The efficacy, side-effects and continuation rates in women with symptomatic endometriosis undergoing treatment with an intra-uterine administered progestogen (levonorgestrel): a 3 year follow-up

Farhana B. Lockhat, Joseph O. Emembolu and Justin C. Konje1

Reproductive Sciences Section, Department of Cancer Studies and Molecular Medicine, Robert Kilpatrick Clinical Sciences Building, Leicester Royal Infirmary, Leicester LE2 7LX, UK

To whom correspondence should be addressed.Email: jck4{at}le.ac.uk


    Abstract
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Side-effects and choice of drugs influence compliance during treatment for endometriosis. Progestogen administered by a device with a 5-year lifespan, has been shown to be an effective medical alternative with several advantages. The aims of this study were to investigate its efficacy, continuation rates and side-effects in women with endometriosis over a 3-year period. METHODS: Thirty-four women with laparoscopically confirmed minimal to moderate symptomatic endometriosis offered insertion of an intrauterine device at diagnostic laparoscopy were followed up at 1, 3 and 6 months, and then every 6 months for 3 years. A symptom diary for side-effects, documentation of symptoms on a visual analogue scale (VAS), a verbal rating scale (VRS) and quantified menstrual loss using the pictorial blood loss chart was used to assess response to treatment. RESULTS: The continuation rates were respectively 85%, 68%, 62% and 56% at, 6, 12, 24 and 36 months. Discontinuation rates were highest at <12 months, and most of these were for irregular and intolerable bleeding and persistent pain. An improvement in symptoms was observed throughout the 36 months. The greatest changes in pain assessed by either the VAS or VRS were between the pretreatment scores and those after 12 months (7.7±1.3 versus 3.5±1.8 for VAS, P<0.001; and 25±13.8 versus 14±9.4 for VRS, P<0.002). The monthly quantified blood loss fell from 204 (196) pretreatment to 60 (50) at 12 months (P<0.001) and then to 70 (30) after 36 months. The most common side-effects were bleeding irregularities (14.7%), one-sided abdominal pain (11.8%) and weight gain (8.8%). CONCLUSIONS: Intrauterine progestogen is effective in symptom control throughout the 3 years on the device, and discontinuation is greatest between 3 and 6 months. For those patients with improvement in symptoms, it is an acceptable long-term alternative.

Key words: endometriosis/levonorgestrel/3 year follow-up


    Introduction
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Endometriosis, an estrogen-dependent condition, affects ~6–20% of women of reproductive age (Winkel, 2003Go). The most common symptoms are pelvic pain, and menstrual and sexual dysfunction. Therapeutic options include antiestrogens (e.g. progestogens, androgens such as danazol) and drugs that induce either a pseudomenopause (e.g. GnRH agonists) or a pseudo-pregnancy state (e.g. combined oral contraceptive pills) (Winkel and Scialli, 2001Go). More recently, the aromatase inhibitor anastrazole has been shown to be effective in women with refractory endometriosis (Takayama et al., 1998Go; Ailawadi et al., 2004Go). Though effective, most of these options are associated with systemic side-effects, which may affect compliance and preclude long-term use. Furthermore, the need for repeated or regular administration compromises compliance and therefore efficacy. This is more so with oral and depot progestogens, which although cheap and efficacious are associated with poorly tolerated side-effects of irregular vaginal bleeding, weight gain, fluid retention, seborrhoea and breast tenderness.

The advent of the levonorgestrel intrauterine system Mirena® provides an alternative route of delivering the 19-C progestogen (levonorgestrel) directly into the uterine cavity at a steady rate of 20 mg/day over a 5-year period (Bilian et al., 1990Go; Andersson et al., 1994Go). Since systemic levels following this route of administration are less than those achieved with oral (Nilsson et al., 1980Go; Luciano et al., 1988Go; Monghissi, 1999Go) or depot (Du et al., 1999Go) progestogens, side-effects should theoretically be less severe (Van de Hurk and O'Brien, 1999Go). This has indeed been demonstrated to be the case when the device is used to treat women with menorrhagia, where it has been shown to be highly effective (Lahteenmaki et al., 1998Go; Stewart et al., 2001Go). Levonorgestrel delivered this way, however, has only recently been demonstrated to alleviate symptoms of pelvic endometriosis during a period of either 6 months (Lockhat et al., 2004Go) or 12 months (Vercelleni et al., 1999Go) and also in women with adenomyosis and rectovaginal endometriosis (Fedele et al., 2001Go).

