlinghurst. 2010.

N.S.W. AUSTRALIA.

Ph: ## 61- 2- 83826983

Fax: ## 61 - 2 - 83826984

bernard@haylen.co

Conflicts of interest: None of the authors have any conflict of interests

AUTHORS

BT HAYLEN: Project development, Data Collection, Manuscript writing, final review

S NAIDOO: Analysis, research, manuscript writing, final review

S KERR: Statistical analysis, manuscript writing, final review

CH YONG: Analysis, research, manuscript writing, final review.

W BIRRELL Research, manuscript writing, final review

Words: Abstract 224 Text 1993

2

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

Anatomical defects at posterior vaginal compartment repairs.

Abstract:

Introduction:

Traditionally, it has been believed that posterior vaginal compartment prolapse was largely

due to defects in the rectovaginal fascia (Level II1), with surgical repairs concentrating on

addressing this defect. We aim to determine the relative size of the defects at the different

vaginal levels (I – III1) at a large number of posterior vaginal compartment repairs (PRs) to

determine if this traditional viewpoint is still appropriate.

Materials and Methods:

In a cross-sectional study of 300 consecutive PRs, mostly following prior or concomitant

hysterectomy, two sets of markers of posterior compartment prolapse were used to measure

anatomical defects at Levels I to III1: (i) from POP-Q2: points C, Ap, Bp and genital hiatus

(GH); from PR-Q3: perineal gap (PG), posterior vaginal vault descent (PVVD), mid vaginal

laxity (MVL) vault undisplaced, rectovaginal fascial laxity (RVFL).

Results: The largest defects were found at level I1 (PVVD - mean 6.0cm; Point C – mean -

minus 0.9cm) and level III1 (PG – mean 2.9cm; GH mean 3.7cm). Level II1 defects (MVL

undisplaced – mean 1.3cm; RVFL – mean 1.1cm; Points Ap, Bp both mean 1.0cm) were

relatively small.

Conclusions: This study suggests that the defects found at surgery for posterior vaginal

compartment prolapse were more at the vaginal vault (Level I1) and at the vaginal introitus

(Level III1) than at the mid-vagina (Level II1). These findings should have implications for

surgical planning.

Keywords:

3

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

Prolapse surgery, posterior vaginal compartment, pelvic organ prolapse, key anatomical

indicators (KAI), anatomical defects, posterior repair quantification (PR-Q)

Summary:

Anatomical defects at posterior compartment repairs appear to be more at the vaginal vault

and introitus rather than the mid-vagina.

4

69

70

71

72

73

74

75

Introduction:

Traditionally, it has been believed that posterior vaginal compartment prolapse was largely

due to defects in the rectovaginal fascia (Level II1). Surgical repairs have tended to

concentrate on addressing these defect4,5,6 using either native tissues or augmentation with the

use of a prosthesis. Attention to the vaginal introitus (Level III1) was generally included

though there were uncertainties4,5 as to whether the repair at this Level should commence at

the hymen or below. There has been some evidence for the contribution of vaginal vault

defects to anterior Level II compartment defects7.

The main anatomical defects at posterior vaginal compartment repairs (PR) have not, to date,

been clearly identified in the literature, leaving uncertainty regarding optimum surgical

approaches. Our recent preliminary report3 of 50 cases examining additional intraoperative

measurements at PR (PR-Q3), suggested that the defects at PR seemed to be more at Levels I1

and III1 than at Level II1.

We aim to employ a larger study to determine with more certainty the relative size of the

anatomical defects at the different vaginal levels (I – III1) at posterior vaginal compartment

repairs (PR). We propose to use two sets of available measurements: (i) POP-Q2; (ii) PR-Q,

the later recently described3. For the benefit of new readers, the core four PR-Q3

measurements will again be outlined in the methodology.

Patients and Methods

This was a cross-sectional study of 300 consecutive PRs, mostly following prior or

concomitant hysterectomy and concomitant anterior colporrhaphy. This was an extension of

the study of 50 PR cases, used in the previous preliminary report3 though focussing on

5

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

anatomical defects. Data collection was from December 2013 to December 2014 at St

Vincent’s and Mater Hospitals, Sydney. Patients were assessed clinically in the rooms using

POP-Q2 to determine the need for prolapse surgery. All POP-Q2 and PR-Q3 measurements

used in the study were taken immediately prior to the PR and at the end of any concomitant

prolapse surgeries. There were no patients excluded from a PR. The postoperative

measurements were not included as the study objective was to identify the relative size of the

preoperative anatomical defects at different vaginal Levels I-III1.

