An Investigation of the Surgical Treatment of Endometriosis

Full text

An Investigation of the Surgical Treatment of
Endometriosis
by
Peter Barton-Smith MB BS MRCOG
Submitted for the Degree of Doctor of Medicine
Postgraduate Medical School
University of Surrey
 Peter Barton-Smith 2010

i
Statement of Originality
This thesis and the work to which it refers are the results of my own efforts. Any
ideas, data, images or text resulting from the work of others (whether published
or unpublished) are fully identified as such within the work and attributed to their
originator in the text, bibliography or in footnotes. This thesis has not been
submitted in whole or in part for any other academic degree or professional
qualification. I agree that the University has the right to submit my work to the
plagiarism detection service TurnitinUK for originality checks. Whether or not
drafts have been so-assessed, the University reserves the right to require an
electronic version of the final document (as submitted) for assessment as
above.

ii
Abstract
Background
The surgical treatment of endometriosis has developed in recent decades
alongside the development of laparoscopic surgery and has been directed
towards understanding and improving the outcomes related to fertility and pain.
Little is known about the techniques and instruments actually used by
gynaecological laparoscopic surgeons in relation to the evidence available.
Furthermore, it is still not clear whether vaporisation or ablation of lesions is as
efficacious as excision.
Objectives
Firstly, to internationally survey the views of gynaecologists to find out how they
surgically treat endometriosis. Secondly, to determine whether excision or
vaporisation is the optimal surgical technique for minimal to moderate
endometriosis.
Methods
A 34 question web-based survey was constructed, piloted and sent out by email
to the members of the BSGE, ESGE and AAGL to get a snapshot of current
practice. A randomised blinded trial of harmonic scalpel excision versus carbon
dioxide laser vaporisation for the treatment of minimal to moderate
endometriosis in women with pelvic pain was carried out, using as a primary
outcome measure EHP-30 Core pain domain, and secondary outcomes for
VAS scores for dysmenorrhoea, dyspareunia, chronic pelvic pain and
dyschezia, EHP-30 HRQoL measures, and HADS.
Results
From the survey, the predominant view is that endometriomas should be
excised and that bowel resection should be avoided if possible in recto-vaginal
disease. For minimal to moderate disease, superficial disease can be treated
with a combination of excision or vaporisation depending on the case, but that
deep disease should be excised.
The trial results show that both excision and vaporisation result in an equally
significant proportion of patients showing some level of pain improvement at 12
months (85.4 v 72.9%). However, excision results in a significantly greater
extent of improvement for both superficial and deep disease than vaporisation
(p=0.008). In addition, for deep disease, the extent of improvement in pain with
vaporisation is not significant (p=0.262). Overall 20% of patients stay the same
or get worse.
Conclusions
Excision results in greater pain and quality of life improvement than vaporisation
for minimal to moderate disease, and is the optimal technique for the surgical
treatment of all types of endometriosis. However it must be taken into account
that the range of improvement is wide and a proportion of patients will continue
to deteriorate.

iii
Table of Contents
STATEMENT OF ORIGINALITY! I!
ABSTRACT!II!
TABLE OF CONTENTS!III!
LIST OF TABLES! IV!
LIST OF FIGURES!VII!
ACKNOWLEDGEMENTS!VIII!
CHAPTER 1 - INTRODUCTION! 1!
CHAPTER 2 - LITERATURE REVIEW! 4!
A!GENERAL!REVIEW! 4!
A!CRITICAL!REVIEW!OF!THE!SURGICAL!TREATMENT!FOR!ENDOMETRIOSIS!19!
CHAPTER 3 - METHODOLOGY!38!
AN!INTERNATIONAL!SURVEY!OF!SURGICAL!TECHNIQUES!USED!IN!THE!TREATMENT!OF!ENDOMETRIOSIS
!38!
A!RANDOMISED!BLINDED!TRIAL!OF!CARBON!DIOXIDE!LASER!VAPORISATION!VERSUS!HARMONIC!
SCALPEL!EXCISION!OF!RASRM!STAGE!1:3!ENDOMETRIOSIS!IN!WOMEN!WITH!PELVIC!PAIN.!50!
CHAPTER 4 - RESULTS!63!
RESULTS!OF!AN!INTERNATIONAL!SURVEY!OF!SURGICAL!TECHNIQUES!USED!IN!THE!TREATMENT!OF!
ENDOMETRIOSIS!63!
CHAPTER 5 - RESULTS!96!
RESULTS!OF!A!RANDOMISED!BLINDED!TRIAL!OF!CARBON!DIOXIDE!LASER!VAPORISATION!VERSUS!
HARMONIC!SCALPEL!EXCISION!OF!RASRM!STAGE!1:3!ENDOMETRIOSIS!IN!WOMEN!WITH!PELVIC!PAIN!96!
CHAPTER 6 - CONCLUSIONS!158!
REFERENCES!162!
APPENDICES!175!
APPENDIX!A!–!REVIEWS!IN!GYNAECOLOGICAL!PRACTICE!REVIEW!ARTICLE!176!
APPENDIX!B!–!ENDOMETRIOSIS!TRIAL!PATIENT!INFORMATION!LEAFLET!185!
APPENDIX!C!–!ENDOMETRIOSIS!TRIAL!CONSENT!FORM!189!
APPENDIX!D!–!ENDOMETRIOSIS!TRIAL!VISUAL!ANALOGUE!SCALE!190!
APPENDIX!E!–!ENDOMETRIOSIS!TRIAL!EHP!30!QUESTIONNAIRE!191!
APPENDIX!F!–!ENDOMETRIOSIS!TRIAL!HADS!QUESTIONNAIRE!197!
APPENDIX!G!–!ENDOMETRIOSIS!TRIAL!12!MONTH!DEMOGRAPHIC!QUESTIONNAIRE!201!
APPENDIX!H!–!RASRM!SCORING!SHEET!205!
APPENDIX!I!!–!TRIAL!FOLLOW!UP!CLARIFICATION!LETTER!206!
APPENDIX!J!!–!INVITATION!EMAIL!FOR!PILOT!SURVEY!207!
APPENDIX!K!–!INVITATION!EMAIL!FOR!MAIN!WEB!SURVEY!208!
APPENDIX!L!–!PILOT!WEB!SURVEY!209!
APPENDIX!M!–!MAIN!WEB!SURVEY!222!

iv
List of Tables
An International Survey of Surgical Techniques used in the Treatment of
Endometriosis
Table 1. A summary of case series for the treatment of recto-vaginal endometriosis.....29
Table 2. Survey response rates by specialist society………………………………………66
Table 3. Number of operations performed for endometriosis by survey responders
per month………………………………....…………………………………………..71
Table 4. Percentages of survey responders that operate on various types
of endometriosis…………..……………………………………………………........72
Table 5. Proportion of surgeons who see and treat patients at the same time for
endometriosis………………………………………………………………………...73
Table 6. Previous and current use of surgical instruments for endometriosis…………..75
Table 7. Number and percentages of responders who use various techniques as
their main method for surgically treating minimal to moderate endometriosis..76
Table 8. Reasons for choosing main technique for treating superficial minimal
to moderate endometriosis………………………………………………………….77
Table 9. Reasons for choosing technique for treating deep minimal to
moderate endometriosis….………………………………………………………....78
Table 10. Responders use of techniques for surgically treating endometriomas………...80
Table 11. Reasons for choosing technique for treating endometriomas of <3cm
diameter……………………………………………………………………………….81
Table 12. Reasons for choosing technique for treating endometriomas of >3cm
diameter……………………………………………………………………………….82
Table 13. Use of pre-operative tests for recto-vaginal nodules…………………………….83
Table 14. Number of responders who operate on recto-vaginal nodules with a
colorectal surgeon.............................................................................................83
Table 15. Surgical technique for treating recto-vaginal nodules……………………………84

v
List of Tables
A Randomised Blinded Trial of Carbon Dioxide Laser Vaporisation versus
Harmonic Scalpel Excision of rASRM Stage 1-3 Endometriosis in Women with
Pelvic Pain
Table 16. Number of participants missing from EHP-30 Core pain score analysis……..101
Table 17. Comparison of baseline variables between patients included in the
main analysis, and missing or excluded patients……………………………….102
Table 18. Numbers analysed at each point of the trial for each outcome measure…….104
Table 19. Comparison of mean starting scores for pain outcomes for superficial
and deep disease…………………………………………………………………..109
Table 20. Comparison of mean baseline variables for excision and vaporisation………109
Table 21. Comparison of mean baseline outcome scores for excision and
vaporisation…………………………………………………………………………111
Table 22. Proportion of patients showing improvement for EHP-30 Core pain score….112
Table 23. Extent of improvement in EHP-30 Core pain score against baseline
for excision and vaporisation alone at 12 months………………………………113
Table 24. Comparative extent of improvement in EHP-30 Core pain score
against baseline for excision and vaporisation at all follow-up points………..114
Table 25. Mean improvement in VAS symptom scores for excision at
12 months against baseline…………………………………………………….…116
Table 26. Mean improvement in VAS symptom scores for vaporisation at
12 months against baseline……………………………………………………….117
Table 27. Extent of comparative improvement for VAS symptom scores for excision
and vaporisation at all follow-up points…………………………………………..118
Table 28. Comparative proportion of improvement for VAS symptom scores for excision
and vaporisation at all follow-up points…………………………………………..120
Table 29. Extent of improvement in EHP-30 HRQoL outcome measures for excision
alone at baseline against 12 months……………………………………………..121

vi
Table 30. Extent of improvement in EHP-30 HRQoL outcome measures for
vaporisation alone at baseline against 12 months……………………………...122
Table 31. Comparative extent of score improvement in EHP-30 HRQoL outcome
measures for excision and vaporisation at all follow-up points………………..123
Table 32. Extent of improvement in anxiety and depression for excision alone at
12 months versus baseline………………………………………………………..126
Table 33. Extent of improvement in anxiety and depression for vaporisation alone at
12 months versus baseline………………………………………………………..126
Table 34. Comparative extent of improvement in anxiety and depression for
excision and vaporisation at all follow-up points………………………………..127
Table 35. Comparison of the extent of improvement in EHP-30 Core pain score for
excision and vaporisation with deep and superficial disease at 12 months
against baseline…………………………………………………………………….130
Table 36. Comparative proportional improvement of patients with deep disease
against those with superficial disease at 6 and 12 months……………………131
Table 37. Comparative proportional improvement of patients with different
rASRM stages at 6 and 12 months……………………………………………….132
Table 38. Extent of improvement in EHP-30 Core pain score for “per protocol”
patients for excision and vaporisation at all follow-up points……………...…..133

vii
List of figures
An International Survey of Surgical Techniques used in the Treatment of
Endometriosis
Fig 1. Distribution of responders by geographical region………………………………………...67
Fig 2. Age distribution of survey responders………………………………………………………68
Fig 3. Year survey responder qualified as a doctor……………………………………………….69
Fig 4. Number of years survey responder has performed independent laparoscopic
surgery………………………………………………………………………………………….70
Fig 5. Pie chart of the distribution of training methods in survey responders…………………..70
Fig 6. Means plot comparing the mean number of years performing laparoscopic surgery
with the number of endometriosis operations performed per month..............................72
Fig 7. Means plot comparing the mean number of years performing independent
laparoscopic surgery with “see and treat” management…………………………………..74
Fig 8. Pie Chart showing the use of 2-stage procedures for endometriomas………………….79
A Randomised Blinded Trial of Carbon Dioxide Laser Vaporisation versus
Harmonic Scalpel Excision of rASRM Stage 1-3 Endometriosis in Women with
Pelvic Pain.
Fig 9. Flow diagram of participants through the trial………………………………………………96
Fig 10. Age distribution of participants……………………………………………………………..105
Fig 11. Distribution of participants by rASRM score………………………………………………106
Fig 12. Distribution of participants by rASRM stage………………………………………………107
Fig 13. Graph of mean improvement in EHP-30 Core pain score against time for
excision and vaporisation…………………………………………………………………...115
Fig 14. Distribution of participants for different grades of anxiety at baseline and 12 months.124
Fig 15. Distribution of participants for different grades of depression at baseline
and 12 months………………………………………………………………………………..125
Fig 16. Graph of the extent of improvement in EHP-30 Core pain score for
“same or worse” pateints at 12 months over time………………………………………..134

viii
Acknowledgements
For their help with academic supervision and statistical analysis:
Dr Karen Ballard PhD
Mr Peter Williams MSc
For their help with patient follow up, data collection and data entry:
Mrs Pat Haines RGN BSc (Hons)
Ms Sue Riva BA (Hons)
For their help with protocol development, surgery and clinical supervision:
Mr Tyrone Carpenter MD MRCOG
Mr Andrew Kent TD MD FRCOG
For their help with funding this thesis:
The Minimal Access Therapy Training Unit, Guildford

1
Chapter 1 - Introduction
In the surgical treatment of endometriosis there are two views that favour either
vaporisation or excision as the optimal means of treatment. There exist various
instruments to carry out these treatments. The “excisers” tend to use
electrosurgical or ultrasound instruments to cut around the lesion and remove it
along with the underlying fibrotic tissue. The “vaporisers” tend to use lasers,
electro-surgery or modified electro-surgical techniques to vaporise the lesion
and underlying fibrosis.
The “excisers” claim that the vaporisation of endometriosis is too superficial and
does not go deep enough to remove infiltrating tissue. The risk of inadequate
treatment or recurrence of symptoms may be greater. The “vaporisers” claim
their technique is faster and more efficient for the removal of widespread
disease and just as effective. It is unclear why surgeons use one approach or
the other and how many are using each approach.
There is only one randomised trial comparing the two approaches. This
compares the treatment of rASRM (revised American Society for Reproductive
Medicine) (1979) stage 1-2 minimal and mild endometriosis by excision or
vaporisation. This trial shows no difference in the two modalities in terms of pain
outcome (Wright et al., 2005), although it did not include stage 3 moderate
disease. The “excisers“ claim that the more advanced the disease, and the
more infiltrating and nodular it becomes, the more important excision becomes.
This disease “progression” culminates in the most extreme form of pelvic

2
endometriosis: the recto-vaginal nodule. This is probably treated by excision,
with or without bowel resection, by all endometriosis surgeons. The
vaporisation/excision debate is not an issue here.
In our unit we use both vaporisation and excision to treat rASRM stage 1-3
endometriosis. By vaporisation, we mean complete destruction or vaporisation
of endometriosis and fibrotic tissue down to underlying normal tissue, not just a
superficial treatment as often seen with ablation. At the end of treatment the
macroscopic appearance is identical for both excision and vaporisation. With
the correct surgical approach, both techniques can be used around sensitive
structures including ureters, bladder and vessels. We use CO2 laser for
vaporisation and the harmonic scalpel for excision.
The CO2 laser can be used both as a vaporisation and excision tool. At
Guildford, Professor Chris Sutton carried out the first randomised double blind
controlled trial of ablative surgery using a CO2 laser (Sutton et al., 1994).
Harmonic (ultrasound generated) energy has been around for more than a
decade as a cutting tool (Feil, 2005). In comparison to electrical energy, it
removes the risk of electrical injury and results in less thermal spread to
adjacent tissues.
A review of the endometriosis literature in chapter 2 reveals gaps in our
knowledge and capabilities in many areas. I have attempted to present a
rationale for surgical therapy as the current optimal treatment for endometriosis
in most cases. 82% of cases of endometriosis fall into stages 1-3 (Redwine,

3
1990b). It therefore seems most logical to mainly concentrate on this group in
my thesis.
In particular I wish to focus on the fundamental question of whether or not there
is a difference between the excision or vaporisation of minimal to moderate
endometriosis in terms of improvement in pain scores. To achieve this I aim to
critically review the literature for the surgical treatment of endometriosis to see if
there is any evidence in favour of one technique over another. I then aim to
compare this with the actual use of techniques and instruments used by
gynaecological laparoscopists. This will be determined by a web-based survey.
My thesis therefore divides into the three areas that are listed below:
1. A literature review of the diagnosis, management and treatment of
endometriosis with particular reference to surgical management (an earlier
version of this review was published in a peer reviewed journal and is at
Appendix A) (Barton-Smith et al., 2006).
2. An international survey of techniques and instruments used in the treatment
of endometriosis: who is using what and why?
3. A randomised blinded trial of carbon dioxide laser vaporisation versus
harmonic scalpel excision of rASRM stage 1-3 endometriosis in women with
pelvic pain.

4
Chapter 2 - Literature Review
A general review
Endometriosis is a common condition of unknown aetiology that can cause pain
and infertility in women. It is defined as the presence of endometrial glands and
stroma present outside of the uterus (Olive and Schwartz, 1993). It is most
commonly found in pre-menopausal women but rarely can also occur in
postmenopausal women or pre-menarchal girls.
Prevalence
The prevalence of endometriosis has probably increased over the last one
hundred years or so for several reasons. Modern women have many more
menses than their predecessors as they spend less of their reproductive lives in
a state of pregnancy. Women now have an estimated 450 menses in their
lifetime compared to the 100 that their Victorian forebears had. Consequently
the frequency of dysmenorrhoea has increased. The introduction of
laparoscopy over the last three decades has allowed us to diagnose
endometriosis more easily, and nowadays most gynaecologists are able to
perform a diagnostic laparoscopy at least. The current prevalence is estimated
to be up to 10% (Vessey et al., 1993, Kjerulff K. H., 1996, Vigano, 2004).
Vercellini stated in his address to the World Congress on Endometriosis in
Maastricht in 2005 that the incidence has not increased in the last thirty years
and remains at 2.37-2.49 per 1000 women per year, equating to an
approximate prevalence of 6-8% (Hummelshoj, 2006, Leibson et al., 2004). A
national epidemiological study of endometriosis within a community based

5
sample, revealed that the prevalence of diagnosed endometriosis in women
aged 18-50 is 1.5% (Ballard et al., 2009). This lower prevalence can be
explained by the sample being drawn from women attending primary rather
than secondary care.
Localisation and Appearance
Endometriosis is most commonly found in the pelvis (Jenkins et al., 1986). More
rarely it is found in remote sites like the lungs and brain (Di Palo et al., 1989,
Thibodeau et al., 1987). It has even been found in men (Oliker and Harris,
1971). In the pelvis, it is more frequently left sided (Vercellini et al., 2004). The
reason for this left sided dominance is unclear, although the direction of flow of
peritoneal fluid been proposed as a possible cause (Chapron et al., 2003).
Endometriotic lesions can infiltrate any structure in the pelvis. They are seen
more commonly on the pelvic peritoneum, utero-sacral ligaments, bladder,
sigmoid colon and rectum, the most severe presentation being the recto-vaginal
nodule. On the ovaries, endometriotic cysts can form that contain a thick
“chocolate-like” substance formed from old blood. These cysts are often
adherent to the pelvic side-wall. Rarely, the ureter can be infiltrated with
endometriosis itself, though stenosis is more commonly due to fibrotic
impingement by a proximal uterosacral nodule (Lucero et al., 1988).
Endometriotic deposits can appear in a range of colours and textures from
transparent vesicles and white fibrotic plaques to red haemorrhagic flares and
blue/black nodular lesions, first systematically described by Martin et al in 1987
(Martin and Vander Zwagg, 1987). These may be superficial or described as

6
deep if they extend more than 5mm beneath the peritoneal surface. This wide
variety of appearance is not always recognised by the general gynaecologist
and the diagnosis may be easily missed. Overall appearance may vary from the
barely visible through to the “frozen pelvis”.
Epidemiology
There are several large epidemiological studies detailed below that give us an
insight into the disease (Hummelshoj, 2006, Treloar, 2005b). The results reveal
the depth of the problem for women, their families and society in general, as
well as highlighting the current inadequacy in management.
The All Party Parliamentary Group in the UK has so far collected
epidemiological data via its on-line questionnaire from 7025 women from 52
countries. It estimates that two million women in the UK suffer from
endometriosis. The results of this survey to date show a delay in diagnosis in
this cohort averaging 8.3 years. 65% of women complained of being wrongly
diagnosed initially. Only a third believed their treatment to be effective. A
startling 78% took an average of 5.3 days off work per month and 72% reported
relationship problems (Hummelshoj, 2006). These results are similar to those
found in a recent Australian genetic epidemiological study of 3895 women
diagnosed with endometriosis where the average age of onset of endometriosis
symptoms was 20.1 ±6.8 years (Treloar, 2005b). The youngest diagnosed
participant in this study was 13 years old. The disease or its antecedent has
been diagnosed in females as young as eight years old (Marsh and Laufer,
2005).

7
Endometriosis is less common in black African women compared with
Caucasians and more common in East Asians, but is found in all ethnic groups
(Sangi-Haghpeykar and Poindexter, 1995, Hasson, 1976).
Endometriosis is found in 40-60% of women with pelvic pain and in 20-30% of
women suffering from infertility (Mahmood and Templeton, 1991, Eskenazi and
Warner, 1997, Ajossa et al., 1994). Women with more advanced disease have
a higher rate of infertility (19.5% for rASRM stage 1 versus 28.7% for rASRM
stage 2-4) (Plumb, 2005).
Risk factors for endometriosis include early age of menarche, short menstrual
cycles, long duration of menstrual flow, a family history of endometriosis, and
there is an inverse relationship with parity (Vigano, 2004).
Aetiology
Many theories of the aetiology of endometriosis have been postulated since
Rokitansky first described the disease in 1860 (Von Rokitansky). Meyer
proposed the theory of coelomic metaplasia in 1909 and postulated that tissue,
with the potential to develop into endometrial-like cells later in life, was laid
down in the trans-embryonic coelom (Meyer, 1909). Halban proposed the
possibility of haematological or lymphatic spread from the endometrium in 1924
(Halban, 1924). Sampson proposed the theory of retrograde menstruation in
1927 (Sampson, 1927) and this has become the “default” explanation.
However, there remain several problems with Sampson’s theory. There is little
doubt that reflux menstruation occurs. Bloody peritoneal fluid is found in 80-

8
90% of menstruating women (Blumenkrantz et al., 1981, Halme et al., 1984)
compared with only 15% of women with occluded fallopian tubes. However,
most women do not develop endometriosis. Moreover, the cells found in
endometriosis are not identical to normal endometrium (Redwine, 2002) and
endometriosis does not generally recur if treated surgically as one might expect
if retrograde menstruation was to continue (Abbott et al., 2003, Redwine, 1991).
Also, it does not explain the small, but nevertheless relevant, occurrence in
men, pre-menarchal girls and post-menopausal women.
Consequently, all of the above theories remain in the frame to this day with no
clear evidence having emerged for either one. Despite being the dominant
theory, it seems likely that Sampson’s view was over simplistic.
More recent advances in technology have permitted the emergence of new
theories. Altered immune function has gained credibility with those seeking to
find a basis for the discrepancy between the frequency of retrograde
menstruation and the infrequency of endometriosis. This theory was first
postulated in 1987 by Gleicher (Gleicher et al., 1987), who suggested that
immune system alterations result in a failure to “mop up” ectopic endometrial
cells and therefore allow them to infiltrate at the site of disease. Immune system
alterations in endometriosis sufferers have been shown in natural killer (NK)
and cytotoxic T cells, and aberrations have been found in immune mediators
such as tumour necrosis factor-α, Interferon-γ and polyclonal B-cell auto
antibodies (Kitawaki et al., 2002).

9
Simpson first suggested a genetic basis for endometriosis in 1980 (Simpson et
al., 1980). This is likely to be complex and polygenic in nature. Linkage study
work has now shown a susceptibility locus on Chromosome 10q26 (Treloar,
2005a). Work is also progressing in other areas including studies of expression
profiling, tumour genetics and functional candidate genes (Barlow and
Kennedy, 2005).
There is evidence suggesting that endometriosis may increase the risk of
cancer. Brinton’s work looking at cancer risk after a diagnosis of endometriosis
shows a relative risk of 1.18 for developing cancer of any form, and a relative
risk of 1.92 for developing ovarian cancer (Brinton et al., 1997). The K-ras
oncogene and P-ten tumour suppressor gene appear to be involved in this
process. Mice with either of these gene mutations developed endometriosis,
and mice with both gene mutations simultaneously produced endometrioid
ovarian adenocarcinoma (Dinulescu, 2005). Clinically, the endometrioma is of
most concern as far as malignancy risk is concerned; a risk of 0.7% has been
suggested (Nishida et al., 2000), and in one more recent report, 7% of
endometriomas contained a neoplastic process (Bedaiwy et al., 2009).
The theory of progesterone resistance is the most recent to emerge. In his
review, Osteen has suggested that the impaired regulation of matrix
metalloproteinases, that has already been shown to increase the implantation
potential of the endometrial tissue (Bruner-Tran et al., 2002), may be due to a
decreased responsiveness to progesterone (Osteen et al., 2005).

10
Other aetiological theories rely on environmental factors. Nutritionally, wheat
has been implicated as a potential source of exacerbation in endometriosis.
Manipulation of the diet in endometriosis sufferers resulted in an improvement
in prospectively collected “measure yourself medical outcome profile scores”
(MYMOP) that measure symptom improvement (Shepperson Mills, 2004). Other
potential environmental agents include organochlorines (PCBs and dioxins). In
rodents and primates these promote the development of endometriosis and
have been found in higher concentrations in human sufferers (Heilier, 2004).
Whilst endometriosis has been shown to be a predominantly oestrogen
dependent disease, the underlying pathophysiology that creates the conditions
for oestrogen to drive the development of endometriosis is likely to be multi-
factorial and include many of the above possibilities.
Diagnosis
The main clinical symptoms of endometriosis are infertility, dysmenorrhoea,
dyspareunia, dyschezia, and chronic pelvic pain (defined as pain of greater than
six months duration and not cyclical in nature) (Jantos, 2007). Other symptoms
seen less commonly include haematuria and rectal bleeding. Making a
diagnosis purely on presenting symptoms is difficult as there is considerable
overlap with other conditions. This often results in a delay in diagnosis of
between five and ten years (Treloar, 2005b, Hadfield et al., 1996, Husby et al.,
2003, Arruda et al., 2003, Ballard et al., 2006). This is an important finding
when one considers that endometriosis is a progressive disease that worsens
with time (Koninckx et al., 1991, Matsuzaki, 2004). The delay in diagnosis may
also result in associated psychological morbidity (Jones et al., 2004b).

11
Clinical signs on examination can also be difficult to elicit. Endometriotic
nodules on the utero-sacral ligaments or in the recto-vaginal septum may be
palpable and are easier to feel on a combined vaginal-rectal examination.
These are most reliably palpated if the examination is undertaken during
menstruation (Koninckx et al., 1996). In some cases endometriosis invading
through the vaginal mucosa may be visible on speculum examination. It
remains to be seen whether symptom profile questionnaires alone, without
invasive testing, will be able to improve diagnosis (Ballard et al., 2009).
The gold standard for diagnosing endometriosis in the abdomen and pelvis is
the visual identification of characteristic lesions at laparoscopy. In one study,
this means of diagnosis was shown to be 97% sensitive and 77% specific
(Buchweitz et al., 2003). Lesions may also be visually confirmed on vaginal
speculum examination, in the bladder at cystoscopy or in the bowel mucosa at
sigmoidoscopy. It is considered good clinical practice that the diagnosis is
confirmed by histology of at least one lesion (Kennedy, 2004).
However, negative histology does not exclude the diagnosis as explained by
Clement who describes the potential alterations in glandular and stromal
components of endometriosis, especially in small biopsy specimens resulting in
false negative results (Clement, 2007, Shafik A, 2000). A correlation of 88%
was found by Ballard between visual inspection and positive histology (Ballard
et al., 2009). It should also be noted that visual inspection alone can produce
false positive diagnoses as lesions may be mimicked by psammoma bodies
caused by old haemorrhage or suture material from previous surgery for
example (Martin and Vander Zwagg, 1987).

12
It is recommended that histology should be obtained for endometriomas of
>3cm diameter and deep disease, to exclude rare cases of malignancy
(Kennedy, 2004). Disease sites and depths should be mapped and recorded
(Kennedy, 2004) to allow adequate reassessment of the disease subsequently.
Many other tests have been employed to aid in diagnosis without resorting to
surgery. CA125 is raised in some cases, but the test remains non-specific. Most
endometriosis sufferers are pre-menopausal and there are many other causes,
both pathological and physiological, for a raised CA125. Therefore it has been
argued that, compared with laparoscopy, CA125 has no value as a diagnostic
tool in endometriosis (Mol et al., 1998). Furthermore, if the level is raised in an
endometriosis case and applied to a risk of malignancy index, then there
remains the possibility of the woman being subjected to an unnecessary
laparotomy for a high-risk score. If a laparoscopy is carried out first then a
laparotomy may be avoided.
New serum tests are being developed. One such test is based on the detection
of autoantibodies against Thomsen-Friedenreich antigen bearing proteins found
in endometriosis. Current sensitivity and specificity results are 80% and work to
improve these is underway (Hummelshoj, 2006). Also, gene expression
analysis is beginning to identify potential markers of endometriosis in peripheral
blood samples as well as in endometriotic lesions (Hornung, 2005, Van
Langendonckt, 2004).
Ultrasound is commonly used as a diagnostic tool in most gynaecological
departments. Trans-vaginaI ultrasound (TVS) is not a useful tool in the

13
diagnosis of peritoneal endometriosis although it is useful for diagnosing
endometriomas, disease infiltrating the bladder (Moore et al., 2002, Bazot et al.,
2004b) and is becoming a more successful tool for the diagnosis of recto-
vaginal disease too (Hudelist and Keckstein, 2009). Also, trans-rectal
ultrasound is a useful tool in diagnosing deep infiltrating disease in the recto-
vaginal septum (Chapron et al., 2004).
Magnetic resonance imaging (MRI) has been shown to be of use in diagnosing
the extent of deep nodular disease, particularly in the recto-vaginal septum
(Bazot et al., 2005, Bazot et al., 2004a). Fat suppression MRI films and the use
of phased array coils appear to offer even better images that can show the level
of invasion into the bowel wall itself. In the future this may help tailor how
radical the excision of recto-vaginal nodules should be prior to surgery. A recent
report from Bazot suggests “MRI provides a more reliable map of DIE than
physical examination, TVS or trans-rectal ultrasound” (Bazot et al., 2009).
CT scans do not offer the required resolution to do this, although spiral CT may
be better. Double contrast barium enema alone has been shown in two recent
studies to have close to 90% sensitivity, and specificity ranging from 54 to 88%
(Ribeiro et al., 2008, Faccioli et al., 2008). For suspected ureteric involvement,
an intravenous urogram (IVU) or MRI may be used and a cystoscopy may be
required to assess the bladder mucosa (Bazot et al., 2008).
Staging the disease in a way that gives useful information about the extent of
pain or fertility, its management and prognosis, has so far eluded us. With
endometriosis, the extent or type of disease is not clearly correlated with pain

14
levels. Several systems have been attempted, often based upon the visual
findings during laparoscopy, histological examination or a combination of the
two (Acosta et al., 1973, Batt and Mitwally, 2003, Adamyan et al., 1993). The
most commonly used one to date is the revised American Society for
Reproductive Medicine (rASRM) originally produced in 1979 (1979) and revised
in 1985 (1985) and 1997, which grades endometriosis as minimal (stage 1),
mild (stage 2), moderate (stage 3) and severe (stage 4) (1997).
Chapron argued that all of these systems fail to correlate well with symptoms of
pain and fertility (Chapron et al., 2003, D'Hooghe et al., 2003) or aid in
prognosis. He argued that including palpation to assess depth clinically was
required. This concept of considering depth further was picked up in the Enzian
system (Tuttlies et al., 2005). Further elucidation of the pathophysiology of the
disease will hopefully facilitate the development of more useful scoring systems.
We have begun to speak more about two types of disease, superficial and deep
infiltrating endometriosis (DIE), as evidence emerges that these two forms of
the disease may behave differently from each other (Garry, 2004). That
superficial endometriosis is less severe in its symptoms and complexity of
management, is one of the subjects to be considered in this thesis. In the light
of all this evidence, future scoring or staging systems may be more
appropriately aimed at predicting surgical morbidity or the appropriateness of
radical surgery as Wright suggested in his unpublished thesis on surgical
endometriosis. That being said, a new validated endometriosis fertility index
(EFI) has been developed by Adamson (Adamson and Pasta, 2009). It is
designed for patients who are attempting spontaneous conception and is an

15
intra-operatively derived score from visual assessment of the ovary, tube and
fimbria, the patient’s age, years of infertility and previous pregnancies to give a
graphical display of the chances of spontaneous conception over time. Its full
evaluation in clinical practice is awaited.
Medical treatment
In most cases, women who reach the gynaecology clinic will already have been
treated with non-steroidal anti-inflammatory drugs (NSAIDs) in primary care.
Although there is some evidence that these drugs reduce endometriosis-related
pain (Kauppila et al., 1979), the majority of women presenting to the
gynaecologist in secondary care will report little benefit from this therapy.
In the first half of the twentieth century the treatment and diagnosis of
endometriosis had been surgical. In this era it became apparent to some
observers, like Meigs, that there was a lower rate of pregnancy amongst
endometriosis sufferers. He theorised that pregnancy itself was a prophylaxis or
even treatment for endometriosis (Meigs, 1922). From this observation there
grew attempts to hormonally create a “pseudo-pregnancy state” as a treatment
(Kistner, 1958). This is the basis for the use of progestogens and combined oral
contraceptives (COCP) in medical therapy.
Similarly, it was noted that endometriosis is a disease of predominantly pre-
menopausal women. Therefore castration was recommended as a therapy to
remove the oestrogen source driving the disease (Cattell R., 1936, Fallon,
1946). From this developed the theory of creating a “pseudo-menopausal state”
to treat endometriosis, and this forms the other cornerstone of current medical

