Sie sind auf Seite 1von 7

REPRODUCTIVE SURGERY

FERTILITY AND STERILITY VOL. 82, NO. 4, OCTOBER 2004

Copyright ©2004 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A.

Laparoscopic excision of endometriosis: a randomized, placebo-controlled trial

Jason Abbott, Ph.D., a Jed Hawe, M.R.C.O.G., b David Hunter, M.R.C.O.G., c Michael Holmes, Ph.D., d Paul Finn, M.Sc., e and Ray Garry, M.D. f

University of New South Wales, Sydney, New South Wales; University of Western Australia, Perth, Western Australia, Australia; Countess of Chester Hospital, Chester; Queens University Belfast, Belfast, Northern Ireland; Leeds Metropolitan University, Leeds; and University of Teesside, Teesside, United Kingdom

Received November 10, 2003; revised and accepted March 9, 2004. Supported by the Academic Department of Gynaecological Surgery, James Cook University Hospital, Teesside, England. Reprint requests: Ray Garry, M.D., University of Western Australia, Department of Obstetrics and Gynaecology, Subiaco Perth, Australia (FAX: 08- 9381-3031; E-mail: rgarry @obsgyn.uwa.edu.au).

a University of New South Wales.

b Countess of Chester Hospital.

c Queens University Belfast.

d Faculty of Health and Environment, Leeds Metropolitan University.

e University of Teesside.

f University of Western Australia.

0015-0282/04/$30.00

doi:10.1016/j.fertnstert.2004.

03.046

878

Objective: To examine the effect on pain and quality of life for women with all stages of endometriosis undergoing laparoscopic surgery compared with placebo surgery. Design: A randomized, blinded, crossover study. Setting: A tertiary referral unit in a district general hospital. Patient(s): Thirty-nine women with histologically proven endometriosis completed the 12-month study. Intervention(s): Women were randomized to receive initially either a diagnostic procedure (the delayed surgical group) or full excisional surgery (the immediate surgery group). After 6 months, repeat laparoscopy was performed, with removal of any pathology present. Main Outcome Measure(s): The end points were changes from baseline values of visual analogue pain scores, validated quality-of-life instruments (EQ-5D and SF-12), and sexual activity questionnaire scores. Patients and assessors of outcomes were blinded to the treatment-group assignment. Result(s): Significantly more of the 39 women operated on according to protocol reported symptomatic improvement after excisional surgery than after placebo: 16 of 20 (80%) vs. 6 of 19 (32%); 2 (1) 9.3. Other aspects of quality of life were also significantly improved 6 months after excisional surgery but not after placebo. Progression of disease at second surgery was demonstrated for women having only an initial diagnostic procedure in 45% of cases, with disease remaining static in 33% and improving in 22% of cases. Nonresponsiveness to surgery was reported in 20% of cases. Conclusion(s): Laparoscopic excision of endometriosis is more effective than placebo at reducing pain and improving quality of life. Surgery is associated with a 30% placebo response rate that is not dependent on severity of disease. Approximately 20% of women do not report an improvement after surgery for endome- triosis. (Fertil Steril 2004;82:878–84. ©2004 by American Society for Reproductive Medicine.) Key Words: Endometriosis, pain, quality of life, randomized controlled surgical trial, laparoscopic surgery

Endometriosis is defined as the presence of endometrial glands and stroma outside of the endometrial cavity. The principal symptoms of endometriosis are pain and infertility (1). The

disease accounts for 10%–15% of new referrals

to gynecologists (2), and its treatment accounts for 25%–35% of laparoscopies and 10%–15%

of hysterectomies each year (2–4). The direct costs of treating endometriosis have been esti-

mated at $5,805 for surgical and $2,418 for medical treatments (5). The indirect costs of

time away from employment, the burden of

pain, and its impact on quality of life are also recognized (6).

The management of endometriosis by lapa-

roscopic excision of disease has been sug-

gested to reduce pain symptoms, though further studies by randomized trials are recommended (7). This study is the first to report the effec- tiveness of excision of all stages of endometri- osis in a blinded, placebo-controlled trial.