The following posterior prolapse markers were measured: (i) from POP-Q2: points C (Point D

not included as vast majority were post-hysterectomy), Ap and Bp, genital hiatus (GH); and

from PR-Q3: perineal gap (PG), posterior vaginal vault descent (PVVD), mid vaginal laxity

(MVL) vault undisplaced, rectovaginal fascial laxity (RVFL). The mid-vaginal laxity (MVL)

vault displaced3 (by traction) was also measured. Each measurement by the primary surgeon

received visual confirmation by an observer/surgical assistant. Bladder was re-emptied prior

to this examination to prevent any limitation of the extent of prolapse aspects by the presence

of a full bladder8.

The four PR-Q3 markers, less familiar to readers, are again described:

1: Perineal Gap (PG)- Level III

The line of the labia minora was visually followed posteriorly till the perineum was reached

on either side. At this point, Moynihan (Littlewood) forceps were applied on each side of the

labia. Gentle bilateral traction will highlight any deficient anterior perineum. Closer

inspection allows marking (by surgical marker) of the junction of the much thinned-out area

medially and the start of thicker perineal tissue laterally, closer to the forceps. This section

between the surgical marks is the perineal gap (PG). It was measured with a surgical ruler in

6

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

centimetres (cm) to one decimal point. Figure 1 shows a diagram of how the PG might be

assessed and its width measured.

2: Posterior Vaginal Vault Descent (PVVD) - Level I1

The total posterior vaginal length (TPVL)3 was measured from the centre of the perineal gap

to the vaginal vault. A Moynihan forceps, oriented horizontally, was applied in the midline

1cm below and posterior to the vaginal vault line (prior hysterectomy) or the junction of

cervix and posterior vaginal vault if the uterus was still in-situ. Gentle inferior traction was

then placed on the Moynihan forceps to assess the vaginal vault descent. The distance

between the perineal gap and the point of attachment of the Moynihan under maximum

displacement was called perineal gap Moynihan (PGMOYN) distance.

The posterior vaginal vault descent (PVVD) was then calculated using the formula below:

PVVD = TPVL – PGMOYN

Figure 2A shows the TPVL from which the PGMOYN distance (Figure 2B) needs to be

subtracted to get the PVVD.

3: Mid-Vaginal Laxity (MVL- vault undisplaced)- Level II1

The above gentle inferior traction on the Moynihan was replaced by slight superior tension

towards the vaginal vault and steadying by a surgical assistant. This served to stabilize the

vaginal vault. The surgeon then placed anterior traction perpendicular to the middle of the

posterior vaginal wall, using Gillies forceps, to determine the mid-vaginal laxity. The MVL

(vaginal vault undisplaced) was the length (cm) of the anteriorly displaced mid-vagina over

the lateral vagina.

Figure 3A demonstrates the measurement of the mid-vaginal laxity (MVL - vault

undisplaced).

7

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

A further measurement of the MVL was performed with the vaginal vault on traction as in

Figure 2B. This was called the mid-vaginal laxity (MVL - vault displaced).

4: Recto-vaginal fascial laxity (RVFL)- Level II

This potential defect was assessed once posterior vaginal surgery was commenced. Once the

perineal gap had been excised, the posterior vaginal wall was opened up in the midline up to

the Moynihan (or vulsellum). Artery forceps were used to support both sides of the posterior

vaginal wall incision. The forcep at the apex of the posterior vaginal wall incision was again

held under slight tension towards the vaginal vault for stabilization. The recto-vaginal fascial

laxity (RVFL) was measured in a similar way to the MVL. A gentle anterior traction on the

fascia was applied and the fascia laxity was measured from the mid-vagina level.

Figure 3B demonstrates the measurement of the recto-vaginal fascial laxity (RVFL) (artery

forceps not shown).

The baseline demographic and surgical factors included were: age; parity; weight; height;

BMI; menopause and prior hysterectomy. The surgical initiatives in response to the defects

found: excision of the perineal defect (PG), vaginal vault suspension with sacrospinous

colpopexy – (SSC), excision of excess posterior vaginal skin and recto-vaginal fascial

plications were recorded. These initiatives were performed as per the previous study3: (a)

excision of the entire perineal gap; (b) SSC if PVVD is greater than 5cm; (c) vaginal skin

excision under half MVL (vault undisplaced); (d) rectovaginal fascial plication if RVFL was

over 0.5cm.