16
therapy. Danazol and gonadotrophin-releasing hormone agonists (GnRH
agonists) are employed along these lines (Audebert et al., 1977, Shaw, 1992).
Cochrane reviews suggest that there is no benefit, in terms of pain relief, in
favour of the pseudo-pregnancy over the pseudo-menopausal treatment
strategy (Moore, 2004, Selak, 2004). All approaches appear equally efficacious
and provide benefit for up to six months following the cessation of treatment.
Consequently, the choice of drug will be driven more by the side effect profile
and contraceptive requirements of the women. The levonorgestrel intrauterine
system (LNG-IUS) also appears to be of equal benefit as well (Vercellini et al.,
1999a, Petta et al., 2005).
The optimum duration of GnRH therapy is unclear. Extending GnRH therapy
beyond three months up to six months does not seem to confer greater benefit
in terms of sustained pain relief (Hornstein et al., 1995). However, extension of
treatment up to two years appears to be safe, in terms of bone protection, if
add-back hormone replacement therapy is used (Surrey et al., 2002). Also,
GnRH analogues used as a post operative adjunct to surgical treatment delay
the recurrence of pain compared with expectant management after surgery
(Vercellini et al., 1999b).
In terms of actual disease regression, medical therapies have been shown to
reduce the extent of disease found at laparoscopy (Telimaa et al., 1987, Fedele
et al., 1989, Cedars et al., 1990), but not to eradicate it entirely.
The European Society of Human Reproduction and Embryology (ESHRE)
guidelines for the management of endometriosis, suggest that it is good clinical

17
practice to use counselling, analgaesia and nutritional therapy combined with
progestogens, the COCP or GnRH analogues, as an empirical treatment for
pelvic pain presumed to be endometriosis (Kennedy, 2004). If a laparoscopy is
carried out, ESHRE guidelines then logically recommend that, as the gold
standard for diagnosis is laparoscopy, ideal clinical practice should be to
surgically remove endometriosis at the same time. Surgery is the only means of
ensuring complete removal of visible disease.
Whilst a laparoscopy undoubtedly carries risks, endometriosis is a progressive
disease and delaying surgical intervention might lead to a greater risk of
compromised fertility as well as pain although there is no direct evidence of this.
Most women have already tried analgaesia and hormonal intervention in
primary care, so in such cases ideally an operative laparoscopy should be
carried out as first line treatment if symptoms have been sufficient enough to
warrant referral to a gynaecologist.
As understanding of the pathophysiology behind endometriosis improves, novel
medical therapies are emerging. These include GnRH antagonists (Kupker et
al., 2002), mifepristone (RU486) (Kettel et al., 1996), TNFα inhibitors (Falconer
et al., 2006), angiogenesis inhibitors (Becker and D'Amato, 2007), matrix
metalloproteinase inhibitors (Osteen et al., 2003), pentoxifylline (Creus et al.,
2008) and aromatase inhibitors (Nawathe et al., 2008). These agents are
mostly in the experimental stage of use.
Where pelvic clearance is performed, and hormone replacement therapy (HRT)
required, then a theoretical benefit exists for including progesterone in the form

18
of combined HRT. Firstly, the addition of progesterone may prevent recurrence
of endometriosis. Secondly, a lack of protective progesterone may result in
endometrial carcinoma arising from endometriotic lesions after total abdominal
hysterectomy and bilateral salpingo-oophorectomy (Ulrich, 2005). However, this
has to be weighed against the possible increased risk of breast cancer in
combined versus Oestrogen only HRT (Beral, 2003).
Measuring outcome
Second look laparoscopy to assess the state of the pelvis requires another
surgical procedure that is most likely unnecessary, and does not give
information on improvement in pain and quality of life issues. Therefore other
measures are required to assess improvement peri-operatively. Visual
analogue scores for pain, quality of life measures, mental state and sexual
function questionnaires have been used. There is an endometriosis-specific
validated questionnaire, called the Oxford EHP30 questionnaire (Jones et al.,
2001), that assesses the impact of endometriosis on different aspects of life.
Multidisciplinary approach
Pelvic pain is not exclusive to endometriosis and there is a degree of crossover
with other conditions like interstitial cystitis and irritable bowel syndrome. There
is also a degree of psychological morbidity. Specific pelvic pain clinics that
adopt a multidisciplinary approach encompassing a wider range of pathologies
and treatments are more likely to offer improved support and relief to women
(Metzger, 1997). They can also develop links with infertility clinics, colorectal
surgeons, urologists, anaesthetist-run pain clinics, psychologists and patient
support groups. Multidisciplinary pain clinics have been shown to be effective in

19
the management of other chronic conditions such as back pain (Guzman et al.,
2001). Endometriosis UK, the patient support group, now helps run a
Department of Health sponsored Expert Patient Programme to teach women
how to live with their symptoms (www.expertpatients.co.uk).
A critical review of the surgical treatment for endometriosis
There are no clinical trials directly comparing surgical and medical treatment for
endometriosis. As no direct comparisons are available we must rely on other
evidence to weigh up the pros and cons of each approach.
Unlike medical therapy, surgery can diagnose and remove all macroscopic
disease at the same procedure in the majority of cases. Even severe
macroscopic disease can be entirely removed surgically. Remember also that
surgery is the gold standard for diagnosis of most endometriosis. Surgical
therapy compared to placebo can result in a continuing positive effect on pain at
6 months after surgery (Sutton et al., 1994, Abbott et al., 2004). Both authors
also have data suggestive of this effect lasting even longer (Sutton et al.,
1997b, Abbott et al., 2003). Treatment at laparoscopy seems to also improve
rates of spontaneous conception for infertility associated with endometriosis
(Marcoux et al., 1997). The evidence for this is analysed in greater detail below.
The development of techniques to remove endometriosis surgically has been
governed by the development of energy sources used in surgery. Most
surgeons would regard complete excision or complete vaporisation as the
preferred techniques for removing endometriosis surgically. These remove the

20
active lesion and the underlying fibrosis, both of which are considered to have a
causal link to pain. Only normal tissue remains. Superficial electro-coagulation
with diathermy may destroy the active lesion but may also leave the treated
tissue and the underlying fibrosis. Applying prolonged diathermy may also
increase the risk of collateral damage.
Cold scissors may be used to excise endometriosis. However, most surgeons
prefer an energy source that gives them a degree of haemostasis as well as the
precision of cold scissors. Electrical energy is most effectively used in its
monopolar form to excise endometriosis. For example, 3mm monopolar
scissors used at a high power of 90 Watts are a precise means of excision
(Redwine, 1993). Electro-surgical energy essentially cuts and coagulates by
applying high levels of heat to the target area. There is a potential of collateral
damage by heat spread proximal to the target area, insulation failure, direct
coupling or capacitive coupling. Bipolar energy is not a useful energy source for
effective excision or vaporisation.
Several modified electro-surgical devices have been developed to reduce the
amount of electrical energy required to cut or coagulate. These devices include
the argon beam coagulator (Daniell et al., 1994) and Helica Thermal Coagulator
(Nardo et al., 2005). Both have been used for the coagulation of superficial
deposits of endometriosis.
Ultrasound generated energy is also a precise means of excision with
simultaneous haemostasis. Ultrasound energy is converted to high-speed
motion in an active blade. This cuts and causes haemostasis, by coaptation

21
more than coagulation, at a lower temperature than electrosurgical energy.
Consequently there is a trend towards less heat damage to adjacent tissues in
human and animal studies (Awadzi, 2005, Meltzer, 1994). Also, ultrasound
energy devices are free from the risks of insulation failure, direct coupling and
capacitive coupling as no electrical energy passes down the instruments.
Haemostasis is good but not as effective as electro-surgery.
Lasers can be used for either the excision or vaporisation of endometriosis
(Bruhat et al., 1989, Sutton, 1989). The common lasers employ carbon dioxide
(CO2 laser) or yttrium aluminium garnet modified with potassium tinanyl
phosphate (KTP laser) to create a highly focused visible light beam that cuts
tissue precisely by heat. Excision is precise and quick however haemostasis is
not as good as electrical or ultrasound energy. Vaporisation is achieved by
moving the beam around over the target area until only an underlying area of
normal tissue remains. With the Swiftlase function on Sharplan CO2 lasers, the
beam is automatically rotated around the aiming point, making vaporisation
easier.
The choice of energy source, instrument and technique is probably controlled
by various factors, including what the surgeon is comfortable with, what they
believe to be best, what they have available, whether the instruments are safe,
and the cost. However, there exists no data on why surgeons use which
technique or which instrument.

22
In addition to different techniques, there are three distinct areas of surgical
treatment for endometriosis: minimal to moderate endometriosis,
endometriomas, and lastly, recto-vaginal nodules.
Surgical treatment of peritoneal endometriosis for Infertility
As yet there has been no direct causal link established between sub-fertility and
endometriosis (Vercellini et al., 2009). According to Hughes Cochrane review of
the subject (Hughes et al., 2003) there is no evidence to suggest that medical
therapy with hormonal drugs is beneficial for women with endometriosis-
associated infertility. Hughes, in a separate paper reviewing 5 cohort studies
and a quasi-randomised study, suggests that laparoscopy is superior to
Danazol by an odds ratio of 2.7 in terms of pregnancy incidence (Hughes et al.,
1993). Evidence does exist to suggest that surgical treatment may be of benefit
in all stages of the disease including endometriomas (Jacobson et al., 2004b,
Adamson et al., 1993, Guzick et al., 1997, Osuga et al., 2002).
Jacobson’s Cochrane meta-analysis of minimal to mild endometriosis,
“Laparoscopic surgery for sub-fertility associated with endometriosis” (Jacobson
et al., 2004b) since updated in 2009 with no significant changes, includes the
EndoCan trial of 341 patients (Marcoux et al., 1997), which shows a beneficial
effect for laparoscopic treatment, and the Italian Group trial which shows no
significant difference and a slight negative effect on live births
(dell'Endometriosi, 1999). The meta-analysis of the two trials shows an
improvement in fecundity with surgical therapy with an odds ratio 1.6 (95%CI
1.05 to 2.57) of pregnancy progressing beyond 20 weeks gestation or live birth.

23
Jacobson has commented that this result should be treated with caution, as the
two results are contradictory.
On further analysing the two trials we find that both had strict eligibility criteria at
the outset. Both had power calculations (although the Italian Group trial ended
up being underpowered as it was attempting to show the 2.7 odds ratio
suggested by Hughes above). The entry criteria for the Italian Group trial were
two years of infertility, versus one year for the EndoCan trial. The Italian Group
subjects probably reflect a group of patients with poorer prognosis at the outset
as a result. There were also more stage 2 cases of disease in the Italian Group
that may have disadvantaged it. The eligibility for the EndoCan trial restricted
the visual diagnosis of endometriosis to blue or black lesions only and so
missed a whole cohort of other appearances. Neither trial confirmed the
presence of endometriosis with histology and so the possibility of false positive
diagnoses exists. Surgical technique is not clearly explained in either trial and
with the number of centres and surgeons involved there is a distinct possibility
that the quality of treatment between the two trials was not consistent. Neither
trial was blinded, so there were potential performance biases involved in that
patient knowledge of which treatment had been received may have affected the
sexual behaviour of participants subsequent to surgery. The two trials had
different follow-up periods: 9 months for the EndoCan trial and 12 months for
the Italian Group. However, about half of the Italian Group patients had 3
months of adjuvant GnRH analogue treatment after surgery, meaning that there
were also 9 months available to them to conceive in half of the cases. This
addition of adjuvant therapy also created new subgroups of treatment though
no significant differences were shown in any of them in the analysis.

24
On balance there appears to be some positive effect on fertility with
laparoscopic surgery for treating endometriosis but it may be less substantial
than originally suggested and recent reviews appear to be downplaying the
positive benefits (Vercellini et al., 2009). In addition, the EndoCan group
calculated from their data that one in eight women should benefit from
laparoscopic treatment of endometriosis. In Jacobson’s meta-analysis
(Jacobson et al., 2004b), the suggestion was that the number needed to treat
for one additional ongoing pregnancy beyond 20 weeks lies somewhere
between 3 and 100. Therefore, this evidence does not clearly indicate that all
women suffering from infertility should undergo laparoscopy to purely look for
endometriosis alone, as potentially only 1 in 8 of the 20% to 68% likely to have
infertility associated with the presence of endometriosis are likely to benefit
(1994, Koninckx et al., 1991, Mahmood and Templeton, 1991, Matorras et al.,
1995). However, in the absence of gamete problems, laparoscopic treatment of
endometriosis can be recommended for endometriosis found at laparoscopy in
association with infertility, especially if it can be removed at the same time as
diagnosis.
Endometriomas
Medical treatment has shown a significant reduction in the size of
endometriomas (Schenken, 1990, Batioglu et al., 1996) but there remains the
risk of ovarian torsion, a small risk of malignancy, continued pain and the effect
on fertility to consider. Surgery had already been touted as beneficial for fertility
as far back as 1957 (Fredrikson, 1957). Furthermore, in modern times
laparoscopy has not been shown to be disadvantageous compared with
laparotomy, as shown by the many reports that considered this during the rise

25
of laparoscopy in the 1990s (Daniell et al., 1991, Donnez et al., 1996, Adamson
et al., 1992, Sutton et al., 1997a). Research then moved on to consider the
technique that should be used to best treat endometriomas considering the
outcomes in respect to pain, recurrence, reoperation, ovarian function and
fecundity. It became clear that fenestration alone was inadequate (Vercellini et
al., 1992) and so the front runners became either straight excision of the
pseudo-cyst capsule or some combination of fenestration and coagulation
involving varying energy sources and the possibility of staged procedures with
adjuvant GnRH analogue therapy. Energy sources for ablation have included
electro-cautery (Beretta et al., 1998, Alborzi et al., 2004b), CO2 laser (Donnez
et al., 1996) and KTP laser (Sutton and Jones, 2002). The two-staged
procedure allows for initial drainage, followed by down regulation for several
months with GNRH analogue, to then permit definitive coagulation of the
capsule and treatment of peritoneal lesions (Donnez et al., 1994).
Randomised evidence appeared in three trials since 1998 (Beretta et al., 1998,
Alborzi et al., 2004b, Alborzi et al., 2007b) that were considered together by
Hart in his Cochrane review (Hart et al., 2008). On reading the methodologies
for the 1998 and 2004 trials by two separate authors they appear to be very
much based on each other, with some small variations. Fecundity seemed to be
the primary outcome variable though this was not stated in either trial and
neither was a power calculation done. Both trials appeared to have rigorous
eligibility for infertility. It should also be noted that in neither trial were patients
blinded from the treatment, and the investigators did not appear to be blinded
from follow-up either. Pain outcomes were only judged by a single VAS for
varying types of pain and were dichotomised into levels of severity. It is not

26
clear from either paper how recurrence of pain was derived and objective pain
results are not described. In both trials fertility was judged by positive
pregnancy on ultrasound, and not by pregnancy outcome. Recurrence of
disease was judged by the recurrence of probable endometriomas greater than
3cm diameter on ultrasound. Pain is a subjective measure, and the lack of
blinding of patients is particularly concerning in this respect despite excision
resulting in a significant reduction in recurrent pain. That being said, the
evidence for fertility improvement and reduced recurrence with excision is
strong in Hart’s meta-analysis: the spontaneous conception rate was OR 5.21,
95%CI 2.04 to 13.29, and the reduced recurrence rate of OR 0.41, 95%CI 0.47
to 4.15. The combination of evidence was compelling in favour of excision when
electro-surgery is used for coagulation. With KTP laser there is a suggestion
that pregnancy rates may be more comparable with excision (57% for KTP
versus 59% in Alborzi’s trial) (Jones and Sutton, 2002, Alborzi et al., 2004a).
This has some logical credibility, though the evidence is not strong enough to
draw firm conclusions, as the KTP laser penetrates well on the wet surface of
an endometrioma, whereas CO2 laser penetration is stopped at a liquid
interface, suggesting that the KTP may be more effective at destroying the
pseudo-cyst.
More recently research has shifted towards concern regarding damage caused
by excision that might compromise ovarian blood supply, ovarian reserve and
the potential success of assisted conception techniques. Consequently Canis
has suggested a combination of excising the outer part, and ablating the deep
inner part of the capsule proximal to the hilum (Canis et al., 2003), but there is
no evidence yet for this approach and coagulating the hilar area may potentially

27
cause more damage than excision of the capsule from this area. Current
research is beginning to be directed towards Anti-Mullerian Hormone as a
marker for ovarian reserve and variations that may be found between excision
and ablation (Lemos et al., 2008). All this being said, the Cochrane evidence
still supports the evidence that higher spontaneous pregnancy rates result from
excision, and even that the follicular response of ovaries to GnRH analogue
after excision is higher for assisted conception (Alborzi et al., 2007a). However
there is no evidence of improved outcomes of IVF following excision of
endometriomas (Garcia-Velasco and Arici, 2004). There probably does remain
a small risk of significant ovarian damage and potential ovarian failure from
over-aggressive excision treatment with a rate of up to 2.6% in a retrospective
review by Busacca (Busacca et al., 2006).
Recto-vaginal nodules
These cases present the ultimate surgical challenge for gynaecological
laparoscopists in centres specialising in the treatment of endometriosis. The
presence of a recto-vaginal nodule is not always obvious, either on recto-
vaginal examination or at laparoscopy. Deep infiltrating disease is under-
diagnosed as it may appear to be minimal disease if one only assesses its
surface appearance (Koninckx et al., 1994). Usually there is obliteration of the
Pouch of Douglas to a greater or lesser extent that can be easily missed by
inexperienced eyes.
Radical excision of recto-vaginal disease was known to carry a high morbidity
and mortality and so was still regarded as less preferential to menstrual
suppression in the mid part of the 20th century. However, evidence of bowel

28
resection being used by some surgeons does exist coming into the 1970s
(McSwain et al., 1974, Gray, 1973, Cromer, 1967). Weed published his series
of 163 cases of resection of bowel endometriosis at laparotomy in 1987, as
laparoscopy began to develop (Weed and Ray, 1987). It is interesting that this
series, and the 72 women undergoing bowel resection in Coronado’s series in
1990 (Coronado et al., 1990), made no report of significant morbidity.
Crosignani and Vercellini reviewed the relative benefits of laparoscopy and
laparotomy in 1995 and found no difference between the two in terms of
endometriosis outcome suggesting that technological advances in laparoscopy
were already advanced enough to allow comparable results for even this
complex area of surgery (Crosignani and Vercellini, 1995). Also, the benefits of
laparoscopy over laparotomy in terms of recovery, pain and hospital stay were
already becoming apparent (Luciano et al., 1992).
The literature then continues in a procession of case series reports to the
present day, here reviewed in a table constructed by Wright and presented at
the ESGE Annual Meeting in Florence in 2009:

29
Table 1. A summary of case series for the treatment of recto-vaginal endometriosis
Authors
Date and Number of cases
Reported Bowel
fistulae/leaks n (%)
95% confidence limits
McSwain
1974/14
0
0-0.215
Weed
1987/53
0
0-0.068
Chen
1989/2
0
0-0.658
Coronado
1990/76
0
0-0.048
Redwine
1999/5
0
0-0.053
Nezhatz
1986/22
0
0-0.068
Bailey
1994/130
0
0-0.029
Donnez
1995/231
0
0-0.016
Koninckx
1994/285
1 (2.85)
0.001-0.02
Jerby
1999/12
1 (8)
0.015-0.354
Verspyck
1997/6
1 (16)
0.030-0.564
English
2004/100
4 (2.5)
0.016-0.098
Darai
2007/71
6 (8.5)
0.039-0.172
Keckstein
2005/202
6 (2.9)
0.014-0.063
Slack
2007/32
3 (9.3)
0.032-0.242
Waters
60
5 (8.3)
0.036-0.181
Possover
2000/34
0
0-0.102
Total
1335
28 (0.21)
0.015-0.03
(Bailey et al., 1994, Chen et al., 1989, Coronado et al., 1990, Darai et al., 2007, Donnez and
Nisolle, 1995, Ford et al., 2004, Jerby et al., 1999, Keckstein and Wiesinger, 2005, Koninckx
and Martin, 1994, McSwain et al., 1974, Nezhat et al., 1994, Possover et al., 2000, Redwine et
al., 1996, Slack et al., 2007, Verspyck et al., 1997, Weed and Ray, 1987).
This review by Wright considers a total of 948 cases of endometriosis
undergoing surgical treatment for their disease. The overall complication rate
was calculated to be 1.7% (95%CI 1 to 2.7). However, the more recent series in
table 1 above seem to be showing up a higher frequency of bowel leaks and
fistulas. This is an interesting finding that may reflect the fact that complication
data was previously under-reported, or that the complexity of cases attempted
has increased. Women presenting with recto-vaginal nodules need to be
carefully counselled and prepared if surgery is being considered, as there now
appears to be a genuine 4-10% risk of major complications associated with
sigmoid and rectal surgery that is confirmed in the experience of general

30
surgeons treating rectal cancer too (Canis, 2005, kenney, 2005). Acutely, these
include anastamotic leaks, fistulas and urinary system damage as well as the
chronic bowel morbidity resulting from strictures and short bowel syndrome. It
appears that these strictures may be more common in endometriosis patients
than in those undergoing rectal resection for rectal cancer (Waters N, 20008)
representing possibly the underlying fibrotic nature of endometriosis. Despite
Dubernard reporting an improvement in all items of SF-36 for a series of 58
colorectal resections of endometriosis, the major complication rate was 15.5%
and tenesmus, constipation and bowel frequency were not improved in the long
term (Dubernard et al., 2006).
Surgically, a combined approach with a laparoscopic colorectal surgeon may be
undertaken. The sigmoid colon and rectum are dissected away from the
ureters, pelvic side-walls, uterus, cervix and vagina. Following this the
endometriotic nodule or nodules are removed from the bowel. In the most
conservative technique this is done by shaving the nodule off the bowel,
thereby avoiding perforation of the bowel mucosa. However, if entering the
bowel is necessary to remove the disease a disc resection may be performed
that is subsequently closed by primary laparoscopic suturing. In cases where a
nodule is deeply infiltrating over a larger area, and it is considered that disc
resection could result in stricture, anterior or segmental resection may be
carried out. Multiple sites of disease, that would otherwise require multiple disc
resections, may also be treated by segmental resection. Recommending
surgery as the main primary treatment is by no means the only option. With the
risk of complications, patients may prefer to consider long-term medical therapy

31
instead. Or, if pregnancy is the main requirement, then IVF may be preferential
as a first line treatment.
We do not know which is the optimum technique to surgically remove recto-
vaginal disease. There are no RCTs and it would be very difficult to produce a
robust trial to investigate this. Some surgeons advocate segmental resection in
all cases as they believe that it is the only way to remove all disease, especially
as there is evidence to suggest that up to 68% of patients have multiple lesions
(Keckstein, 1999). Others believe that there is gene-profiling evidence emerging
to suggest that the nodules are cervical in origin, and so surgery should be
more radical on the cervico-vaginal side and less radical on the rectal one in
order to decrease bowel complications and risk of recurrence (Van
Langendonckt, 2004). Overall success rates of surgery for recto-vaginal
nodules in terms of pain relief are in the region of 85-95% with recurrence rates
of 5-15% (Canis, 2005) though the positive effects on fertility are less clear.
Pain recurrence was estimated at 28% in Dubernard’s series despite colonic
resection (Dubernard et al., 2006). In addition to this, Abbott showed in his 2-5
year follow up of women undergoing laparoscopic excision for all stages of
endometriosis that the risk of requiring subsequent surgery was 33%, estimated
at 36% over 5 years, and that women with deep infiltrating stage 4 disease
were significantly more likely to be in this group showing that their risk of re-
operation was even higher (Abbott et al., 2003).
In summary, despite favourable improvement in pain symptoms after radical
surgery, the acute complication risk, long-term bowel morbidity and the risk of
recurrence tend to favour a more “conservative” radical approach, whereby as

32
much endometriosis as possible is removed with the least risk of sustaining
bowel complications. The level of pain improvement has not been shown to be
any worse for a more “conservative” approach. This seems particularly
important in view of the fact that endometriosis is not lethal and has a
substantial likelihood of recurrence despite how aggressive the resection is. In
addition to this, the progress in robotic surgical techniques, allowing state-of-
the-art 3D vision and unparalleled precision, may allow for more accurate
excision of recto-vaginal endometriosis with reduced bowel morbidity (Magrina,
2007).
Surgical treatment of minimal to moderate endometriosis for pelvic pain
In the laparoscopic era of surgery for endometriosis, various techniques for
operating on minimal to moderate endometriosis developed side by side, but
can broadly be put into the categories of excision, vaporisation or
ablation/coagulation. Excision is in a clear category of its own with no doubt
over the technique of staying in normal tissue and excising around the
suspected lesion. Ablation and coagulation are terms that have been used
frequently to describe the application of heat energy to a lesion to result in its
destruction. These terms reflect a wide range of resulting effects from just
applying heat to the surface of a lesion leaving a charred area of tissue that
probably represents sub optimal treatment, through to full destruction of tissue
resulting is an appearance of normal tissue similar to excision. For that reason
the use of the word “vaporisation” is perhaps more appropriate to describe the
technique where energy is applied to a lesion where the outcome appears
similar to excision. These varying approaches were reported in various case

33
series throughout the 80s and 90s (Redwine, 1996, Candiani et al., 1986,
Lomano, 1987, Martin and Vander Zwagg, 1987, Redwine, 1991).
Sutton then produced the first RCT looking at the surgical treatment of minimal
to moderate endometriosis for relief of pelvic pain in 1994 with a series of 63
patients (Sutton et al., 1994). Patients underwent laser vaporisation of all
lesions with or without the addition of adhesiolysis and utero sacral nerve
transection or had diagnostic laparoscopy alone. Patients completed VAS pain
scores based on their worst symptom from dysmenorrhoea, dyspareunia and
pelvic pain (which was dysmenorrhoea in all cases), at 0, 3 and 6 months.
Scores were not significantly different at 3 months but became significant at 6
months with 62.5% of the treatment group improving. There was no power
calculation included in the report, no information on allocation concealment and
no data on quality of life, yet this gave the first substantial evidence that surgical
treatment was effective. However, it has subsequently been established that
there is no evidence that laparoscopic uterine nerve ablation (LUNA) by itself or
combined with surgical removal of endometriosis is of benefit for women with
pelvic pain and endometriosis (Vercellini et al., 2003, Johnson et al., 2004,
2003, Daniels, 2009).
However, many gynaecologists continued with excision, presumably because
they felt they were achieving the same effect, if not better, and were also saving
the substantial cost of investing in a laser. More critical evidence for the effect of
excision, including quality of life data, did not really arrive until Abbott’s paper of
2003, prospectively looking at the outcome of patients for pain and quality of life
at 2-5 years post laparoscopic surgical excision. This was a more substantial

34
look at the evidence than their group’s previous report by Garry in 2000 (Garry
et al., 2000). The findings showed that the most common symptom at
presentation was non-menstrual pain (74%), 67% were improved (though a
disturbing 25% were worse), there was a significant risk of re-operation (33%),
analgaesia was still required for 35%, and hormonal therapy in 26%. Median
pain scores were significantly improved for all pain modalities and quality of life
was significantly improved though not to normal population levels.
Dysmenorrhoea remained the most common symptom at follow up. It must be
noted that this sample of women appears to be skewed towards more
advanced disease (41% had stage 4 disease) suggesting that these outcomes
are those one may expect in a tertiary referral practice and not necessarily the
general population of endometriosis sufferers.
This prospective study was followed up with the first RCT for excision therapy
by the same group published in 2004 (Abbott et al., 2004). This included 39
women (just missing their power calculation of 40), again with all stages of
disease that seems once more to be biased towards more complex disease
compared with the general endometriosis population, with 17/39 (43.6%) of
patients having stage 4 disease, in what was presumably a tertiary referral
practice. Comparative data for excision versus control is available at 6 months,
as in the Sutton trial, and showed that 80% (versus 32% in the control group) of
patients in the immediate surgery group had some form of pain improvement,
with a median improvement of 30 points on a 100-point VAS of overall pain
improvement. Interestingly, for individual pain modalities of dysmenorrhoea,
dyspareunia, non-menstrual pain and dyschezia, there was no significant
difference in direct score comparison between excision and controls groups at 6

35
months. However, within each group, excision significantly improved for all 4
modalities and the controls for dysmenorrhoea and CPP only. The placebo
effect noted by Sutton at 3 months appears to extend, to some extent, to 6
months in the Abbott trial. That the comparatively significant improvement in
overall VAS for pain did not correlate with the individual pain modalities is
disappointing and casts some doubt on the overall result at 6 months. The
authors themselves call into question the potential limitations of VAS pain
scores alone to interpret outcomes as they are probably subject to significant
intra-observer error. Ideally, additional validated outcome measure instruments
need to be used looking at HRQoL as was done in Abbot’s trial with EQ-5D and
SF-12. This trial raised the question of whether 6 months follow up was really
long enough to pick up a significant difference as suggested by Sutton’s trial.
Sutton’s follow up report on his trial at 12 months also suggested that 90% of
those who responded at 6 months had continued symptom relief, although the
blinding was now broken (Sutton et al., 1997b). 12 months of direct comparative
follow up was the logical next step.
Jarrell et al attempted to look at 12 month follow up and simultaneously
overcome the VAS problem above by taking repeated daily VAS pain scores for
a month pre-operatively and at 3 monthly intervals up to one year (Jarrell et al.,
2005). Sadly only 16/29 women completed the rigorous follow up protocol,
resulting in a significantly underpowered trial. The other two potential trials, by
Tutunaru and Lachlandani, considered by Jacobson in his updated Cochrane
review (Jacobson et al., 2009) were data presented from conferences, are
unpublished in full form, and should be treated with caution. The meta-analysis
of the advantage of laparoscopic treatment versus diagnostic laparoscopy alone
at 6 months suggests an odds ratio of 5.72 (95%CI 3.09 to 10.60).

36
These trials did not attempt to compare excision and vaporisation, the two
leading forms of surgical treatment for minimal to moderate endometriosis. Only
one published trial of 24 women exists from Wright, comparing excision and
ablation with monopolar electro-surgery for superficial minimal to mild
endometriosis (Wright et al., 2005). A significant improvement was found for
symptoms and signs in both groups but no significant difference between the
two groups at 6 months. This trial was small and confined to only rASRM stage
1 and 2 superficial disease, and perhaps the sensitivity of the 1 to 5 ordinal
scale used is limited in being able to differentiate between treatments.
There is currently insufficient evidence to differentiate between excision and
vaporisation in the treatment of minimal to moderate endometriosis. It would
seem logical to assume that, provided all endometriosis and fibrosis is removed
leaving only normal tissue, then there should be no difference between the two.
Conclusion
Endometriosis is almost certainly under diagnosed and under treated and much
more research is needed. The development of the World Endometriosis Society
(WES), the availability of Internet information from quality websites like
endometriosis.org, and the rapid development of patient-support groups are all
helping to raise the profile of the disease. This in turn will hopefully lead to a
better understanding of the aetiology and effectiveness of treatments. Ultimately
however, it will be informed patients demanding high quality evidence-based
treatment that will drive progress. The British Society of Gynaecological
Endoscopy (BSGE) is also developing a national treatment database and
network of accredited centres with the aim of improving standards of treatment

37
and collecting data for research. Ten centres are currently accredited on the
BSGE website (www.bsge.org.uk).
150 years after the disease was first described, we are still debating its
aetiology. The rise of laparoscopic surgery has raised the profile of the disease
over the past 30 years and this has led to a corresponding increase in our
understanding of the underlying processes. As the evidence currently stands,
laparoscopic surgery appears to be the most logical approach to treatment
provided women accept the risks of surgery especially for complex disease.
There is little evidence to support the use of one technique or instrument over
another, and there are insufficient centres and surgeons capable of dealing with
the problem. This thesis will hopefully be able to assess more clearly the
currents trends in practice for all forms of the disease, and perhaps throw more
light upon whether one technique or other is advantageous for minimal to
moderate disease.

38
Chapter 3 - Methodology
An International Survey of Surgical Techniques used in the
Treatment of Endometriosis
Introduction
When the patient comes to the operating theatre there are many factors that
have influenced the procedure that they will actually undergo. These include:
1. Factors related to the surgeon themselves including their training,
experience and character.
2. Factors related to the evidence for the procedure including the
medical literature and guidelines.
3. Factors related to economics and availability of instrumentation
pertaining to the hospital in which the surgery is being carried out.
This complex inter-relationship is often not developed in a systematic way and
instead is the result of a series of coincidences influenced in varying degrees by
the factors listed above. Dealing with each of these factors in turn we can see
more clearly the complexity of the issue.
Part of why surgeons choose to use different instruments or techniques will be
based upon their own character. Some may be more risk averse and more
inclined to carry out less invasive and radical techniques than others who are
more aggressive in nature. Some may be more driven by ambition and

39
discovery and inclined to explore and develop their own new techniques that
result in a change in widespread practice. The training that a surgeon has
received will vary considerably from a standard-type training path to a specialist
Fellowship or postgraduate degree. Their theoretical and practical ability will
vary widely as a result. The influence of the teacher in this process is likely to
be significant in affecting the method that the surgeon ultimately applies on his
or her own patient. Experienced surgeons are likely to be more confident and
may be more likely to be more aggressive and innovative, or conversely may be
stuck in their ways in outdated techniques.
The medical literature for the surgical treatment has been outlined in detail in
Chapter 2 but in summary suggests that the gold standard for the diagnosis of
endometriosis is laparoscopy (Kennedy et al., 2005). This is not necessarily
required if medical therapy controls pain symptoms, fertility is not an issue at
that moment, and the woman is comfortable without a definitive diagnosis. This
guideline was developed on the basis of RCT evidence summarized in the
Cochrane database (Hart et al., 2008, Jacobson et al., 2009, Jacobson et al.,
2004b) as well as large numbers of non-randomised studies. There is no clear
evidence that one technique or instrument is better than any other, and so
factors 1 and 3 above play a larger part in influencing what happens to the
patient on the operating table than they would if there was a definitive
procedure with overwhelming evidence supporting it.
Economical issues in health care play a large role in determining what is
available to the surgeon. There are many instruments capable of carrying out
excision, vaporisation or ablation of endometriosis with varying costs and

40
characteristics. Less economically restricted hospitals, with visionary chief
executives may be more likely to invest in the latest developments. Others will
wait until there is more evidence, or even till there is no choice once the
evidence for a procedure is overwhelmingly in favour of it.
This study looks at which instruments and techniques are currently being used
and why. It is therefore an exploratory study to attempt to understand more
about surgeons choices in instrument and technique in relation to the three
factors listed above.
Study design
The survey was administered and designed to collect reliable, valid and
unbiased data from a representative sample, in a timely manner and within the
given resource constraints.
In questionnaire design, certain methods have been identified that have been
shown to help achieve this (Boynton, 2004, R Nakash, 2006). For
administration these include:
1. Saliency of the questionnaire to the responder
2. Linguistic ability of the responders
3. Efficient means to complete the questionnaire
4. Will to complete the questionnaire
5. Clear aim of survey conveyed to responders by covering letter
6. Pre notification
7. Reminders

41
8. Use of incentives
9. Available researcher to answer questions
10. Pilot study
11. Efficient data entry and cleaning
For design these include:
1. Clear design and layout
2. Visually appealing
3. Short questionnaires
4. Question wording
5. Question ordering
6. Question form
Objectives
Primary endpoints
To discover the type of surgical techniques and instruments used in treating
different types of endometriosis, including superficial minimal to moderate
disease, deep minimal to moderate disease, endometriomas and recto-vaginal
nodules, within the study population.
Secondary endpoints
1.The characteristics of the laparoscopic surgeons using these techniques and
instruments defined by region, experience and training.
2.The geographical distribution of the use of these techniques and instruments,
and why the surgeons are choosing them.