MATERIALS AND METHODS

Women for the study were recruited from the outpatient department of The James Cook University Hospital in Middlesbrough, United Kingdom. Approval for the study was obtained from the institution’s ethics com- mittee. Inclusion criteria were clinical symp- toms and signs suggestive of endometriosis, such as dysmenorrhea, nonmenstrual pelvic

pain, dyspareunia or dyschezia, and pelvic abnormality on examination, in association with histologic evidence of endometriosis at the time of surgery. Women were ex- cluded if they had suspected gynecologic malignancy or its precursors, current or chronic pelvic inammatory disease, or became pregnant preoperatively.

Randomization was by computer-generated randomiza- tion blocks in balanced groups of 10, with concealment achieved by third-party allocation to one of two groups. In the delayed surgery group (DSG), women had a staging laparoscopy performed at the time of surgery 1, with note made of revised American Fertility Society score, and a detailed laparoscopic assessment of endometriosis. At sur- gery 2, 6 months later, surgical excision of endometriosis was undertaken by a method previously reported (8), with tissue specimens sent to conrm disease histologically.

In the immediate surgery group (ISG), women had exci- sion of endometriosis by laparoscopy performed at surgery 1. Histologic diagnosis of endometriosis was conrmed. At surgery 2, 6 months later, a laparoscopy was performed with ndings noted and recurrent or residual disease documented in a systematic manner. If endometriosis was evident or suspected, these areas were surgically excised and the spec- imen again sent for histologic analysis.

All women had a size 14 Foley bladder catheter, a size 10 Bellovac closed suction drainage system, a patient-controlled analgesia machine for postoperative analgesia, and intravenous uids continued. At the end of all procedures in both groups, 100 mL of uid was instilled into the peritoneal cavity to ensure some drainage from all patients. Patients were kept in the recovery room, so that their minimum time away from the postoperative ward was 2 hours, to maintain blinding of the postoperative staff.

Medical and nursing staff not involved with the research team undertook pre- and postoperative care. Surgery was undertaken by one of ve surgeons: two consultant gynecol- ogists and three fellows training in minimal-access surgery. The operative notes were kept within the operating suite and could be accessed in case of emergency. Separate notes accompanied the patient to the recovery areas and postoper- ative care ward. No information was given to staff respon- sible for postoperative care, to the patient, to the patients family, or to general practitioners regarding the exact pro- cedure that had been performed. Surgeons and other staff present during the surgical procedure did not interact with the patients postoperatively, to minimize potential inuence.

Before the rst surgery, women completed a visual ana- logue scale (VAS) assessment for dysmenorrhea, nonmen- strual pelvic pain, dyspareunia, and dyschezia. They also completed three quality-of-life instruments: [1] the EQ-5D, which is a measure of patientsself-rated health status and a weighted utility index based on ve different components of healthmobility, self-care, usual activities, pain, anxiety, or

FERTILITY & STERILITY

depression (9), [2] the SF-12, a measure of physical and mental health (10), and [3] a sexual activity questionnaire, which measures sexual pleasure, habit, and pain in a struc- tured and repeatable manner (11). There are normal values reported for the EQ-5D and SF-12 instruments (9, 10). All of the above measures were readministered immediately before the second surgery at 6 months, and on follow-up at 12 months after the rst surgery. Blinded research personnel performed all assessments.

Statistical Analysis

Sample size calculations were based on the assumption that there would be a 50% reduction in pain after excision of endometriosis (8) and that a 22% placebo response to sur- gery would be reported (12). Based on these factors, 40 women were required in the study (Epi Info 6; CDC Epide- miology Ofce, Atlanta, GA). Signicance was set at the 5% level. To allow for a 20% rate due to dropout or exclusion for a diagnosis other than endometriosis, 50 women were sought for the study.

Comparison between the groups was undertaken with the Mann-Whitney test for nonparametric data and the independent-samples t-test, as appropriate. Within-group comparison was undertaken with the Wilcoxon rank sum test for nonparametric data. For dichotomous results, the 2 test was used. For multiple-point testing within groups, Freidmans test was used. For analyses comparing results with a known population mean, a single-sample t-test was performed. All analyses were two-tailed and undertaken by intention to treat.

