Post-operative GnRH analogue treatment after conservative surgery for symptomatic endometriosis stage III–IV: a randomized controlled trial

M. Busacca,1, E. Somigliana, S. Bianchi, S. De Marinis, C. Calia, M. Candiani and M. Vignali

II Department of Obstetrics and Gynecology, University of Milano, Milan, Italy1To whom correspondence should be addressed at: II Department of Obstetrics and Gynecology, Clinica `L. Mangiagalli', Via Commenda 12, 20122, Milano, Italy.


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: In order to decrease endometriosis recurrence after surgical therapy, it has been proposed to use a post-surgical oestrogen-lowering medical treatment. Results from previous trials on this topic are contradictory. METHODS: A total of 89 women were randomized, by computer-generated list, after laparoscopic conservative surgery for symptomatic endometriosis stage III–IV to receive monthly i.m. injections of gonadotrophin-releasing hormone (GnRH) analogue, leuprolide acetate depot (3.75 mg) for 3 months (n = 44) or to an expectant management (n = 45). All patients were followed up every 6 months for evaluation of pain symptoms, fertility and objective disease recurrence. RESULTS: During the follow-up, which ranged from 6–36 months, five (33%) of the 15 women who wanted children and who were allocated the GnRH analogue and six (40%) of the 15 given no treatment became pregnant (not significant). Moderate/severe pelvic pain recurred during the follow-up in 10 (23%) of the women allocated the GnRH analogue and 11 (24%) of those allocated no treatment; the cumulative pain recurrence rates at 18 months were 23 and 29% respectively (not significant). Four women (9%) treated with GnRH analogue and four women (9%) who received no treatment had objective disease recurrence as demonstrated by gynaecological examination and/or pelvic ultrasonography. CONCLUSIONS: This study does not support the routine post-operative use of a 3 month course of GnRH analogue in women with symptomatic endometriosis stage III–IV.

Key words: endometriosis/gonadotrophin-releasing hormone analogue/laparoscopy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
An unsolved important priority for those caring for women with endometriosis is disease recurrence after surgical therapy. Although conservative surgery at laparoscopy is nowadays considered the treatment of choice for stage III–IV symptomatic endometriosis (Cook and Rock, 1991Go; Bateman et al., 1994Go; Adamson and Nelson, 1997Go), this approach does not seem to be highly effective in preventing recurrence of the disease (Adamson and Nelson, 1997Go). Specifically, ~20% of patients have recurrences over 5 years following complete extirpation of endometriotic lesions (Wheeler and Malinak, 1983Go; Redwine, 1991Go). In this context, the use of a short course of post-surgical oestrogen-lowering medical therapy to increase the efficacy of treatment (Telimaa et al., 1987Go; Hornstein et al., 1997Go; Heinrichs and Henzl, 1998Go) has been suggested. The rationale of this approach is to accomplish complete resection of lesions that could not be surgically removed, to treat microscopic foci and to prevent iatrogenic dissemination of endometriotic cells. However, considering costs and harmful side-effects related to these medical therapies, definitive evidence supporting a routine clinical use of this surgical/medical combined regimen should be mandatory.