42
Population
The survey was sent to members of the British Society of Gynaecological
Endoscopy (BSGE), European Society of Gynaecological Endoscopy (ESGE),
and the American Association of Gynecologic Laparoscopists (AAGL). The
Australian Gynaecological Endoscopy Society (AGES) was approached but
declined to participate for data protection reasons. These societies had email
addresses for 4878 medical members in total.
Type of survey
This was an international web-based self-completed survey with a large target
population of nearly 5000. Clearly a self-administered questionnaire was
preferable to an interviewer-administered approach for numerical and
geographical reasons. An electronic web-based rather than postal distribution
was chosen for several reasons:
1. An electronic survey would be less expensive, saving on postal costs.
2. Answering is controlled, by stopping responders from moving on until
they have completed certain questions.
3. Out of range answers are prevented from being entered in error.
4. Response collation and transfer to a statistical database is electronic.
5. A sampling frame existed, as each society maintained a member
email address database.
6. Electronic surveys are easy to distribute with less risk of postal failure
provided that email addresses are accurate.
7. Electronic surveys are efficient to complete in the presence of a good
Internet connection.

43
8. Responders, being doctors in mainly developed countries, were likely
to have easy access to computer and Internet facilities.
Survey Administration
Covering letter and invitation to participate
Each Society agreed to distribute the survey by an invitation covering letter,
stating the clear aim of the survey, with an electronic link into the questionnaire.
This confirmed the bona fide credentials of the survey, and was presented with
headings and logos. It also guaranteed confidentiality. The BSGE and ESGE
agreed to do this by an individual email invitation to each member on their
databases. The AAGL would only agree to do this by including the invitation in
the monthly newsletter. No other pre-notification technique was used. A copy of
the invitation letter for the main survey is at appendix K.
Use of incentives
All responders email addresses were included into a post-survey raffle. The
winner received an iPod Nano. This was stated clearly on the invitation email.
Available researcher to answer questions
On the bottom of each questionnaire page the researcher’s email address was
included so that responders could ask questions.
Pilot Study
A pilot study was sent to a representative sample of 12 gynaecologists from the
main study population. They were also sent a covering invitation email directly
to their address, which allowed them to click on an electronic link into the pilot

44
questionnaire. The results of the pilot study were analysed to assess issues
regarding layout and design, non-responses due to poor language or
signposting, relevance, ambiguities and its ability to hold interest. An open free
text box was included for responders to comment on the survey. A copy of the
pilot invitation and pilot questionnaire are at appendices J and L.
Data handling and record keeping
All data responses were transferred to a secure central database operated by
the commercial web survey system until the survey was closed. This was then
transferred electronically into an Excel database. From there, the Excel
database was copied manually into SPSS for further analysis. The SPSS
database was held on a password protected laptop. No paper records of
responders’ personal details were kept. All data was handled in accordance
with data protection legislation.
Reminders
Both the BSGE and ESGE agreed to send out a reminder one month after the
initial send out of the invitation email. The same invitation covering letter as
before was emailed out to each individual member. The AAGL declined to send
out a reminder, or to include the invitation in their next monthly newsletter.
Questionnaire software
The initial pilot study survey was set up using the Panorama© online survey
system that was developed at the School of Management, University of Surrey.
The system was registered for copyright in 2004. It was compatible for PC and
Apple Mac users and worked in common web browser systems including

45
Internet Explorer. For the pilot survey it was very successful, but unfortunately
restrictions forced upon the IT Department at the University of Surrey in 2007
resulted in this study needing to find a new distribution system. The survey was
switched to the commercial system Zoomerang, which was recommended by
the Sociology Department. It was simple, intuitive and straightforward to transfer
the questionnaire from Panorama©. Consequently, the main survey was sent
out using Zoomerang.
In Zoomerang, results are collated for each question and simple descriptive
statistical outcomes are tabulated automatically. Responses are broken down
by date and hour of the day allowing analysis of response patterns. Unlike
Panorama, data was not broken down by the browser system that the
responder uses, to rule out browser system bias. Also unavailable in
Zoomerang were:
1. Drop out analysis showing data on whether particular parts of the
survey were repeatedly not completed to help identify problem areas.
2. Duration analysis showing the time taken to complete the survey.
3. Automatic coding for export into SPSS for further statistical analysis.
Questionnaire development
Initially a list of questions was generated from using the literature and the
known instruments and techniques used for the surgical treatment of
endometriosis in order to cover all of the areas for which information was
required. Meetings were held with supervisors to ensure the validity of the
coverage. The ideas and topics were then developed into open and closed

46
questions of varying types. Initially these were pre-piloted on members of the
research group, and then formally piloted in electronic form as described above.
They were then refined, and a questionnaire was created that took about 10
minutes to complete and was subdivided into the following subsections:
1. A profile of the respondent.
2. The responders’ management of minimal to moderate endometriosis.
3. The responders’ management of endometriomas.
4. The responders’ management of recto-vaginal nodules.
Clear design and layout and visual appeal
The Zoomerang survey system produced a very clear, professional design and
layout. It had varying designs that could be selected from, and had enough
flexibility in question generation to be able to produce a simple, visually
appealing result. An introductory welcome page gave clear instructions and a
thank you statement was included at the end. Any filter questions were clearly
marked. Lower case letters were generally used to avoid the appearance of
shouting. Numbering was clear, and response categories were never split over
two pages. A paper copy of the main questionnaire is at appendix M (however it
was not possible to print this out from the Internet with the page breaks in the
correct place).
Questionnaire length
The questionnaire was kept a short as possible to get the required information,
and was 34 questions long. No questions were added or removed after the pilot
study and the main questionnaire remain the same length.

47
Question form
The questionnaire mainly employed closed type questions with pre-coded
scaled or dichotomous response choices. Some questions had small open
response text boxes to account for response alternatives missed in the pre-
coded list. Closed type questions were used as they are simpler and quicker to
administer and code, as well as being appropriate to the information required
from the survey. However, it was appreciated that they can also give clues
about the type of response expected, and this was minimised where possible.
Response scales in Likert format (5 point scale, strongly agree to strongly
disagree) were used for questions about why responders used the techniques
they chose. They are commonly used, easily understood and analysed, and are
more sensitive and precise than dichotomous responses. Creating categorical
scales, by using words in the Likert scale, shows a similar responsiveness to
visual analogue scales (Jaeschke, 1990) with less ceiling and floor effects than
dichotomous responses. Neutral responses in the scale were felt to be
important and so were included.
Question ordering
An advantage of electronic questionnaires is that responders are unable to read
through them and assess questions in the light of ones further on in the
questionnaire. A funnel approach was taken to question order, starting with
broad questions in logical categories, filtering out responders who these
questions were irrelevant to, and then progressing onto more detailed questions
within that category. Responders who were filtered out were automatically
taken onto the next question that applied to them.

48
Question wording
The wording, format and direction of response were varied where possible to
make responders think about the question and avoid a “response set”. Wording
was also aimed to be as non-confrontational as possible. Double-negatives and
colloquialisms were avoided. Complex questions about the management of
recto-vaginal nodules were broken up into a series of simpler questions that
were more easily understood. There were no double-barrelled questions (two
questions in one sentence) and ambiguity was avoided where possible.
Wording was aimed to draw specific answers rather than general ones, and was
not “loaded”.
Ethics
Ethical approval was gained from the University of Surrey Ethics Committee.
Responders were emailed in their independent capacities as members of
Gynaecological Societies. No resources or information directly related to the
NHS were used and so ethical approval from the NHS was not required.
Statistical analysis
Simple descriptive statistics were used for response rate and demographic
factors with means and standard deviations used where appropriate for
continuous variables, and median and range used for skewed continuous, or
discrete variables. To test for differences between proportions Pearson’s Chi-
square test was used and, where the groups were small, Fisher’s Exact test
was employed. For testing differences in mean values between multiple groups
an Anova was used. For testing differences in median values between multiple
groups Kruskal-Wallis test was used.

49
Financing and insurance
The only financial outlay for the study was the subscription for the Zoomerang
survey system. This was $198 for 6 months use and was funded from a
gynaecological endoscopy fund based at the Royal Surrey County Hospital.
The author’s salary was 50% funded by the Minimal Access Therapy Training
Unit with funds provided by Ethicon Endo-Surgery from October 2004 until April
2008.

50
A Randomised Blinded Trial of Carbon Dioxide Laser
Vaporisation versus Harmonic Scalpel Excision of rASRM
Stage 1-3 Endometriosis in Women with Pelvic Pain.
Participants
Patients were recruited from General Practitioner NHS referrals to the Royal
Surrey County Hospital (RSCH) and private referrals to a Surrey based private
practice. The sample was therefore not self-selected. RSCH is a UK District
General Hospital that has a history of performing tertiary referral-type
gynaecological laparoscopy. The other site, The Guildford Nuffield Hospital, is a
small private institution attached to the RSCH. Although consultation and
surgery took place in more than one hospital, it was under the care of only one
Consultant. From these, patients with pelvic pain who were booked for a
laparoscopy for known or suspected endometriosis were judged against the
inclusion-exclusion criteria and were then verbally invited to take part in the trial
and given an information leaflet explaining the trial (appendix B). This was done
either in the outpatient setting or on admission to the hospital prior to
laparoscopy by one of the three surgeons taking part in the trial. The County of
Surrey, in which the practice is based, has an “average” population with little
evidence of poverty or other major public health problems, and is well
connected by public transport.
Eligibility
Inclusion criteria
• Patients in whom surgical treatment of the endometriosis is considered
the treatment of choice.

51
• Patients consented to participate in the trial.
• Patients who are 18 years old or older.
• Patients who have no contraindications to either of the treatment
modalities proposed.
• Endometriosis stage I-III according to the revised American Society for
Reproductive Medicine scoring system (rASRM) found on visual
inspection at laparoscopy (Appendix H).
Exclusion criteria
• Patients who do not wish to participate or have not signed the informed
consent form.
• Pregnancy or breastfeeding.
• Patients who are unable or unwilling to discontinue hormonal treatment
for six months post-operatively.
• Patients who have received additional treatment for their endometriosis
within three months of surgery.
• Patients with other known conditions causing pelvic pain other than
endometriosis. For example, conditions of the gastrointestinal or genito-
urinary system.
If patients agreed to participate they signed consent forms for admission to the
trial (appendix C). All pre-op questionnaires were completed in hospital on the
day of surgery and kept in an individual trial folder for each patient. The trial
folder contained their personal details, consent forms, questionnaire replies,
rASRM score sheet and any correspondence generated during the trial. The
Senior Research Sister held all files secured in a filing cabinet in the team

52
office. Patients returned at 3, 6, and 12 months for follow-up appointments and
were seen by the research Sister, or a deputising clinic, to fill in their post-op
questionnaires. The research Sister also oversaw the administration of the
private patients questionnaires. Patients who did not attend were contacted by
mail or telephone and requested to return a postal questionnaire.
Interventions
The diagnosis and treatment intervention in this trial was laparoscopy on an
intention to treat basis. These were carried out in either the Day Surgery Unit or
Main Theatres at the RSCH or Guildford Nuffield Hospital. Three surgeons were
involved in the interventions; one Consultant, and two experienced senior
gynaecological laparoscopic surgery Fellows, who were trained in the operative
technique by the same Consultant mentioned above. All cases were undertaken
with the intention of being day cases.
Laparoscopic entry was by the recognised Middlesbrough technique (1999). At
laparoscopy patients were visually assessed for the presence of endometriosis
by diagnostic laparoscopy. An assistant by the bedside completed the rASRM
scoring sheet (Appendix H) with information supplied by the operating surgeon.
All types of endometriosis were included including red, vesicular, blue, black
and white lesions. The score was added up, the stage derived, and reported
back to the surgeon. Disease was judged to be deep if subjectively on palpation
it appeared to infiltrate >5mm below the peritoneal surface. Women who were
found to have stage 1-3 endometriosis, according to the rASRM score were
then randomised to complete destruction of all visible endometriotic lesions
either by vaporisation or excision (see randomisation paragraph).

53
Sharplan carbon dioxide laser at 300mm focal length, 30w power with a 2.5mm
Swiftlase spot, down a Stortz 10mm operating endoscope until the entire lesion
was destroyed and normal underlying tissue was visible, carried out
vaporisation. Excision was by Ethicon Endo-Surgery LCS-C5 or ACE Harmonic
Scalpel staying in normal tissue at the edges to ensure total removal of lesions.
Histology was not routinely collected from the beginning of the trial. This was
changed and routine collection began in June 2005 with case number 46 when
it was recognised that was desirable. On completion of the treatment the
procedure was concluded by our standard closure technique.
The rASRM scoring sheet was reassessed post-operatively for depth score
since actual excision or vaporisation of the lesion gives a more accurate
subjective view of depth compared with the original assessment by palpation.
This did not result in any score adjustments leading to trial exclusion. The
procedures were carried out or directly supervised by the three surgeons in the
trial. Patients were generally discharged home the same day in most cases.
Patients were unable to tell from the appearance of wounds after surgery which
treatment they had received and were effectively blinded.
Objectives
The hypothesis of this trial was that there is no difference between excision and
vaporisation in the treatment of rASRM stage 1-3 minimal to moderate
endometriosis using the CO2 laser for vaporisation and Harmonic Scalpel for
excision where pain is the primary outcome measure.

54
Outcomes
Primary outcome measure
The primary outcome measure with respect to improvement in endometriosis
was pain as recorded by Endometriosis Health Profile-30 Core Pain domain
(Jones et al., 2001). This gave a score between 0-100 where 0 is the best
possible state and 100 the worst (Appendix E).
Secondary outcome measures
Pain
• Visual Analogue Scales (VAS)
10 cm ungraded line for dysmenorrhoea, dyspareunia, chronic pelvic
pain (CPP, pain not associated with menstruation persisting for greater
than 6 months), and dyschezia. Patients marked off a point on the line,
which was measured and recorded as a score out of 10 to one decimal
place (Appendix D).
Health-related quality of life (HRQoL)
• Endometriosis Health Profile Questionnaire (EHP-30)
Core domains for control and powerlessness, emotional wellbeing, social
support, self-image, and the intercourse module each resulting in a score
out of 100 for each, where 0 is the best and 100 is the worst possible
case (Appendix E).
• Hospital Anxiety & Depression Scale (HADS) (Appendix F).
Separate scores are derived for Anxiety and Depression with a maximum
of 21 where 0 is the best state and 21 the worst. Scores are then
classified into 0-7 normal, 8-10 mild, 11-14 moderate and 15-21 severe.

55
Demographic
At 12 months a demographic questionnaire was given to all patients to collect
data on whether or not they went onto hormonal treatment for endometriosis or
had subsequent surgery or became pregnant in the 12 month follow up period
(Appendix G).
Rationale for choosing outcome measures
A single VAS pain score may be subject to significant large day-to-day variation
and so a more sensitive, responsive measure of pain incorporating the quality
of life concerns of patients would be a more powerful tool. This resulted in the
choice of EHP30 Core pain domain as the primary outcome measure. Mr
Barton-Smith spent a day with the author of the Endometriosis Health Profile to
learn its strengths and limitations to be sure that we applied it effectively. Also
permission was granted to use EHP-30. EHP-30 is likely to be a more sensitive
means of measuring endometriosis pain than a VAS because it has a validated
bell shaped probability distribution, showing good construct validity, enabling a
moderate shift in mean score to be easily detected with two moderately sized
treatment groups. The pain domain was developed to “evaluate the outcomes
of conservative surgical treatment for women with endometriosis” and has the
following attributes (Jones et al., 2001):
1. Good construct validity against SF36 with a bell shaped probability
distribution, showing good construct validity, enabling a moderate shift in
mean score to be easily detected with two moderately sized treatment
groups

56
2. Good internal reliability reflected by high alpha co-efficient values of
>0.7 as required in standardised tools based upon established
psychometric test theory (Nunnally 78).
3. Good test-test reliability with intra class correlations exceeding 0.8.
4. Good content validity as items are generated from the concerns of
patients rather than medical signs and symptoms and give a more
powerful indicator of patient benefit.
5. Good responsiveness showing that the test is sensitive to subtle
changes. It shows greater responsiveness than the SF36 bodily pain
scale.
The 11 questions in the EHP-30 Core pain domain give a total score of 44. The
subjects’ score is calculated by dividing their score into 44 and multiplying by
100 to give a score where 0 is the best health state and 100 is the worst. Any
missing variable results in the whole score for that subject being discounted.
Therefore there is a risk in choosing EHP-30 Core pain domain as the primary
outcome, as it could have resulted in a large number of missing data.
Consequently, the VAS scores were included for two reasons:
1. As a back up for the risk of losing large amounts of EHP-30
Core pain score data.
2. As a means of being able to directly compare with previous
trials that used VAS scores (see sample size below).
In addition to pain data, it was considered important to collect HRQoL
information that was relevant to patients’ ability to function on a day-to-day

57
basis. The remainder of EHP-30 Core questions and the sexual intercourse
module fulfilled this criteria. HADS was added in order to evaluate levels of
anxiety and depression in patients, as endometriosis is known to have a
significant psychological effect in sufferers (Jones et al., 2004b). The HADS
shows internal consistency of 0.93 and 0.90 for the subscales of anxiety and
depression, as well as good face, concurrent and construct validity (Zigmond,
1983). It is short, acceptable, easy to administer and also splits subjects into
definable clinical groups.
Outcome collection during the trial
The following data was collected at the indicated time points during the trial.
The primary time point of interest was 12 months.
• Pre-operative
Age
EHP-30 Core and Intercourse Module
VAS scores for Dysmenorrhoea, Dyspareunia, CPP, and Dyschezia
HADS
• Intra-operative
rASRM score and stage
Depth and location of disease
Surgeon
Instrument
Histology from case 46 onwards

58
• 3, 6 and 12 months post-operative
EHP-30 Core and Intercourse Module
VAS scores for Dysmenorrhoea, Dyspareunia, CPP, and Dyschezia
HADS
• 12 months post-operative
Pregnancy data
Adjuvant medical therapy
Subsequent surgical intervention
Sample size
The power calculation for this trial was based upon two requirements:
1. To compare against previous trials.
2. To use a validated primary outcome measure for pain that is most
pertinent to patients’ quality of life.
To calculate a sample size against previous trials, there was only one RCT to
use at the time in 2002. In the 1994 landmark Sutton vaporisation study (Sutton
et al., 1994), 62.5% of patients showed an improvement in pain symptoms at 6
months, based on a visual analogue scale of 0-10 asking patients to score
according to their worst symptom, which was dysmenorrhoea in all cases. Of
the case studies that existed for excision of endometriosis, David Redwine’s
report of 400 consecutive women who had excision, showed a 75% complete
relief, and 20% improvement in pain (Redwine, 1996). This suggested a
potential difference of up to 32.5% in pain outcome. However, it was decided to

59
be more conservative with the difference, especially as the hypothesis was that
there would be no difference. Being more conservative would result in an
increase in the numbers required for the trial, but being able to show up to a
20% difference gave a sample size that was manageable for a single centre as
shown below.
The hypothesis of this study was that there is no difference between excision
and vaporisation, and so the power calculation to show that neither treatment
was worse than the other was as follows. The Food and Drug Administration
80/20 rule for bioequivalence was used to calculate the sample size required to
achieve 80% power. Thus ε=0.20, ∝=0.2 (as equivalence trial), 1-β=0.8. This
will need to be a 2-sided test as it is possible the harmonic scalpel is better than
the laser. If we are assuming equivalence, then π1=π2=0.625 (original study was
62.5%). On this basis we would require 53 patients in each arm if we were to
have an 80% chance of showing a 20% difference in treatment outcomes
(Statistical procedures for bioequivalence using a standard two treatment cross
over design. US Dept of Health and Human Services, Public Health Service.
Food and Drug Administration 1992).
As it turned out, the subsequent 2004 Abbott excision RCT (Abbott et al., 2004)
used a 0-100 “no change in pain to complete relief of pain” visual analogue
scale as the primary outcome measure, showing an improvement in 80% of
patients at 6 months, and also VAS scores for all four pain modalities. This
suggested a potential difference between excision and vaporisation in the order
of around 20%, and endorsed the selection of 20% difference chosen for the
power calculations.

60
The two primary outcome measures in the Sutton and Abbott RCTs are different
and not easily directly comparable with each other. By using VAS scores in this
study for dysmenorrhoea, dyspareunia, CPP and dyschezia, it will be possible
to correlate the data directly to the dysmenorrhoea score in the Sutton trial and
to all 4 modalities in the Abbott trial. There is no valid basis to combine the
individual 4 pain modalities used for the VAS to create a global score,
especially since some of the components are often not relevant to the patient
(i.e. dyspareunia).
The primary outcome measure in this study is EHP-30 Core pain score, derived
and recorded on a continuous scale, and has a previously reported maximum
SD for change in pain score of 26.0 (Jones et al., 2004a). In order to detect an
underlying difference of 20 points out of 100, 28 subjects are needed in each
treatment group (2-sided test with size = 5% and power = 80%). The number in
the bioequivalence power calculation above comfortably satisfies this
requirement.
Randomisation
Sequence generation
Restricted randomisation was used to generate blocks of 10 random treatments
where odd numbers 1,3,5,7,9 defined allocation to excision and even numbers
2,4,6,8,10 defined allocation to vaporisation. This improved the chances of a
balanced outcome at the cost of reducing the unpredictability of the sequence.
The block size was not randomly varied. No stratification was used.

61
Allocation concealment
Sealed opaque envelopes containing a number from 1-10, in blocks of ten,
were then randomly sorted and added to the trial folders, one per folder. The
sequence was therefore concealed until the intervention was assigned.
Implementation
The trial folders and envelopes containing the concealed treatment were
produced by Mr Carpenter and Mr Barton-Smith and kept locked in a separate
filing cupboard to the active and completed trial folders. The operating surgeons
administered the pre-op questionnaires. Once in theatre, if the patient fulfilled
the entry criteria by rASRM score, then the envelope was opened during the
surgery. The surgeon was told the allocated treatment and carried this out
immediately. Mr Carpenter and Mr Barton-Smith took part in both the
generation and implementation of the random sequence of treatments.
Blinding
Both the patients and the research nurse who performed the post-operative
assessments were blinded from the treatment received. The surgeons were
blinded to the post-operative questionnaire replies. The success of the blinding
was not evaluated to assess for performance or detection bias.
Statistical Methods
The study analysis was designed on an intention to treat basis. Descriptive
analysis was generally presented as proportional numerator/denominator
comparisons with percentages. Chi-square analyses were performed to analyse
proportional differences. Primary and secondary outcome measures were

62
analysed, in terms of proportional improvements, with Chi-square tests. The
extent of improvement for each technique at 12 months against baseline was
analysed by paired t tests. Comparative analysis between treatment techniques
of the extent of improvement was analysed with unpaired t tests. Standard
deviations and 95% confidence intervals were quoted where appropriate.
Pearson’s Correlation Test was used to examine association between
continuous variables. The Mann-Whitney U Test was used to compare two
groups for ordinal variable scores. Exploratory analysis of major findings, to
identify demographic and other factors influencing results, was performed with
backward stepwise linear regression. All statistical analysis was performed
using SPSS v17 (SPSS inc Chicago IL, USA; 2008).
Ethical approval
Ethical approval for this trial was obtained from the Surrey Local Research
Ethics Committee, and Research Governance was obtained from the Royal
Surrey County Hospital Research and Development Committee.
Cost of Study
There were no additional costs over and above that entailed in current
treatment schedules. The laser and harmonic scalpel are routinely used in the
operating theatre. The author’s salary was 50% funded by the Minimal Access
Therapy Training Unit with funds provided by Ethicon Endo-Surgery from
October 2004 until April 2008.

63
Chapter 4 - Results
Results of an International Survey of Surgical Techniques
used in the Treatment of Endometriosis
Introduction
The survey was developed to gain a snapshot of what equipment and
techniques gynaecological surgeons were using and why, as many devices and
energy forms are available. From this a picture would hopefully emerge of the
current views of gynaecological laparoscopists in relation to the evidence, or
lack of it. Implementing this trial was difficult in so far that obtaining agreement
from the three societies proved to be very challenging. It took over a year to
resolve data protection and distribution issues, and even then the plan for the
AAGL was suboptimal, and the Australian Gynaecological Endoscopy Society
had refused to participate. That being said, the survey was eventually sent out
with optimism that it would achieve its aim.
Pilot Study Results
A Pilot Study of the Panorama© based survey took place between 10 January
and 8 February 2006. The survey was sent to 12 laparoscopic gynaecologists
who act as Faculty members at the Minimal Access Therapy Training Unit at the
University of Surrey. There were 7 completed responses received between 10-
31 January and one further response at the second attempt on 8 February
2006.
Total response after 2 emails was 8/12 (66.7%). The response rate was two
thirds and so the length and feasibility of the survey seemed reasonable.

64
Certain changes were made to the questionnaire resulting from the pilot. Some
of these were triggered from our own observations and some from the
responders who had an extra question in their surveys asking about their views
on the survey structure. The following adjustments were made:
1. A question was added asking about whether responders try to treat
their patients surgically at the time of laparoscopic diagnosis.
2. A question about 2 stage procedures for endometriomas was
included.
3. More text boxes after possibly ambiguous questions were added to
allow greater ability for responders to explain their practice.
4. Minor wording changes were made to avoid ambiguity in some
questions.
Main Survey Results
The main survey was initiated on 18 April 2008 and the last response before the
survey was closed was recorded on 23 September 2008.
Data cleaning
The database was checked for errors by carrying out simple descriptive
statistics and looking for potentially erroneous outliers. 17 responders were
excluded because they failed to answer any questions once they had passed
beyond the introduction page. 13 responders at question 4 said that they did not
treat endometriosis surgically and so were automatically taken to the end of the

65
survey and excluded from the analysis. 18 responders partially completed the
survey but were kept in the dataset. Therefore a total of 30 cases were
excluded leaving a data set of 339 of which 321 were correctly completed and
18 were partially completed. There were a number of correct partial responses
with 5 who did not treat endometriomas, and 134 not treating recto-vaginal
nodules, which meant that they were automatically filtered to the end of the
survey at questions 15 and 21 respectively. Zoomerang had electronically
coded some answers as “no” after partial responders had ceased to answer
further questions. These were recoded as missing data.
The wide variety of countries from which responses originated, coupled with the
low response rate, meant that it was sensible to recode these as geographical
regions of UK, mainland Europe, USA and the rest of the World. In addition to
this, the free text responses were coded into either already existing codes or
into new codes. New codes were created for the following:
Q9: Argon beam coagulator, Bipolar-sealing systems.
Q10: Argon beam coagulator.
Q17: May observe endometriomas <3cm.
Q22: Intravenous Urogram

66
Response Rate
Table 2. Survey response rates by specialist society.
No on database
Successfully sent
No of Replies (%)
BSGE
489
390
92 (23.6)
ESGE
1623
1481
175 (10.8)
AAGL
3259
3007
102 (3.1)
Overall
5371
4878
369 (7.5)
There were 489 total visits to the survey, although as table 2 shows, 369
actually started it which means that 120 people did not get beyond the
introduction page. Once started, only 35 responders of those 369 (9.5%) failed
to complete the questionnaire.
General characteristics of responders
General demographic data questions were asked for age, region of principal
practice, number of years performing laparoscopic surgery independently, main
form of training in laparoscopic surgery, years qualified as a doctor and whether
they were a specialist or trainee.
Specialist versus trainee
267 (88%) responders were specialists whilst 37 (12%) were trainees with 35
missing data. BSGE response was 23.6% of total membership but 194 (45%)
are trainees and 237 (55%) specialists. When excluding trainee responses
there were 54 responses from specialists, which is 54/237 = 22.8% of the
consultant population.

67
Region
The majority of responses came from mainland Europe and the UK as is seen
in Fig 1 below.
Fig 1. Distribution of responders by geographical region.
Mainland Europe 49%
UK 23.9%
Rest of World 19.2%
USA 7.7%

68
Experience
As fig 2 below shows, the average age of responders is 46.9 years (+/-9.14).
Although the age data are slightly skewed, the skewness is within acceptable
limits and therefore parametric statistics could be used (skewness = 0.214, SE
0.140).
Fig 2. Age distribution of survey responders.
As can be seen from fig 3 below, the time qualified as a doctor is normally
distributed (skewness = -0.330 SE 0.140) with an average of 37.4 years (+/-
9.44) since qualification.

69
Fig 3. Year survey responder qualified as a doctor.
The data for the number of years that responders have been performing
independent laparoscopic surgery is skewed (skewness 0.584 SE 0.133) and
shows a median value of 11 years (range 1-35). The skewness seen in fig 4
below, illustrates more gynaecologists joining the specialty from about 20 years
ago. As this skewness is minimal, it will be treated as normally distributed. This
variable was used as the best indicator of experience in subsequent analysis.

70
Fig 4. Number of years survey responder has performed independent laparoscopic surgery.
Training
Fig 5. Pie chart of the distribution of training methods in survey responders.

71
As seen in fig 5 above, 39% of responders had trained through a specific
laparoscopic training programme or fellowship. This was more likely if you were
less experienced in terms of number of years independently performing
laparoscopic surgery. Responders who had had formal training had a mean of
9.2 years experience versus 14.0 years for those who were self-taught, giving a
mean difference of 4.6 years (p<0.0005, 95%CI 2.9 to 6.4). Those who had
undergone a specific training program and fellowships also had a trend towards
doing more complex surgery in terms of their likelihood to be doing surgery for
recto-vaginal nodules (61% compared to 54%, although this difference was not
statistically significantly different, p=0.251).
Number of operations performed per month for endometriosis
Table 3. Number of operations performed for endometriosis by survey responders per month.
Operations
<10 per month
10-20 per month
>20 per month
n
(%)
Responders
202 (59.6)
102 (30.1)
35 (10.3)
In table 3 above, the majority of responders are performing <10 procedures for
endometriosis per month. A sub-analysis of these results showed that training
method does not influence the number of procedures carried out per month is
(p=0.227). However, gynaecologists with more experience in independent
laparoscopic surgery undertook more surgical cases per month than those with
less experience (Anova Test F=7.69, p=<0.001). This is well illustrated in fig 6
below:

72
Fig 6. Means plot comparing the mean number of years performing laparoscopic surgery with
the number of endometriosis operations performed per month.
Types of endometriosis operated on
Table 4. Percentages of survey responders that operate on various types of endometriosis.
Type of endometriosis
Responders n (%)
Minimal to moderate
275 (81.1)
Endometriomas
300 (88.5)
Severe
176 (51.9)
Recto vaginal nodules
141 (41.6)

73
Do responders treat at the time of laparoscopic diagnosis?
Table 5. Proportion of surgeons who see and treat patients at the same time for endometriosis.
Treat at Diagnosis
n (%)
No
13 (3.8)
Sometimes
44 (13.0)
Yes, except severe
153 (45.1)
Yes, always
129 (38.1)
The data in table 5 above shows that the majority of patients are seen and
treated at the same time. The means plot below in fig. 7 shows that more
experienced surgeons treated patients at the same time as the diagnostic
procedure, when compared to less experienced surgeons (Anova Test F=3.18,
p = 0.024).

74
Fig 7. Means plot comparing the mean number of years performing independent laparoscopic
surgery with “see and treat” management.

75
Changes in surgical instruments used to treat endometriosis
Responders replied by ticking all instruments that they may currently use and
have previously used to treat endometriosis as seen in table 6 below.
Table 6. Previous and current use of surgical instruments for endometriosis.
Type of instrument
previous use
current use
P value
n
(%)
Cold scissors
207 (61.1)
198 (58.4)
*0.481
Monopolar diathermy
231 (68.1)
193 (56.9)
*0.006
Bipolar diathermy
260 (76.7)
293 (86.4)
*0.001
Ultrasound technology
87 (25.8)
139 (41.1)
*<0.001
CO2/YAG/KTP laser
101 (29.8)
73 (21.5)
*0.014
Helica
24 (7.1)
19 (5.6)
*0.431
Argon beam coagulator
4 (1.2)
10 (2.9)
**0.105
Bipolar sealing devices
0 (0.0)
16 (4.7)
NA
* Pearson’s Chi-square Test **Fisher’s Exact Test
Electro-surgery is still the most frequently used energy source in the surgical
treatment of endometriosis though the use of no energy with cold scissors only
is still very common. Bipolar and ultrasound energy use have significantly
increased from 77% to 86% and from 26 to 41% respectively.

76
Management of minimal to moderate endometriosis
Table 7. Number and percentages of responders who use various techniques as their main
method for surgically treating minimal to moderate endometriosis.
Superficial n (%)
Deep n (%)
*P value
Excision
90 (27.4)
158 (49.2)
P<0.0005
Vaporization
15 (4.6)
8 (2.5)
P=0.137
Combination
161 (48.9)
123 (38.3)
P=0.002
Coagulation
62 (18.8)
20 (6.2)
P<0.0005
*Pearson Chi-Square Test
In table 7 above, there is a significantly larger proportion of surgeons using
excision as the main technique for excising deep minimal to moderate
endometriosis compared with superficial.
There is a regional trend towards greater use of coagulation (22.7% v 11.7) and
vaporisation (6.7% v 3.9%) in mainland Europe versus the UK (p=0.169). This
seems to be reflected in the instrument use in Europe where a significantly
greater use of bipolar diathermy exists (p<0.001). Interestingly, experience in
terms of number of years performing independent laparoscopic surgery does
not affect choice of technique (Anova test p=0.496).