RESULTS

Between January 1999 and August 2000, 168 women seen in the outpatient department were considered eligible to enter the study: 52 women (31%) consented to partic- ipate and were randomized in the study. Thirteen subse- quently withdrew or were excluded: three women became pregnant before their rst surgery, three women withdrew without explanation, and seven women underwent surgery and had diagnoses other than endometriosis (three had pelvic inammatory disease, three had no pathology, and one had endosalpingiosis).

The remaining 39 women were randomized to delayed surgery (n 19) and immediate surgery (n 20). The mean (SD) age for women in the study was 32.1 (5.8) years. Fifty-one percent of women had previous medical treatment, and 17% had previous surgical treatment for endometriosis. There were no signicant differences between the groups at baseline for any demographic parameter, pain or quality of life measure, or previous treatment for endometriosis.

Operative Findings From Surgery 1

All women had surgery performed as per randomization, with the exception of one woman in the DSG who was noted to have rapidly progressive cystic ovarian disease, who had

879

TABLE 1
TABLE
1

Operative findings at surgeries 1 and 2.

 

Surgery 1

Surgery 2

DSG

ISG

DSG vs. ISG

DSG

ISG

DSG vs. ISG

 

(n

19)

(n

20)

P

(n

18)

(n

16)

P

No disease

 

——

0

9 (56)

2 (1) 13.77,

 

P

.0001

rAFS Stage I, n (%)

0

1 (5)

1

2 (11)

3 (18)

1

rAFS

Stage

II, n (%)

8 (42)

9 (45)

0.85

4 (22)

1 (8)

.34

rAFS Stage III, n (%) rAFS Stage IV, n (%) Median rAFS score (range)

2 (10)

2 (10)

1

2 (11)

0

.49

9 (48)

8 (40)

0.75

10 (66)

3 (18)

2 (1) 4.85, P .02

27 (6142)

16 (3142)

0.84

46 (3142)

0 (0142)

 

.0001

Note: rAFS revised American Fertility Society; Stage I minimal disease; Stage II mild disease; Stage III moderate disease; Stage IV severe disease.

Abbott. Excision of endometriosis for pain: RCT. Fertil Steril 2004.

full excision of endometriosis performed. Operative ndings by stage are detailed in Table 1. There were two complica- tions in the ISG; one patient had disease of the posterior cervix that required a laparotomy to repair, and a second patient had a 2-U blood transfusion postoperatively for symptomatic surgical anemia.

Postoperative Pain and Quality-of-Life Outcomes From Surgery 1

A signicantly greater number of women reported an

improvement in their symptoms in the ISG (80%) compared with women in the DSG (32%). Table 2 summarizes the outcomes for response to surgery 1. Table 3 summarizes individual pain symptoms, and Tables 4 and 5 summarize the quality-of-life outcomes after surgery 1.

Operative Findings From Surgery 2

In the DSG, 18 of 19 patients and in the ISG 16 of 20

patients had surgery 2 according to protocol. In the ISG, 3

pregnant women and 1 with a complication at surgery 1 did not have surgery 2. Comparing revised American Fertility Society scores before rst and second surgery in the DSG, 8 of 18 patients (45%) had disease that was worse, 6 (33%) had disease that was unchanged, and 4 (22%) had disease that was improved from their original surgery.

In the ISG at surgery 2, 5 of 16 patients (31%) had no evidence of endometriosis, 4 of 16 (25%) had areas that were suspicious of disease but proved biopsy negative, and 7 of 16 (43%) had biopsy-proven disease. Overall, 14 of 16 patients (88%) had disease that was improved, and 1 (6%) had worse disease than at her rst surgery. The ndings at the time of laparoscopy are summarized in Table 1.

Postoperative Pain and Quality-of-Life Outcomes From Surgery 2

In the DSG at surgery 2, 83% of women reported an improvement in their symptoms after surgery. More than

TABLE 2
TABLE
2

Change in overall level of pain reported after surgery.