At present, little is known about the post-operative effects of these medical therapies and results from the few randomized controlled trials are often contradictory (Telimaa et al., 1987Go; Parazzini et al., 1994Go; Hornstein et al., 1997Go; Bianchi et al., 1999Go; Vercellini et al., 1999Go). Specifically, two studies have emphasized a potential role of a short-term therapy with gonadotrophin-releasing hormone (GnRH) analogue after surgery in improving pain relief (Hornstein et al., 1997Go; Vercellini et al., 1999Go) while others failed to observe such an improvement (Telimaa et al., 1987Go; Parazzini et al., 1994Go; Bianchi et al., 1999Go). In the present study, we have evaluated the effects of a post-operative regimen of GnRH analogue in women with pelvic pain symptoms who underwent laparoscopic conservative surgery for endometriosis stage III–IV. This was studied in a randomized trial in which no post-surgical therapy was compared with a protocol consisting of a post-operative course of 3 months of GnRH analogue. The primary objective was to assess whether this surgical/medical combined approach could produce a significantly longer relief of pain symptoms. The between-group differences in both objective disease recurrence and pregnancy rates among women wanting children were also evaluated.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The study population was selected from women with endometriosis who attended the Endoscopic Surgical Service of the II Department of Obstetrics and Gynecology to undergo gynaecological laparoscopy between July 1997 and December 1999. The criteria for inclusion were: (i) that the patients were of reproductive age and not >40 years old; (ii) that these women had a laparoscopic diagnosis of endometriosis stage III–IV according to the revised American Society for Reproductive Medicine (ASRM) classification (ASRM, 1997). Exclusion criteria were previous medical or surgical therapy for endometriosis, the presence of other diseases that might affect fertility or cause pelvic pain and diagnosis of liver, endocrine or neoplastic diseases. A total of 89 women was selected for this study.

The laparoscopic procedures were performed according to the technique described by Cook and Rock (Cook and Rock, 1991Go). All patients were under the care of three of the authors (M.B., S.B. and M.C.). Approval for this study was granted by the local Human Institutional Investigation Committee. Written consent was obtained from each woman before study entry.

Randomization was achieved at the time of post-operative control (7 days after surgery) so that a definitive histological diagnosis of endometriosis was available. This randomization was performed according to a computer-generated list unknown to the physicians. Eight out of 97 women who were eligible for randomization refused to enter the study. Forty-four patients were randomized to receive an i.m. injection of GnRH analogue leuprolide acetate depot (3.75 mg) every 4 weeks for 8 weeks (three injections). Treatment began on day 1 or 2 of the first menstrual flow after surgery. Forty-five women were allocated to receive no post-operative treatment and were used as controls (Figure 1Go).



View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Protocol of the study. Women of reproductive age with a laparoscopic diagnosis and conservative surgical treatment of endometriosis stage III–IV were eligible for randomization. Eight out of 97 women did not accept to enter the study design and were excluded leaving 89 women to be randomized into two groups. Forty-four patients received an i.m. injection of GnRH analogue leuprolide acetate depot (3.75 mg) every 4 weeks for 8 weeks (three injections). Forty-five women were allocated to receive no post-operative treatment and were used as controls.

 
Before operation, all women completed a questionnaire to investigate the presence and severity of dysmenorrhoea, pelvic pain and deep dyspareunia. All these symptoms were graded according to two scales. On one linear scale, pain severity was scored from 0–10, with 0 indicating the absence of pain, and scores 1–4, 5–7 and 8–10 indicating mild, moderate and severe pain respectively. The other scale, used to evaluate the severity of dysmenorrhoea and pelvic pain, was multidimensional and considered any limitation of daily activities, the coexistence of systemic symptoms and the need for analgesics. Conversely, the Biberoglu and Behrman grading scale was used to assess the severity of deep dyspareunia (Biberoglu and Behrman, 1981Go). Finally, all infertile women underwent a standard diagnostic work-up before operation, including hysterosalpingography, hormone profile [FSH (twice), LH and oestradiol assays in the follicular phase, and progesterone (three times) and prolactin assays in the luteal phase] and postcoital test. Moreover, all partners had two semen analyses.

After surgery, patients underwent a standard gynaecological and pelvic ultrasonography every 6 months. On each occasion, the occurrence of any pregnancy was recorded and any pain symptoms were evaluated using the two scales.