77
Why do responders choose their preferred technique for treating superficial and
deep minimal to moderate endometriosis.
Responders filled in a Likert scale question prompting them to reply to a set of
statements as shown in the full survey (appendix M), along a spectrum where
1= strongly disagree, 2= disagree, 3= neutral, 4= agree, 5= strongly agree. The
results are shown in tables 8 and 9 below.
Table 8. Reasons for choosing main technique for treating superficial minimal to moderate
endometriosis.
Excision
Vaporization
Combination
Coagulation
*P value
Median
(IQR)
Trained
4 (3-4)
4 (3-4.25)
4 (3-4)
4 (3-4)
0.299
Self-taught
3 (2-4)
3.5 (2.75-4.25)
3 (2-4)
4 (3-4)
0.752
Best way
4.5 (4-5)
4 (3-4.25)
4 (4-5)
4 (3-4)
<0.0005
Easier
3 (2-4)
3.5 (2.75-4)
3 (2-4)
4 (3-4)
<0.0005
Tried other
4 (3-4)
3.5 (3-4.25)
3 (2-4)
3 (2-4)
0.003
No preference
2 (1-3)
2 (1-3.25)
2 (2-3.5)
3 (2-4)
<0.0005
Lack of equip
2 (1-4)
3 (2-4)
3 (1.5-4)
4 (3-5)
<0.0005
Lack of skill
2 (1-3)
2.5 (1.75-4)
3 (2-4)
4 (3-4)
<0.0005
Cost effective
3 (2.25-4)
3 (2.75-4.25)
3 (3-4)
4 (3-4)
0.194
*Kruskal –Wallis Test
As table 8 above illustrates, techniques for treating superficial endometriosis
were used because the surgeon was trained in the particular procedure.
Similarly, surgeons agreed that the technique they used was the best and
disagreed that they had no preference over technique.

78
Table 9. Reasons for choosing main technique for treating deep minimal to moderate
endometriosis.
Excision
Vaporization
Combination
Coagulation
*P value
Median
(IQR)
Trained
4 (3-4)
4 (3-4)
4 (3-4)
4 (4-4)
0.782
Self-taught
3 (2-4)
3 (2-4)
4 (3-4)
4 (3-4)
0.436
Best way
5 (4-5)
3 (3-4)
4 (3-4.5)
3 (3-4)
<0.0005
Easier
2.5 (2-3)
3 (3-4)
3 (3-4)
4 (3-4)
<0.0005
Tried other
3 (2-4)
3 (3-4)
3 (2-4)
3 (2-4)
0.939
No preference
2 (1-3)
3 (2-3)
3 (2-3)
3 (2-3)
<0.0005
Lack of equip
2 (1-4)
4 (3-4)
4 (2-4)
4 (3-4)
<0.0005
Lack of skill
2 (1-3)
3 (3-4)
4 (2-4)
4 (3-4)
<0.0005
Cost effective
3 (2-4)
4 (3-4)
3 (3-4)
4 (3-4)
0.210
*Kruskal-Wallis Test
In table 9 for deep disease, the users of all techniques agreed that they used
their preferred technique because that is how they were trained or taught
themselves to do it, though the feeling was stronger in favour of how they were
trained and more coagulators appear to use their technique because they were
self-taught. The feeling for all technique users was somewhere between neutral
and agreement over concern over cost effectiveness. A further cross-tabulation
of region against concern over cost effectiveness revealed no significant
difference (p=0.292).
The excisers agreed more strongly than other technique users that they used
their technique because they thought it was the best and feel this more strongly
when treating deep disease. They also seemed to have tried more techniques

79
than the others for treating superficial endometriosis. The coagulators felt more
strongly that they used their technique because they thought it was easier.
However they also felt that they lacked the skill or equipment to use another
technique (there are however only 19 responders in the deep disease response
group). Despite this they still believed that their technique was the best. The
vaporizers and combined exciser/vaporizers seem to fall somewhere in the
middle between the excisers and coagulators in their views.
Management of Endometriomas
98.4% of responders said that they surgically treated endometriomas. Initially
they were asked to say whether or not they used a 2-stage technique of
drainage, hormonal down regulation and subsequent definitive procedure. The
results of this are shown in the pie chart in fig 8 below.
Fig 8. Pie Chart showing the use of 2-stage procedures for endometriomas.
Yes=26 (8%) Sometimes=92 (29.9%) No=190 (61.7%)

80
Table 10. Responders use of techniques for surgically treating endometriomas.
<3cm
>3cm
*P value
Excision
206 (66.2)
223 (72.2)
P=0.110
Ablation
24 (7.7)
19 (6.1)
P=0.580
Combination
69 (22.2)
61 (19.7)
P=0.455
*Pearson Chi-square Test
There is no significant difference in technique whether endometriomas are
either less or greater than 3cm diameter however excision is clearly the
commonest technique overall. 5 (1.5%) respondents commented that they
observe endometriomas that are less than 3cm diameter and do not surgically
treat them.

81
Why do responders choose their preferred technique for treating
endometriomas of different size?
The results for this are divided into the techniques used for endometriomas of
<3cm and >3cm diameter.
Table 11. Reasons for choosing technique for treating endometriomas of <3cm diameter.
Excision
Ablation
Combination
*P value
Median
(25-75%)
Trained
4 (3-4)
4 (3-4)
4 (3-4)
0.862
Self-taught
4 (3-4)
4 (3-4)
4 (3-4)
0.500
Best
4 (4-5)
4 (3-4)
4 (4-5)
<0.0005
Easier
3 (2-4)
3 (3-4)
3 (3-4)
0.002
Tried other
3 (2-4)
4 (2-4)
3 (2-4)
0.939
No preference
2 (2-3)
3 (2-4)
3 (2-4)
0.001
Lack of equipment
2 (1-3)
4 (2-5)
3 (2-4)
<0.0005
Lack of skill
2 (1-3)
4 (2-5)
3 (2-4)
0.001
Cost effective
3 (3-4)
3 (3-4)
3 (3-4)
0.945
*Kruskal-Wallis Test
In table 11 above, for <3cm diameter endometriomas the technique groups do
not vary in their view in that they agree that they carry out their technique
because they were either trained or taught to do it. They are all neutral in their
concern over cost effectiveness and whether they have tried other techniques
or not. The excisers do not feel like they lack equipment or skill where as the
ablators seem to, and the combined group are neutral on the issue.

82
Table 12. Reasons for choosing technique for treating endometriomas of >3cm
diameter.
Excision
Ablation
Combination
*P value
Median
(25-75%)
Trained
4 (3-4)
4 (3.25-4)
4 (3-4)
0.351
Self-taught
4 (3-4)
3 (2.25-4)
4 (3-4)
0.078
Best
4 (4-5)
3.5 (3-4)
4 (4-5)
<0.0005
Easier
3 (2-3)
3 (3-3.75)
3 (2-4)
0.021
Tried other
3 (2-4)
3 (2-4)
3.5 (2.75-4)
0.352
No preference
2 (1-3)
3 (2.25-3)
3 (2-4)
<0.0005
Lack of equip
2 (1-3)
4 (3-4)
3 (2-4)
<0.0005
Lack of skill
2 (1-3)
3.5 (3-4)
3 (2-4)
<0.0005
Cost effective
3 (3-4)
4 (3-4)
3 (2-4)
0.245
*Kruskal-Wallis Test
In table 12 above, similar trends are apparent in the reasons for choosing
treatment technique for >3cm endometriomas, suggesting that the approach to
endometriomas does not seem to differ much no matter what the size.
Management of recto-vaginal nodules
175 responders said that they surgically treat recto-vaginal nodules.
Responders were more likely to do this if they were more experienced in terms
of number of years performing independent laparoscopic surgery (mean of 13.9
v 10.1 years p<0.0005, mean diff 3.848, 95%CI 1.976-5.721). 78.6% of
responders surgically treat recto-vaginal nodules more often by laparoscopy

83
compared with 10.7% who more often use laparotomy. A further 10.7% use
laparoscopy and laparotomy equally.
Preoperative investigations
As table 13 below illustrates, trans-vaginal ultrasound scan (TVS) is the most
common pre operative test for recto vaginal nodules followed by MRI.
Table 13. Use of pre-operative tests for recto-vaginal nodules.
n (%)
TVS
142 (81.1)
TRS
55 (31.4)
CT
15 (8.6)
MRI
105 (60)
Colon/sig
80 (45.7)
Ba enema
25 (14.3)
IVU
7 (4.0)
None
4 (2.3)
Operating with a colorectal surgeon
Table 14. Number of responders who operate on recto-vaginal nodules with a colorectal
surgeon.
n (%)
Never
9 (5.4)
Rarely
66 (39.3)
Mostly
62 (36.9)
Always
31 (18.5)

84
In table 14 above, only 5.4% of responders never operate with a colorectal
surgeon and 18.5% always operate with a colorectal surgeon. The remaining
responders are split between rarely or mostly. There are no significant
differences when comparing training method (p=0.512) or experience (p=0.360)
against likelihood of operating with a colorectal surgeon.
Surgical technique for removing recto-vaginal nodules
Table 15 below, shows surgeons’ surgical technique for removing recto-vaginal
nodules in terms of whether an attempt is made to initially shave recto-vaginal
nodules off the bowel to avoid opening it, those who try to use disc resection
rather than segmental resection if shaving fails, those who are ever prepared to
do a segmental resection, and those who always aim to do a segmental
resection for all cases.
Table 15. Surgical technique for treating recto-vaginal nodules.
Shave
Disc
Ever segment
Always segment
n
(%)
No
22 (13.2)
57 (33.7)
48 (28.7)
124 (73.8)
Yes, by myself
56 (33.5)
28 (16.6)
4 (2.4)
3 (1.8)
Yes, with surgeon
27 (16.2)
56 (33.1)
103 (61.7)
32 (19.0)
Yes, either/or
61 (36.5)
27 (16.0)
12 (7.2)
9 (5.4)
144 (86.2%) responders were in favour of attempting an approach that tried to
avoid opening the bowel in the first instance. 111 (65.7%) responders were
prepared to carry out disc resections to avoid segmental resection if shaving

85
failed. 119 (71.3%) responders were prepared to use segmental resections. 44
(26.2%) responders will always aim to carry out a segmental resection.
Discussion
The validity of the data is immediately called into question by the low response
rate achieved, despite efforts to follow recognised methods of questionnaire
construction in the methodology (Nakash R, 2006, Boynton, 2004).
In an attempt to explain this, one can consider the strengths and limitations of
the methodology. This study had certain strengths when considering survey
administration. The questionnaire was salient to its population and should have
generated an interest. An efficient means of completing the questionnaire was
provided with an electronic link directly to it. A clear covering invitation letter
confirming confidentiality was given. An incentive was provided in the form of a
prize for one responder. A reminder was included, a researcher was available
by email to answer questions, and a pilot study was performed.
However, there were some limitations in administration also. There is a
possibility that some of the responders in mainland Europe or non-English
speaking countries may have been put off from completing the questionnaire in
English. There was no pre-notification of the survey to the population. It would
have been nice to have an advert in each society a month or so before sending
out the survey to warn people that it was coming and generate some interest.
However, bearing in mind the complexities involved in getting societies to agree
to even sending out the survey, this would likely have been difficult. In addition,

86
the incentive was probably not strong enough. However this was developed
within the economic restraints of the study. The lack of personal invitation email
to AAGL members, and the refusal to send a reminder, undoubtedly limited the
response from the AAGL. Many AAGL members would have missed the
invitation in the newsletter or perhaps never read the newsletter at all. There
also remains the possibility of over exposure in the population and a feeling of
“oh no….not another questionnaire!”.
From the questionnaire design perspective, the study was strong in that it did
produce a clear design and layout that was visually appealing. At 34 questions
long it was short according to Nakash’s meta-analysis, which defined shorter
questionnaires as 7 to 47 questions (Nakash R, 2006). As has been described
in the methodology section, a significant amount of effort went into question
form, wording and ordering, that was refined following the pilot study. The main
failure appeared to be getting people to start the survey as 90.5% did complete
it if they went beyond the invitation page.
Despite the low response rate, the characteristics of those returning the
questionnaire appear to reflect those expected in the general population of
laparoscopic gynaecologists.
The age of responders is normally distributed and the number of years
practising independently as a laparoscopic surgeon is only mildly negatively
skewed. Assuming the historical growth of laparoscopic surgery over the last 20
years, then it is not unreasonable to expect this skewness in the population, as

87
gynaecological laparoscopic surgeons join the sub-specialty each year in
increasing numbers.
Although only 10% of responders were trainees, they were analysed within the
whole group together with qualified specialists. This was because they were
considered as legitimate members of the specialist societies in their own right
and there were not enough of them to excessively duplicate the answers of the
specialist they worked with.
The regional divide of responses was not equally representative as has been
explained above by the poor response rate especially from the AAGL (3.1%).
Consequently comparative regional analysis was not able to be carried out.
41.6% of responders said that they operated on recto-vaginal endometriosis
and this appears higher that one might expect. It seems unlikely that nearly half
of gynaecological laparoscopic surgeons are genuinely tackling the most
complex endometriosis. This suggests that, whilst we seem to be looking at a
sample that is fairly normally distributed by age and experience, the sample
population contains a greater proportion of surgeons undertaking complex
endometriosis surgery than would be expected in the general population of
gynaecological laparoscopic surgeons. It is perhaps not surprising that those
who do tackle recto-vaginal endometriosis will have a greater vested interest in
the disease, and therefore are more likely to respond to a survey such as this.
Within the limits of the sample population described above, the analysis
revealed that formal training appears to be frequent, and a surprisingly high

88
percentage (39%) of responders had undergone some form of specialist
laparoscopic training. The finding that more experienced surgeons carried out a
greater number of endometriosis cases per month, suggests that they are
becoming more specialised with a greater proportion of endometriosis cases in
their referrals. It is also possible that they are operating more quickly and
efficiently.
It is interesting that 19% of responders do not operate on minimal to moderate
endometriosis. 78% of these 19% however do operate on endometriomas.
Therefore, it is likely that this group consists mainly of infertility specialists who
may be only dealing with endometriomas prior to IVF, and not dealing with any
coexisting disease despite it existing in pretty much all cases (Banerjee et al.,
2008). Also 31% of the 19% stated that they do operate on recto-vaginal
disease, and so this may represent a super specialised group who are dealing
with only more complex cases.
It is recommended that if a laparoscopy is performed and endometriosis
diagnosed, then a see and treat approach be taken provided that adequate
consent has been obtained (Kennedy et al., 2005). It is reassuring to see that
the majority of surgeons take this approach. It perhaps is not surprising that if a
patient sees a more experienced surgeon then they are more likely to get a see
and treat approach.
In terms of instruments used to treat endometriosis, a substantial number of
surgeons will avoid energy use altogether, and just use cold scissors for
removing presumably more straightforward lesions in the peritoneum that may

89
lie over sensitive structures like the ureters. This reduces the risk of thermal
damage to these structures and the amount of bleeding from the peritoneum
whilst using cold scissors is small, negating the need for a haemostatic
instrument.
When energy is used, electro-surgery remains the most frequently used energy
source for treating endometriosis. This probably reflects its great flexibility as
either an excision or vaporative tool in monopolar form, despite which
surprisingly monopolar use has fallen significantly. The increase in bipolar use
is potentially concerning as coagulating lesions not only fails to remove all
disease and has been shown to result in a worse outcome for endometriomas
(Hart et al., 2008), but also fails to allow full assessment and treatment of the
full depth of lesions. However some bipolar use may be for haemostasis
following excision by cold scissors or another energy source.
Ultrasound devices have significantly increased in popularity possibly as a
result of successful marketing claiming advantages of increased safety and
user friendliness. Despite laser technology being supported by the RCT
evidence for surgically treating endometriosis behind it (Sutton et al., 1994), its
use seems to have significantly fallen away which may be due to the specific
training or substantial initial financial outlay required.
Minimal to moderate endometriosis
In terms of technique used for minimal to moderate endometriosis, the
significant increase in excision use for deep versus superficial disease is logical.
In deep disease, excision alone is more popular than a combined approach of

90
excision/vaporization, which was the most popular approach for superficial
disease. That the “excisers only” are a group who “strongly agreed” in their
approach over others, was not a surprise. At conference, the excisers appear to
defend their technique with more fervour than others. The majority view that a
combination of true vaporization, resulting in the same visual effect as excision,
or excision alone for superficial endometriosis is reassuringly correlative with
existing randomized data (Jacobson et al., 2009, Wright et al., 2005).
Coagulation by itself is a minority option compared with excision, vaporization
or a combined excision/vaporization approach. This again seems to fit the
hypothesis that ablation/coagulation is suboptimal in assessment and
treatment. As argued above, the data may be biased in favour of experts as
shown by the high percentage of responders that treat recto-vaginal
endometriosis. This may mean that there is possibly an even higher percentage
in the true population who use a superficial coagulation/ablation technique who
lack the skills or instruments to carry out full excision or vaporization. This
statement is supported by the fact that surgeons who are self-taught or feel they
lack skills and equipment, were more likely to use coagulation. Despite
admitting to a lack of equipment and skill, the coagulators had a tendency to
agree that their technique was still the best, which may be influenced by a
difficulty to admit that what you are doing is not the best option for the patient.
Endometriomas
The treatment of endometriomas is currently done by excision by the majority of
responders whether or not they are > or < 3cm diameter. It is currently
recommended in ESHRE guidelines that excision is the preferred technique in

91
endometriomas of >3cm to obtain the cyst wall for histology to exclude
malignancy. If this fails then a two-stage procedure, with hormonal down
regulation between operations, may be tried to improve the chances of ablation
being successful at the second procedure. Only 30% of the responders in this
study said that they sometimes use this technique, suggesting that they feel
they are, more often than not, successful with a one-stage procedure, most
likely using excision for this. It may also imply that the concern over potential
ovarian damage from excision (Loh 99) is generally less important than the
perceived improved pain, fertility and recurrence outcomes shown in several
studies of endometrioma treatment (Beretta et al., 1998, Alborzi et al., 2004b).
There is no randomised evidence commenting on the best treatment for
endometriomas measuring <3cm diameter.
Recto-vaginal endometriosis
Laparoscopy is the predominant method to tackle recto-vaginal endometriosis
amongst the responders. However, the finding that 10.7% mainly took a
laparotomy approach may be erroneously low when considering the whole
gynaecological surgical population. The question here remains whether or not
women are being referred to this expert group for laparoscopic treatment and
potentially fertility sparing surgery, or whether “non-experts” are still operating
on a significant number with laparotomy and pelvic clearance.
For advocates of a “conservative” approach to surgically removing recto-vaginal
endometriosis, the ideal aim of preoperative testing would be to accurately
predict that a lesion could be shaved off or removed by disc resection of the
bowel without segmental resection. This would require accurate measurement

92
of the depth of invasion into the bowel wall and being able to correlate this
information to correctly identify the surgical procedure required to remove the
disease. However, the accuracy of such a pre-operative test remains unproven.
For those surgeons who advocate segmental resection in all cases of bowel
involvement, the logic of this approach relates to the finding of macroscopic and
microscopic satellite lesions of endometriosis in the rectum and sigmoid colon in
68% of cases (Keckstein, 1999). Thereby, removing the segment of bowel may
reduce the risk of recurrence. This is however at the expense of significant
chronic bowel morbidity in young patients (Dubernard et al., 2006).
Preoperative tests like TVS, MRI and TRS have been effective at predicting
bowel involvement in recto-vaginal endometriosis (Dessole et al., 2003,
Hudelist et al., 2009, Fedele et al., 1998, Chapron et al., 2004, Bazot et al.,
2009) and so their common use in modern practice seen in the survey results is
understandable if only to alert the surgeon to potential bowel involvement and
consequently to be able to adequately inform and consent the patient prior to
surgery. In Hudelist’s report, involvement of the serosal and smooth muscle
layers of the bowel was positively predicted in 98% of cases, and the PPV for
mucosal involvement was 53%. These results are perhaps not immediately or
easily reproducible in most units.
Barium enema or sigmoidoscopy/colonoscopy are still used as adjuncts to the
more effective imaging modalities outlined above. Double contrast barium
enema alone has been shown in two recent studies to have close to 90%
sensitivity, and specificity ranging from 54 to 88% (Ribeiro et al., 2008, Faccioli
et al., 2008).

93
In modern practice where litigation is becoming increasingly predominant it is
not surprising to find that only 5.4% of gynaecological laparoscopists never
operate with a colorectal surgeon for recto-vaginal endometriosis.
The argument of whether to shave, disc resect or segmental resect for recto-
vaginal endometriosis is one that has been a main focus in conferences over
the last few years with surgeons offering their strong opposing opinions based
on large case series (see table 1). There is no high quality evidence to support
one procedure over the other and due to the methodological difficulties of
designing an RCT, it is unlikely that evidence will emerge in the near future.
Prior to this survey little was known about the extent to which opinion is divided
in this contentious area. The majority of responders (86.2%) in the survey are in
favour of a conservative shaving approach if possible in the first instance, with
no significant regional variation in this opinion (p=0.291). Whilst 144 surgeons
stated that they would attempt a shave first, there are only 111 and 119 in
favour of ever performing discs resections or segmental resections respectively.
This suggests a group of responders who either believe that one should avoid
opening the bowel at all costs, or will refer the patient onto someone who can.
At the other end of the spectrum lie the 26.2% of responders who will always
aim to do a segmental resection. Nevertheless, the results show that the
majority of responders take a pragmatic conservative approach to the problem
of opening the bowel or performing segmental resections where they have to.

94
Conclusion
Taking into account the poor response rate, though seemingly fairly
representative sample nevertheless, this study reveals for the first time the
attitudes and practice of gynaecological laparoscopists dealing with
endometriosis who belong to specialist societies. In the main the findings
support previous study results, but we must be guarded in this statement as the
responders may well be a biased group of experts, and these results represent
practice within the specialist societies, not what is likely to be happening
amongst the general population of gynaecologists.
It is reassuring to see that formal training is becoming commonplace especially
as responders generally agreed that they carried out particular procedures
because that is “how they had been taught to do it” implying that training is a
major influence on practice. Practice was not dictated generally by concerns
over cost, but some felt they lacked the skill or equipment to perform their
optimal technique and this is something that is likely to improve over time. It is
encouraging that the majority of responders favour a see and treat approach.
For minimal to moderate endometriosis treatment, the majority of laparoscopic
gynaecologists are electro-surgery users with a tendency to excise, or take a
combined excision/vaporization approach, in line with current evidence, with a
greater tendency to excise deep disease. Questions remain over endometrioma
treatment where laparoscopic gynaecologists are generally excisers for all sizes
of cyst, and possibly influenced by the current evidence, believing that on
balance they are doing more good than harm. By and large, laparoscopic

95
gynaecologists are taking a conservative approach to recto-vaginal disease
where possible, to avoid the potential complications of bowel surgery. However,
when required, the study shows an acceptance of a multi-disciplinary approach
involving colorectal colleagues in the surgical management.

96
Chapter 5 - Results
Results of a randomised blinded trial of carbon dioxide laser
vaporisation versus harmonic scalpel excision of rASRM
stage 1-3 endometriosis in women with pelvic pain
Participant Flow
Fig 9. Flow diagram of participants through the trial.
!527!eligble!participants!!
!133!randomly!allocated!
66!allocated!to!
Excision!!
66!received!allocated!
intervention!!
!Followed!up!at!
3!months!n=62!
6!month!n=58!
12!months!n=49!
48!in!Analysis!at!12!months!!
1!missing!baseline!score,16!
lost!to!follow!up,!1!excluded!
for!pain!!score!zero!at!
baseline!
67!allocated!to!
Vap or isati on !
67!received!allocated!
intervention!
Followed!up!at!
3!months!n=67!
6!months!n=61!
12!months!n=53!
47!in!Analysis!at!12!months!
1!missing!baseline!score,!17!
lost!to!follow!up,!1!excluded!
for!pain!score!zero!at!
baseline,1!excluded!for!
histology!showing!foreign!
body!reaction!!
394!excluded!
!(Due!to!current!or!recent!
hormonal!treatment,!not!
wishing!to!participate!or!no!
endometriosis!at!
laparoscopy)!!

97
The study analysis was designed on an “intention to treat basis” but in the
event, all of the patients received the surgical procedure that they were
allocated to. Data for the reasons why eligible patients were excluded from
entering the trial prior to surgery was not collected. However, subjectively, the
majority of patients were not eligible because they had either received hormonal
therapy in the previous 3 months, or were currently receiving it for pain control
or contraception. Additional, though less common reasons, were because they
did not wish to participate or because they were found to not have
endometriosis on visual inspection at laparoscopy.
Interim analysis of recruiting and follow up
Protocol deviations were identified in April 2006 when it was discovered that
some patients were not actively being followed up because they had become
pregnant. An interim analysis of recruiting and follow-up was therefore
undertaken to analyse the effect of this. Dr Haider Jan independently reviewed
the patient folders so that none of the researchers were unblinded. The
completeness of the data and the reasons for failure to follow-up were
reviewed. The findings were as follows:
Overall Result
79 women were in the database, 70 women were at 6 months or more
post treatment. 56 of those women had completed 0,3 and 6 months
questionnaires. Of the 14 who had not completed all questionnaires, 6
had not done so because they had become pregnant and 8 had not done
so because they had failed to attend follow-up appointments.

98
Drop out rate
This reflected an overall drop out rate of 14/70 = 20%. Drop out rate due
to reasons other than pregnancy was 8/70 = 11%. This was generally
due to women not coming to appointments that had been sent to them
through the post.
Trial state
106 women were required in total from the original power calculation. At
the audit point, the trial was 56/106 = 53% complete.
9 women had not completed all questionnaires. Given the drop out rate
of 20%, then 7 of those should complete the trial.
Therefore the current estimated number of trial completions was 63/106
(56+7) or was 59% complete.
Therefore 106-63 more women were needed = 43. Accounting for the
estimated 20% projected drop out seen so far, then 9 patients out of 43
would be expected to drop out.
Therefore the total number required to complete the trial was 52 (43+ 9)
in April 2006. With 79 already recruited, this implied that the total sample
would now need to be 131 (79+52).

99
Recommendations of interim analysis:
1. The follow-up protocol needed to be enhanced to improve the drop out
rate of patients (see clarification letter at Appendix).
2. Complete questionnaires for all those that missed their follow-up at 6
or 12 months for a snapshot of how they are now, no matter the time
post procedure. This should be attempted by telephone if other means
fail as described in the clarification letter.
3. Begin to transfer data from paper questionnaires onto an SPSS
database so an electronic copy is formed.
4. Repeat mail-shot to GPs to improve recruitment.
5. OASIS training for the Nurse Researcher to improve her ability to
independently contact women by telephone or mail.
6. Adjustment of the originally intended sample size to compensate for
the drop out rate from 106 to 132.
7. Re-audit in 9 months.
Final Audit of Database
A 10% audit check of the entire SPSS database was carried out on 17/4/09 to
confirm its accuracy. Incorrect data was found in the HADS 3 month and VAS 6
month files. Instead they contained EHP-30 intercourse module data. Therefore
a previously backed up version of the files was recovered and re-entered into
HADS 3 month and VAS 6 month fields. Following this 10% of cases were
randomly selected by random number generation for assessment on 6/5/09.
The SPSS data for each case was completely checked against the original
paper questionnaires.
The selected cases were 11,17,33,37,48,53,64,67,77,88,102,121,127.

100
The following errors were discovered:
Case 11 – 1 error in HADS 3, 3 missing boxes in EHP-30 12 month.
Case 17 – 1 error EHP-30 6 month, 2 errors HADS 6 month.
Case 102 – 1 error in EHP-30 intercourse module 12 month.
The total number of data boxes per case is 220: (EHP-30=120, Intercourse=20,
VAS=44, HADS=56). Therefore the total number of data boxes in the SPSS
database is 220x133 patients = 29260. The total number of data box errors in
the audit was 8 in 10% of the total dataset. Therefore 80 data box errors were
estimated overall. As a percentage, the overall estimated data box error was
80/29260 = 0.27%. This was judged to be acceptable.
The final SPSS database was then analysed simply to look for outlying results
suggestive of possible data entry errors. Two errors were found from this: Case
62 had missing data for 3, 6, and 12 month EHP-30 Core questions as she had
failed to attend follow up due to pregnancy. This had been entered erroneously
as -9 (not relevant) instead of 99 (missing). A similar problem was found for
case 85 who had missing data that had been entered erroneously as -9 instead
of 99.
Numbers analysed
Recruitment
133 Patients were recruited into the trial between 15 November 2002 and 30
May 2008. Case number 59 was excluded from the trial at the end of her follow-
up period because her histology showed a foreign body reaction (she had had a

101
previous laparoscopy in 1990), suggesting that she did not have endometriosis.
The follow-up period was completed in June 2009. The recruitment rate over
66.5 months was 2.03 patients per month.
Patients who did not attend for follow up
The patients who did not attend for follow-up at each time point were analysed
to ensure that they were not a common subgroup that could bias the results of
the overall study. There were 4 patients who did not attend follow-up at 3
months (2.3%), 14 at 6 months (10.5%) and 31 at 12 months (23.3%).
Therefore 102 (76.7%) were followed up at 12 months.
However, to establish whether there was a bias created by missing variables as
a whole, the total number of missing scores for EHP-30 Core pain domain was
calculated for each time point. This included those with a missing or zero pain
score at baseline, as well as those who completely failure to attend follow-up.
This is displayed in table 16 below:
Table 16. Number of participants missing from EHP-30 Core pain score analysis.
Technique
Pre-op
n/n
3 months
n/n
6 months
n/n
12 months
n/n (%)
Missed >/= 1
follow up
Excision
2/66
5/66
11/66
18/66 (27.3)
19/66
Vaporisation
3/67
8/67
13/67
20/67 (29.9)
24/67
At 12 months there is no difference, in terms of group balance, between the
18/66 (27.3%) for excision and the 20/67 (29.9%) for vaporisation (p=0.742, chi-
squared test). Therefore the number of patients in the main analysis for the
primary outcome was well balanced between the two groups with 95 patients

102
analysed: 48 in the excision arm and 47 in the vaporisation arm. The
characteristics of the 38 patients who had missing or excluded scores at 12
months were analysed to see if they were not representative of the whole
population in any way. As seen in fig 17 below, there were no findings
suggestive that the group of missing or excluded results for EHP-30 Core pain
domain were greatly different from the analysed results.
Table 17. Comparison of baseline variables between patients included in the main analysis, and
missing or excluded patients.
Characteristic Analysed Missing/Excluded
(n=95) (n=38)___________
Mean age +/- SD 33.69+/-7.57 31.18+/-7.90
Median rASRM score (IQ Range) 6 (4-11) 5 (3-8)
Positive Histology n/n (%) 44/55 (80%) 10/12 (83.3%)
Deep disease n/n (%) 52/93 (55.9%) 20/37 (54.1%)
____________________________________________________________________
Other factors that were considered when deciding on numbers analysed
1. Pregnancy, adjuvant hormonal therapy and subsequent surgery.
2. Patients who scored zero for the variable at outset.
1. Pregnancy, adjuvant hormonal therapy and subsequent surgery.
There were 22 pregnancies in 19 patients (3 patients had 2 pregnancies each),
5 patients who underwent further surgery (including one hysterectomy and
bilateral salpingo-oophorectomy, and four laparoscopies), and 16 patients who
had adjuvant hormonal therapy at some point during the 12 month follow-up

103
period. These can be considered to be protocol deviations during the follow-up
period. None of these patients were excluded from the main analysis in order to
produce a result reflecting the natural evolution of the disease following surgery
on an intention to treat basis. However, a “per protocol” sub-analysis of the
main group without these patients was also carried out.
2. Zero scores for outcome modalities pre-op
Some patients had scores of zero for pain or QOL outcomes at the pre-op visit
showing that they did not suffer from that particular symptom as a result of their
endometriosis. These were excluded from the analysis of improvement for that
particular modality as it was felt that they were unable to improve and only able
to stay the same or get worse. The stay the same or get worse patients for each
modality were subsequently analysed separately to ensure that we were not
missing an effect of treatment causing symptoms to deteriorate. Similarly the
same was applied to this group in that the patients who had the worst possible
score at baseline were excluded as they were unable to get any worse.
Summary of numbers analysed at each time point
For the primary outcome measure EHP-30 Core pain score, the follow-up rates
at 3, 6, and 12 months were 129/133 (97.0%), 119/133 (89.5%) and 102/133
(76.7%). Taking further into account missing data and exclusions, the analysis
rates were 122/133 (91.7%), 111/133 (83.5%) and 96/133 (72.2%). Table 18
below summarises the numbers analysed for each outcome measure at the
main analysis points in the trial:

104
Table 18. Numbers analysed at each point of the trial for each outcome measure.
Analysis point: 0 months 3 months 6 months 12 months
Technique: Exc Vap Exc Vap Exc Vap Exc Vap
Outcome n=66 n=67 n=66 n=67 n=66 n=67 n=66 n=67
EHP30
Pain 65 65 62 60 56 55 48 48
Control & powerlessness 64 64 62 61 57 55 47 48
Emotional wellbeing 66 64 62 61 57 56 50 48
Social support 64 65 62 60 56 56 49 49
Self image 66 65 62 62 57 56 50 50
Intercourse 61 54 57 51 53 47 45 44
VAS
Dysmenorrhoea 61 64 62 56 56 51 40 47
Dyspareunia 59 59 57 47 52 45 40 44
CPP 60 58 62 58 55 51 44 45
Dyschezia 62 60 56 57 56 47 45 45
HADS
Anxiety 66 65 62 61 48 50 48 50
Depression 66 63 60 60 49 50 49 50
___________________________________________________________________________
Baseline Data
Baseline data for the 132 patients included in the analysis was calculated for
Age, rASRM score and stage, histology, location and depth of disease.

105
Overall baseline statistics for continuous variables
Age
As fig 10 below shows, the mean age of patients was 32.9 years (+/-7.7
range 19-50) and was normally distributed.
Fig 10. Age distribution of participants.

106
rASRM Score
In fig 11 below, the distribution has a long tail showing that patients in the
study are more frequently found in the lower stages of the disease and
confirms the finding that endometriosis is not normally distributed
according to rASRM score. Therefore, the descriptive analysis for rASRM
score was calculated by median and interquartile range. Median rASRM
score was 6 (interquartile range 3-10).
Fig 11. Distribution of participants by rASRM score.

107
Overall baseline statistics for discrete variables
rASRM Stage
Fig 12 below shows that 87.9% patients are equally distributed between
rASRM Stages 1 and 2. The remaining 12.1% fall into stage 3. This
reflects the natural distribution of the disease between stages (Redwine,
1990a, 1994).
Fig 12. Distribution of participants by rASRM stage.