DSG

ISG

DSG vs. ISG

Surgery 1 Any improvement in pain, n (%) No change/worse pain, n (%) VAS a change in pain, score (range) Surgery 2 b Any improvement in pain, n (%) No change/worse pain, n (%) VAS change in pain, score (range)

n 19 6 (32) 13 (68) 0 (0100) n 18 15 (83) 3 (17) 82.5 (0100)

n 20 16 (80) 4 (20) 30 (095) n 15 8 (53) 7 (47) 50 (0100)

2 9.3, P .002 Z 2.5, P .012

2 3.88, P .13 Z 1.22, P .26

Note: Patients were asked to report on their pain relief 6 mo after surgery. For surgery 1, this was immediately before surgery 2 and for surgery 2 this was 12 mo from surgery 1. a Visual analogue scale, where 0 no change in pain and 100 complete relief of pain. b Patients were asked to report on the improvement in overall level of pain after surgery 2, not compared with baseline.

Abbott. Excision of endometriosis for pain: RCT. Fertil Steril 2004.

880 Abbott et al.

Excision of endometriosis for pain: RCT

Vol. 82, No. 4, October 2004

TABLE 3
TABLE
3

Comparison of VAS scores over the 12-month study period.

DSG

ISG

DSG vs. ISG

Within-group comparison

 

Mean (SD)

Mean (SD)

Mean difference (95% condence interval)

DSG

ISG

Dysmenorrhea

 

86.5 (16.7)

62.9 (27.4)

78.1 (23.7)

52.1 (25.6)

8.3 ( 5.0, 21.6), P .21 10.8 ( 7.4, 29.1), P .24 1.1 ( 20.8, 18.6), P .91

 

a

Baseline

6

mo

P .0001 a P .002 b P .0001 c

P .01 P .02 P .0001 c

b

12

mo

28.1 (32.8)

29.2 (26.9)

Nonmenstrual pelvic pain Baseline

58.8 (28.1)

62.3 (24.4)

3.4 ( 20.5,13.6), P .68

P .01

a

P .006 a

6

mo

34.9 (31.5)

43.5 (29.3)

8.5 ( 29.5, 12.4), P .41 3.4 ( 11.8, 18.7), P .65

P .03 P .003 c

b

P .02 b P .002

c

12

mo

20.1 (24.2)

16.6 (22.2)

Dyspareunia

Baseline

6

mo

52.7 (36.5)

42.2 (37.1)

65.4 (26.2)

48.6 (31.3)

12.6 ( 33.1, 7.9), P .22 6.4 ( 29.9, 17.2), P .58 6.5 ( 24.7, 11.5), P .47

P .09 P .02 P .001 c

a

b

a

P .002 P .003 P .0001 c

b

12

mo

15.7 (24.9)

22.3 (29.9)

Dyschesia

 

Baseline

50.8 (39.9)

44.6 (30.1)

6.2 ( 16.7, 29.1), P .59 2.5 ( 21.5, 26.6), P .83 3.1 ( 20.6, 14.5), P .72

P .10 P .64 P .02

a

b

c

P .03 P .60 P .01

a

b

c

6

mo

32.7 (38.3)

30.1 (30.5)

12

mo

22.1 (25.2)

25.1 (28.1)

Note: Visual analogue scale scores are reported between 0 100, where 0 no pain and 100 the worst pain imaginable. a Baseline vs. 6 mo. b 6 mo vs. 12 mo. c Baseline, 6-mo, and 12-mo comparison (Friedman test).

Abbott. Excision of endometriosis for pain: RCT. Fertil Steril 2004.

50% of women in the ISG reported an improvement. This was not signicantly different. The results for response, pain, and quality-of-life after surgery 2 are detailed in Tables 25.

Pregnancy

Of the 12 women trying for a pregnancy, 6 (50%) con- ceived during the study period. All pregnancies followed excisional surgery. Of these 6, 5 occurred within 6 months of this surgery. All pregnancies continued to live births.

Co-Interventions

During the 12-month study period, one woman in each group commenced taking the oral contraceptive pill for con- traceptive purposes. No other hormonal treatment was pre- scribed or used by any of the participants. Six women in the DSG and ve women in the ISG reported using analgesics during the study period. These might have been prescribed by their general practitioner or bought without prescription.