Statistical analysis was performed using Student's t-test for independent samples and {chi}2 analysis to compare basal characteristics of the two groups. Recurrence of pain symptoms and recovery of fertility were analysed by the product-limit method and the curves of the two groups were compared with the log-rank test. The event dates considered were the date of operation and the date of first menstruation associated with moderate or severe pain or the last menstruation before a positive pregnancy test, since the operation. P < 0.05 was considered significant in all comparisons. The sample size used in our study allowed us to exclude, at the usual level of study power (80% chance at a 5% level of significance), a two-fold lower pain recurrence rate in the treated group.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Table IGo shows patient's age, parity, stage of endometriosis and desire for pregnancy as divided into groups according to treatment. The two groups were similar with respect to these baseline characteristics.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of patients
 
No patients were lost to follow-up, which ranged from 6–36 months. The mean follow-up interval in women who received GnRH analogue and in controls did not differ (19.6 ± 7.7 and 18.8 ± 6.3 months respectively). During the course of GnRH analogue treatment all patients became amenorrhoeic and most of them experienced menopausal symptoms. These side-effects were generally well tolerated; only one woman was withdrawn after 1 month of treatment due to reported unacceptable side-effects. This case was not excluded from the treated group in order to perform an intention-to-treat analysis.

Frequencies of pregnancies, pain and clinical recurrence and the need for second surgery are shown in Table IIGo. During the follow-up, five (33%) of the 15 women who wanted children and who were allocated the GnRH analogue and six (40%) of the 15 given no treatment became pregnant (not significant): the respective cumulative pregnancy rates at 18 months were 38 and 40% (log rank test, not significant). Moderate/severe pelvic pain recurred during the follow-up in 10 (23%) of the 44 women with pelvic pain allocated the GnRH analogue and 11 (24%) of the 45 allocated no treatment; the cumulative pain recurrence rates at 18 months were 23 and 29% respectively (log rank test, not significant) (Figure 2Go). No statistically significant differences in time to recurrence of pain symptoms between the two groups were observed. Four women (9%) treated with GnRH analogue and four (9%) who received no treatment had objective disease recurrence as demonstrated by gynaecological examination and/or pelvic ultrasonography (not significant). Finally, two patients allocated to the GnRH analogue group, underwent repeat operations (not significant). In these cases, surgery was decided on due to both the ultrasonographic evidence of an endometriotic cyst and the recurrence of pain symptoms. The diagnosis of endometriosis was confirmed at histological examination.


View this table:
[in this window]
[in a new window]
 
Table II. Clinical outcome of patients
 


View larger version (10K):
[in this window]
[in a new window]
 
Figure 2. Cumulative 30 month pain recurrence rate in 89 women undergoing laparoscopic conservative surgery for symptomatic endometriosis stage III–IV according to post-operative treatment (log rank test, not significant). Women treated with GnRH analogue (n = 44) and controls (n = 45) are depicted with solid and dotted lines respectively. Vertical tick marks represent censored observations due to pregnancies occurring or follow-up availability.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The results of our trial do not support the routine post-operative use of a 3 month course of GnRH analogue in women with symptomatic endometriosis stage III–IV. To date, five randomized controlled trials have been published regarding the effects of oestrogen-lowering medical therapies after surgery (Telimaa et al., 1987Go; Parazzini et al., 1994Go; Hornstein et al., 1997Go; Bianchi et al., 1999Go; Vercellini et al., 1999Go). All these previous studies agreed that post-operative medical treatment does not confer significant additional benefit in improving pregnancy rates. Data on recovery of fertility in our study, although based on a relatively small sample size, are in line with these previous observations. In this context it should be noted that the immediate post-operative period is thought to be particularly favourable for conception; therefore, suppressing ovulation for some months after surgery has been claimed to be detrimental in infertile women (Buttram and Reiter, 1985Go). Based on these observations, we believe that a post-surgical medical regimen should be avoided whenever possible in women wanting children; indeed, additional benefit has never been demonstrated and these therapies lead to a useless delay because of the patient's inability to conceive while taking the medication.