108
Histology
Histology was taken in 65 of 133 cases (48.9%), 49 from the excision
group and 16 from the vaporisation group. It was not always possible to
get histology from the vaporisation group as biopsy may have resulted in
excision of all or most of some lesions. Taking histology was not started
till case 46 in June 2005. Overall 54 of the 65 cases had histology
positive for endometriosis, showing a successful correlation between
visual inspection and histological analysis in 83.1% of cases.
Predominance of left sided disease
The number of times that left sided and right sided disease was recorded
in any location was analysed. There were three possible locations on
each side of the pelvis where disease could be recorded for the rASRM
score; ovary, side-wall and uterosacral ligament. It was found that there
was more left sided disease overall in the trial population in that there
were 175 areas of left sided disease v 125 areas of right sided disease
(p=0.0005 paired t-test) as one would expect from previous observers
(Vercellini et al., 2004).
Did the patients with deep disease have a higher starting score
The baseline statistics were analysed to see whether or not deep
disease had a higher mean baseline score (table 19 below), suggesting
they suffer from greater levels of pain compared with superficial disease:

109
Table 19. Comparison of mean starting scores for pain outcomes for superficial and deep
disease.
Outcome Superficial Deep *p value
Mean +/-SD (n=66) (n=66)__________ ___________
EHP30 pain 40.47+/-33.49 41.10 +/-22.28 p=0.899
Dysmenorrhoea 6.53 +/-2.69 6.65 +/- 2.71 p=0.809
Dyspareunia 4.15 +/-3.35 4.71 +/-3.20 p=0.36
CPP 4.77 +/-3.31 4.93 +/-3.17 p=0.785
Dyschezia 2.88 +/-3.44 3.29 +/-3.22 p=0.500
________________________________________________________________________
*unpaired t test
Comparative baseline statistics for patient characteristics
Table 20 below summarises the comparative statistics between the two
treatment groups for general baseline characteristics:
Table 20. Comparison of mean baseline variables for excision and vaporisation.
Characteristic Excision Vaporisation
(n=66) (n=66)______
Mean age +/- SD 33.05+/-6.69 32.74+/-8.65
Median rASRM score (IQ Range) 6(4-10) 6(3-9)
Positive Histology n/n (%) 41/49 (83.7) 13/17 (76.5%)
Deep disease n (%) 44 (66.7%) 28 (43.8%)
________________________________________________________________

110
The CONSORT Statement warns against using statistical tests to adjust for
baseline variable differences as this can bias the estimated treatment effect
(Altman et al., 2001). Deep disease has been claimed to show a greater extent
of improvement after surgical treatment than superficial disease (Banerjee et
al., 2008) and so imbalanced groups for depth of disease may bias the results.
By chance there were 44 (66.7%) deep cases in the excision group compared
with 28 (43.8%) in the vaporisation group (p=0.009, Chi-sq test).
However, no attempt was made to adjust for this difference in depth as this may
have resulted in a bias in the estimated treatment effect. It has already been
shown above that women with deep disease do not start on a significantly
higher baseline pain score (table 20) and therefore uneven distribution in
disease depth between the groups does not affect the baseline primary
outcome measure of pain. To decide whether imbalanced depth of disease
between the groups has biased outcomes, depth was sub analysed later in this
chapter.
Comparative baseline statistics for outcome measures
Firstly, the data were analysed to see whether a higher baseline pain score
resulted in a greater probability of a larger fall in score for the primary outcome
measure EHP-30 Core pain domain if the patient improved. The implication of
this is that if the groups are not balanced in baseline scores then the one with
the significantly higher score has a greater chance of showing more
improvement. There is a highly significant positive correlation between baseline
pain score and amount of improvement in pain for EHP30 Core pain score
(Pearson’s Correlation test r=+0.535, p<0.0005). However in the table 21

111
below, the groups are well balanced for mean starting scores so this effect is
not likely to affect the results in this study:
Table 21. Comparison of mean baseline outcome scores for excision and vaporisation.
Outcome Excision Vaporisation
Mean +/-SD (n=66) (n=66)______
EHP30
Pain 42.3 +/-21.6 43.3 +/-21.7
Control & powerlessness 51.0 +/-27.3 53.9 +/-21.3
Emotional wellbeing 48.6 +/-21.4 47.3 +/-19.6
Social support 40.4 +/-30.9 41.4 +/-24.8
Self image 33.7 +/-28.2 38.1 +/-26.8
Intercourse 48.9 +/-29.2 55.7 +/-28.6
VAS
Dysmenorrhoea 6.8 +/-2.8 6.5 +/-2.6
Dyspareunia 4.2 +/-3.3 4.8 +/-3.3
CPP 4.8 +/-3.3 4.9 +/-3.1
Dyschezia 3.4 +/-3.5 2.8 +/-3.1
HADS
Anxiety 9.6 +/-4.2 9.6 +/-4.0
Depression 4.1 +/-3.3 4.5 +/-3.7
_______________________________________________________________________

112
Outcomes and Estimation
Primary outcome measure: EHP30 Core Pain
In the whole study group 76 (80%) patients improved and 19 (20%) patients
stayed the same or got worse at 12 months.
Comparative pain was analysed using two outcomes:
1. The proportion of patients who improved or “stayed the same or got
worse”, to see if one technique was managing to improve a greater
number of people.
2. The extent of improvement for each technique to see if one technique
gave a greater amount of improvement.
Proportional improvement in primary outcome measure
Table 22 below shows the proportion of patients showing improvement for the
primary outcome variable EHP-30 Core pain score at all the analysis time
points:
Table 22. Proportion of patients showing improvement for EHP-30 Core pain score.
Analysis point Excision Vaporisation Difference p value*
n/N (%) n/N (%) % ________________________
Improved
3 months 46/62 (74.2) 44/60 (73.3) 0.9 0.954
6 months 41/56 (73.2) 40/55 (72.7) 0.5 0.954
12 months 41/48 (85.4) 35/48 (72.9) 12.5 0.132
____________________________________________________________________________
*Chi-square test

113
Both forms of surgical treatment are effective in reducing endometriosis-
associated pain to any degree at 12 months (85.4% for excision against 72.9%
for vaporisation). Comparing the two treatment arms against each other does
not produce statistical significance (p=0.132) showing that both treatment
modalities result in improvement in line with that found by previous observers
(Sutton et al., 1994, Abbott et al., 2004).
Extent of improvement in primary outcome measure
Table 23 below shows that both treatment modalities alone result in significant
improvement in absolute pain score at 12 months compared with baseline:
Table 23. Extent of improvement in EHP-30 Core pain score against baseline for excision and
vaporisation alone at 12 months.
Mean change 0 v 12 months +/- SD *p value (95%CI)__________
Excision -23.9 +/-26.2 p<0.0005 (-31.5 to -16.3)
Vaporisation -10.7 +/-20.8 p=0.001 (-16.7 to -4.7)
____________________________________________________________________________
*paired t test
Comparative extent of score improvement in primary outcome measure
Table 24 below summarises the changes in EHP-30 Core pain score (score at x
months – score at baseline) for each of the analysis points. Since a higher
score on the EHP-30 Core pain scale represents a higher level of pain, a
negative change indicates an improvement (nb: this rule applies to score results
for all questionnaires in the analysis):

114
Table 24. Comparative extent of improvement in EHP-30 Core pain score against baseline for
excision and vaporisation at all follow-up points.
Analysis point Excision Vaporisation Difference p value*
mean score improvement +/-SD mean (95% CI)____________
3 months -15.8 +/-21.1 -11.0 +/-15.8 -4.7(-11.5 to 2.0) 0.168
6 months -14.4 +/-23.3 -12.0 +/-17.9 -2.4(-10.3 to 5.5) 0.544
12 months -23.9 +/-26.2 -10.7 +/-20.8 -13.2 (-22.8 to -3.5) 0.008
____________________________________________________________________________
*unpaired t test
Excision exhibits a greater improvement in pain score than vaporisation at 3, 6
and 12 months and reaches statistical significance at 12 months with a greater
improvement by -13.2 points (p=0.008, 95%CI -22.8 to -3.5, unpaired t-test).

115
As can be seen in the graph in fig 13 below, over time the two treatment
modalities appear to be fairly equal up to 6 months. The significant difference
between excision and vaporisation appears to develop between 6 and 12
months.
Fig 13. Graph of mean improvement in EHP-30 Core pain score against time for excision and
vaporisation.
Overall 14.6% of the excision group and 27.2% of the vaporisation group were
“the same or worse” at 12 months, however there is not a statistically significant
difference between the two groups (p=0.132, Chi-square test).

116
Secondary outcome measures
Extent of score improvement in VAS
VAS pain scores were analysed independently for each of the 4 symptoms of
dysmenorrhoea, dyspareunia, CPP and dyschezia. As both treatment arms,
and superficial and deep disease, started with similar pre-op scores, the results
are given as absolute changes in pain score out of 10.
Individual improvement in VAS pain modalities, by treatment arm, comparing
baseline against 12 month follow up results (score at 12 months – score at
baseline), are shown in tables 25 and 26 below.
Excision
Table 25. Mean improvement in VAS symptom scores for excision at 12 months against
baseline.
Mean change 0 v 12 months +/- SD *p value (95%CI)___________
Dysmenorrhoea -2.94 +/-3.65 <0.0005 (-4.22 to -1.67)
Dyspareunia -1.98 +/-3.34 0.006 (-3.33 to -0.63)
CPP -3.50 +/-3.62 <0.0005 (-4.91 to -2.10)
Dyschezia -2.73 +/-4.18 0.007 (-4.64 to -0.83)
____________________________________________________________________________
*paired t test

117
Vaporisation
Table 26. Mean improvement in VAS symptom scores for vaporisation at 12 months against
baseline.
Mean change 0 v 12 months +/- SD *p value (95%CI)___________
Dysmenorrhoea -1.50 +/-2.82 0.001 (-2.40 to -0.61)
Dyspareunia -1.27 +/-3.42 0.040 (-2.49 to -0.06)
CPP -0.93 +/-2.78 0.060 (-1.90 to 0.04)
Dyschezia -1.11 +/-3.42 0.143 (-2.62 to 0.41)
*paired t test
There are statistically significant improvements in all VAS pain score symptoms
in the excision arm at 12 months compared with baseline. For vaporisation,
there are statistically significant improvements in dysmenorrhoea and
dyspareunia.

118
Comparative improvement for VAS pain modalities
Table 27 below summarises the extent of comparative improvement by
treatment arm in VAS for dysmenorrhoea, dyspareunia, CPP and dyschezia for
each of the trial follow up points:
Table 27. Extent of comparative improvement for VAS symptom scores for excision and
vaporisation at all follow up points.
Analysis point Excision Vaporisation Difference p value*
mean score improvement +/-SD mean (95%CI)_______________________
Dysmenorrhoea
3 months -1.26 +/-2.85 -1.33 +/-2.68 0.07(-1.02 to 1.15) 0.904
6 months -1.48 +/-2.80 -1.60 +/-3.01 0.12(-1.08 to 1.32) 0.841
12 months -2.94 +/-3.65 -1.50 +/-2.82 -1.44(-2.93 to 0.05) 0.059
CPP
3 months -1.64 +/-3.05 -1.31 +/-2.83 -0.33(-1.61 to 0.95) 0.612
6 months -2.38 +/-3.10 -0.99 +/-2.37 -1.38(-2.67 to -0.10) 0.035
12 months -3.50 +/-3.62 -0.93 +/-2.78 -2.57(-4.20 to -0.95) 0.002
Dyspareunia
3 months -2.42 +/-3.29 -1.97 +/-3.53 -0.45(-2.00 to 1.10) 0.564
6 months -1.91 +/-3.07 -2.36 +/-3.64 0.44(-1.17 to 2.05) 0.584
12 months -1.98 +/-3.34 -1.27 +/-3.42 -0.71(-2.49 to 1.07) 0.429
Dyschezia
3 months -1.58 +/-3.09 -2.32 +/-3.54 0.73(-0.96 to 2.43) 0.391
6 months -1.87 +/-2.99 -1.74 +/-3.10 -0.14(-1.78 to 1.50) 0.866
12 months -2.73 +/-4.18 -1.11 +/-3.42 -1.62(-3.97 to 0.72) 0.170
*unpaired t test
In terms of the extent of score improvement, there was no significant difference
between excision and vaporisation for dyspareunia and dyschezia. For
dysmenorrhoea, there was an indication of excision being better than

119
vaporisation at 12 months but this did not reach statistical significance
(p=0.059). However, there was a significantly greater improvement in CPP in
the excision group at 6 and 12 months compared with vaporisation (p=0.035
and p=0.002).

120
Proportion of improvers for VAS pain scores
Table 28 below shows the comparative proportion between treatment arms of
patients showing improvement for the 4 modalities of pain measured by VAS at
all the analysis time points:
Table 28. Comparative proportion of improvement for VAS symptom scores for excision and
vaporisation at all follow up points.
Analysis point Excision Vaporisation Difference p value*
n/n (%) n/n (%) % _______________________
Dysmenorrhoea
3 months 31/52(59.6) 32/51(62.7) -2.9 0.745
6 months 27/47(57.4) 27/46(58.7) -1.3 0.903
12 months 26/34(76.5) 26/41(63.4) 13.1 0.222
Dyspareunia
3 months 27/39(69.2) 25/38(65.8) 3.4 0.747
6 months 24/34(70.6) 23/36(63.9) 6.7 0.551
12 months 18/26(69.2) 19/33(57.6) 11.6 0.358
CPP
3 months 27/43(62.8) 26/41(63.4) -0.6 0.953
6 months 27/38(71.1) 18/35(51.4) 19.7 0.085
12 months 22/28(78.6) 20/34(58.8) 19.8 0.098
Dyschezia
3 months 22/32(68.8) 23/29(79.3) -10.5 0.349
6 months 21/31(67.7) 19/25(76.0) -8.3 0.496
12 months 14/21(66.7) 14/22(63.6) 3.1 0.835
____________________________________________________________________________
*Chi-square test

121
There was no significant difference between excision and vaporisation in terms
of the proportion of patients who improved at each follow up point.
Individual extent of score improvement in EHP-30 HRQoL measures
EHP-30 HRQoL scores were analysed independently for each of the domains.
As both treatment arms, and superficial and deep disease, started with similar
pre-op scores, the results are given as absolute changes in score out of 100.
Individual improvement in EHP-30 HRQoL domains, by treatment arm,
comparing baseline against 12 month follow up results (score at 12 months –
score at baseline), are shown in tables 29 and 30 below.
Excision
Table 29. Extent of improvement in EHP-30 HRQoL outcome measures for excision alone at
baseline against 12 months.
Mean change 0 v 12 months +/- SD *p value (95%CI)___________
Control & Powerlessness -30.56 +/-29.56 <0.0005 (-39.77 to -21.34)
Emotional wellbeing -24.74 +/-20.65 <0.0005 (-30.68 to -18.81)
Social support -31.25 +/-30.12 <0.0005 (-41.29 to -21.21)
Self image -21.37 +/-25.13 <0.0005 (-29.51 to -13.22)
Intercourse -21.11 +/-32.86 <0.0005 (-31.91 to -10.31)
____________________________________________________________________________
*paired t test

122
Vaporisation
Table 30. Extent of improvement in EHP-30 HRQoL outcome measures for vaporisation alone
at baseline against 12 months.
Mean change 0 v 12 months +/- SD *p value (95%CI)___________
Control & Powerlessness -12.86 +/-28.74 0.004 (-21.40 to -4.33)
Emotional wellbeing -13.95 +/-23.82 <0.0005 (-21.02 to -6.88)
Social support -12.35 +/-28.91 0.008 (-21.36 to -3.34)
Self image -5.75 +/-28.24 0.194 (-14.56 to 3.05)
Intercourse -22.42 +/-30.18 <0.0005 (-32.07 to -12.76)
____________________________________________________________________________
*paired t test
There were highly significant improvements in all domains at 12 months
compared with baseline for excision patients. This was also the case with
vaporisation other than for self-image.

123
Comparative extent of score improvement in EHP-30 HRQoL measures
Table 31 below summarises the extent of comparative improvement by
treatment arm in EHP-30 HRQoL domains for each of the trial follow up points:
Table 31. Comparative extent of score improvement in EHP-30 HRQoL outcome measures for
excision and vaporisation at all follow up points.
Analysis point Excision Vaporisation Difference p value*
mean score improvement (+/-SD) mean(95%CI)_______
Control & powerlessness
3 months -22.73(+/-23.87) -11.94(+/-23.55) -10.80(-19.52 to -2.08) 0.0166
6 months -18.75(+/-25.86) -11.48(+/-28.82) -7.27(-17.88 to 3.33) 0.177
12 months -30.56(+/-29.56) -12.86(+/-28.74) -17.69(-30.05 to -5.33) 0.006
Emotional wellbeing
3 months -16.53(+/-22.13) -9.04(+/-20.03) -7.49(-15.13 to 0.15) 0.055
6 months -17.19(+/-20.58) -10.11(+/-22.91) -7.08(-15.31 to 1.15) 0.091
12 months -24.74(+/-20.65) -13.95(+/-23.82) -10.80(-19.86 to -1.73) 0.020
Social support
3 months -16.41(+/-28.90) -2.55 (+/-24.75) -13.86(-24.40 to -3.32) 0.010
6 months -16.62 (+/-31.87) -2.17(+/-29.77) -14.45(-27.15 to -1.75) 0.026
12 months -31.25(+/-30.12) -12.35(+/-28.91) -18.90(-32.13 to -5.66) 0.006
Self image
3 months -4.83(+/-28.08) -4.17(+/-25.07) -0.67(-11.00 to 9.67) 0.898
6 months -9.63(+/-29.46) -5.27(+/-29.94) -4.36(-16.12 to 7.40) 0.464
12 months -21.37(+/-25.13) -5.75(+/-28.24) -15.61(-27.47 to -3.75) 0.011
Intercourse
3 months -20.44(+/-27.48) -14.33(+/-29.83) -6.11(-17.79 to 5.56) 0.301
6 months -15.77(+/-26.52) -14.41(+/-27.09) -1.36(-12.72 to 10.01) 0.813
12 months -21.11(+/-32.86) -22.42(+/-30.18) 1.31(-12.91 to 15.53) 0.855
*Unpaired t test

124
At 12 months excision was statistically significantly better than vaporisation at
improving scores in all of the EHP-30 HRQoL domains other than for sexual
intercourse where no difference was found.
Change in Anxiety & Depression Categories between baseline and 12 months
Fig 14 below shows the number of patients falling into the different categories of
normal (0-7 points), mild (8-10 points), moderate (11-14 points) or severe (15-
21 points) for anxiety and depression at baseline compared with 12 months for
all patients not differentiating by treatment arm.
Fig 14. Distribution of participants for different grades of anxiety at baseline and 12 months.
In the anxiety graphs comparing baseline and 12 months for anxiety, visually
there appears to be a shift of patients to the left, or more normal side of the
graph at 12 months.

125
Fig 15. Distribution of participants for different grades of depression at baseline and 12 months.
This shift is not apparent in the depression graph as the majority of people were
not depressed at the beginning and the two graphs remain very similar.
Individual extent of score improvement in HADS for treatment arms
HADS scores for anxiety and depression were analysed independently for
changes. As both treatment arms, and superficial and deep disease, started
with similar pre-op scores, the results are given as absolute changes in score
out of 21.
Individual improvement in EHP-30 HRQoL domains, by treatment arm,
comparing baseline against 12 month follow up results (score at 12 months –
score at baseline), are shown in tables 32 and 33 below.

126
Excision
Table 32. Extent of improvement in anxiety and depression for excision alone at 12 months
versus baseline.
Mean change 0 v 12 months +/- SD *p value (95%CI)__________
Anxiety -2.88 +/-4.84 <0.0005 (-4.28 to -1.47)
Depression -0.96 +/-3.46 0.068 (-1.99 to 0.07)
____________________________________________________________________________
*paired t test
Vaporisation
Table 33. Extent of improvement in anxiety and depression for vaporisation alone at 12 months
versus baseline.
Mean change 0 v 12 months +/- SD *p value (95%CI)___________
Anxiety -1.64 +/-3.27 0.001 (-2.57 to -0.71)
Depression -0.66 +/-3.21 0.165 (-1.60 to 0.28)
____________________________________________________________________________
*paired t test
There was a significant improvement in anxiety levels for both excision and
vaporisation at 12 months compared with baseline, but not for depression.

127
Comparative extent of score improvement in HADS for treatment arms
Table 34 below summarises the comparative extent of improvement for excision
and vaporisation in scores for anxiety and depression:
Table 34. Comparative extent of improvement in anxiety and depression for excision and
vaporisation at all follow up points.
Analysis point Excision Vaporisation Difference p value*
mean score improvement +/-SD mean(95%CI)_____________
Anxiety
3 months -1.94(+/-3.68) -0.80(+/-3.35) -1.13(-2.39 to 0.13) 0.077
6 months -2.96(+/-4.89) -1.64(+/-3.27) -1.32(-2.98 to 0.34) 0.119
12 months -2.88(+/-4.84) -1.64(+/-3.27) -1.24(-2.89 to 0.42) 0.141
Depression
3 months -0.74(+/-3.46) -0.61(+/-3.07) -0.13(-1.35 to 1.09) 0.833
6 months -1.02(+/-3.52) -0.66(+/-3.21) -0.36(-1.75 to 1.02) 0.605
12 months -0.96(+/-3.46) -0.66(+/-3.21) -0.30(-1.67 to 1.08) 0.669
*unpaired t test
Comparatively, there was no difference between excision and vaporisation for
improvement of anxiety.

128
Exploratory Sub analysis
Linear regression analysis to investigate possible predictors of improvement in
primary outcome measure EHP-30 Core pain score
Backward stepwise linear regression analysis was performed to investigate
whether any of the demographic factors (technique, age, surgeon, rASRM
score, depth of disease, location of disease) influence change in EHP-30 Core
pain score at 12 months over baseline. Three statistically significant factors
were identified as reducing improvement in pain (r squared for final model =
0.154):
1. Vaporisation (p=0.012)
estimated loss of pain reduction = 12.24/100 (95%CI 2.81 to 21.67)
2. Being operated on by Surgeon 1 (p=0.035)
estimated loss of pain reduction = 10.19/100 (95%CI 0.76 to 19.62)
3. Patient being older (p=0.035)
estimated loss of pain reduction = 0.68/100 per year (95%CI 0.047 to
1.306)
Linear regression analysis to investigate possible predictors of CPP
improvement
CPP was found to have statistically significantly improved to a greater extent
than excision in the main analysis (table 28). Therefore backward stepwise
linear regression analysis was performed to investigate whether any of the
demographic factors (technique, age, surgeon, rASRM score, depth of disease,
location of disease) influence change in CPP at 12 months over baseline. There

129
were two statistically significant factors identified as influencing improvement in
CPP (r squared for final model = 0.195):
1. Vaporisation (p=0.014)
estimated loss of pain reduction = 0.27/10 (95%CI 0.06 to
0.48)
2. Disease in uterovesical fold (p=0.027)
estimated gain in pain reduction for superficial disease
compared with no disease in uterovesical fold = -0.31/10
(95%CI -0.58 to -0.04)
estimated gain in pain reduction for deep disease compared
with no disease in uterovesical fold = -0.62/10 (95%CI -1.16 to
-0.08)
Depth and rASRM Stage of disease
Sub analysis of deep against superficial disease for each treatment modality at
12 months for the primary outcome measure, EHP Core pain score, was
performed. Excision results in significant improvement in pain score for both
deep (mean point reduction of -23.86 out of 100, +/-26.59, p<0.0005, paired t
test) and superficial disease (mean point reduction of -23.99, +/-26.36, p=0.001,
paired t test). Whilst vaporisation shows significant pain score improvement for
superficial disease (mean point reduction of -17.89 out of 100, +/-19.78,
p<0.0005, paired t test), it fails to result in significant improvement for deep
disease (mean point reduction of -4.44, +/-18.07, p=0.262, paired t test).

130
On direct comparison between the two treatment arms, as seen in table 35
below, there is no significant difference in outcome for the treatment of
superficial disease. However, excision performs equally well for deep disease
as it does for superficial, whereas vaporisation performs significantly worse than
excision for the treatment of deep disease.
Table 35. Comparison of the extent of improvement in EHP-30 Core pain score for excision and
vaporisation with deep and superficial disease at 12 months against baseline.
Excision Vaporisation Difference p value*
mean score improvement +/-SD mean (95%CI) ____________
Deep disease -23.86+/-26.59 -4.44+/-18.07 -19.42(-32.61 to -6.23) 0.005
Superficial disease -23.99+/-26.36 -17.89 +/-19.78 -6.10(-20.67 to 8.46) 0.402
____________________________________________________________________________
*unpaired t test

131
Is deep disease or more severe disease (by rASRM stage) more likely to
improve EHP-30 Core pain score at 12 months versus baseline?
The results in tables 36 and 37 below show a trend towards a greater chance of
improvement at 6 and 12 months if endometriosis at baseline was superficial
rather than deep, and a statistically significant chance of greater improvement
for patients with a lower rASRM stage.
Table 36. Comparative proportional improvement of patients with deep disease against those
with superficial disease at 6 and 12 months.
6 months 12 months
n/n (%) n/n (%) ___________
Excision
Deep 25/38 (65.8) 25/30 (83.3)
Superficial 16/18 (88.9) 16/18 (88.9)
*p value 0.106 0.696
Vaporisation
Deep 19/27 (70.4) 15/23 (65.2)
Superficial 20/26 (76.9) 19/23 (82.6)
*p value 0.589 0.179
Overall
Deep 44/65 (67.7) 40/53 (75.5)
Superficial 36/44 (81.8) 35/41 (85.4)
*p value 0.102 0.236
_______________________________________________________________________
*Chi-square test

132
Table 37. Comparative proportional improvement of patients with different rASRM stages at 6
and 12 months.
6 months 12 months
n/n (%) n/n (%) ___________
Excision
rASRM 1 19/25 (76) 20/22 (90.9)
rASRM 2 18/25 (72) 18/21 (85.7)
rASRM 3 4/6 (66.7) 3/5 (60)
*p value 0.631 0.170
Vaporisation
rASRM 1 21/27 (77.8) 18/21 (85.7)
rASRM 2 17/21 (81.0) 15/21 (71.4)
rASRM 3 2/7 (28.6) 2/6 (33.3)
*p value 0.114 0.024
Overall
rASRM 1 40/52 (76.9) 38/43 (88.4)
rASRM 2 35/46 (76.1) 33/42 (78.6)
rASRM 3 6/13 (46.2) 5/11 (45.5)
*p value 0.137 0.008
________________________________________________________________________
*Mann-Whitney U test
Per protocol analysis for primary outcome measure EHP-30 Core pain score
Analysis was carried out on the sub sample whose benefits appear to be solely
related to surgery, excluding the patients who became pregnant during follow
up, took hormonal medication for pain or contraception in the follow up period,
or had subsequent surgery within the follow up period. There were still 35
patients in each group at 12 months, and still a difference was found of -21.04

133
(+/-23.97) for excision against -9.42 (+/-19.36) for vaporisation (p=0.029, 95%CI
-22.02 to -1.23). The results are shown in table 38 below.
Table 38. Extent of improvement in EHP-30 Core pain score for “per protocol” patients for
excision and vaporisation at all follow up points.
Analysis point Excision Vaporisation Difference p value*
mean score improvement +/-SD mean (95% CI)__________________
3 months -14.64(+/-21.12) -10.96(+/-16.06) -3.68 (-10.64 to 3.28) 0.297
6 months -12.75(+/-22.20) -11.50(+/-17.41) -1.25 (-9.08 to 6.59) 0.753
12 months -21.04(+/-23.97) -9.42(+/-19.36) -11.62 (-22.02 to -1.23) 0.029
____________________________________________________________________________
*unpaired t test
Was there a placebo effect in the “same or got worse” subgroup
Fig 16 below shows a graph for patients who at 12 months “stayed the same or
got worse” for EHP-30 Core pain score over time. Interestingly these patients
appear to improve initially at 3 months and recover to baseline at 6 months
before then deteriorating. This may result from an initial placebo effect following
treatment rather than a positive effect of the treatment itself. However, in this
small sub group of patients (n=20), the only statistically significant difference
was observed between 3 months and 12 months. In order to identify a
statistically significant difference over the shorter time span of 3 months, from 0-
3 months and 3-6 months, given the level of variation shown by the confidence
limit intervals in the graph below accompanying this small sub group sample
size, the underlying improvement at 3 months would have had to be extremely
dramatic.

134
Fig 16. Extent of improvement in EHP-30 Core pain score for “same or worse” pateints at 12
months over time.
Adverse events
There were no major surgical complications reported in any of the trial patients
in either group.

135
Discussion
Synopsis of the key findings
At the outset of this trial it was a firmly held belief that there would be no
difference between excision and vaporisation for the treatment of minimal to
moderate endometriosis. The major concern was that the sample size would be
too small to show up the suspected small difference that might be found
between the two treatment arms, and the bioequivalence power calculation
allowed for this in that, with sufficient patients recruited, it would be possible to
claim that one treatment was no worse than the other. The unexpected result,
that excision performed better than vaporisation in this trial, was consequently a
surprise that prompted an analysis that was often directed towards finding
statistical evidence that the outcome was incorrect.
Primary outcome measure – EHP-30 Core pain score
For the primary outcome measure, EHP-30 Core pain score, there was a
significant improvement in score at 12 months versus baseline whether excision
or vaporisation was used as the treatment, where excision showed a mean -
23.9 drop (p<0.0005), and vaporisation a mean -10.7 drop on the 100 point
scale (p=0.001). Also, there was no significant difference in the proportion of
patients who improved in each treatment arm (85.4% for excision and 72.9% for
vaporisation, p=0.132). However, at 12 months, the extent of score
improvement above was statistically significantly greater in the excision versus
the vaporisation group (p=0.008, 95%CI -22.8 to -3.5). In addition, the excision

136
group appear to be still improving at 12 months, whereas the vaporisation group
improvement did not continue after 6 months.
In terms of the treatment of deep and superficial disease, excision resulted in
statistically significant improvement in pain score at 12 months versus baseline
in both superficial (mean -23.99 +/-26.36, p<0.001) and deep disease (mean -
23.86 +/-26.59, p<0.0005). However, vaporisation only showed statistically
significant improvement for superficial disease (mean -17.89 +/-19.78,
p<0.0005) and not for deep (mean -4.44 +/-18.07, p=0.262). Comparatively,
there was no difference in the extent of improvement between excision and
vaporisation for treating superficial disease (mean difference -6.10, p=0.402).
However, the improvement in pain score for excision was highly statistically
significantly better than vaporisation for the treatment of deep disease (mean
difference -19.42, p=0.005).
Of interest is that 14.6% of the excision group and 27.2% of the vaporisation
group were either the same or worse at 12 months compared with baseline.
The comparative difference between excision and vaporisation does not reach
statistical significance (p=0.132). Those patients who stayed the same or got
worse appear to initially improve possibly showing a placebo effect, though it is
not statistically possible to show this in this study.
VAS scores for Dysmenorrhoea, Dyspareunia, CPP and Dyschezia
Excision resulted in a statistically significant score improvement in all four pain
modalities at 12 months versus baseline scores. This was true for vaporisation
for dysmenorrhoea (p=0.001) and dyspareunia (p=0.040), but not for CPP

137
(p=0.060) and dyschezia (0.143). However, the proportion of improvers was not
statistically different for the treatment arms at any time point in the trial. The
extent of improvement in CPP in the excision group versus vaporisation was
statistically significant at 6 (p=0.035) and 12 months (p=0.002). As this trial
included only stage 1-3 rASRM scores one would expect dyschezia to be a less
significant symptom in this group of patients and even question whether it is an
appropriate question to include. In support of this Table 21 shows that
dyschezia is the lowest scored symptom at baseline and appears less of a
problem than dysmenorrhoea (the most significant symptom), CPP and
dyspareunia. However patients still scored dyschezia between 2.8-3.4 out of 10
at baseline and it is interesting to note that excision still resulted in a statistically
significant improvement in dyschezia in these stage 1-3 patients (p=0.007). The
conclusion is that it was correct to include it as a symptom in the VAS pain
scores.
EHP-30 HRQoL outcomes
Excision resulted in highly statistically significant improvements in all EHP-30
HRQoL domains at 12 months versus baseline (all p values <0.0005). Results
for vaporisation were also statistically significant other than for self-image
(0.194). However, when considering the extent of improvement, excision
resulted in statistically significantly greater improvement than vaporisation at 12
months for all domains other than the intercourse module, for which there was
no significant difference (p=0.855). For the social support domain, excision was
significantly better at all follow up points in the trial. At no follow-up point, for
any domain, was vaporisation significantly better than excision.