Mirena confers several advantages over the oral or depot routes of administration. Apart from the fewer side-effects, there is no need for repeated administration, thus overcoming the problem of remembering to take the tablets regularly or the depot injection every 3 months, and there is no need for contraception. Although the device, which has a life span of 5 years, has been shown to control menorrhagia for this duration, such information is unavailable for its use in endometriosis.

We recently published our pilot data from an observational study of the effectiveness of this device on symptom control over the first 6 months (Lockhat et al., 2004Go). We present here follow-up data on the evaluation of the efficacy of Mirena as a long-term (3 years) option for endometriosis concentrating on symptom control, side-effects and continuation rates.


    Subjects and methods
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
The subjects were mainly women aged between 18 and 42 years who underwent a diagnostic laparoscopy for either known endometriosis, who had been previously treated but were representing with recurring symptoms, or suspected endometriosis based on a history of cyclical lower abdominal pain associated with deep dyspareunia and/or menorrhagia. The inclusion criteria included no hormonal therapy in the preceding 3 months, no immediate desire to conceive, no clinical history of pelvic inflammatory disease, no contra-indications to the use of an intrauterine contraceptive device and laparoscopically diagnosed minimal to moderate endometriosis (American Fertility Society, 1985Go). Those with severe and extensive disease, including those with ovarian endometriotic cyst, dense adhesions and deep rectovaginal disease, were excluded as these are commonly treated by surgery in our unit. No additional surgical procedure was undertaken at the laparoscopy.

For the month preceding the Mirena insertion, each patient completed a diary for the generation of baseline variables, which were used for the assessment of response to treatment. Response to treatment was assessed by changes in the variables, which included the patient's perception of pelvic pain severity using a visual analogue scale (VAS), her rating of both types of pelvic pain (dysmenorrhoea and/or non-cyclical pelvic pain) on a verbal rating scale (VRS), a monthly pelvic pain and bleeding score.

The VAS was a subjective assessment of the pain on a scale of 0 (no pain) to 10 (most severe pain). It was recorded on a 10-cm ruler in the diary at each follow-up visit and reflected the severity of this symptom as perceived by the patient in the preceding month. A 4-point scale (0–3; where 0=no pain, 1=mild pain, 2=moderate pain, 3=severe pain) was used to rate dysmenorrhoea and/or non-cyclical pain on a daily basis. A monthly score was then generated from the summation of the daily VRS over a 28-day period (0=no pain, 96=maximum pain). Once again, this VRSmonthly was only determined for the 28 days prior to the follow-up visit. Menstrual loss was quantified using the pictorial blood loss assessment chart (PBAC) described by Higham et al. (1990)Go. A score of >100 was used to define menorrhagia. Only the loss in the month prior to a follow-up visit was quantified.

Follow-up visits after the Mirena insertion were initially after 1, 3 and 6 months, and thereafter every 6 months. A month before each visit, the patients completed a diary of their pain scores and menstrual loss. These were collected at the follow-up visit and new diaries given for the next visits. Each subject signed an informed consent to partake in the study, which was also approved by the local ethics committee.

Data analysis
SPS version 11.0 was used to record and statistically analyse the data. Values at the time of the insertion of the intrauterine system (i.e. time zero) were compared with those at different time points after insertion using the paired t-test or Wilcoxon rank analysis as appropriate. Additionally, perception by the patient of the efficacy of the device in pain control was evaluated at 1, 3, 6, 12, 18, 24, 30 and 36 months using a VAS, as well as overall satisfaction with the treatment (taking into account the undesirable side-effects) as indicated on a 4-point VRS.