In regard to pain symptoms and disease recurrences, data from the randomized trials on this topic are more controversial. Indeed, in a previous study, we failed to demonstrate additional benefit of a short course of post-operative danazol treatment in preventing pain recurrence (Bianchi et al., 1999Go); similarly, Parazzini et al. reported that medical treatment with 400 µg/day nasal naferelin for 3 months after surgery did not markedly improve short-term pelvic pain prognosis (Parazzini et al., 1994Go). On the other hand, results from three other groups documented some kind of benefit from using an oestrogen-lowering medical regimen after surgery (Telimaa et al., 1987Go; Hornstein et al., 1997Go; Vercellini et al., 1999Go). Specifically, these therapies have been claimed to significantly delay the return of endometriosis symptoms. Hornstein et al. found that 15 out of 49 women treated with GnRH analogue for 6 months required an alternative therapy for recurrent pain compared with 25 out of 44 women allocated to post-operative placebo (Hornstein et al., 1997Go). Vercellini et al. using survival analysis, reported that time to symptom recurrence was significantly longer in the GnRH analogue group (Vercellini et al., 1999Go). To the contrary, in our study, we failed to document a significant longer relief of pain symptoms in women with symptomatic endometriosis stage III–IV treated with a 3 month course of post-operative GnRH analogue. The relatively small sample size of our study did not allow the identification of less important effects of treatment. However, advantages associated with this therapy have to be weighed with costs and harmful side-effects; in this context, the use of GnRH analogue should be considered only if it provides a marked improvement in short-term pelvic pain prognosis. The sample size used in our study allows the exclusion, at the usual level of study power (80% chance at a 5% level of significance), a two-fold lower pain recurrence in the GnRH analogue-treated versus the untreated women. Interestingly, since the day of surgery was chosen as a starting point in both groups of women, patients who were allocated to the GnRH analogue group have symptom relief for at least the first 3 months compared with women given no treatment. However, this bias would most likely further increase rather than decrease the benefits of treatment.

Discrepancies among studies on this topic are difficult to explain; three reasons can be postulated. First, therapeutic protocol used in these studies varies and this may account at least for some differences. In our study, an i.m. injection of GnRH analogue depot every 4 weeks was chosen since this protocol appeared to be very simple, and adherence with this method of administration seems high; moreover, a 3 month therapy instead of a 6 month was decided on since it has been reported that a shorter protocol of this drug could be as effective as a longer one, whereas costs and side-effects could be reduced (Hornstein et al., 1995Go). In this context, it has to be noted that this study is the first report on this topic using this protocol. Second, another controversy may be related to the study design since our trial was not blinded and we did not use placebo. However, it should be noted that none of the available studies have conducted a true double-blinded, placebo-controlled trial; indeed, although such a study design would have been preferable, the frequent side effects observed with oestrogen-lowering medical therapies do not allow its application. At present, we are unable to assess the importance of this aspect in determining pain recurrence and in explaining differences between studies. Nevertheless, considering that we failed to identify a beneficial effect of GnRH analogue in lowering pain recurrence, it is unlikely that this bias may play an important role in our study. Third, the comparison with the other studies may be difficult because of differences in patient's characteristics, disease extension (only cases with advanced stages of the disease were considered in this study), surgical procedures and surgeon's skill. In our study, compliance with the study protocol was complete, surgery was always performed by only three experienced surgeons and the two study groups were comparable in terms of age, reproductive history, pain symptoms and disease stage. Interestingly, pregnancy rates as well as pain recurrence rates are comparable with those reported in the largest review of the literature (Hughes et al., 1993Go; Adamson and Pasta, 1994Go; Candiani et al., 1995Go). Therefore, we estimate that important sources of bias in our trial can be excluded.

In conclusion, the present study does not support the routine post-operative use of a 3 month course of GnRH analogue in women with symptomatic endometriosis stage III–IV. However, larger series and longer follow-ups are required to identify less important effects of treatment, in particular on the objective disease recurrence rate. Moreover, these data could not rule out that post-surgical GnRH analogue or other oestrogen-lowering medical therapies may be of value in selected patients, particularly those in whom disease has not been completely extirpated. Further trials to identify women who may benefit from these therapies are hence also required.