The surgical materials and techniques used were identical for the 300 patients. The posterior

vaginal skin was sutured down to the surgical marks for the perineal gap using a 1 Vicyl

8

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

suture. For the perineorrhaphy, 2/0 vicryl rapide was used with 0 vicryl used for the

rectovaginal fascial plication if indicated. Dissection in the recto-vaginal space was

minimalized. Levator or levator fascial plication was not used in this study. A Capio device

(Boston Scientific) was used to insert the size O-Ethibond sacrospinous colpopexy sutures.

Institutional ethical approval was received for the study. All patients were subject to informed

consent as part of this audit of surgical practice. Methods, definitions and units conform to

standards jointly recommended by the International Continence Society and the International

Urogynecological Society, except where specifically noted9. Statistical analysis was

conducted with Stata version 13.1 (College Station, TX, USA).

Results

Table 1 shows the parameter distribution summary. Of the 300 women, (i) 39 (13.0%) were

premenopausal whilst 261 (87.0%) were menopausal; (ii) 139 (46.3%) had a prior

hysterectomy, 128 (42.7%) has a concomitant hysterectomy and 33 (11.0%) had uterus-in-

situ.

The largest defects were found at level I (PVVD - mean 6.0cm; Point C – mean minus 0.9cm)

and level III (PG – mean 2.9cm; GH mean 3.7cm). Level II defects (MVL undisplaced –

mean 1.3cm; RVFL – mean 1.1cm; Points Ap, Bp both mean 1.0cm) were relatively small.

Mean preoperative MVL displaced was 2.8cm. The mean preoperative MVL undisplaced was

1.3cm. From this, it can be interpreted that 1.5cm (55%) of preoperative MVL displaced was

due to vaginal vault laxity.

9

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

In this study, SSC was required in 84% cases to restore Level I support. Vaginal skin excision

occurred in 96% cases with 67% requiring only up to 0.5cm bilateral excision. Fascial

suturing occurred in 76% cases.

Discussion

This study has confirmed and strengthened our preliminary findings3 that the anatomical

defects found at surgery for posterior vaginal compartment prolapse were more at the vaginal

vault (Level I1) and at the vaginal introitus (Level III1) rather than at the mid-vagina (Level

II1). A majority (55%) of mid-vaginal laxity was found to be due to vaginal vault descent.

These results indicate the presence of significant posterior vaginal vault laxity (Level I

defect1) in the majority of the women undergoing the PR. The majority of vaginal vault

support obtained during hysterectomy (prior or concomitant) by using the two uterine

supportive ligaments, the uterosacral (USL) and cardinal, has been shown to be directed

towards the anterior vaginal vault and wall with very little influence on the posterior vaginal

vault and wall10,11,12. Additional posterior vaginal vault support, here by the SSC, was required

in 84% cases. We acknowledge that an SSC is not the only vaginal vault supportive

procedure13,14.

We have adopted a “cutoff” of 5.0cm for the level of PVVD requiring posterior vaginal vault

fixation, e.g. by a SSC. Our rationale for this arbitrary figure is that: (i) above 5.0cm PVVD,

an SSC is more anatomically and surgically desirable; (ii) below that figure, the posterior

vaginal vault support is less in question and it is much harder for the vaginal vault to reach

the sacrospinous ligament.

10

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

The results tend to challenge the traditional concept of a “rectocele” as a Level II 1 defect of

rectovaginal fascia (septum)15,16,17,18 and accompanying vaginal skin laxity. Overall mean

MVL (undisplaced) was only 1.3cm and the mean RVFL was only 1.1cm (Ap and Bp both

1.0cm). These results have also highlighted the “bigger” issues in the posterior vaginal

compartment were, in fact, at the perineum (Level III1 – overall mean PG of 2.9cm; GH mean

3.7cm ) and the vaginal vault (Level I1 – overall mean PVVD of 6.0cm; point C mean -

0.9cm). The surgical implication of this, in the traditional posterior repair, is that the mid-

vagina (Level II1) may have perhaps received to date more intervention than required with the

vaginal vault and perineum perhaps receiving less surgical attention than needed. MRI

evidence19 also supports the view that there is overall weakening and generalized deformation

of pelvic floor tissues rather than specific fascial defects, pointing to this as a reason for the

“relatively modest success rates” of posterior vaginal repairs16-19.