    Results
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
A total of 45 patients underwent the initial laparoscopy to diagnose and stage the endometriosis. Fifteen of these were known cases of endometriosis. Out of the remaining 30 undergoing the procedure on the basis of symptoms, 22 (73.3%) were newly diagnosed cases. Of the 37 women with endometriosis, three were excluded at the initial laparoscopy for the following reasons: Fitz–Hughes–Curtis syndrome (which was taken to indicate previous Chlamydia trachomatis infection) (one patient) and severe (American Fertility Society, 1985Go) disease requiring surgical intervention (two patients). The mean age of the remaining 34 women forming the subjects of the study was 31±7.2 years (range 18–42). There were 13 patients with minimal disease, 15 with mild and six with moderate disease. The 6-month data on the effect of Mirena on endometriosis have already been published (Lockhat et al., 2004Go).

The continuation rates and reasons for requesting for removal over the 3-year period are shown in Table I. The device was retained by 29 (85%) women at 6 months, 23 (68%) at 12 months, 21 (62%) at 24 months and 19 (56%) at 36 months. There were five discontinuations before 6 months for personal reasons (one), pelvic pain (three) and acne (one). After 6 months, six requested removal because of weight gain and acne (one), migraine headaches, weight gain and irregular periods (one), and no improvement in symptoms and irregular bleeding (four). There were only four patients who requested the device to be removed after 12 months, the reasons being bleeding problems (two), weight gain and persistent abdominal pain that was thought to be related to a functional ovarian cyst (one), and planning to start a family (one). Out of a total of 15 discontinuations, five (33%) were for unacceptable irregular bleeding (unscheduled bleeding), most of which were within the first 6 months. Pelvic pain (20.6%) and weight gain (8.8%) were the second and third most common reason for requesting for removal, respectively. There were no expulsions over the 3-year period.


View this table:
[in this window]
[in a new window]
 
Table I. Indications for and timing of removal of Mirena over a period of 3 years

 
The changes in the VAS over the 3 years are shown in Table II. This fell continuously from a pre-insertion score of 7.7 (1.3) to a nadir of 2.4 (1.3) at 30 months and then rose slightly to 2.7 (1.5) at 36 months. The differences between VAS pre-insertion and subsequent follow-up scores after 6 months were all statistically significant (P<0.05). The VRS for dysmenorrhoea fell throughout the 3 years on Mirena. The greatest change in VRS was achieved during the first 18 months on Mirena. Of the 27 patients experiencing moderate to severe dysmenorrhoea pre-insertion, 19 (70%) continued to have similar pains at 3 months (P=0.001), 15 (51%) at 6 months (P<0.001), seven (26%) at 12 months (P<0.001) and only three (11.1%) at 18 months (P<0.0001). At 24 months, only one patient continued to experience moderate, rather than severe, dysmenorrhea, which she found acceptable and hence retained the device. There were five (18.5%) discontinuations in these 27 cases, one each at 1, 3, 4, 6 and 18 months after insertion. The main reason for discontinuation was persistent pain. Overall, there was no consistent response pattern based on the stage or score of the disease; in the group of seven patients who experienced no effect of the treatment on pain, three had minimal, three mild and one minimal disease.


View this table:
[in this window]
[in a new window]
 
Table II. Changes in the VAS, VRSmonthly pain and menstrual scores during 3 years on Mirena

 
The total number of days of pain experienced during a 28-day period were determined from the diaries returned at the follow-up visits. This fell from a mean of 15.0 days (6.9) to 10.7 days (8.7) after 6 months therapy (P<0.05) and to 6 days (3.4) after 12 months (P<0.001).The mean score for pain per month (out of a total of 84) dropped from 25 (13.8) pre-insertion to 19 (18.9) 6 months later; this change was not statistically significant (P=0.076). However, after 12 months, the mean score per month had dropped to 14 (9.4) and this change was statistically significant compared with pretreatment values (P<0.05). There was a further statistically significant (P<0.001) fall to 11 (5.2) after 18 months, but thereafter the fall was small and not statistically significant. The changes in the quantified menstrual loss from the PBAC chart are shown in Table II. This fell from 204 (196) pre-insertion to 90 (157) after 6 months of treatment (P<0.001). There was further statistically significant (P<0.05) reduction in the total blood loss after 12 months. Although this number did not change significantly over the remaining 2 years, most of them were reported as spotting rather than heavy bleeding.