    Notes
 
1 E-mail: auro.busacca{at}unimi.it Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Adamson, G.D. and Nelson, H.P. (1997) Surgical treatment of endometriosis. Obstet. Gynecol. Clin. North. Am., 24, 375–409.[ISI][Medline]

Adamson, G.D. and Pasta, D.J. (1994) Surgical treatment of endometriosis-associated infertility: meta-analysis compared with survival analysis. Am. J. Obstet. Gynecol., 171, 1488–1505.[ISI][Medline]

American Society for Reproductive Medicine (1997) Revised American Society for Reproductive Medicine classification of endometriosis: 1996. Fertil. Steril., 67, 817–821.[ISI][Medline]

Bateman, B.G., Kolp, L.A. and Mills, S. (1994) Endoscopic versus laparotomy management of endometriosis. Fertil. Steril., 62, 690–695.[ISI][Medline]

Bianchi, S., Busacca, M., Agnoli, B. et al. (1999) Effects of 3 month therapy with danazol after laparoscopic surgery for stage III/IV endometriosis: a randomized study. Hum. Reprod., 14, 1335–1337.[Abstract/Free Full Text]

Biberoglu, K.O. and Behrman, S.J. (1981) Dosage aspects of danazol therapy in endometriosis: short-term and long-term effectiveness. Am. J. Obstet. Gynecol., 139, 645–654.[ISI][Medline]

Buttram, V.C. and Reiter, R.C. (1985) Treatment of endometriosis with danazol: report of a 6-years prospective study. Fertil. Steril., 43, 353–360.[ISI][Medline]

Candiani, G.B., Fedele, L. and Bianchi, S. (1995) Recurrent endometriosis. In Nezhat, C.R., Berger, G.S., Nezhat, F.R. and Buttram, V.C. Jr (eds) Endometriosis: Advanced Management and Surgical Techniques. Springer-Verlag, New York, pp. 159–171.

Cook, A.S. and Rock, J.A. (1991) The role of laparoscopy in the treatment of endometriosis. Fertil. Steril., 55, 663–680.[ISI][Medline]

Heinrichs, W.L. and Henzl, M.R. (1998) Human issues and medical economics of endometriosis. J. Reprod. Med., 43, 299–308.[ISI][Medline]

Hornstein, M.D., Yuzpe, A.A., Burry, K.A. et al. (1995) Prospective randomized double-blind trial of 3 versus 6 months of naferelin therapy for endometriosis associated pelvic pain. Fertil. Steril., 63, 959–962.

Hornstein, M.D., Hemmings, R., Yuzpe, A.A. and Heinrichs, W.L. (1997) Use of nafarelin versus placebo after reductive laparoscopic surgery for endometriosis. Fertil. Steril., 68, 860–864.[ISI][Medline]

Hughes, E.G., Fedorkow, D.M. and Collins, J.A. (1993) A quantitative overview of controlled trials in endometriosis-associated infertility. Fertil. Steril., 59, 963–970.[ISI][Medline]

Parazzini, F., Fedele, L., Busacca, M. et al. (1994) Postsurgical medical treatment of advanced endometriosis: results of a randomized clinical trial. Am. J. Obstet. Gynecol., 171, 1205–1207.[ISI][Medline]

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, 628–634.[ISI][Medline]

Telimaa, S., Ronnberg, L. and Kaupilla, A. (1987) Placebo-controlled comparison of danazol and high-dose medroxyprogesterone acetate in the treatment of endometriosis after conservative surgery. Gynecol. Endocrinol., 1, 363–371.[Medline]

Vercellini, P., Crosignani, P.G., Fadini, R. et al. (1999) A gonadotrophin-releasing hormone agonist compared with expectant management after conservative surgery for symptomatic endometriosis. Br. J. Obstet. Gynecol., 106, 672–677.[ISI][Medline]

Wheeler, J.M. and Malinak, L.R. (1983) Recurrent endometriosis: incidence, management and prognosis. Am. J. Obstet. Gynecol., 146, 247–253.[ISI][Medline]

Submitted on February 2, 2001; accepted on July 18, 2001.