The results for GH between this series and the preliminary report were essentially similar,

3.7cm vs 3.6cm3. GH has been noted to have a strong correlation with prolapse severity20.

The different anatomical benefits of perineal gap (PG) excision have been previously

outlined21. These include: (i) 100% excision of thinned out perineal skin; (ii) 24% increase in

vaginal length over that if the PR was commenced at the hymen; (iii) mean 31% decrease in

GH; (iv) mean 28% increase in perineal length; (v) mean 57% increase in mid-perineal

thickness. Postoperative anatomical results were discussed in the preliminary report3 though

they are not a specific part of this particular report.

We believe that the strengths of this study were to focus attention on the defects present at the

different levels1 of the posterior vaginal compartment. We have shown that the defects are

different from those traditionally thought present. These may then be used to guide

11

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

appropriate surgical measures on a case-specific basis, rather than employing a “one-size-fits-

all” standardized repair. The latter, by not being individualized to the patient’s anatomy,

might leave some defects unresolved. It also has the potential to create new defects5. A

weakness of the study is that it does not, at this stage, include further short and longer term

follow-up data. Part of the future difficulty in creating follow-up data is the need for PR-Q

measurements to be performed under anaesthetia. We did not want to confuse or compromise

the key message from the paper. Other validation studies including inter-observer reliability

studies are, however, also planned for the PR-Q prolapse markers.

Acknowledgements

The authors thank the theatre teams at St Vincent’s and the Mater Hospitals, Sydney in

providing support and facilitating this study. We acknowledge Surgical Assistant, Dr John

McNamara as an additional observer to measurements and the support and patience of

Anesthetists, Associate Professor Alan Molloy and Drs Colleen Kane, Simon Adamo, Luke

Vyvyan and Alex Wang, who recorded most of the measurements.

12

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

References

1: DeLancey JOL (1992) Anatomic aspects of vaginal eversion after hysterectomy. Amer J

Obstet Gynecol 166:1717-1728.

2: Bump RC, Mattiasson A, Bo K, Brubaker LP, et al. (1996) The standardization of female

pelvic organ prolapse and pelvic floor dysfunction. Amer J Obstet Gynecol, 175(1):10 -11.

3: Haylen BT, Avery D, Chiu TL, Birrell W (2014) Posterior repair quantification (PR-Q)

using key anatomical indicators (KAI) – Preliminary Report. International Urogynecological

Journal, 2014(25): 1665-1672.

4: Fowler GE, Richmond DH (2011) Operations for pelvic organ prolapse. In: Lopes T,

Spirtos NM, Naik R, Monaghan JM Bonney’s Gynaecological Surgery. 11 th Edition. Wiley-

Blackwell London. Chapter 16. P154.

5: Nichols DH, Randall CL (1996) Posterior colporrhaphy and perineorrhaphy. In Nichols

DH, Randall CL Vaginal Surgery 4th Edition. Williams & Wilkins. Baltimore. Chapter 11

p279.

6: Karram MM, Maher C (2013) Surgery for posterior vaginal wall prolapse. Int Urogynecol

J 24:1835-1 841.

7: Summers A, Winkel LA, Hussain HK, DeLancey JOL (2006) The relationship between

anterior and apical compartment support. Am J Obstet Gynecol 194:1438-1443.

8: Yang, A., Mostwin, J., Genadry, R., Sanders, R. (1993) Patterns of prolapse demonstrated

with dynamic fastscan MRI; reassessment of conventional concepts of pelvic floor

weaknesses. Neurourol Urodyn, 12(4): 310-311.

9: Haylen BT, Freeman RM, de Ridder D, Swift SE, Berghmans B, Lee J, Monga A, Petri E,

Rizk D, Sand P, Schaer G (2010). An International Urogynecological Association (IUGA) –

International Continence Society (ICS) Joint Report into the Terminology for Female Pelvic

13

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

Floor Dysfunction. Neurourology & Urodynamics, 29:4-20. International Urogynecology J,

21:5-26.