Table III demonstrates the side-effects reported by the patients during the study period. The most common was irregular vaginal bleeding. There was a single case of severe depression, which responded well to a 3-month course of antidepressants. She was not willing to discontinue the device as it had significantly improved the quality of her life. The partners of two patients complained of discomfort during sexual intercourse caused by the strings of the device. Although these were shortened, the discomfort persisted but in a mild form, hence the patients retained the device. There were three cases of simple functional ovarian cysts diagnosed at laparoscopy and confirmed on ultrasound scan. Two of these presented with one-sided abdominal pain. All cases were monitored with serial ultrasound scan only.


View this table:
[in this window]
[in a new window]
 
Table III. Side-effects reported during the 3-year period on Mirena

 

    Discussion
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Our results confirm that this device is effective in symptom control for at least 3 years. The symptoms continued to improve, albeit less dramatically after the first 12–18 months on the device. Although a few of the patients experienced the typical systemic side-effects of progestogens, these were mild and tolerable. These results provide ample support for the use of this device in those women who are not immediately desirous of pregnancy. The benefits of using Mirena on even a short-term basis significantly outweigh the cost of a 6 months course of a GnRH agonist. The continuation rate after 3 years was 55.9%, similar to that in women using the device for contraception (Andersson et al., 1994Go; Faculty of Family Planning and Reproductive Health Care Guidance, 2004Go) and as treatment of menorrhagia (Hurskainen et al., 2004Go). It is, however, difficult to compare this with other medical treatment options, as these often involve a much shorter duration. An advantage of Mirena over other medical treatment options, which are usually short-term, is obviation of the need for repeated treatment (often with alternate regimens) where symptoms recur (in up 75% of patients within 5 years; Surrey and Hornstein, 2002Go) and the lack of need for additional contraception.

The main reasons for discontinuation were menstrual irregularities, persistent pelvic pain and weight gain. The menstrual problems were predominantly during the first 6 months on the device. These findings are not dissimilar to those in women on the device for contraception and the treatment of menorrhagia. In this study only two patients requested that the device be removed, for unacceptable spotting after the first 6 months. If this information is provided to patients during counselling and at follow-up visits, we would anticipate the continuation rates to remain high. Although there have been anectodal approaches to managing irregular bleeding in such women, such as the short-term use of the combined oral contraceptive pill, there are no studies on the best approach to treatment. Until this evidence is available, we would advocate a careful selection of patients, pre-insertion counselling supported by information leaflets and regular re-assuring follow-up visit during the first 6 months.

The most dramatic improvements in symptoms as determined by the VAS, VRS and quantified blood loss occurred during the first 12 months of therapy. Thereafter, there were no significant changes in the variables used to assess response to treatment over the remaining 24 months. Whether these improvements in symptoms will persist for the entire 5 years remains to be determined.

The proportion of women experiencing unacceptable unscheduled bleeding was similar to that reported by Vercellini and colleagues (Vercellini et al., 1999Go; 2003Go). However, the proportion of those continuing with the device at 12 months was lower in our study compared with that reported previously (Vercellini et al., 1999Go; 2003Go; Fedele et al., 2001Go). The drop-out rates between 12 and 36 months were, however, higher in patients having the device as adjunct to surgery (Vercellini et al., 2003Go) compared with our study, where it was used as the only treatment modality. These differences could partly be explained by the patients enrolled in these studies: a large proportion of our patients were undergoing treatment for the first time, while those in the other studies had already had other forms of treatment and were therefore more likely to tolerate side-effects, provided the treatment resulted in an improvement in their symptoms. Additionally, the numbers in our study are small.

The precise mechanism by which this device acts in pelvic endometriosis is uncertain. Several suggestions have been proposed including systemic (Luukkainen, 2000Go) and local (Pakarinen et al., 1995Go; Jones and Critchely, 2000Go; Hurskainen et al., 2000Go) actions. The systemic effects are most likely to be mediated via the suppression of ovulation. The critical systemic level of levonorgestrel required to achieve this is 200 pg/ml (Nilsson and Lukkainen, 1977Go). This level is achieved in most women during the first 3 months on the device, and falls thereafter. In fact, anovulatory rates of between 71% and 85% have been reported during the first 3 months on the device. It is perhaps only during this time that this mechanism influences patient's symptoms, since the ovulatory rate remains low in a large number of women (Lahteenmaki et al., 2000Go).