One woman in the DSG with stage II disease had excision of endometriosis performed at surgery 2 and 4 months later had a hysterectomy at another hospital for severe dysmen- orrhea. There was no macroscopic or microscopic endome- triosis or other pathology evident at this time.

DISCUSSION

Endometriosis is reported to have signicant impact on quality of life, and its treatment can be either medical or surgical. Surgical excision of endometriosis has been re-

FERTILITY & STERILITY

ported to decrease pain, although unlike medical treat- ments, there has been only one previous placebo-con- trolled study, with limited patient entry, in which this outcome was examined (12).

In this study, we demonstrated a signicant difference in the patientssymptoms in the two groups, on both dichoto- mous and continuous variables over the study period. This is similar to previously reported rates of response in women with minimal and mild endometriosis (12). The initial as- sessment of pain scores in both groups was similar to those reported in other prospective, longitudinal studies (8, 13, 14). When individual pain scores were assessed at 6 months, there was a reduction in all scores by all patients, indepen- dent of the group to which they were randomized.

For women in the ISG, this translated to a signicant reduction for all four pain parameters measured. For women in the DSG, dysmenorrhea and pelvic pain were signicantly reduced, but dyspareunia or dyschezia were not. It has been reported that deep endometriosis might be responsible for dyspareunia and dyschezia and that its excision might relieve these symptoms (15, 16), which might account for the nd- ings in our study.

For women in the DSG, all pain scores decreased after placebo surgery 1. This might indicate that the VAS alone provides limited information on pain and should be used in conjunction with other instruments that measure psychoso- cial inuences on pain perception. It might also be that when

881

TABLE 4
TABLE
4

Quality of life outcomes.

 

DSG

DSG vs. normal

ISG

ISG vs. normal

Comparison

 

Mean

Mean

DSG

(SD)

Mean difference (CI), P

(SD)

Mean difference (CI), P

vs. ISG

DSG

ISG

EQ-5D index summary, mean normal score 0.91 (SD 0.15)

 

Baseline

0.68 (0.28)

0.22 ( 0.35, 0.08), P .003 0.68 (0.28)

0.21 ( 0.35, 0.07), P .004 0.13 ( 0.26, 0.01), P .03

P .88 P .07

P .31 P .18

a

P .21 P .24

a

b

6

mo

0.74 (0.23)

0.16 ( 0.28, 0.04), P .01

0.77 (0.25)

b

12

mo

0.82 (0.351)

0.08 ( 0.20, 0.02), P .10

0.85 (0.731)

0.03 ( 0.08, 0.02), P .34

P .51

P .05

c

P .06

c

EQ-5D VAS summary, mean normal score 85.3 (SD 14.8)

 

Baseline

66.1 (19.5)

19.1 ( 28.5, 9.8), P .0001 77.5 (14.9)

7.8 ( 15.2, 0.4), P .04 4.3 ( 8.8, 0.22), P .54

P .07 P .01

P .95 P .03

a

P .05 P .16

a

b

6

mo

65.9 (21.3)

19.3 ( 32.2, 6.4), P .006

83.6 (10.8)

b

12

mo

82.7 (16.2)

2.6 ( 10.3, 5.4), P .51

88.6 (10.4)

3.3 ( 1.4, 8.1), P .15

P .23

P .06

c

P .05

c

SF-12 physical component score, mean normal score 52.8 (SD 10)

 

Baseline

40.1 (8.1)

12.3 ( 16.5, 8.2), P .0001 43.5 (8.1)

9.3 ( 13.2, 5.4), P .0001 4.6 ( 8.1, 0.9), P .02

P .27 P .36

P .08 P .02

a

P .01 P .31

a

b

6

mo

45.5 (10.0)

7.2 ( 12.4, 1.9), P .01

48.2 (7.6)

b

12

mo

52.4 (4.9)

0.35 ( 2.7, 2.0), P .35

51.2 (6.1)