138
HADS Anxiety and Depression
For the sample as a whole, patients did not generally appear to be depressed at
baseline, with the majority of patients falling into the normal category. For
anxiety though, there was a spread of patients throughout the categories with
most falling into the normal to moderate categories. At 12 months less patients
appear to be anxious, with a shift into the normal and less severe groups. The
lack of depression remains similar. The extent of improvement in anxiety was
statistically significant for both excision (p=<0.0005) and for vaporisation
(p=0.001). However, there was no difference in the extent of change in anxiety
(p=0.141) or depressive scores (p=0.669) comparing between excision and
vaporisation.
Summary
Contrary to expectation, for the primary outcome, VAS scores and HRQoL
outcomes, whilst both techniques resulted in significant improvement in scores
at 12 months, excision outperformed vaporisation statistically at 12 months in
the extent of improvement for all outcome measures, suggesting that it is the
best technique for the surgical treatment of minimal to moderate endometriosis.
However it should be noted that the absolute improvement was no better than a
mean -23.9 point drop on the 100 point EHP-30 Core pain scale at 12 months
(from a baseline mean of 42.3 to 18.4, or a 56.5% drop in pain score), and a
concerning number of patients reported being either the same or worse.
Possible mechanisms and explanations
The primary outcome for this study was pain as measured by EHP-30 Core pain
score. This was selected as it was judged to be the most sensitive measure and

139
the one that would relate most directly to the impact of pain on women’s quality
of life. Based on this outcome measure the data show that excision is better
than vaporisation at reducing pain at 12 months post treatment. It has been
argued that ablation/vaporisation does not remove all of the disease particularly
with superficial ablation (Wright et al., 2005). Wright theorised in his randomised
trial of excision versus ablation that ablative techniques have the potential
disadvantage of leaving a greater area of necrotic tissue behind, with increased
inflammatory action and a higher propensity to develop adhesive disease.
In our study laser vaporisation was used to destroy all visible lesions down to
normal tissue until it looked like an excision has been carried out. This can be
done quickly and efficiently for widespread superficial disease and appears to
be a useful technique. However our study shows that even this technique of
vaporisation performs statistically significantly worse than excision. The reasons
are not clear although several possible mechanisms can be suggested.
The implication of the results is that in some way the operative field is not the
same for each treatment at the end of the operation despite it visually appearing
so. It may be that there is more residual disease in the vaporisation cases than
the excision ones. There are several possible ways to explain this. It may be
that by vaporising a lesion, margins may be left that are not easily visible and
are missed by the surgeon. In contrast, excision aims to remain in the normal
tissue margin from the beginning of the dissection and so may be better at
clearing marginal disease. Alternatively, possibly the power of the laser beam is
driving off active endometriotic tissue particles into adjacent areas that then
implant and recur. Full vaporisation of deep lesions is technically more difficult

140
as an eschar tends to build up on the lesion surface that has to be washed
away to continue the vaporisation of the lesion. It is easier to imagine
incomplete vaporisation with these deep lesions.
Our study also shows that excision is significantly better for treating deep
disease than vaporisation. Vaporisation is not only significantly comparatively
worse than excision, but also gave insignificant improvement in EHP-30 Core
pain score at 12 months versus baseline for deep disease. This could have
been explained by the fact that there were more deep cases in the excision
group, if it had also been found that patients with deep disease had a higher
mean starting score, and that a higher mean starting score resulted in a greater
chance of having a larger score reduction. Both of these possibilities were
tested for. It was indeed found that having a higher starting score resulted in a
greater probability of a larger fall in score at 12 months for EHP-30 Core pain
score (Pearson’s Correlation test r=+0.535, p=<0.0005). However, mean scores
at baseline were the same for superficial and deep disease for EHP-30 Core
pain score and VAS scores, and also for excision and vaporisation arms for all
outcome measures at baseline. This also fits with the finding that the extent or
depth of the disease does not correlate with pain (Crosignani et al., 1996, Garry
et al., 2000) Therefore the possible bias resulting from unbalanced baseline
scores did not occur in favour of the excision group. Furthermore, the superficial
group had an advantage by having less deep disease for two reasons. Firstly it
performed significantly worse for deep disease and secondly, superficial
disease was more likely to improve overall in our study.

141
Backward stepwise linear regression analysis was also carried out on the EHP-
30 Core pain score results to see whether technique, age, surgeon, rASRM
score and depth or location of disease influenced the change in EHP-30 Core
pain score at 12 months over baseline. This confirmed the finding of
vaporisation as a significant factor in reducing pain improvement (p=0.012,
95%CI 2.81 to 21.67). It has already been shown above that a sample with a
lower mean baseline score has less capacity to improve. Mean baseline scores
for instrument and technique use were similar and so the finding that
vaporisation is genuinely performing worse was reinforced.
However, it was hypothesised that a significantly lower mean baseline score for
the less successful surgeon and older patients could explain the finding that
they appear to have resulted in a significantly lower improvement. Indeed, this
was found to be the probable cause, as the surgeon identified had less capacity
for his patients to improve. His 68 cases had a significantly lower mean
baseline score of 34.79 (+/-31.47). This is compared with the mean baseline
score for the other surgeons of 46.84 (+/-21.23) (difference of 12.05, p=0.012,
unpaired t-test). At 12 months his patients mean improvement in score was -
12.64 against the other surgeons mean improvement of -21.78 (difference of -
9.14, p=0.066, unpaired t-test). Neither did the surgeon have significantly more
vaporisation cases than the other surgeons, which would have biased his
performance in favour of less improvement (he performed 34 cases with each
technique). Older patients also had less capacity to improve, as there was a
statistically significant negative correlation between age and pre-op pain score
(Pearson’s correlation test: r=-0.320, p<0.0005).

142
VAS scores for dysmenorrhoea, dyspareunia, CPP and dyschezia mirrored the
findings in EHP-30 Core pain score in that there was no difference in the
proportion of patients who improved at any follow-up point for excision and
vaporisation. Vaporisation analysed alone did not result in a statistically
significant improvement at 12 months versus baseline for CPP and dyschezia,
though significance was almost reached for CPP. There seems to be no
obvious explanation for this other than the general trend in this study that
vaporisation performs less well at 12 months. Comparative scores of the extent
of improvement show up a significant difference for excision performing better
than vaporisation in reducing CPP at 6 and 12 months. Significance is almost
reached for dysmenorrhoea too, and the remaining two symptoms of
dyspareunia and dyschezia show a trend towards excision performing better.
Thus, as with EHP-30 Core pain score, the results illustrate that the proportion
of patients improving after surgery is similar for vaporisation and excision, but
the extent to which they improve is greater for excision compared with
vaporisation. It may be that the failure to show significance in all symptoms is
related to limitations in the performance of visual analogue scales as Abbott
found and commented on in his trial (Abbott et al., 2004). Alternatively, it may
be that women are able to relate better to some questions than others. Perhaps
the notion of CPP is more meaningful than the notion of dyschezia. There
appears to be no clear trend of one symptom improving to a greater extent than
another from the results.
The backward stepwise linear regression analysis that was performed to see if
any factors influenced the change in CPP score at 12 months versus baseline

143
again confirmed that vaporisation was predictive of a lesser improvement in the
EHP-30 Core pain score, although this was only by 0.27 points on the 10 point
scale. The finding that disease in the uterovesical fold results in a greater
likelihood of score improvement cannot be explained, and is more likely to be
due to chance.
The trend of excision outperforming vaporisation continued into the EHP-30
HRQoL parameters. Alone, though generally resulting in improvement,
vaporisation did not produce results that were as significant. In comparative
terms, the significantly better improvement in quality of life domains for excision,
mirroring the reduction in pain, presumably reflects the profound effect that
chronic pain has on quality of life. It stands to reason that, as the quality of life
questions in EHP-30 are ones specifically identified by endometriosis sufferers,
then they will improve if the treatment is effective. Again, there appeared to be
no trend of any one domain improving more than any other, and it is not clear
why sexual intercourse was the only domain that did not show up a
comparatively statistically significant difference between excision and
vaporisation. It is possible that women don’t like responding to questions about
sexual intercourse and are therefore less likely to provide responses that reflect
reality. It is also possible that a considerable proportion of women feel that
sexual intercourse is not such a prominent aspect of their lives as being able to
go to work or socialise for example.
Study reports of medical therapy for endometriosis have commented that
starting scores for HRQoL scores are less than population norms and do not
return to normal levels following treatment (Miller, 2000, Zhao et al., 1999,

144
Bergqvist and Theorell, 2001), and only recently have HRQoL measures like
SF-12 and EQ-5D been used in surgical studies (Garry et al., 2000, Abbott et
al., 2003, Abbott et al., 2004). However, it is not possible to comment on
whether scores started below or returned to that found in general population
norms in our study, as EHP-30 is a disease specific measure and there are no
data relating to the general population. Nonetheless, the EHP-30 has been
externally validated against SF-36 and been shown to be more sensitive to
changes (Jones et al., 2004a).
The HADS questionnaire attempts to find anxiety and depression in subjects,
not in the sense of the extent of their predisposition to these stresses, but in the
transient amount of stress they actually have. In contrast to the EHP-30 HRQoL
domains, it does not deal with factors that moderate the impact of these
stresses like coping or social support. The results show that depression is not a
component of psychological stress in patients with minimal to moderate
endometriosis. Most probably, those that lie outside of the normal range are
depressed for reasons other than their endometriosis. Anxiety ratings did
improve significantly for both treatments between baseline and 12 months. As a
result, it appears to compare with the EHP-30 Core pain, HRQoL and VAS
score outcomes. Contrary to the EHP-30 data, the HADS anxiety scale is
judged against general population norms, suggesting that anxiety returns to
normal levels in many cases.
Comparison with other studies
In the report of the trial by Sutton et al (Sutton et al., 1994), it is not clear in the
original paper whether the linear analogue scale for pain as a primary outcome

145
measure related to overall pain, or for each of the symptoms of dysmenorrhoea,
dyspareunia or pelvic pain. In their follow up paper in 1997, they comment that:
“in the original study, no attempt was made to separately assess the
symptoms….in all cases, the most severe symptom was dysmenorrhoea,
which was therefore the symptom assessed when the patients were
asked to quantify their worst symptom by visual analogue score…”
(Sutton et al., 1997b).
In the original study by Sutton and colleagues (2004) there was a 62.5%
improvement in pain for laser vaporisation at 6 months. Our trial shows a 72.7%
improvement in EHP-30 Core pain, and a 58.7% improvement in
dysmenorrhoea at 6 months, and consequently appears to have a similar
outcome at 6 months, particularly when comparing dysmenorrhoea.
The Sutton trial reported a greater proportion of pain improvement of 73.7% for
the mild and moderate patients analysed alone compared with the overall
improvement of 62.5%, suggesting that more severe disease results in a
greater chance of improvement. The statistical robustness of this finding is open
to criticism as the data were broken down into stage 1-3 versus stage 2-3, and
this kind of sub-analysis is likely to create bias in treatment outcomes. However,
a view appears to have developed since this paper was published that
superficial disease is more difficult to treat than deep disease. In contrast, in our
study we found a definite trend that patients were more likely to have improved
EHP-30 Core pain scores if they had superficial disease at baseline, and a
statistically higher chance of improvement for patients with lower rASRM stage
at baseline. Also this study did not support the finding by Banerjee that better

146
relief of pain was found in women who had deep disease rather than superficial
(Banerjee et al., 2008). Their study had a poor follow-up rate of near 50% and it
may be that the non-responders were those patients with deep disease who did
not improve. Also their study includes those with stage 4 disease that are a sub-
group not included in our study.
In our study we showed that having a higher baseline score results in a greater
likelihood of a greater drop in pain score. It may be that these other studies had
higher baseline scores in their more severe and deep disease biasing the
outcome. In our study the baseline scores were not significantly different
between superficial and deep disease, and it seems logical that less severe
disease should be easier to treat and clear. Abbott’s findings support this to the
extent that the higher the stage of disease, the greater likelihood of requiring re-
operation (Abbott et al., 2003).
The Abbott trial reports an 80% improvement for the immediate surgery group
that underwent excision at 6 months, though it is not clear how the change in
the overall level of pain was calculated (Abbott et al., 2004). Baseline
demographic parameters could not be compared, as they are not detailed in the
Abbott paper. In our study, the proportion of improvers for excision by EHP-30
Core pain score at 6 months was 73.2 %. This again appears comparable,
although as previously mentioned, the sample in the Abbott trial differs in that it
included severe disease, and had a median rASRM score of 16 in the group
that underwent immediate excision, compared with a median score of 6 in our
study. For HRQoL outcomes, our trial is unable to shed light on Abbott’s finding
of a return to statistically insignificant differences compared with population

147
norms, found in contrast to their previous investigations (Garry et al., 2000,
Abbott et al., 2003).
In the Abbott 2003 prospective observational cohort study of excision treatment,
there appears to be an improvement in pain sustained out to 2-5 years for 67%
of patients. Our study confirms that there is no drop off in effect at 12 months for
improvers who underwent excision, and in fact, for EHP-30 Core pain score,
both the proportion and extent of improvement still appears to be increasing at
12 months post surgery for excision and not for vaporisation, in keeping with a
trend that adds some weight to Abbott’s 2-5 year findings.
In contrast, Sutton et al followed up on their 1994 trial and reported that 90% of
those patients who had shown pain improvement at 6 months had reported
continued benefit at 12 months, suggesting a small drop off in the effect of
vaporisation, though this was no longer part of the blinded trial (Sutton et al.,
1997b). In this study, the proportion of patients with improved dysmenorrhoea
score (the best direct comparison with the Sutton results) increased from 58.7
at 6 months to 63.4% at 12 months, but stayed the same for EHP-30 Core pain
score (72.7% versus 72.9%). This does not confirm this possible drop off effect.
However, whilst the extent of score improvement in EHP-30 and
dysmenorrhoea continued to improve in the excision group from 6 months to 12
months, this did not occur in the vaporisation group, again suggesting that
vaporisation is struggling to maintain its effect at 12 months compared with
excision.

148
In our study it is not the proportion of patients showing any improvement that
shows the difference between excision and vaporisation, but the extent of
improvement at 12 months. In Sutton’s trial the median VAS score for
vaporisation dropped by 2.85/10 points from a baseline median score of 8.5
(based mainly on dysmenorrhoea) between baseline and 6 months. This is
comparable with the 30/100 (range 0-95 on a 100 point linear scale) noted for
excision in Abbot’s trial for the immediate surgery group. In our study, VAS
dysmenorrhoea score for vaporisation reduced by a more conservative 1.60/10,
from a mean baseline score of 6.5, at 6 months. For excision this improved to
2.94/10 at 12 months versus baseline. Alarmingly, what the data from all these
results suggests is that at best on average, pain is improved by only
somewhere between a quarter and a half.
Added to this seemingly moderate improvement in pain scores, that are similar
for EHP-30 HRQoL outcomes, is the concerning finding that 20% of patients in
this study stayed the same or got worse. Abbott also found 20% of the patients
in his study stayed the same or got worse (Abbott et al., 2004) all of whom had
positive histology for endometriosis. It is probable in these cases that it is not
the surgery per se that is causing symptoms to worsen. Recurrence of new, or
progression of residual lesions are both possible culprits. However, Sutton
found that only 2 of the 5 women with continued pain, who had laser treatment
initially and then had a second look laparoscopy, were found to have visually
confirmed endometriosis (Sutton et al., 1994). The other possibility is that pelvic
pain in these women is not related to endometriosis at all, and is caused by
other pathologies with very similar symptoms like adenomyosis, irritable bowel
syndrome or interstitial cystitis.

149
If the range of improvement in pain scores was narrow then the moderate
extent of pain improvement achieved would call into question whether
subjecting patients to surgery is indeed worth it. However, the ranges of
improvement are wide, with some patients showing extensive improvement and
others showing very little, none, or becoming worse; this implies that we need to
have better prognostic indicators to select out those who will truly benefit from
surgical treatment of endometriosis.
There remains one other possibility suggested by the findings of Abbott’s
prospective cohort (Abbott et al., 2003) looking at follow-up at 2-5 years after
excision. It may be that the benefits of surgery are in fact much more gradual
than we believe, as the level of pain improvement in this study measured by
drop in median VAS pain score appears greater than that seen at 6-12 months
(dysmenorrhoea 9 to 3, non-menstrual pain 8 to 3, dyspareunia 7 to 0, and
dyschezia 7 to 2). However, at such follow-up periods, many confounding
variables are likely to be introduced including recurrent surgery or
hysterectomy, hormonal therapy and intervening pregnancies, all of which are
present in these results.
Limitations of the study
Methodological limitations
As with all surgical trials, there are many pitfalls that can cast doubts upon the
validity of results. In this trial every attempt was made to be a rigorous as

150
possible over methodology. However, retrospectively it is always possible to
find limitations, and that was no different with this trial.
Patient selection
One must first consider the type of patients that were entered into this trial.
Surrey is a relatively wealthy area, with a significant proportion of educated,
“middle class” patients that have a keen interest in their own health care.
Therefore they may be, to some extent, a biased set of patients in comparison
to the general population. However, they will of course be a similar set of
patients to the Sutton trial that was performed in the same unit 10 years
previously (Sutton et al., 1994). All geographical locations are likely to have
their own climates, economics, physical geography, social and cultural
peculiarities, all of which can affect a studies external validity. However, none of
the factors in this trial area or population are so extreme that they are likely to
be having a major influence on external validity.
Eligibility criteria
The other factors influencing the type of patients in the trial, and consequently
the external validity, are the eligibility criteria. In this trial, as in the other major
trials reported by Jacobson in his Cochrane review (Jacobson et al., 2009), we
excluded patients aged under 18, pregnancy, pelvic inflammatory disease and
inflammatory bowel conditions of the pelvis, and patients not willing to comply
with the trial protocol. However, there were several other exclusions mentioned
in these other trials that we did not use:

151
Sub-fertility patients were not excluded, as in this trial they all received
treatment and so had the possible benefit of improved fertility as a result
(Jacobson et al., 2004b). Those women who were included with sub-fertility
were not assessed for normal ovulatory cycles, partner’s semen sample, recent
infertility treatment or previous fertility surgery. Therefore, we were not able to
assess whether or not surgical treatment for endometriosis resulted in an
improvement in spontaneous conception.
Suspected gynaecological malignancy was not stated as an exclusion in the
protocol. No patients had suspected gynaecological malignancy, although one
turned out to have a borderline tumour of the ovary following ovarian
cystectomy, which was managed conservatively for fertility reasons.
Urgent patients were not included in the exclusion criteria, but no emergency
patients were included in the trial. All patients were admitted for elective
surgery.
Severe adhesions were not included in the list of exclusion criteria. If patients
fell within the rASRM score for stages 1-3 then they were included in the trial no
matter what the distribution of disease was.
Previous abdominal surgery was not listed as a reason for exclusion. There is a
possible limitation with this trial as a result of this. Some patients may have had
multiple unsuccessful laparoscopies for pelvic pain previously, with perhaps
little improvement. This may have biased the result against the positive effect of
surgical treatment. Conversely it may have biased the result in favour of

152
treatment, where “expert patients” had received repeated benefit from previous
surgeries. The other problem arising from previous surgery is that of creating
lesions containing psammoma bodies that mimic endometriosis leading to false
positive assessment of the pelvis (Martin and Vander Zwagg, 1987). In fact, the
sole exclusion from this trial before analysis was due to this.
This potential limitation of including patients who had undergone previous
abdominal surgery could have been prevented by taking histology in all cases,
and only including those with positive histology in the analysis as Abbott
rigorously did in his studies (Abbott et al., 2003, Abbott et al., 2004). However,
in those trials, excision was the only treatment modality, and so the problem of
biopsying small lesions without excising them before vaporisation did not exist.
Further to this, inflammation in lesions and traumatic biopsy can affect the
architecture of endometrial glands and stroma, making histological examination
difficult, resulting in false negative results (Shafik A, 2000), meaning that
patients are erroneously excluded. As has already been described, in this study
histology was taken in 65 of 133 cases (48.9%), partly because taking histology
was not begun until case 46 and partly because it was not always possible to
get a biopsy with vaporisation, as it may have resulted in excision of all or most
of the lesion. Overall 54 of the 65 cases had histology positive for
endometriosis, showing a successful correlation between visual inspection and
histological analysis in 83.1% of cases. A similar correlation of 88% was found
by Ballard between visual inspection and positive histology (Ballard et al.,
2009).

153
Randomisation
Proper randomisation should be able to eliminate selection bias from a sample
and is very important to achieve a high quality result (Altman, 1991). In this trial
the generation of the randomisation sequence was not computer generated.
However, the generated sequence was truly random in blocks of ten and there
was no way of telling which treatment was contained in the sealed opaque
envelope in each unallocated folder. There is a potential argument that, as the
generation of the sequence was done by two of the surgeons, this could have
affected the randomisation concealment at allocation at the time of surgery.
However, neither surgeon felt that this was a problem in this trial, especially as
the recruitment rate was about 2 patients per month, and each block of ten
folders took, on average, 5 months to allocate. This meant that mentally
keeping track of randomisation generation and allocations was unfeasible.
Recruitment
The slow recruitment rate of two patients per month mainly arose from the
exclusion of potentially eligible patients because they were on hormonal
medication for contraception rather than patients not wishing to participate.
Those who were on hormonal therapy for symptomatic relief only were asked if
they would stop it 3 months prior to surgery so as to fit the inclusion criteria.
However, by no means all patients agreed to do this. Additionally, there was
also an increased risk that such patients may have been more likely to restart
hormonal medication within the follow up period. Recruitment and follow up
management of patients in surgical trials is particularly difficult as strong
eligibility criteria are required to reduce the risk of confounding variables,
resulting in long recruitment periods that subsequently require long term follow-

154
up. It is not surprising, therefore, that it took nearly 66 months to recruit for this
trial, followed by another 1 year for the follow-up of the final patient, to complete
the trial in a total of six and a half years, during which time the trial standards
need to be maintained.
Blinding
Strict double blinding was not possible as this is a surgical trial and so there
was no way of blinding the surgeon from the procedure he was performing.
However, the surgeon was blinded from the outcomes, the patient was blinded
from the procedure, and the evaluator was also blinded from the procedure.
Although there was no evaluation carried out of the success of blinding, the
importance of blinding was strongly appreciated, particularly as this trial had the
subjective measure of pain as its primary outcome measure. An additional
question in the 12 month demographic questionnaire asking whether patients
were able to guess the treatment they received would have been useful.
Theoretically, they should have been no more likely to guess correctly than
chance. That being said, the fact that all patients were treated with what had
been hypothesised as equally good treatments, and there was no placebo
group, reduced the risk of performance bias also. There were no specific
adverse effects from treatment that could lead the patient or evaluator to
ascertain the treatment group. Equally, a positive outcome at evaluation did not
suggest one treatment or other as both treatments were thought to be equally
efficacious.

155
Protocol deviations
The protocol deviations discovered in April 2006, whereby pregnant women
were not being followed up, resulted in a mid term audit that proved to be very
useful in assessing the recruitment and drop out rate into the trial as well as
tightening any areas of the protocol that had become loose. This seems
particularly important in trials that take a long time to recruit.
Lack of control group
Both Sutton and Abbott reported placebo effects at 3 months (Sutton et al.,
1994, Abbott et al., 2004), and possibly up to 6 months in Abbott’s trial. The
follow up in this study was continued up to 12 months for this reason. Since
there is no control group, there was potential placebo interference that would
result in criticism of 6 months results. It seems more likely that the placebo
effect has worn off at 12 months and the results show the benefits of surgical
treatment.
Outcome limitations
Follow up rates and missing data
The loss of 27.8% of patients at 12 months for the analysis of the primary
outcome measure is a potential limitation of this trial. However, the lost sub
group was analysed in two ways to see if their absence from the main analysis
may have had a significant effect on the result. Firstly, it was determined
whether or not this had created imbalanced groups. This was not found to be
the case as the excision arm lost 27.3% of cases and the vaporisation arm lost
29.9% (p=0.742). Secondly, the lost sub group was analysed against the

156
remaining patients in the main analysis to see whether they differed in age,
median rASRM score, histology finding and depth of disease. No difference was
found here either; suggesting that the lost sub group did not differ significantly
from the main analysis group, and consequently had not biased the result.
Balanced arms in the main analysis group
From the descriptive statistics, that otherwise showed well balanced groups for
age, rASRM score and positive histology, there was a potential limitation found
in that there were more deep cases in the excision group (44/66, 66.7%)
compared with the vaporisation group (28/66, 43.8%). As the results show that
vaporisation performed significantly less well for deep disease, both
independently and in comparison with excision, then less deep cases in the
vaporisation group gave it an advantage. Consequently, vaporisation appeared
to be performing better than it would have done had the number of deep cases
been evenly distributed. The comparative advantages of excision over
vaporisation may well be underestimated by this study as a result.
A further limitation relating to depth of disease exists here, as its assessment
was by palpation the visual appearance of the lesions, and so was a subjective
decision by the surgeon that it appeared to be infiltrating >5mm below the
surface of the peritoneum. This may have resulted in assignment errors to the
wrong depth classification, casting doubt on the results relating to depth of
disease. However, all the surgeons were experienced at diagnosing
endometriosis and the depth adjusted on the rASRM score sheet at the end of
the case once a better assessment of depth was available following treatment
of the lesions. If any bias resulted, it is likely that there would have been an over

157
diagnosis of deep disease, rather than an under diagnosis. This would have
diluted the deep category with more superficial cases, again allowing the
vaporisation group an advantage.
Protocol deviations and “per protocol” analysis
The patients with protocol deviations during the follow up period were included
in the main analysis and result in a possible trial limitation. There were 19
patients who became pregnant, 5 who underwent further surgery, and 16 who
had some form of adjuvant hormonal therapy. A “per protocol” analysis for the
primary outcome measure EHP-30 Core pain was carried out on those patients
who had not deviated from the protocol. The improvements found in these
patients are more likely to be related to surgical treatment. There were still 35
patients in each arm at 12 months (more than the 28 patients required in the
power calculation for EHP-30 Core pain to have an 80% chance of showing a
20% difference). Excision (-21.04 +/- 23.97) still outperformed vaporisation (-
9.42 +/-19.36) at 12 months (p=0.029, 95%CI -22.02 to -1.23).

158
Chapter 6 - Conclusions
This thesis set out to improve our understanding of the surgical treatment of
endometriosis. Much of the early evidence in the literature was related to fertility
outcomes rather than pelvic pain and clinical practice thus emphasised this
component of the problem. Larger case series and randomised trials have
subsequently begun to focus on the more subjective issue of pelvic pain. Two
clear technique categories developed, vaporisation/ablation and excision, both
with randomised data supporting their use. However, there has been no clear
data about who is really using which techniques, and whether any one is really
any better than the other. In addition, the hypothesis of this thesis was that
there would be no difference between the two techniques in terms of pain
outcome if the lesions were truly vaporised by CO2 laser down to normal tissue.
The lack of information regarding the extent to which gynaecological
laparoscopic surgeons are using varying techniques for treating pelvic
endometriosis of all types, was investigated with an international web-based
survey to address this. This survey has some definite limitations in terms of its
sample size, however it does suggest that surgeons who belong to the
specialist laparoscopic societies appear to be aware of the evidence currently
available and are trying to practise in an evidence-based manner. Some
surgeons would change their practice if they had the access to training and
equipment to allow them to do so. In general, there exists a view that
endometriomas should be excised and that bowel resection should be avoided
if possible for recto-vaginal disease. For minimal to moderate disease,

159
superficial disease can be treated with a combination of excision or vaporisation
depending on the case, but that deep disease should be excised.
This very question for minimal to moderate disease was addressed in the
randomised blinded trial reported in this thesis. It was clearly found that,
contrary to expectation, although the proportion of patients who improve is no
different for either vaporisation or excision, the extent of improvement in both
pain and quality of life is significantly better for excision than vaporisation at 12
months post operation. In addition to this, for deep disease vaporisation does
not convey significant improvement at 12 months. The statistical analysis very
much set out to try and disprove this finding as it was contrary to the hypothesis
at the outset. However, it became apparent that the finding was genuine and
the limitations were sufficiently small to accept the result. Since diagnosing
deep endometriosis is essentially an intra-operative one, and since excision will
give the best results in all forms of minimal to moderate disease, it makes
logical sense to approach treatment intending to excise in every case in the
light of these findings.
The implications of this view must be considered in the context of clinical
practice. There is little doubt that training doctors to laparoscopically excise
endometriosis is not a straightforward one. A sound knowledge of the anatomy
of the pelvic side-wall and the skill to avoid damage in this location is integral.
Not all gynaecologists are likely to be able to achieve the skills required or have
the training opportunity to achieve this safely, in order to stop the culture of not
surgically excising at the time of diagnosis or dusting the surface with bipolar
diathermy. The implication of this is that most patients with endometriosis may

160
need referral to a specialist gynaecologist, which is probably currently
unattainable. This does however seem to present more evidence that we
should be moving towards a culture of greater subspecialisation, and away from
the old fashioned “jack of all trades” Obstetrician and Gynaecologist.
Furthermore it must be added that a significant proportion of patients will
deteriorate even with surgery and that the extent of improvement is
disappointing on average and widely variable in practice. To this end, much
more work is required to ascertain the symptoms or signs that will allow us to
direct our surgical treatment towards those patients who are most likely to
benefit, and avoid operating unnecessarily on those who will not.
The current randomised data only tells of improvement for a year, which brings
into question the cost-benefit of endometriosis surgery, especially when there is
at least a third risk of recurrent surgery within five years. The extent of
improvement resulting from excision beyond 12 months still remains unclear as
no randomised data exists, and is likely to be confounded by other variables
introduced over time following the index surgery.
The use of EHP-30 as a validated tool to evaluate pain and quality of life in
endometriosis appears to have been successful in this study as it attempts to
address issues directly related to the patient and their quality of life. The data
gained here will be pooled with existing data to further improve its use as a tool
for endometriosis, though ideally population norm data is required for EHP-30.

161
In conclusion, whilst it is recognised that all studies have their limitations, and
that is no different in this case, it appears that for the surgical treatment of
endometriosis, excision is the way forward and that training and referral
strategies need to be put in place to ensure that women have access to the
optimal treatment.

162
References
1979. Classification of endometriosis. The American Fertility Society. Fertil Steril, 32,
633-4.
1985. Revised American Fertility Society classification of endometriosis: 1985. Fertil
Steril, 43, 351-2.
1994. Prevalence and anatomical distribution of endometriosis in women with selected
gynaecological conditions: results from a multicentric Italian study. Gruppo
italiano per lo studio dell'endometriosi. Hum Reprod, 9, 1158-62.
1997. Revised American Society for Reproductive Medicine classification of
endometriosis: 1996. Fertil Steril, 67, 817-21.
1999. A consensus document concerning laparoscopic entry techniques:
Middlesbrough, March 19-20 1999. Gynaecol Endosc, 8, 403-6.
2003. A randomised controlled trial to assess the efficacy of Laparoscopic Uterosacral
Nerve Ablation (LUNA) in the treatment of chronic pelvic pain: The trial
protocol [ISRCTN41196151]. BMC Womens Health, 3, 6.
ABBOTT, J., HAWE, J., HUNTER, D., HOLMES, M., FINN, P. & GARRY, R. 2004.
Laparoscopic excision of endometriosis: a randomized, placebo-controlled trial.
Fertil Steril, 82, 878-84.
ABBOTT, J. A., HAWE, J., CLAYTON, R. D. & GARRY, R. 2003. The effects and
effectiveness of laparoscopic excision of endometriosis: a prospective study
with 2-5 year follow-up. Hum Reprod, 18, 1922-7.
ACOSTA, A. A., BUTTRAM, V. C., JR., BESCH, P. K., MALINAK, L. R.,
FRANKLIN, R. R. & VANDERHEYDEN, J. D. 1973. A proposed
classification of pelvic endometriosis. Obstet Gynecol, 42, 19-25.
ADAMSON, G. D., HURD, S. J., PASTA, D. J. & RODRIGUEZ, B. D. 1993.
Laparoscopic endometriosis treatment: is it better? Fertil Steril, 59, 35-44.
ADAMSON, G. D. & PASTA, D. J. 2009. Endometriosis fertility index: the new,
validated endometriosis staging system. Fertil Steril.
ADAMSON, G. D., SUBAK, L. L., PASTA, D. J., HURD, S. J., VON FRANQUE, O.
& RODRIGUEZ, B. D. 1992. Comparison of CO2 laser laparoscopy with
laparotomy for treatment of endometriomata. Fertil Steril, 57, 965-73.
ADAMYAN, L. V., FANCHENKO, N. D., ALEXEYEVA, M. L., ANDREYEVA YE,
N., NOVIKOV YE, A. & JAHAN, I. 1993. Hormonal and immunologic
methods in the diagnosis and treatment of patients with benign ovarian tumors
and endometriotic cysts. Int J Fertil, 38, 92-8.
AJOSSA, S., MAIS, V., GUERRIERO, S., PAOLETTI, A. M., CAFFIERO, A.,
MURGIA, C. & MELIS, G. B. 1994. The prevalence of endometriosis in
premenopausal women undergoing gynecological surgery. Clin Exp Obstet
Gynecol, 21, 195-7.
ALBORZI, S., MOMTAHAN, M., PARSANEZHAD, M. E., DEHBASHI, S. &
ZOLGHADRI, J. 2004a. A prospective, randomized study comparing
laparoscopic ovarian cystectomy versus fenestration and coagulation in patients
with endometriomas. Fertil Steril, 82, 1633-7.
ALBORZI, S., MOMTAHAN, M., PARSANEZHAD, M. E., DEHBASHI, S.,
ZOLGHADRI, J. & ALBORZI, S. 2004b. A prospective, randomized study
comparing laparoscopic ovarian cystectomy versus fenestration and coagulation
in patients with endometriomas. Fertil Steril, 82, 1633-7.
ALBORZI, S., RAVANBAKHSH, R., PARSANEZHAD, M. E., ALBORZI, M.,
ALBORZI, S. & DEHBASHI, S. 2007a. A comparison of follicular response of

163
ovaries to ovulation induction after laparoscopic ovarian cystectomy or
fenestration and coagulation versus normal ovaries in patients with
endometrioma. Fertil Steril, 88, 507-9.
ALBORZI, S., RAVANBAKHSH, R., PARSANEZHAD, M. E., ALBORZI, M. &
DEHBASHI, S. 2007b. A comparison of follicular response of ovaries to
ovulation induction after laparoscopic ovarian cystectomy or fenestration and
coagulation versus normal ovaries in patients with endometrioma. Fertil Steril,
88, 507-9.
ALTMAN, D. G. 1991. Randomisation. Bmj, 302, 1481-2.
ALTMAN, D. G., SCHULZ, K. F., MOHER, D., EGGER, M., DAVIDOFF, F.,
ELBOURNE, D., GOTZSCHE, P. C. & LANG, T. 2001. The revised
CONSORT statement for reporting randomized trials: explanation and
elaboration. Ann Intern Med, 134, 663-94.
ARRUDA, M. S., PETTA, C. A., ABRAO, M. S. & BENETTI-PINTO, C. L. 2003.
Time elapsed from onset of symptoms to diagnosis of endometriosis in a cohort
study of Brazilian women. Hum Reprod, 18, 756-9.
AUDEBERT, A., BERNARD, I. & EMPERAIRE, J. C. 1977. [Treatment of
endometriosis using Danazol: preliminary report on clinical tolerance].
Gynecologie, 28, 29-34.
AWADZI, G., FRAPPELL, J., ORIOLOWO, A., SIBANDA, T. 2005. A clinical
evaluation of the LOTUS ultrasonic shears in gynaecological surgery Gynecol
Surg, 2.
BAILEY, H. R., OTT, M. T. & HARTENDORP, P. 1994. Aggressive surgical
management for advanced colorectal endometriosis. Dis Colon Rectum, 37, 747-
53.
BALLARD, K., LANE, H., HUDELIST, G., BANERJEE, S. & WRIGHT, J. 2009. Can
specific pain symptoms help in the diagnosis of endometriosis? A cohort study
of women with chronic pelvic pain. Fertil Steril.
BALLARD, K., LOWTON, K. & WRIGHT, J. 2006. What's the delay? A qualitative
study of women's experiences of reaching a diagnosis of endometriosis. Fertil
Steril, 86, 1296-301.
BANERJEE, S. K., BALLARD, K. D. & WRIGHT, J. T. 2008. Endometriomas as a
marker of disease severity. J Minim Invasive Gynecol, 15, 538-40.
BARLOW, D. H. & KENNEDY, S. 2005. Endometriosis: new genetic approaches and
therapy. Annu Rev Med, 56, 345-56.
BARTON-SMITH, P., BALLARD, K. & KENT, A. S. H. 2006. Endometriosis: A
general review and rationale for surgical therapy. Reviews in Gynaecological
and Perinatal Practice, 6, 168-176.
BATIOGLU, S., CELIKKANAT, H., UGUR, M., MOLLAMAHMUTOGLU, L.,
YESILYURT, H. & KUNDAKCI, M. 1996. The use of GnRH agonists in the
treatment of endometriomas with or without drainage. J Pak Med Assoc, 46, 30-
2.
BATT, R. E. & MITWALLY, M. F. 2003. Endometriosis from thelarche to midteens:
pathogenesis and prognosis, prevention and pedagogy. J Pediatr Adolesc
Gynecol, 16, 337-47.
BAZOT, M., DARAI, E., HOURANI, R., THOMASSIN, I., CORTEZ, A., UZAN, S.
& BUY, J. N. 2004a. Deep pelvic endometriosis: MR imaging for diagnosis and
prediction of extension of disease. Radiology, 232, 379-89.
BAZOT, M., LAFONT, C., ROUZIER, R., ROSEAU, G., THOMASSIN-NAGGARA,
I. & DARAI, E. 2009. Diagnostic accuracy of physical examination,