10: Haylen BT, VU D, Birrell W, Vashevnik S, Tse K (2012) A preliminary anatomical basis

for dual (uterosacral and sacrospinous ligaments) vaginal vault support at colporrhaphy. Int

Urogynecol J 23: 879-882.

11: Samaan A, Vu D, Haylen BT, Tse K (2014) Cardinal ligament surgical anatomy: cardinal

points at hysterectomy. Int Urogynecol J 25(2):189-195.

12: Vu D, Haylen BT, Tse K, Farnsworth A (2010) Surgical anatomy of the uterosacral

ligament. Int Urogynecol J. 21: 1123-1128.

13: McCall ML (1957) Posterior culdoplasty: surgical correction of enterocoele during

vaginal hysterectomy; a preliminary report. Obstet Gynecol 10:595-602.

14: Montella JM, Morrill MY (2005) Effectiveness of the McCall culdeplasty in maintaining

support after vaginal hysterectomy. Int Urogynecol J 16:226-229.

15: Francis WFA, Jeffcoate TNA (1961) Dyspareunia following vaginal operations. J Obstet

Gynaecol Br Commonw 68:1-10

16: Karram MM, Maher C (2013) Surgery for posterior vaginal wall prolapse. Int Urogynecol

J 24:1835-1841.

17: Richardson AC (1993) The rectovaginal septum revisited: its relationship to rectocele and

its importance in rectocele repair. Clin Obstet Gynecol 36:976-983.

18: Richardson AC (1995) The anatomic defects in rectocele and enterocele. J Pelv Surg

1:214-221.

19: Lewicky-Gaupp C, Yousuf A, Larson KA, Fenner DE, DeLancey JOL (2010) Structural

position of the posterior vagina and pelvic floor in women with and without posterior vaginal

prolapse. Am J Obstet Gynecol 202(5) 497.

14

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

307

308

309

310

311

20: Ghetti C, Grerory WT, Edwards SR, Otto LN, Clark AL (2005) Severity of pelvic organ

prolapse associated with measurements of pelvic floor function. Int Urogynecol J 16:432-

436.

21:Haylen BT, Birrell W, Naidoo S, Younis M (2014) Perineorrhaphy quantitative

assessment (Pe-QA) International Urogynecology J 26(4):539-544.

15

312

313

314

315

316

317

Table 1: Pre-operative baseline patient characteristics and the two sets of posterior prolapse

markers: (i) PR-Q; (ii) POP-Q

Variable Mean SD Min Max

Age (years) 63.6 11.8 31 91

Weight (kg) 71.1 14.6 44 141

Height (cm) 162.9 7.1 142 187

BMI (kg/m2) 26.7 5.0 18.6 46.3

Parity 2.6 1.2 0 8

PR-Q POSTERIOR PROLAPSE MARKERS

Perineal gap - PG (cm) 2.9 1.0 0.3 6.0

PVVD (cm) 6.0 2.0 0.3 15.0

MVL, undisplaced (cm) 1.3 0.6 0 3.5

Recto-vaginal fascial laxity - RVFL (cm) 1.1 0.7 0 4.0

POP-Q POSTERIOR PROLAPSE MARKERS

Pre-op point C (cm) -0.9 2.3 -8.0 8.0

Pre-op point Ap (cm) 1.0 1.4 -3.0 5.0

Pre-op point Bp (cm) 1.0 1.5 -3.0 6.0

Genital Hiatus (GH) pre-op (cm) 3.7 0.9 1.5 6.5

16

318

319

320

321

Figures:

Figure 1: Perineal Gap (PG): Thinned out medial area (cm) between Moynihan forceps

placed bilaterally where the line of the labia minora meets the perineum

Figures 2A & 2B: Posterior Vaginal Vault descent (PVVD): Subtract distance from inferiorly

displaced vaginal vault and PG ( Figure 2B) from total posterior vaginal length (TPVL -

Figure 2A)

Figures 3A – Mid-Vaginal laxity (MVL- vault undisplaced) – Vault is held is undisplaced

position by Moynihan whilst anterior traction on the midpoint of the vaginal supero-inferiorly

is used to demonstrate the MVL (cm).

Figure 3B – Rectovaginal fascial laxity (RVFL): This is similar to 3A though the

measurement of the RVFL occurs once the posterior vaginal wall has been opened to the

apical Moynihan.

17

322

323

324

325

326

327

328

329

330

331

332

333

334