1.5 ( 4.5, 1.4), P .36

P .60

P .008 c

P .02

c

SF-12 mental component score, mean normal score 51.9 (SD 10)

 
 

43.5 (12.9)

45.3 (11.8)

8.3 ( 14.9, 1.7), P .01

6.5 ( 12.8, 0.18), P .04

42.8 (9.1)

47.6 (9.7)

9.1 ( 13.5, 5.6), P .0001 4.3 ( 8.8, 0.22), P .06 1.2 ( 2.7, 5.1), P .44

P .84 P .55 P .19

P .77 P .04 P .07

a

a

P .03 P .04 P .003 c

b

Baseline

6

mo

b

12

mo

49.5 (9.8)

2.4 ( 7.1, 2.3), P .22

53.1 (8.2)

c

Note: CI 95% condence interval.

a Baseline vs. 6 mo.

b 6 mo. vs. 12 mo.

c Baseline, 6-mo, and 12-mo comparison (Friedman test).

Abbott. Excision of endometriosis for pain: RCT. Fertil Steril 2004.

women are confronted with the uncertainty of whether they received excisional surgery, they are more likely to be con- servative when estimating their pain. This is in keeping with

the placebo model, whereby reduction in symptoms seems to occur in approximately 30% of patients when an intervention has taken place, regardless of its efcacy (17). In this study,

TABLE 5
TABLE
5

Sexual activity questionnaire results.

DSG

ISG

Within-group

Within-group

 

Mean (SD)

Mean (SD)

DSG vs. ISG

comparison, DSG

comparison, ISG

Pleasure

 

10.4 (4.1)

10.2 (5.7)

8.8 (3.7)

10.4 (4.8)

 

a

P .82 P .01 P .008 c

b

 

a

Baseline

6

mo

P

P .16

.98

 

P

P

.23

.20

b

12

mo

14.1 (4.9)

13.3 (5.1)

P

.42

P

.02

c

Habit

Baseline

0.86 (0.74)

0.64 (1.08)

P .51

P .48 P .10 P .01

a

b

c

 

P

.09

a

6

mo

1.07 (0.91)

1.12 (1.02)

P

.93

P

.11

b

12

mo

1.68 (0.87)

1.12 (0.88)

P

.09

P

.02

c

Discomfort

 

Baseline

3.5 (1.6)

3.3 (1.9)

P .81

a

P .79 P .003 P .001

 

P

.19

a

6

mo

3.1 (1.9)

2.4 (1.9)

P

.33

b

P

.52

b

12

mo

1.8 (1.5)

1.8 (1.7)

P .90

c

P .05

c

Note: Pleasure is scored to a maximum of 18; habit to a maximum of 3; and discomfort to a maximum of 6.

a Baseline vs. 6 mo.

b 6 mo. vs. 12 mo.

c Baseline, 6-mo, and 12-mo comparison (Friedman test).

Abbott. Excision of endometriosis for pain: RCT. Fertil Steril 2004.

882 Abbott et al.

Excision of endometriosis for pain: RCT

Vol. 82, No. 4, October 2004

a decrease in the reporting of individual symptoms did not

necessarily correspond with the reporting of improved symp- toms overall.

At the 12-month assessment, women in both groups reported

a signicant reduction in all of the pain parameters from their

6-month assessment, with the exception of dyschezia. This further pain reduction in the ISG might be due to the fact that some women had further excisional surgery. Additionally, pain reduction might continue for up to 12 months for some women. Finally, there might have been a placebo response in patients who did not have excision of disease at surgery 2.

It should be emphasized that the number of women in the study who were available for analysis by intention to treat was just below that of the power calculation (39 instead of 40). In addition, owing to three pregnancies and one patient with- drawal, the number of patients available for some analyses was even lower. As a result, negative results should be treated with caution because a larger sample size might have produced additional statistically signicant differences. It also suggests that the positive ndings from this study are robust.

Both groups had signicantly reduced quality of life compared with a normal sample without endometriosis at baseline, which has been previously reported (8, 18 20). After surgery 1, in the ISG there was a signicant improve- ment in the EQ-5D VAS and both mental and physical components of the SF-12. No such change was found in the DSG after surgery 1.