164
transvaginal sonography, rectal endoscopic sonography, and magnetic resonance
imaging to diagnose deep infiltrating endometriosis. Fertil Steril, 92, 1825-33.
BAZOT, M., NASSAR, J., DARAI, E., THOMASSIN, I., CORTEZ, A., BUY, J. N.,
UZAN, S. & MARSAULT, C. 2005. [Value of sonography and MR imaging for
the evaluation of deep pelvic endometriosis]. J Radiol, 86, 461-7.
BAZOT, M., THOMASSIN, I., HOURANI, R., CORTEZ, A. & DARAI, E. 2004b.
Diagnostic accuracy of transvaginal sonography for deep pelvic endometriosis.
Ultrasound Obstet Gynecol, 24, 180-5.
BAZOT, M., THOMASSIN-NAGGARA, I., DARAI, E. & MARSAULT, C. 2008.
[Imaging of chronic pelvis pain.]. J Radiol, 89, 107-14.
BECKER, C. M. & D'AMATO, R. J. 2007. Angiogenesis and antiangiogenic therapy in
endometriosis. Microvasc Res, 74, 121-30.
BEDAIWY, M. A., HUSSEIN, M. R., BISCOTTI, C. & FALCONE, T. 2009. Pelvic
Endometriosis is Rarely Associated with Ovarian Borderline Tumours,
Cytologic and Architectural Atypia: A Clinicopathologic Study. Pathol Oncol
Res, 15, 81-8.
BERAL, V. 2003. Breast cancer and hormone-replacement therapy in the Million
Women Study. Lancet, 362, 419-27.
BERETTA, P., FRANCHI, M., GHEZZI, F., BUSACCA, M., ZUPI, E. & BOLIS, P.
1998. Randomized clinical trial of two laparoscopic treatments of
endometriomas: cystectomy versus drainage and coagulation. Fertil Steril, 70,
1176-80.
BERGQVIST, A. & THEORELL, T. 2001. Changes in quality of life after hormonal
treatment of endometriosis. Acta Obstet Gynecol Scand, 80, 628-37.
BLUMENKRANTZ, M. J., GALLAGHER, N., BASHORE, R. A. & TENCKHOFF, H.
1981. Retrograde menstruation in women undergoing chronic peritoneal
dialysis. Obstet Gynecol, 57, 667-70.
BOYNTON, P. 2004. Administering, analysing, and reporting your questionnaire. BMJ,
328, 1372-1375.
BRINTON, L. A., GRIDLEY, G., PERSSON, I., BARON, J. & BERGQVIST, A. 1997.
Cancer risk after a hospital discharge diagnosis of endometriosis. Am J Obstet
Gynecol, 176, 572-9.
BRUHAT, M. A., WATTIEZ, A., MAGE, G., POULY, J. L. & CANIS, M. 1989. CO2
laser laparoscopy. Baillieres Clin Obstet Gynaecol, 3, 487-97.
BRUNER-TRAN, K. L., EISENBERG, E., YEAMAN, G. R., ANDERSON, T. A.,
MCBEAN, J. & OSTEEN, K. G. 2002. Steroid and cytokine regulation of
matrix metalloproteinase expression in endometriosis and the establishment of
experimental endometriosis in nude mice. J Clin Endocrinol Metab, 87, 4782-
91.
BUCHWEITZ, O., POEL, T., DIEDRICH, K. & MALIK, E. 2003. The diagnostic
dilemma of minimal and mild endometriosis under routine conditions. J Am
Assoc Gynecol Laparosc, 10, 85-9.
BUSACCA, M., RIPARINI, J., SOMIGLIANA, E., OGGIONI, G., IZZO, S.,
VIGNALI, M. & CANDIANI, M. 2006. Postsurgical ovarian failure after
laparoscopic excision of bilateral endometriomas. Am J Obstet Gynecol, 195,
421-5.
CANDIANI, G. B., FEDELE, L. & VERCELLINI, P. 1986. Conservative surgical
treatment of endometriosis. Acta Eur Fertil, 17, 173-80.
CANIS, M., MAGE, G., WATTIEZ, A., POULY, J. L. & BRUHAT, M. A. 2003. The
ovarian endometrioma: why is it so poorly managed? Laparoscopic treatment of

165
large ovarian endometrioma: why such a long learning curve? Hum Reprod, 18,
5-7.
CANIS, M., TAMBURRO, S., STERKERS, N., BOTCHORISHVILI, R.,
RABISCHONG, B., JARDON, K., POULY, J., MAGE, G. 2005. Complications
and long term follow up after laparoscopic treatment of deep infiltrating
endometriosis. A series of 406 cases with 10 years follow up. Eu J Obstet
Gynecol Reprod Biol, 123, S17.
CATTELL R., S. N. 1936. Endometriosis. With particular reference to conservative
treatment. . N Engl J Med, 214, 341-6.
CEDARS, M. I., LU, J. K., MELDRUM, D. R. & JUDD, H. L. 1990. Treatment of
endometriosis with a long-acting gonadotropin-releasing hormone agonist plus
medroxyprogesterone acetate. Obstet Gynecol, 75, 641-5.
CHAPRON, C., FAUCONNIER, A., VIEIRA, M., BARAKAT, H., DOUSSET, B.,
PANSINI, V., VACHER-LAVENU, M. C. & DUBUISSON, J. B. 2003.
Anatomical distribution of deeply infiltrating endometriosis: surgical
implications and proposition for a classification. Hum Reprod, 18, 157-61.
CHAPRON, C., VIEIRA, M., CHOPIN, N., BALLEYGUIER, C., BARAKAT, H.,
DUMONTIER, I., ROSEAU, G., FAUCONNIER, A., FOULOT, H. &
DOUSSET, B. 2004. Accuracy of rectal endoscopic ultrasonography and
magnetic resonance imaging in the diagnosis of rectal involvement for patients
presenting with deeply infiltrating endometriosis. Ultrasound Obstet Gynecol,
24, 175-9.
CHEN, T. F., STJOHN COLLIER, D., EVERETT, W. G. & DIXON, A. K. 1989.
Endometriosis of the recto-vaginal septum treated by anterior resection. Ann
Chir Gynaecol, 78, 324-6.
CLEMENT, P. B. 2007. The pathology of endometriosis: a survey of the many faces of
a common disease emphasizing diagnostic pitfalls and unusual and newly
appreciated aspects. Adv Anat Pathol, 14, 241-60.
CORONADO, C., FRANKLIN, R. R., LOTZE, E. C., BAILEY, H. R. & VALDES, C.
T. 1990. Surgical treatment of symptomatic colorectal endometriosis. Fertil
Steril, 53, 411-6.
CREUS, M., FABREGUES, F., CARMONA, F., DEL PINO, M., MANAU, D. &
BALASCH, J. 2008. Combined laparoscopic surgery and pentoxifylline therapy
for treatment of endometriosis-associated infertility: a preliminary trial. Hum
Reprod, 23, 1910-6.
CROMER, J. K. 1967. Endometriosis. With special emphasis on lesions of the colon
(sigmoid, rectosigmoid and rectovaginal septum). Med Ann Dist Columbia, 36,
100-1.
CROSIGNANI, P. G. & VERCELLINI, P. 1995. Conservative surgery for severe
endometriosis: should laparotomy be abandoned definitively? Hum Reprod, 10,
2412-8.
CROSIGNANI, P. G., VERCELLINI, P., BIFFIGNANDI, F., COSTANTINI, W.,
CORTESI, I. & IMPARATO, E. 1996. Laparoscopy versus laparotomy in
conservative surgical treatment for severe endometriosis. Fertil Steril, 66, 706-
11.
D'HOOGHE, T. M., DEBROCK, S., HILL, J. A. & MEULEMAN, C. 2003.
Endometriosis and subfertility: is the relationship resolved? Semin Reprod Med,
21, 243-54.
DANIELL, J. F., KURTZ, B. R. & GURLEY, L. D. 1991. Laser laparoscopic
management of large endometriomas. Fertil Steril, 55, 692-5.

166
DANIELL, J. F., MCTAVISH, G., KURTZ, B. R. & TALLAB, F. 1994. Laparoscopic
Use of Argon Beam Coagulator in the Management of Endometriosis. J Am
Assoc Gynecol Laparosc, 1, S9.
DANIELS, J. G., R. HILLS, R.K. 2009. Laparoscopic Uterosacral Nerve Ablation for
Alleviating Chronic Pelvic Pain: A Randomized Controlled Trial. JAMA, 302,
955-961.
DARAI, E., BAZOT, M., ROUZIER, R., HOURY, S. & DUBERNARD, G. 2007.
Outcome of laparoscopic colorectal resection for endometriosis. Curr Opin
Obstet Gynecol, 19, 308-13.
DELL'ENDOMETRIOSI, G. I. P. L. S. 1999. Ablation of lesions or no treatment in
minimal-mild endometriosis in infertile women: a randomized trial. Hum
Reprod, 14, 1332-4.
DESSOLE, S., FARINA, M., RUBATTU, G., COSMI, E., AMBROSINI, G. &
NARDELLI, G. B. 2003. Sonovaginography is a new technique for assessing
rectovaginal endometriosis. Fertil Steril, 79, 1023-7.
DI PALO, S., MARI, G., CASTOLDI, R., STAUDACHER, C., TACCAGNI, G. & DI
CARLO, V. 1989. Endometriosis of the lung. Respir Med, 83, 255-8.
DINULESCU, D. M., INCE, T.A., QUADE, B.J., SHAFER, S.A., CROWLEY, D.,
JACKS, T. 2005. Role of K-ras and Pten in the development of mouse models
of endometriosis and endometrioid ovarian cancer. Nat Med, 11, 63-70.
DONNEZ, J. & NISOLLE, M. 1995. Advanced laparoscopic surgery for the removal of
rectovaginal septum endometriotic or adenomyotic nodules. Baillieres Clin
Obstet Gynaecol, 9, 769-74.
DONNEZ, J., NISOLLE, M., GILLEROT, S., ANAF, V., CLERCKX-BRAUN, F. &
CASANAS-ROUX, F. 1994. Ovarian endometrial cysts: the role of
gonadotropin-releasing hormone agonist and/or drainage. Fertil Steril, 62, 63-6.
DONNEZ, J., NISOLLE, M., GILLET, N., SMETS, M., BASSIL, S. & CASANAS-
ROUX, F. 1996. Large ovarian endometriomas. Hum Reprod, 11, 641-6.
DUBERNARD, G., PIKETTY, M., ROUZIER, R., HOURY, S., BAZOT, M. &
DARAI, E. 2006. Quality of life after laparoscopic colorectal resection for
endometriosis. Hum Reprod, 21, 1243-7.
ESKENAZI, B. & WARNER, M. L. 1997. Epidemiology of endometriosis. Obstet
Gynecol Clin North Am, 24, 235-58.
FACCIOLI, N., MANFREDI, R., MAINARDI, P., DALLA CHIARA, E., SPOTO, E.,
MINELLI, L. & MUCELLI, R. P. 2008. Barium enema evaluation of colonic
involvement in endometriosis. AJR Am J Roentgenol, 190, 1050-4.
FALCONER, H., MWENDA, J. M., CHAI, D. C., WAGNER, C., SONG, X. Y.,
MIHALYI, A., SIMSA, P., KYAMA, C., CORNILLIE, F. J., BERGQVIST, A.,
FRIED, G. & D'HOOGHE, T. M. 2006. Treatment with anti-TNF monoclonal
antibody (c5N) reduces the extent of induced endometriosis in the baboon. Hum
Reprod, 21, 1856-62.
FALLON, J., BROSNAN, J., MORAN, W. 1946. Two hundred cases considered from
the viewpoint of the practitioner. N Engl J Med, 235, 669-73.
FEDELE, L., BIANCHI, S., PORTUESE, A., BORRUTO, F. & DORTA, M. 1998.
Transrectal ultrasonography in the assessment of rectovaginal endometriosis.
Obstet Gynecol, 91, 444-8.
FEDELE, L., BIANCHI, S., VIEZZOLI, T., ARCAINI, L. & CANDIANI, G. B. 1989.
Gestrinone versus danazol in the treatment of endometriosis. Fertil Steril, 51,
781-5.
FEIL, W. 2005. Ultrasonic Energy for Cutting, Coagulating, and Dissecting, Stuttgart,
Georg Thieme Verlag.

167
FORD, J., ENGLISH, J., MILES, W. A. & GIANNOPOULOS, T. 2004. Pain, quality
of life and complications following the radical resection of rectovaginal
endometriosis. Bjog, 111, 353-6.
FREDRIKSON, H. 1957. Pregnancy after conservative surgery for ovarian
endometriosis. Acta Obstet Gynecol Scand, 36, 468-80.
GARCIA-VELASCO, J. A. & ARICI, A. 2004. Surgery for the removal of
endometriomas before in vitro fertilization does not increase implantation and
pregnancy rates. Fertil Steril, 81, 1206.
GARRY, R. 2004. The endometriosis syndromes: a clinical classification in the
presence of aetiological confusion and therapeutic anarchy. Hum Reprod, 19,
760-768.
GARRY, R., CLAYTON, R. & HAWE, J. 2000. The effect of endometriosis and its
radical laparoscopic excision on quality of life indicators. Bjog, 107, 44-54.
GLEICHER, N., EL-ROEIY, A., CONFINO, E. & FRIBERG, J. 1987. Is endometriosis
an autoimmune disease? Obstet Gynecol, 70, 115-22.
GRAY, L. A. 1973. Endometriosis of the bowel: role of bowel resection, superficial
excision and oophorectomy in treatment. Ann Surg, 177, 580-7.
GUZICK, D. S., SILLIMAN, N. P., ADAMSON, G. D., BUTTRAM, V. C., JR.,
CANIS, M., MALINAK, L. R. & SCHENKEN, R. S. 1997. Prediction of
pregnancy in infertile women based on the American Society for Reproductive
Medicine's revised classification of endometriosis. Fertil Steril, 67, 822-9.
GUZMAN, J., ESMAIL, R., KARJALAINEN, K., MALMIVAARA, A., IRVIN, E. &
BOMBARDIER, C. 2001. Multidisciplinary rehabilitaion for chronic low back
pain: systematic review. BMJ, 322.
HADFIELD, R., MARDON, H., BARLOW, D. & KENNEDY, S. 1996. Delay in the
diagnosis of endometriosis: a survey of women from the USA and the UK. Hum
Reprod, 11, 878-80.
HALBAN, J. 1924. Hysteroadenosis metstica. Wien Klin Wochenschr, 37, 1205-28.
HALME, J., HAMMOND, M. G., HULKA, J. F., RAJ, S. G. & TALBERT, L. M.
1984. Retrograde menstruation in healthy women and in patients with
endometriosis. Obstet Gynecol, 64, 151-4.
HART, R. J., HICKEY, M., MAOURIS, P. & BUCKETT, W. 2008. Excisional surgery
versus ablative surgery for ovarian endometriomata. Cochrane Database Syst
Rev, CD004992.
HASSON, H. M. 1976. Incidence of endometriosis in diagnostic laparoscopy. J Reprod
Med, 16, 135-8.
HEILIER, J., GRANDJEAN, F., DONNEZ, O., DE BUYL, L., VEROUGSTRAETE,
V., NACKERS, F., TONGLET, R., LISON, D., DONNEZ, J. 2004. Comparison
of environmental and personal risk factors in women with endometriosis or deep
endometriotic nodules. Eur J Obstet Gynecol Reprod Biol, 123, S6.
HORNSTEIN, M. D., YUZPE, A. A., BURRY, K. A., HEINRICHS, L. R.,
BUTTRAM, V. L., JR. & ORWOLL, E. S. 1995. Prospective randomized
double-blind trial of 3 versus 6 months of nafarelin therapy for endometriosis
associated pelvic pain. Fertil Steril, 63, 955-62.
HORNUNG, D., XU, H., REHBEIN, M., EBERT, A., WOELFLER, M., DIEDRICH,
K. 2005. A new diagnostic tool? CCR1 mRNA expression in peripheral blood
leucocytes of patients with and without endometriosis is significantly differently
expressed. Eu J Obstet Gynecol Reprod Biol, 123, S13.
HUDELIST, G. & KECKSTEIN, J. 2009. [The use of transvaginal sonography (TVS)
for preoperative diagnosis of pelvic endometriosis]. Praxis (Bern 1994), 98,
603-7.

168
HUDELIST, G., TUTTLIES, F., RAUTER, G., PUCHER, S. & KECKSTEIN, J. 2009.
Can transvaginal sonography predict infiltration depth in patients with deep
infiltrating endometriosis of the rectum? Hum Reprod, 24, 1012-7.
HUGHES, E., FEDORKOW, D., COLLINS, J. & VANDEKERCKHOVE, P. 2003.
Ovulation suppression for endometriosis. Cochrane Database Syst Rev,
CD000155.
HUGHES, E. G., FEDORKOW, D. M. & COLLINS, J. A. 1993. A quantitative
overview of controlled trials in endometriosis-associated infertility. Fertil Steril,
59, 963-70.
HUMMELSHOJ, L., PRENTICE, A., GROOTHUIS, P. 2006. Update on
endometriosis. Women's Health, 2, 53-56.
HUSBY, G. K., HAUGEN, R. S. & MOEN, M. H. 2003. Diagnostic delay in women
with pain and endometriosis. Acta Obstet Gynecol Scand, 82, 649-53.
JACOBSON, T. Z., BARLOW, D. H., KONINCKX, P. R., OLIVE, D. &
FARQUHAR, C. 2004b. Laparoscopic surgery for subfertility associated with
endometriosis. Cochrane Database Syst Rev, CD001398.
JACOBSON, T. Z., DUFFY, J. M., BARLOW, D., KONINCKX, P. R. & GARRY, R.
2009. Laparoscopic surgery for pelvic pain associated with endometriosis.
Cochrane Database Syst Rev, CD001300.
JAESCHKE, R., SINGER, J, GUYATT, G.H. 1990. A comparison of seven-point and
visual analogue scales: data from a randomised trial. Controlled Clinical Trials,
11, 43-51.
JANTOS, M. 2007. Understanding Chronic Pelvic Pain. Pelviperineology, 26, 66-69.
JARRELL, J., MOHINDRA, R., ROSS, S., TAENZER, P. & BRANT, R. 2005.
Laparoscopy and reported pain among patients with endometriosis. J Obstet
Gynaecol Can, 27, 477-85.
JENKINS, S., OLIVE, D. L. & HANEY, A. F. 1986. Endometriosis: pathogenetic
implications of the anatomic distribution. Obstet Gynecol, 67, 335-8.
JERBY, B. L., KESSLER, H., FALCONE, T. & MILSOM, J. W. 1999. Laparoscopic
management of colorectal endometriosis. Surg Endosc, 13, 1125-8.
JOHNSON, N. P., FARQUHAR, C. M., CROSSLEY, S., YU, Y., VAN
PEPERSTRATEN, A. M., SPRECHER, M. & SUCKLING, J. 2004. A double-
blind randomised controlled trial of laparoscopic uterine nerve ablation for
women with chronic pelvic pain. Bjog, 111, 950-9.
JONES, G., JENKINSON, C. & KENNEDY, S. 2004a. Evaluating the responsiveness
of the Endometriosis Health Profile Questionnaire: the EHP-30. Qual Life Res,
13, 705-13.
JONES, G., JENKINSON, C. & KENNEDY, S. 2004b. The impact of endometriosis
upon quality of life: a qualitative analysis. J Psychosom Obstet Gynaecol, 25,
123-33.
JONES, G., KENNEDY, S., BARNARD, A., WONG, J. & JENKINSON, C. 2001.
Development of an endometriosis quality-of-life instrument: The Endometriosis
Health Profile-30. Obstet Gynecol, 98, 258-64.
JONES, K. D. & SUTTON, C. J. 2002. Pregnancy rates following ablative laparoscopic
surgery for endometriomas. Hum Reprod, 17, 782-5.
KAUPPILA, A., PUOLAKKA, J. & YLIKORKALA, O. 1979. Prostaglandin
biosynthesis inhibitors and endometriosis. Prostaglandins, 18, 655-61.
KECKSTEIN, J. 1999. [Surgical therapy of endometriosis]. Wien Med Wochenschr,
149, 366-71.

169
KECKSTEIN, J. & WIESINGER, H. 2005. Deep endometriosis, including intestinal
involvement--the interdisciplinary approach. Minim Invasive Ther Allied
Technol, 14, 160-6.
KENNEDY, S., BERGQVIST, A., CHAPRON, C., D'HOOGHE, T., DUNSELMAN,
G., GREB, R., HUMMELSHOJ, L., PRENTICE, A. & SARIDOGAN, E. 2005.
ESHRE guideline for the diagnosis and treatment of endometriosis. Hum
Reprod, 20, 2698-704.
KENNEDY, S., BERGQVIST, A,. CHAPRON, C., D'HOOGHE, T., DUNSELMAN,
G., GREB, R., HUMMELSHOJ, L., PRENTICE, A., SARIDOGAN, E. 2004.
ESHRE guideline for the diagnosis and treatment of endometriosis. Hum
Reprod, 20, 2698-704.
KENNEY, N., ENGLISH, J., TAYLOR, A., HOLE, L., NORBROOK, M. 2005.
Anterior rectal resection to treat rectovaginal endometriosis. Median term
follow-up and complications. Eu J Obstet Gynecol Reprod Biol, 123, S17.
KETTEL, L. M., MURPHY, A. A., MORALES, A. J., ULMANN, A., BAULIEU, E. E.
& YEN, S. S. 1996. Treatment of endometriosis with the antiprogesterone
mifepristone (RU486). Fertil Steril, 65, 23-8.
KISTNER, R. 1958. The use of newer progestins in the treatment of endometriosis. Am
J Obstet Gynecol, 75, 264-278.
KITAWAKI, J., KADO, N., ISHIHARA, H., KOSHIBA, H., KITAOKA, Y. &
HONJO, H. 2002. Endometriosis: the pathophysiology as an estrogen-dependent
disease. J Steroid Biochem Mol Biol, 83, 149-55.
KJERULFF K. H., E. B. A., LANGENBERG P.W. 1996. Chronic gynaecological
conditions reported by US women: findings from the National Health Interview
Survey. 1984 to 1992. Am J Public Health, 86, 195-9.
KONINCKX, P. R. & MARTIN, D. 1994. Treatment of deeply infiltrating
endometriosis. Curr Opin Obstet Gynecol, 6, 231-41.
KONINCKX, P. R., MEULEMAN, C., DEMEYERE, S., LESAFFRE, E. &
CORNILLIE, F. J. 1991. Suggestive evidence that pelvic endometriosis is a
progressive disease, whereas deeply infiltrating endometriosis is associated with
pelvic pain. Fertil Steril, 55, 759-65.
KONINCKX, P. R., MEULEMAN, C., OOSTERLYNCK, D. & CORNILLIE, F. J.
1996. Diagnosis of deep endometriosis by clinical examination during
menstruation and plasma CA-125 concentration. Fertil Steril, 65, 280-7.
KONINCKX, P. R., OOSTERLYNCK, D., D'HOOGHE, T. & MEULEMAN, C. 1994.
Deeply infiltrating endometriosis is a disease whereas mild endometriosis could
be considered a non-disease. Ann N Y Acad Sci, 734, 333-41.
KUPKER, W., FELBERBAUM, R. E., KRAPP, M., SCHILL, T., MALIK, E. &
DIEDRICH, K. 2002. Use of GnRH antagonists in the treatment of
endometriosis. Reprod Biomed Online, 5, 12-6.
LEIBSON, C. L., GOOD, A. E., HASS, S. L., RANSOM, J., YAWN, B. P.,
O'FALLON, W. M. & MELTON, L. J., 3RD 2004. Incidence and
characterization of diagnosed endometriosis in a geographically defined
population. Fertil Steril, 82, 314-21.
LEMOS, N. A., ARBO, E., SCALCO, R., WEILER, E., ROSA, V. & CUNHA-FILHO,
J. S. 2008. Decreased anti-Mullerian hormone and altered ovarian follicular
cohort in infertile patients with mild/minimal endometriosis. Fertil Steril, 89,
1064-8.
LOMANO, J. M. 1987. Nd:YAG laser ablation of early pelvic endometriosis: a report
of 61 cases. Lasers Surg Med, 7, 56-60.

170
LUCERO, S. P., WISE, H. A., KIRSH, G., DEVOE, K., HESS, M. L.,
KANDAWALLA, N. & DRAGO, J. R. 1988. Ureteric obstruction secondary to
endometriosis. Report of three cases with a review of the literature. Br J Urol,
61, 201-4.
LUCIANO, A. A., LOWNEY, J. & JACOBS, S. L. 1992. Endoscopic treatment of
endometriosis-associated infertility. Therapeutic, economic and social benefits. J
Reprod Med, 37, 573-6.
MAGRINA, J. F. 2007. Robotic surgery in gynecology. Eur J Gynaecol Oncol, 28, 77-
82.
MAHMOOD, T. A. & TEMPLETON, A. 1991. Prevalence and genesis of
endometriosis. Hum Reprod, 6, 544-9.
MARCOUX, S., MAHEUX, R. & BERUBE, S. 1997. Laparoscopic surgery in infertile
women with minimal or mild endometriosis. Canadian Collaborative Group on
Endometriosis. N Engl J Med, 337, 217-22.
MARSH, E. E. & LAUFER, M. R. 2005. Endometriosis in premenarcheal girls who do
not have an associated obstructive anomaly. Fertil Steril, 83, 758-60.
MARTIN, D. C. & VANDER ZWAGG, R. 1987. Excisional techniques for
endometriosis with the CO2 laser laparoscope. J Reprod Med, 32, 753-8.
MATORRAS, R., RODIQUEZ, F., PIJOAN, J. I., RAMON, O., GUTIERREZ DE
TERAN, G. & RODRIGUEZ-ESCUDERO, F. 1995. Epidemiology of
endometriosis in infertile women. Fertil Steril, 63, 34-8.
MATSUZAKI, S., CANIS, M., POULY, J., BOTCHORISHVILI, R., HOULLE, C.,
JARDON, K., RABISCHONG, G., MAGE, G. 2004. Relationships between
delay before sugical diagnosis and severity of disease in patients with
symptomatic deep infiltrating endometriosis. Eu J Obstet Gynecol Reprod Biol,
123, S9.
MCSWAIN, B., LINN, R. J., HALEY, R. L., JR. & FRANKLIN, R. H. 1974.
Endometriosis of the colon: report of 14 patients requiring partial colectomy.
South Med J, 67, 651-8.
MEIGS, J., . 1922. Endometrial hematomas of the ovary. Bost Med Surg J, 187, 1-13.
MELTZER, R., HOENIG, D., CHROSTEK, C., AMARAL, J. 1994. Porcine
seromyotomies using an ultrasonically activated scalpel. Surg Endosc, 8, 253.
METZGER, D. A. 1997. Treating endometriosis pain: a multidisciplinary approach.
Semin Reprod Endocrinol, 15, 245-50.
MEYER, R. 1909. Uber entzundliche neterope epithelwucherungen im weiblichen
Genetalg ebeit und uber eine bis in die Wurzel des Mesocolon ausgedehnte
benigne Wucherung des Dar mepithel. Virch Arch Pathol Anat, 195, 487.
MILLER, J. D. 2000. Quantification of endometriosis-associated pain and quality of life
during the stimulatory phase of gonadotropin-releasing hormone agonist
therapy: a double-blind, randomized, placebo-controlled trial. Am J Obstet
Gynecol, 182, 1483-8.
MOL, B. W., BAYRAM, N., LIJMER, J. G., WIEGERINCK, M. A., BONGERS, M.
Y., VAN DER VEEN, F. & BOSSUYT, P. M. 1998. The performance of CA-
125 measurement in the detection of endometriosis: a meta-analysis. Fertil
Steril, 70, 1101-8.
MOORE, J., COPLEY, S., MORRIS, J., LINDSELL, D., GOLDING, S. & KENNEDY,
S. 2002. A systematic review of the accuracy of ultrasound in the diagnosis of
endometriosis. Ultrasound Obstet Gynecol, 20, 630-4.
MOORE, J., KENNEDY, S., PRENTICE, A. 2004. Modern oral contraceptives for pain
associated with endometriosis (Cochrane Review). In the Cochrane Library.

171
NAKASH R, J. H., E JORSTAD-STEIN, S GATES, S LAMB 2006. Maximising
response to postal questionnaires - A systematic review of randomised trials in
health research. BMC Med Res Methodol, 6.
NARDO, L. G., MOUSTAFA, M. & GARETH BEYNON, D. W. 2005. Laparoscopic
treatment of pelvic pain associated with minimal and mild endometriosis with
use of the Helica Thermal Coagulator. Fertil Steril, 83, 735-8.
NAWATHE, A., PATWARDHAN, S., YATES, D., HARRISON, G. R. & KHAN, K.
S. 2008. Systematic review of the effects of aromatase inhibitors on pain
associated with endometriosis. Bjog, 115, 818-22.
NEZHAT, C., NEZHAT, F., PENNINGTON, E., NEZHAT, C. H. & AMBROZE, W.
1994. Laparoscopic disk excision and primary repair of the anterior rectal wall
for the treatment of full-thickness bowel endometriosis. Surg Endosc, 8, 682-5.
NISHIDA, M., WATANABE, K., SATO, N. & ICHIKAWA, Y. 2000. Malignant
transformation of ovarian endometriosis. Gynecol Obstet Invest, 50 Suppl 1, 18-
25.
OLIKER, A. J. & HARRIS, A. E. 1971. Endometriosis of the bladder in a male patient.
J Urol, 106, 858-9.
OLIVE, D. L. & SCHWARTZ, L. B. 1993. Endometriosis. N Engl J Med, 328, 1759-
69.
OSTEEN, K. G., BRUNER-TRAN, K. L. & EISENBERG, E. 2005. Reduced
progesterone action during endometrial maturation: a potential risk factor for the
development of endometriosis. Fertil Steril, 83, 529-37.
OSTEEN, K. G., YEAMAN, G. R. & BRUNER-TRAN, K. L. 2003. Matrix
metalloproteinases and endometriosis. Semin Reprod Med, 21, 155-64.
OSUGA, Y., KOGA, K., TSUTSUMI, O., YANO, T., MARUYAMA, M., KUGU, K.,
MOMOEDA, M. & TAKETANI, Y. 2002. Role of laparoscopy in the treatment
of endometriosis-associated infertility. Gynecol Obstet Invest, 53 Suppl 1, 33-9.
PETTA, C. A., FERRIANI, R. A., ABRAO, M. S., HASSAN, D., ROSA, E. S. J. C.,
PODGAEC, S. & BAHAMONDES, L. 2005. Randomized clinical trial of a
levonorgestrel-releasing intrauterine system and a depot GnRH analogue for the
treatment of chronic pelvic pain in women with endometriosis. Hum Reprod, 20,
1993-8.
PLUMB, K., SINAII, N., COTTON, L., ZONDERVAN, K., TRELOAR, S.,
LAMBERT, A., STRATTON, P., KENNEDY, S. 2005. Differences in
characteristics of 1000 women with endometriosis based on disease severity. Eu
J Obstet Gynecol Reprod Biol, 123, S5.
POSSOVER, M., DIEBOLDER, H., PLAUL, K. & SCHNEIDER, A. 2000.
Laparascopically assisted vaginal resection of rectovaginal endometriosis.
Obstet Gynecol, 96, 304-7.
R NAKASH, J. H., E JORSTAD-STEIN, S GATES, S LAMB 2006. Maximising
response to postal questionnaires - A systematic review of randomised trials in
health research. BMC Med Res Methodol, 6.
REDWINE, D. 1996. Database management on a personal computer.
j.Am.Assoc.Gynecol.Laparosc, 3, 643-650.
REDWINE, D. B. 1990a. American Fertility Society classification of endometriosis--
the last word? Fertil Steril, 54, 180-1.
REDWINE, D. B. 1990b. The visual appearance of endometriosis and its impact on our
concepts of disease. Prog Clin Biol Res, 323, 393-412.
REDWINE, D. B. 1991. Conservative laparoscopic excision of endometriosis by sharp
dissection: life table analysis of reoperation and persistent or recurrent disease.
Fertil Steril, 56, 628-34.