Although scores in the sexual activity questionnaire after surgery 1 did not change signicantly, individual changes were reported in the ISG. The three women who were apareunic owing to pain were able to return to sexual activity because their pain was reduced. In addition, three women in this group conceived and were pregnant at the time of the 6-month assessment. In the DSG after surgery 1, there was one woman who remained apareunic, and one womans pain during sexual activity increased to the level that she stopped having sexual intercourse.

The quality-of-life analyses show statistically signicant improvements over baseline for both groups in all measures except the mental component of the SF-12 for the DSG. None of these scores at 12 months were signicantly differ- ent when compared with those of a normal sample for the same instrument at 12 months; and for the ISG this was the case for the EQ-5D VAS and the mental component of the SF-12 at 6 months. An improvement in quality of life has been previously reported, but levels have not been reported to return to normal in either the short term (8) or longer term (14). Reasons for this might include a higher motivation among women participating in this placebo-controlled surgical trial, and the Hawthorne effect (21), relating to the attention associ- ated with multiple points of assessment in this study.

The placebo response of 32% in our study is similar to the 22% reported in the expectant group, as reported by Sutton

FERTILITY & STERILITY

et al. (12). This gure includes one woman in the DSG who had surgery for rapidly advancing disease. The response rate to placebo surgery has few comparators, because most con- trolled surgical trials conclude that the surgical intervention is not more efcacious than the placebo surgical treatment (2224). The placebo response in this study is similar to that in other nonsurgical trials (17, 2528).

The nonresponse rate of 20% in women who had sur- gery is markedly less than the 38% reported in Suttons study (12) and might reect the study population. The group that was least likely to respond in the Sutton study were women with stage I disease, whereas women with higher stages of disease had a greater response to laparo- scopic excision. There were very few women with stage I disease in our studythe stage most common in Suttons study. Additionally, because Suttons study did not re- quire histologic conrmation of disease, it is possible that the diagnosis of endometriosis was incorrect, and there- fore a response would not be expected.

A signicant advantage of the crossover design used in this study is that women in the DSG were able to act as their own controls. It is demonstrated that after surgery 1, there was no signicant perception of improvement overall, although there were decreases reported for two of the four pain scores in this group. After surgery 2, however, there was a symptomatic improvement for most women in this group. These results mirror the ndings from the ISG after surgery 1 and are further evidence of the efcacy of surgery for pain. Women reported a signicant improvement in overall symptoms for each pain score and in quality-of-life outcomes after surgery 2.

There were minimal co-interventions, with no woman re- ceiving specic hormonal treatment for endometriosis during the follow-up period. Two women received the combined oral contraceptive pill, and 28% reported taking oral analgesics in the follow-up period. It is unlikely that these interventions would have made a signicant impact on the results, given that similar numbers of patients in both groups took them during the study. The only surgical co-intervention was the woman who had a hysterectomy and was an inappropriate responder to surgery.

This study provides evidence that surgery is successful in treating the symptoms of pain and improving quality of life for women with endometriosis. This study also provides evidence that pelvic pain is not always due to endometriosis when it is present. Possible patient proling in the preoper- ative setting might assist in selecting patients likely to re- spond to surgery.

Acknowledgments: The authors thank the nursing and medical staff, who made signicant contributions to this study through their diligence, attention

883

to detail, and belief in its importance. They also thank the patients who participated in this study and accepted the importance of scientically valid research to illuminate womens health.

References

1. Burns W, Schenken R. Pathophysiology of endometriosis-associated infertility. Clin Obstet Gynecol 1999;42:586 610.

2. Reiter R. A prole of women with chronic pelvic pain. Clin Obstet Gynecol 1990;33:130 6.

3. Lee N, Dicker R, Rubin G, Ovy H. Conrmation of the pre-operative diagnoses for hysterectomy. Am J Obstet Gynecol 1984;150:1619.

4. Gambone J, Reiter R, Lench J, Moore J. Impact on quality assurance process on the frequency and conrmation rate of hysterectomy. Am J Obstet Gynecol 1990;163:54550.