172
REDWINE, D. B. 1993. Laparoscopic excision of endometriosis with 3-mm scissors:
comparison of operating times between sharp excision and electro-excision. J
Am Assoc Gynecol Laparosc, 1, 24-30.
REDWINE, D. B. 2002. Was Sampson wrong? Fertil Steril, 78, 686-93.
REDWINE, D. B., KONING, M. & SHARPE, D. R. 1996. Laparoscopically assisted
transvaginal segmental resection of the rectosigmoid colon for endometriosis.
Fertil Steril, 65, 193-7.
RIBEIRO, H. S., RIBEIRO, P. A., ROSSINI, L., RODRIGUES, F. C., DONADIO, N.
& AOKI, T. 2008. Double-contrast barium enema and transrectal endoscopic
ultrasonography in the diagnosis of intestinal deeply infiltrating endometriosis. J
Minim Invasive Gynecol, 15, 315-20.
SAMPSON, J. A. 1927. Peritoneal endometriosis due to the menstrual dissemination of
endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol, 14, 422-69.
SANGI-HAGHPEYKAR, H. & POINDEXTER, A. N., 3RD 1995. Epidemiology of
endometriosis among parous women. Obstet Gynecol, 85, 983-92.
SCHENKEN, R. S. 1990. Gonadotropin-releasing hormone analogs in the treatment of
endometriomas. Am J Obstet Gynecol, 162, 579-81.
SELAK, V., FARQUHAR, C., PRENTICE, A. 2004. Danazol for pelvic pain associated
with endometriosis. The Cochrane library.
SHAFIK A, R. N., WRIGHT JT. 2000. The importance of histological diagnosis in
patients with chronic pelvic pain and laproscopic evidence of endometriosis.
Gynaecol Endosc, 9, 301-304.
SHAW, R. W. 1992. The role of GnRH analogues in the treatment of endometriosis. Br
J Obstet Gynaecol, 99 Suppl 7, 9-12.
SHEPPERSON MILLS, D. 2004. Nutritional counselling and therapy provides a cost
effective method of improvement in the pain scores and the sense of wellbeing
in women with endometriosis. Eu J Obstet Gynecol Reprod Biol, 123, S42.
SIMPSON, J. L., ELIAS, S., MALINAK, L. R. & BUTTRAM, V. C., JR. 1980.
Heritable aspects of endometriosis. I. Genetic studies. Am J Obstet Gynecol,
137, 327-31.
SLACK, A., CHILD, T., LINDSEY, I., KENNEDY, S., CUNNINGHAM, C.,
MORTENSEN, N., KONINCKX, P. & MCVEIGH, E. 2007. Urological and
colorectal complications following surgery for rectovaginal endometriosis. Bjog,
114, 1278-82.
SURREY, E. S., SILVERBERG, K. M., SURREY, M. W. & SCHOOLCRAFT, W. B.
2002. Effect of prolonged gonadotropin-releasing hormone agonist therapy on
the outcome of in vitro fertilization-embryo transfer in patients with
endometriosis. Fertil Steril, 78, 699-704.
SUTTON, C. 1989. CO2 laser laparoscopy in the treatment of endometriosis. Baillieres
Clin Obstet Gynaecol, 3, 499-523.
SUTTON, C. J., EWEN, S. P., JACOBS, S. A. & WHITELAW, N. L. 1997a. Laser
laparoscopic surgery in the treatment of ovarian endometriomas. J Am Assoc
Gynecol Laparosc, 4, 319-23.
SUTTON, C. J., EWEN, S. P., WHITELAW, N. & HAINES, P. 1994. Prospective,
randomized, double-blind, controlled trial of laser laparoscopy in the treatment
of pelvic pain associated with minimal, mild, and moderate endometriosis. Fertil
Steril, 62, 696-700.
SUTTON, C. J. & JONES, K. D. 2002. Laser laparoscopy for endometriosis and
endometriotic cysts. Surg Endosc, 16, 1513-7.
SUTTON, C. J., POOLEY, A. S., EWEN, S. P. & HAINES, P. 1997b. Follow-up report
on a randomized controlled trial of laser laparoscopy in the treatment of pelvic

173
pain associated with minimal to moderate endometriosis. Fertil Steril, 68, 1070-
4.
TELIMAA, S., PUOLAKKA, J., RONNBERG, L. & KAUPPILA, A. 1987. Placebo-
controlled comparison of danazol and high-dose medroxyprogesterone acetate in
the treatment of endometriosis. Gynecol Endocrinol, 1, 13-23.
THIBODEAU, L. L., PRIOLEAU, G. R., MANUELIDIS, E. E., MERINO, M. J. &
HEAFNER, M. D. 1987. Cerebral endometriosis. Case report. J Neurosurg, 66,
609-10.
TRELOAR, S., WICKS, J., NYHOLT, D., MONTGOMERY, G., BAHLO, M.,
KENNEDY, S. 2005a. Genomewide linkage study in 1,176 affected sister pair
families identifies a significant susceptibility locus for endometriosis on
chromosome 10q26. Am J Hum Gen 77, 365-76.
TRELOAR, S. A., MARTIN N.G., KENNEDY S.H., MONTGOMERY G.W. Year.
Characteristics and symptoms in 3985 women diagnosed with endometriosis in
an Australian genetic epidemiological study. In: 9th World Congress on
Endometriosis, 2005 2005b Maastricht, Holland. Elsevier, S5.
TUTTLIES, F., KECKSTEIN, J., ULRICH, U., POSSOVER, M., SCHWEPPE, K. W.,
WUSTLICH, M., BUCHWEITZ, O., GREB, R., KANDOLF, O., MANGOLD,
R., MASETTI, W., NEIS, K., RAUTER, G., REEKA, N., RICHTER, O.,
SCHINDLER, A. E., SILLEM, M., TERRUHN, V. & TINNEBERG, H. R.
2005. [ENZIAN-score, a classification of deep infiltrating endometriosis].
Zentralbl Gynakol, 127, 275-81.
ULRICH, U., RHIEM, K., KAMINSKI, M., WARDELMANN, E., TROG, D.,
VALTER, M., RICHTER, O. 2005. Parametrial and rectovaginal
adenocarcinoma arising from endometriosis. Int J Gynaecological Cancer, 15,
1206-9.
VAN LANGENDONCKT, A., MARQUES DE SAFE, G., GONZALEZ, D.,
SQUIFFLET, J., DONNEZ, J. 2004. HOXA-10 and HOXA-13 gene expression
in endometriotic nodules of the vaginal septum. . Eu J Obstet Gynecol Reprod
Biol, 123, S35.
VERCELLINI, P., AIMI, G., BUSACCA, M., APOLONE, G., UGLIETTI, A. &
CROSIGNANI, P. G. 2003. Laparoscopic uterosacral ligament resection for
dysmenorrhea associated with endometriosis: results of a randomized, controlled
trial. Fertil Steril, 80, 310-9.
VERCELLINI, P., AIMI, G., PANAZZA, S., DE GIORGI, O., PESOLE, A. &
CROSIGNANI, P. G. 1999a. A levonorgestrel-releasing intrauterine system for
the treatment of dysmenorrhea associated with endometriosis: a pilot study.
Fertil Steril, 72, 505-8.
VERCELLINI, P., CHAPRON, C., FEDELE, L., GATTEI, U., DAGUATI, R. &
CROSIGNANI, P. G. 2004. Evidence for asymmetric distribution of lower
intestinal tract endometriosis. Bjog, 111, 1213-7.
VERCELLINI, P., CROSIGNANI, P. G., FADINI, R., RADICI, E., BELLONI, C. &
SISMONDI, P. 1999b. A gonadotrophin-releasing hormone agonist compared
with expectant management after conservative surgery for symptomatic
endometriosis. Br J Obstet Gynaecol, 106, 672-7.
VERCELLINI, P., SOMIGLIANA, E., VIGANO, P., ABBIATI, A., BARBARA, G. &
CROSIGNANI, P. G. 2009. Surgery for endometriosis-associated infertility: a
pragmatic approach. Hum Reprod, 24, 254-69.
VERCELLINI, P., VENDOLA, N., BOCCIOLONE, L., COLOMBO, A., ROGNONI,
M. T. & BOLIS, G. 1992. Laparoscopic aspiration of ovarian endometriomas.

174
Effect with postoperative gonadotropin releasing hormone agonist treatment. J
Reprod Med, 37, 577-80.
VERSPYCK, E., LEFRANC, J. P., GUYARD, B. & BLONDON, J. 1997. Treatment of
bowel endometriosis: a report of six cases of colorectal endometriosis and a
survey of the literature. Eur J Obstet Gynecol Reprod Biol, 71, 81-4.
VESSEY, M. P., VILLARD-MACKINTOSH, L. & PAINTER, R. 1993. Epidemiology
of endometriosis in women attending family planning clinics. Bmj, 306, 182-4.
VIGANO, P., PARAZZINI, F., SOMIGLIANA, E., VERCELLINI, P. 2004.
Endometriosis: epidemiology and aetiological factors. Best Pract Res Clin
Obstet Gynaecol, 18, 177-200.
VON ROKITANSKY, C. Ueber Uterusdruden-Neubildung in Uterus und Ovarial
Sarcomen. Zesch. Gesselsch. Aertz., 16, 577.
WATERS N, M. G., ROCKALL T, KENT A 20008. Rectal stenosis after laparoscopic
excision of rectovaginal endometriosis. Journal of Gynaecological Surgery, 5,
34-35.
WEED, J. C. & RAY, J. E. 1987. Endometriosis of the bowel. Obstet Gynecol, 69, 727-
30.
WRIGHT, J., LOTFALLAH, H., JONES, K. & LOVELL, D. 2005. A randomized trial
of excision versus ablation for mild endometriosis. Fertil Steril, 83, 1830-6.
ZHAO, S. Z., KELLERMAN, L. A., FRANCISCO, C. A. & WONG, J. M. 1999.
Impact of nafarelin and leuprolide for endometriosis on quality of life and
subjective clinical measures. J Reprod Med, 44, 1000-6.
ZIGMOND, A. S. S., R.P. 1983. The Hospital Anxiety and Depression Scale. Acta
Psychiatrica Scandinavica, 67, 361-370.
!

175
!
Appendices

176
Appendix A – Reviews in Gynaecological Practice Review
Article

177

178

179

180

181

182

183

184

185
Appendix B – Endometriosis Trial Patient Information
Leaflet
Patient information
A study comparing:
Laser surgery with the harmonic scalpel
In the treatment of pelvic pain due to endometriosis.
THANK YOU FOR READING THIS.
Introduction
You are being invited to take part in a research study. Before you decide it is important
for you to understand why the research is being done and what it will involve. Please
take time to read the following information carefully and discuss it with friends,
relatives and your GP if you wish. Ask us if there is anything that is not clear or if you
would like more information. Take time to decide whether or not you wish to take part.
Consumers for Ethics in Research (CERES) publish a leaflet entitled ‘ Medical
Research and You ‘. This leaflet gives more information about medical research and
looks at some questions you may want to ask. A copy may be obtained from CERES,
PO Box 1365, London N16 OBW.
Purpose of Study
Endometriosis is a common and debilitating disorder that diminishes the quality of
women’s lives in their reproductive years. It is associated with pelvic pain, infertility
and menstrual disorders. It remains a difficult condition to treat. Drugs are ineffective in
the long term and have unacceptable side effects. However, laparoscopic (key hole)
surgery has been shown to be an effective form of treatment.
Keyhole laser surgery was first performed in this country at St. Luke's Hospital in
Guildford in l982, so we have 20 years experience of this technique and an excellent
safety record on more than 4,500 patients. We conducted the world’s first scientific
study of laser surgery, which clearly showed that laser treatment is an effective
treatment in the majority of patients and has considerable advantages over medical
therapy with anti-endometriosis drugs.
It is now our normal practice is to use a carbon dioxide laser to vaporise the
endometriotic deposits.
Recently, the UltraCision Harmonic Scalpel has been developed. It was introduced
commercially in 1993, and since then it has been increasingly used in surgery. We
would like to lead the way in developing the use of the UltraCision Harmonic Scalpel in
the treatment of endometriosis.

186
The UltraCision Harmonic Scalpel converts electrical energy, to mechanical motion at
the instrument’s titanium tip, which vibrates at a speed of 55,500 cycles per second.
This mechanical energy allows both cutting of tissue and coagulation (the prevention of
bleeding) at the precise point of impact. For this reason, we feel that the UltraCision
Harmonic Scalpel offers a safe alternative to the carbon dioxide laser. Further more,
keyhole laser surgery is not widely available in this country. The UltraCision Harmonic
Scalpel is easier to use than a laser, and considerable cheaper to purchase. It is
potentially much more widely available in other hospitals, and it also has multiple
applications in gynaecological surgery, whereas the lasers have much more limited
range of uses.
Why have I been chosen?
Patients with pelvic pain and a known or possible diagnosis of endometriosis are being
asked to take part in the study. We plan to recruit around 100 patients over the space of
one year.
Do I have to take part?
It is up you to decide whether or not to take part. If you do decide to take part you will
be given this information sheet to keep and be asked to sign a consent form. If you
decide to take part you are still free to withdraw at any time and without giving a
reason. This will not affect the standard of care you receive. If you do not wish to take
part in the study, then you will receive keyhole laser surgery as per our usual practice.
What will happen to me if I take part?
Your symptoms will be recorded before surgery in the usual way. Standard
questionnaires will be used to assess the effect your symptoms have on the quality of
your life and on your psychological state.
The comparison of laser and the UltraCision Harmonic Scalpel has never been
subjected to a scientific study. Because we do not know which way of treating patients
is best, we need to make such a comparison. People will be put into two groups. The
groups will be selected by a computer, which has no information about the individual
(i.e. by chance). Patients in each group will then have a different treatment and these
treatments are then compared. Because this is a double blind trial neither you nor the
research nurse who will follow you up in the clinic will know in which treatment group
you are (although, if your doctor needs to find out he / she can do so).
The operation (keyhole surgery) is performed under general anaesthesia through a
telescope called a laparoscope, which is introduced inside the abdomen via a tiny cut
within the umbilicus. You will also have 2 or 3 small cuts just above the hairline for
additional instruments to be inserted into your abdomen. The remaining endometriotic
deposits will then be destroyed by means of the UltraCision Harmonic Scalpel or the
laser.

187
The research nurse will also record any changes in your symptoms at 3, and 6 months
following surgery. The need for any medical therapy or further surgical treatment for
pelvic pain will also be recorded.
What do I have to do?
You will not have to restrict or change your life style in any way. This study will not
affect your fertility intentions at all. Those women who want to get pregnant will be
encouraged to do so. Those women who do not will be asked to use a barrier
contraceptive (cap, condom or coil) or be sterilised as they wish. We will ask you to
avoid using an oral or an injectable hormonal contraceptive for the 6 months following
surgery. This is because hormonal contraceptives affect the degree of pain and the
amount of bleeding during a period.
What are the additional risks of taking part?
There are no additional risks beyond that of routine laparoscopic (keyhole) surgery. The
risks include injury to the bowel, bladder, or blood vessels, which occurs in about 1 in
1000 cases and may need a laparotomy (a surgical procedure to open the abdomen and
repair the injuries). These issues are routinely discussed with the patients at the time of
informed consent.
What are the possible benefits of taking part?
We hope that both the treatments will help you. However, this cannot be guaranteed.
The information we get from this study may help us to treat future patients
endometriosis better.
What if new information becomes available?
If new information about the treatment being studied becomes available during the
course of a research project, you will be told about it. If this happens, your research
doctor will discuss with you whether you want to continue in the study. If you decide to
withdraw your research doctor will make arrangements for your care to continue. If you
decide to continue in the study you will be asked to sign an updated consent form.
What if something goes wrong?
If taking part in this research project harms you, there are no special compensation
arrangements. If you are harmed due to someone’s negligence, then you may have
grounds for legal action. Regardless of this, if you wish to complain about any aspect of
the way you have been approached or treated during the course of this study, the normal
National Health Service complaints mechanisms are available to you.

188
Will my taking part in this study be kept confidential?
All information that is collected about you during the course of the research will be kept
strictly confidential. Any information about you that leaves the hospital will have your
name and address removed so that you cannot be recognised from it. Your GP or the
doctor who referred you to Mr Kent for treatment will be kept informed of your
treatment and your participation in the study by hospital discharge summary in the usual
way.
What will happen to the results of the study?
After your 6 months follow up appointment the research doctor will tell you which arm
of the study you were in. The results of this study will be published in a medical
journal, and they will be presented to other doctors at scientific meetings. You will not
be identified in any report or publication.
Who is organising and funding the research?
The project is being organised by Mr Peter Barton-Smith and Mr Andrew Kent. We do
not receive any specific funding for this project at the present time. Ethicon-endo
Surgery, who make the UltraCision Harmonic Scalpel do however contribute to the
research budget of the department.
Who has reviewed the study?
The South West Surrey LREC has reviewed and approved this study.
Contact for further information
The persons to contact in the event of an unexpected reaction or incident will be
Mr Peter Barton-Smith, 01483 571122 page 4210 or Mrs Pat Haines 01483 571122 ext
4569.
Please keep this information sheet for your own records, together with the copy of
the consent form.

189
Appendix C – Endometriosis Trial Consent Form
Study Number:
Patient Identification Number for this trial
CONSENT FORM
Title of Project: Carbon dioxide-laser versus UltraCision Harmonic Scalpel in the
treatment of pelvic pain due to endometriosis.
Name of Researcher: Mr Peter Barton-Smith, Clinical Research Fellow in
Gynaecological Endoscopy.
1. I confirm that I have read and understand the information sheet dated
…………………. (Version) for the above study and have had the opportunity to
ask questions
2. I understand that my participation is voluntary and that I am free to withdraw at
any time, without giving any reason, without my medical care or legal rights
being affected
3. I agree to take part in the study
Name of Patient Date Signature
Name of Person Taking Consent Date Signature
(If different from researcher)
Researcher Date Signature
1 for patient; 1 for researcher; 1 to be kept in hospital notes

190
Appendix D – Endometriosis Trial Visual Analogue Scale
Visual Analogue Score for Pelvic Pain due to Endometriosis
Hospital No:
Dysmenorrhoea Yes / No / Not Sure
(painful periods)
No pain _______________________________________________ Worst pain
Chronic Pelvic Pain Yes / No / Not Sure
(pain not linked to periods going on for >6 months)
No pain _______________________________________________ Worst pain
Dyspareunia Yes / No / Not Sure
(painful sexual intercourse)
No pain _______________________________________________ Worst pain
Dyschezia Yes / No / Not Sure
(pain on bowel opening)
No pain _______________________________________________ Worst pain
Additional information:

191
Appendix E – Endometriosis Trial EHP 30 Questionnaire
University of Oxford National Endometriosis Society
The Endometriosis Health Profile Questionnaire (EHP 30)
© 2001 Nuffield Department of Obstetrics & Gynaecology & Health Services Research
Unit
University of Oxford
In collaboration with The National Endometriosis Society U.K.
___________________________________________________________________________
 This questionnaire has been developed to measure the effect endometriosis has upon
a woman's quality of life
 To complete the questionnaire please would you answer;
Part 1: All 30 questions
Part 2: All sections that apply to you
 We are aware that you may have had endometriosis for a long time. We also
understand that how you feel now may be different to how you have felt in the past.
However, please would you answer the questions only in relation to the effect that
endometriosis has had on your life during the last 4 weeks
 There are no right or wrong answers, so please tick the answers which best
represent your feelings and experiences.
 Due to the personal nature of some of the questions please understand that you do
not have to answer any questions if you would prefer not to.
 The information and answers you give will be treated with the utmost
confidentiality.
 If you have any problems or would like any help or assistance with the completion
of this questionnaire please contact Pat Haines on 01483 406797 who will be
happy to help you.
 Once you have completed the questionnaire please could you return it in the pre-
paid envelope provided.
 We would like to thank you very much in anticipation for taking the time to help us
with this important research and we look forward to receiving your answers.
• This research was funded with an educational grant from Pharmacia Corporation,
USA

192
_________________________________________________________
PART 1: CORE QUESTIONNAIRE
DURING THE LAST 4 WEEKS, HOW OFTEN
BECAUSE OF YOUR ENDOMETRIOSIS HAVE YOU…..
Never Rarely Sometimes Often Always
______________________________________________________________________________________________
1. Been unable to go to social events     
because of the pain?
_____________________________________________________________________________
2. Been unable to do jobs around the     
home because of the pain?
_____________________________________________________________________________
3. Found it difficult to stand because     
of the pain?
_____________________________________________________________________________
4. Found it difficult to sit because of     
the pain?
_____________________________________________________________________________
5. Found it difficult to walk because     
of the pain?
_____________________________________________________________________________
6. Found it difficult to exercise or do the     
leisure activities you would like to do
because of the pain?
_____________________________________________________________________________
7. Lost your appetite and/or been unable    
to eat because of the pain?
_____________________________________________________________________________
Please check that you have ticked one box for each question
before moving onto the next page

193
DURING THE LAST 4 WEEKS, HOW OFTEN
BECAUSE OF YOUR ENDOMETRIOSIS HAVE YOU…..
Never Rarely Sometimes Often Always
_____________________________________________________________________________
8. Been unable to sleep properly because     
of the pain?
_____________________________________________________________________________
9. Had to go to bed/lie down because of     
the pain?
_____________________________________________________________________________
10. Been unable to do the things you want    
to do because of the pain?
______________________________________________________________________________________________
11. Felt unable to cope with the pain?     
_____________________________________________________________________________
12. Generally felt unwell?     
_____________________________________________________________________________
13. Felt frustrated because your symptoms    
are not getting better?
_____________________________________________________________________________
14. Felt frustrated because you are not     
able to control your symptoms?
_____________________________________________________________________________
Please check that you have ticked one box for each question
before moving onto the next page

194
DURING THE LAST 4 WEEKS, HOW OFTEN
BECAUSE OF YOUR ENDOMETRIOSIS HAVE YOU…..
Never Rarely Sometimes Often Always
_____________________________________________________________________________
15. Felt unable to forget your symptoms?     
_____________________________________________________________________________
16. Felt as though your symptoms are     
ruling your life?
_____________________________________________________________________________
17. Felt your symptoms are taking away     
your life?
______________________________________________________________________________________________
18. Felt depressed?     
_____________________________________________________________________________
19. Felt weepy/tearful?     
_____________________________________________________________________________
20. Felt miserable?     
_____________________________________________________________________________
21. Had mood swings?     
_____________________________________________________________________________
22. Felt bad tempered or short tempered?     
_____________________________________________________________________________
Please check that you have ticked one box for each question
before moving onto the next page

195
DURING THE LAST 4 WEEKS, HOW OFTEN
BECAUSE OF YOUR ENDOMETRIOSIS HAVE YOU…..
Never Rarely Sometimes Often Always
_____________________________________________________________________________
23. Felt violent or aggressive?     
_____________________________________________________________________________
24. Felt unable to tell people how you feel?    
_____________________________________________________________________________
25. Felt others do not understand what you     
are going through?
_____________________________________________________________________________
26. Felt as though others think you are     
moaning?
_____________________________________________________________________________
27. Felt alone?     
_____________________________________________________________________________
28. Felt frustrated as you cannot always     
wear the clothes you would choose?
_____________________________________________________________________________
29. Felt your appearance has been affected?    
_____________________________________________________________________________
30. Lacked confidence?     
_____________________________________________________________________________
Please check that you have ticked one box for each question
before moving onto Part 2

196
Part 2: Modular Questionnaire
Section C: These questions concern the effect endometriosis has had on your sexual
relationships during the last 4 weeks
How often during the last 4 weeks
BECAUSE OF YOUR ENDOMETRIOSIS HAVE YOU…..
Never Rarely Sometimes Often Always
_____________________________________________________________________________
1. Experienced pain during or after     
intercourse?
If not relevant please tick here 
2. Felt worried about having intercourse     
because of the pain?
If not relevant please tick here 
3. Avoided intercourse because of the     
pain?
If not relevant please tick here 
4. Felt guilty about not wanting to have    
intercourse?
If not relevant please tick here 
5. Felt frustrated because you cannot     
enjoy intercourse?
If not relevant please tick here 
_____________________________________________________________________________

197
Appendix F – Endometriosis Trial HADS Questionnaire
Hospital Anxiety and Depression Scale (HADS)
Patients are asked to choose one response from the four given for each
interview. They should give an immediate response and be dissuaded
from thinking too long about their answers. The questions relating to
anxiety are marked "A", and to depression "D". The score for each
answer is given in the right column. Instruct the patient to answer how it
currently describes their feelings.
A
I feel tense or 'wound up':


Most of the time
3

A lot of the time
2

From time to time,
occasionally
1

Not at all
0
D
I still enjoy the things I used
to enjoy:


Definitely as much
0

Not quite so much
1

Only a little
2

Hardly at all
3
A
I get a sort of frightened
feeling as if something awful
is about to happen:


Very definitely and quite
badly
3

Yes, but not too badly
2

A little, but it doesn't worry
me
1

Not at all
0

198
D
I can laugh and see the
funny side of things:


As much as I always could
0

Not quite so much now
1

Definitely not so much now
2

Not at all
3
A
Worrying thoughts go
through my mind:


A great deal of the time
3

A lot of the time
2

From time to time, but not
too often
1

Only occasionally
0
D
I feel cheerful:


Not at all
3

Not often
2

Sometimes
1

Most of the time
0
A
I can sit at ease and feel
relaxed:


Definitely
0

Usually
1

Not Often
2

Not at all
3

199
D
I feel as if I am slowed
down:


Nearly all the time
3

Very often
2

Sometimes
1

Not at all
0
A
I get a sort of frightened
feeling like 'butterflies' in
the stomach:


Not at all
0

Occasionally
1

Quite Often
2

Very Often
3
D
I have lost interest in my
appearance:


Definitely
3

I don't take as much care as I
should
2

I may not take quite as much
care
1

I take just as much care as
ever
0
A
I feel restless as I have to be
on the move:


Very much indeed
3

Quite a lot
2

Not very much
1

Not at all
0

200
D
I look forward with
enjoyment to things:


As much as I ever did
0

Rather less than I used to
1

Definitely less than I used to
2

Hardly at all
3
A
I get sudden feelings of
panic:


Very often indeed
3

Quite often
2

Not very often
1

Not at all
0
D
I can enjoy a good book or
radio or TV program:


Often
0

Sometimes
1

Not often
2

Very seldom
3

Scoring (add the As =
Anxiety. Add the Ds =
Depression). The norms
below will give you an idea of
the level of Anxiety and
Depression.


0-7 = Normal


8-10 = Borderline abnormal


11-21 = Abnormal




Reference: Zigmond and Snaith (1983)

201
Appendix G – Endometriosis Trial 12 month Demographic
Questionnaire
Endometriosis Trial Demographic Questionnaire 12 months
…………Months since initial operation
You were recruited into our trial comparing the surgical treatment of endometriosis by
vaporisation with a carbon dioxide laser or excision with harmonic scalpel on:
………………………....
We would now like to ask you to answer these questions about what has happened to
you since your operation.
Further surgery
Have you undergone any of the following procedures since your initial operation?
1. Repeat laparoscopy and treatment of endometriosis WITHOUT removal of the uterus
or ovaries
YES NO
2. Hysterectomy
YES NO
3. Removal of one or both ovaries
YES NO
Details
…………………………………………………………………………………………..
…………………………………………………………………………………………..

202
Medical treatment for endometriosis
Have you received any of the following treatments FOR ENDOMETRIOSIS since
your initial operation?
Combined oral contraceptive pill YES NO
Progesterone pill YES NO
Mirena coil YES NO
GnRH analogues YES NO
Other……………….............. YES NO
Details (e.g.: how soon after operation/dates/ongoing?):
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..

203
Fertility
1. How many times had you been pregnant BEFORE your initial operation?
………………times
Details of dates and outcome (live births/top/miscarriage/IVF etc)….…………………
…………………………………………………………………………………………
2. Were you actively trying to become pregnant BEFORE your initial operation?
YES NO
If YES, how long had you been actively trying for?
……………..Months
3. Have you had difficulty in becoming pregnant AFTER your initial operation?
YES NO
If YES, how long have you been actively trying for?
………………Months
4. How many times have you been pregnant AFTER you had your initial
operation?
……………..times
Details of dates and outcome (live births/TOP/miscarriage etc)….……………………
…………………………………………………………………………………………
…………………………………………………………………………………………

204
5. Have you had infertility treatment since your initial operation?
YES NO
If YES, give details (IVF/IUI, outcome etc)
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
…………………………………………………………………………………………..
Overall
If you had pain from your endometriosis, the pain has changed in the following
way:
Pain free much improved improved same worse much worse
Overall my operation resulted in my quality of life being:
much improved improved same worse much worse

205
Appendix H – rASRM Scoring Sheet
Harmonic / Laser Study
Randomisation No:
Hospital ID no:
Laparoscopic staging of endometriosis
Revised American Fertility Society Classification
Endometriosis
< 1 cm
1 – 3 cm
>3 cm
Pelvic
Superficial
1
2
4
Peritoneum
Deep
2
4
6
Right
Superficial
1
2
4
Ovary
Deep
4
16
20
Left
Superficial
1
2
4
Ovary
Deep
4
16
20
Posterior cul-de-sac
Partial
Complete
obliteration
4
40
Adhesions
< 1/3
1/3 – 2/3
> 2/3
Right
Flimsy
1
2
4
Ovary
Dense
4
8
16
Left
Flimsy
1
2
4
Ovary
Dense
4
8
16
Right
Flimsy
1
2
4
Tube
Dense
4*
8*
16
Left
Flimsy
1
2
4
Tube
Dense
4*
8*
16
*If the fimbrial end of the fallopian tube is completely enclosed, change the point to 16
Additional
endometriosis……………………………………………………………………..............
……………………………………………………………………………………………
Stage 1: 1 - 5 Stage II : 6 - 15 Stage III : 16 - 40 Stage IV : > 40

206
Appendix I – Trial Follow up Clarification letter
Clarification on follow-up protocol for women taking part in the
carbon dioxide laser vs. harmonic scalpel endometriosis study.
March 2006
1. ALL women who are entered into the study must be followed up according to
the protocol at 3, 6 and 12 months post-surgical treatment. This includes women
who have become pregnant and women who have started on other treatments for
endometriosis since being entered into the trial. These women may form
interesting subgroups at later analysis.
2. Should the woman be unavailable at exactly 3,6,or 12 months then every attempt
should be made to ensure that the questionnaires are completed as close to the
correct time as possible.
3. Every attempt within reason should be made to locate women for follow up if
they do not attend for their follow up appointment. Normally this would entail a
polite telephone call to the patient if they have not responded to contact by letter.
4. They must complete ALL of the questionnaires as normal.
5. In cases where the woman is not menstruating due to pregnancy, or medical
treatment inducing a “pseudo menopausal” or “pseudo pregnancy” state, then
they should be directed by the person conducting the questionnaires to omit
answering questions relating to menstruation. The person conducting the
questionnaires should record the reason why these answers have been omitted.
This will include what medication is taken, why, its duration and who started it.
In the case of pregnancy, the LMP and EDD must be recorded. These subgroups
will be analysed at the end of the trial.
6. Initially, ALL women should be invited to the hospital to fill in the
questionnaires to ensure that as many as possible are completed in the same
environment. This will reduce the risk of bias especially in the anxiety
questionnaire, as patients are less likely to be anxious in their home
environments. If they are unwilling and this is not possible, then responses by
telephone or postal questionnaire may be carried out and the method used must
be recorded. This must be first discussed with the research fellow, or if he is
unavailable then AK.
7. Once the patient has completed the questionnaires then the responses should be
also entered onto the SPSS database as soon as possible so that an electronic
record is also held. At present there is only one copy of the results and a
duplicate is required in case of loss of data for any reason. This database is
currently being produced.
Mr Andrew Kent Mr Peter Barton-Smith
Investigator Investigator
13 March 2006

207
Appendix J – Invitation email for Pilot Survey
An International Survey of Surgical Techniques in the
Treatment of Endometriosis – Pilot Study
Compiled by the Minimal Access Therapy Training unit, Guildford, UK.
Approved by the British Society of Gynaecological Endoscopy.
Dear MATTU Faculty,
We are conducting a web survey of the surgical techniques used worldwide by
laparoscopists in the treatment of endometriosis. We would like to invite you to take
part in the pilot study of this survey. It should only take between 5 and 10 minute to fill
in!
The results of this pilot study will be used to improve the survey for general release in
2006. We hope to collate UK data first and then expand the survey to other countries
including the USA, European countries and Australia.
Please now click on the following link to start the survey or come back to this email
later when you have a few moments free to fill it in:
http://surveys.som.surrey.ac.uk/survey?code=32425470
Many thanks for sparing a few minutes to help out
Peter Barton-Smith Andrew Kent Karen Ballard
Research Fellow Gynaecology Director Lecturer

208
Appendix K – Invitation email for Main Web Survey
An International Survey of Surgical Techniques in the Treatment of Endometriosis
Compiled by the Minimal Access Therapy Training unit, Guildford, UK.
In Affiliation with the
American Association of Gynecologic Laparoscopists
European Society of Gynaecological Endoscopy
British Society of Gynaecological Endoscopy
Dear Colleague,
We are conducting a web-survey of the surgical techniques used worldwide by laparoscopists
in the treatment of endometriosis. This email has been sent to members of gynaecological
endoscopy societies around the world. We are interested in how you trained in laparoscopic
surgery, how you treat endometriosis and which instruments you use.
We would like to invite you to take part in this survey. It consists of 34 mainly multiple
choice type questions. It should only take between 5 and 10 minutes to fill in!
Clicking on the link to the survey below will constitute your consent to participate in the
study. However you may withdraw at any time by contacting us (p.barton-
smith@surrey.ac.uk). Your responses will then be removed from the database if you so wish.
The anonymous responses you give are confidential and held on a secure database.
If you complete the survey and enter your email address (for prize draw purpose
only) then you will be automatically entered into a prize draw and one lucky
respondent will win an Apple iPod nano!
YOU MAY RECEIVE THIS EMAIL MORE THAN ONCE BY VIRTUE OF BEING ON THE MAILING
LIST OF MORE THAN ONE SOCIETY. HOWEVER PLEASE ONLY COMPLETE IT ONCE.
Many thanks for sparing a few minutes to help out.
Peter Barton-Smith Andrew Kent Karen Ballard
Research Fellow Director of Gynaecological Surgery Senior Lecturer
Please now click on the following link to start the survey or come back to this email later
when you have a few moments free to fill it in.
http://www.zoomerang.com/survey.zgi?p=WEB227E6KPU7AD

209
Appendix L – Pilot Web Survey

210

211

212

213

214

215

216

217

218

219

220

221

222
Appendix M – Main Web Survey