5. Winkel C. Modeling of medical and surgical treatment cost of chronic pelvic pain: new paradigms for making clinical decisions. Am J Manag Care 1999;5 Suppl:S276 90.

6. Boling R, Abbasi R, Ackerman G, Schipul A, Chaney S. Disability from endometriosis in the United States Army. J Reprod Med 1988;33:49 52.

7. Jacobson T, Barlow D, Garry R, Koninckx P. Laparoscopic surgery for pelvic pain associated with endometriosis. Cochrane Database Syst Rev

2001;(4):CD001300.

8. Garry R, Clayton R, Hawe J. The effect of endometriosis and its radical laparoscopic excision on quality of life indicators. Br J Obstet Gynaecol 2000;107:44 54.

9. Brooks R, EuroQOL Group. EuroQOL: the current state of play. Health Policy 1996;37:5372.

10. Ware J, Kosinski M, Keller S. A 12-item short-form health survey. Construction of scales and preliminary tests of reliability and validity. Med Care 1995;34:220 33.

11. Thirlaway K, Falloweld L, Cuzick J. The sexual activity question- naire: a measure of womens sexual functioning. Qual Life Res 1996;

5:8190.

12. Sutton C, Ewen S, Whitelaw N, Haines P. Prospective, randomized, double-blind trial of laser laparoscopy in the treatment of pelvic pain associated with minimal, mild and moderate endometriosis. Fertil Steril 1994;62:696 700.

13. Redwine D, Wright J. Laparoscopic treatment of complete obliteration of the cul-de-sac associated with endometriosis: long-term follow-up of en bloc resection. Fertil Steril 2001;76:358 65.

884 Abbott et al.

Excision of endometriosis for pain: RCT

14. Abbott J, Hawe J, Clayton R, Garry R. The effects and effectiveness of laparoscopic excision of endometriosis: a prospective study with 2-5 years follow-up. Hum Reprod 2003;18:19227.

15. Koninckx P, Martin D. Treatment of deeply inltrating endometriosis. Curr Opin Obstet Gynecol 1994;6:23141.

16. Koninckx P, Martin D. Surgical management of deeply inltrating endometriosis. In: Shaw R, ed. Endometriosiscurrent understanding and management. Oxford: Blackwell Scientic, 1995:264 81.

17. Heeg M, Deutsch K, Deutsch E. The placebo effect. Eur J Nucl Med 1997;24:143340.

18. Miller J. Quantication 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 2000;182:14838.

19. Zhao S, Kellerman L, Francisco C, Wong J. Impact of nafarelin and leuprolide for endometriosis on quality of life and subjective clinical measures. J Reprod Med 1999;44:1000 6.

20. Bergqvist A, Theorell T. Changes in quality of life after hormonal treatment of endometriosis. Acta Obstet Gynecol Scand 2001;80:628

37.

21. de Craen A, Kaptchuk T, Tijssen J, Kleijnen J. Placebos and placebo effect in medicine: a historical overview. J R Soc Med 1999;92:5115.

22. Cobb L, Thomas G, Dillard D, Merendino K, Bruce R. An evaluation of internal mammary artery ligation by a double-blind technic. N Engl J Med 1959;260:11158.

23. Freeman TB, Vawter DE, Leaverton PE, Godbold JH, Hauser RA, Goetz CG, et al. Use of placebo surgery in controlled trials of a cellular-based therapy for Parkinsons disease. N Engl J Med 1999; 341:988 92.

24. Moseley J, OMalley K, Petersen N, Menke TJ, Brody BA, Kuykendall DH, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N Engl J Med 2002;347:818.

25. Brody H. The placebo response. J Fam Pract 2000;49:649 54.

26. Brown W. Harnessing the placebo effect. Hosp Pract (Off Ed) 1998;

33:10716.

27. Hrobjartsson A, Gotzsche P. Is the placebo powerless? An analysis of clinical trials comparing placebo with no treatment. N Engl J Med 2001;344:1594 602.

28. Margo C. The placebo effect. Surv Ophthalmol 1999;44:3144.

Vol. 82, No. 4, October 2004