Various potential local mechanisms have been proposed for the action of this device in women with endometriosis, but an exact understanding of these mechanisms remains unclear. We measured serum and peritoneal fluid levonorgestrel levels in this cohort during the second-look laparoscopy and demonstrated that the peritoneal fluid levels were approximately two-thirds of the serum levels. These laboratory data and discussions about the possible mechanisms of action of the device will be published separately (Lockhat et al., 2005Go). A combination of a reduction in peritoneal fluid volume (Drake et al., 1980Go; Khorram et al., 1993) and a low concentration of peritoneal fluid macrophages (Ramey and Archer, 1993Go) and inflammatory markers (Haney and Weinberg, 1988Go; Kupker et al., 1998Go) may be one of the mechanisms by which this device alters the symptom of pelvic pain. Whatever the case, it is likely that the device is effective through a combination of systemic and local mechanisms. The changes in the staging we demonstrated at 6 months (Lockhat et al., 2004Go) suggest that there is likely to be a local effect on the condition, although this is by no means conclusive, since systemic levonorgestrel may also induce the same changes. Whatever the exact mechanism, the local effect of the progestogen on the endometrium resulting in hypomenorrhoea or amenorrhoea significantly improves the pain of dysmenorrhoea and menorrhagia.

Although it is perhaps not unreasonable to adduce from these findings that the improvement in symptoms was primarily a result of the levonorgestrel device, the placebo effect of the diagnostic laparoscopy cannot be eliminated. The only way to answer this question will be a randomized trial where one group receives the levonorgestrel intrauterine system and another a placebo intrauterine system. Since such a placebo device is currently not available, the copper intrauterine device may be used. We are currently undertaking such a randomized controlled trial with the gold standard for the medical treatment of endometriosis.

The results presented provide the first evidence to support the long-term use of this form of therapy in women presenting with symptoms of endometriosis. A better tool for the assessment of its overall effectiveness in these patients would have been a quality of life analysis. Unfortunately this was not used in this study, but will now be used in subsequent studies. Whether the symptom control observed over the 3-year period reported here will persist for the remaining 2 years of the lifespan of the device remains unanswered. Since there are no forms of medical therapy that offer such long-term relief of symptoms, it will be difficult to undertake randomized, controlled and comparable studies. However, there is the need to gather more information on the long-term use of this device, especially in the older patient who may ultimately avoid the need for a hysterectomy.


    Acknowledgements
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
We are grateful to all the women who participated in this study and the support we received from the local branch of the National Endometriosis Society and the SHE trust. We would like to thank Schering Healthcare for funding the study, and are especially grateful to Drs Sheila Thorpe and Peter Longthorne for their support.


    References
 Top
 Abstract
 Introduction
 Subjects and methods
 Results
 Discussion
 Acknowledgements
 References
 
Ailawadi AK, Jobanputra S, Kataria M, Gurates B and Bulun S (2004) Treatment of endometriosis and chrinic pelvic pain with letrazolew and norethindrone: a pilot study. Fertil Steril 81, 290–296.[CrossRef][ISI][Medline]

American Fertility Society (1985) Revised American Fertility Society classification of endometriosis. Fertil Steril 43, 351–352.[Medline]

Andersson K, Odlind V and Rybo G (1994) Levonorgestrel-releasing and copper-releasing (Nova T) IUDs during five years of use: a randomized comparative trial. Contraception 49, 56–72.[ISI][Medline]

Bilian X, Liying Z, Xuling Z, Mengchun, Luukkainen T and Allonen H (1990) Pharmacokinetic and pharmacodynamic studies of levonorgestrel-releasing intrauterine device. Contraception 41, 353–362.[CrossRef][ISI][Medline]

Drake TS, Grunnert GM and O'Brien WF (1980) Peritoneal fluid volume in endometriosis. Fertil Steril 34, 280–281.[ISI][Medline]

Du M, Shao O and Zhou X (1999) Serum levels of levonorgestrel during longterm use of Norplant [in Chinese]. J Zhonghua Fu Chan Ke Za Zhi 34, 363–365.

Faculty of Family Planning and Reproductive Heath Care Guidance (2004) The levonorgestrel-releasing intrauterine system (LNG-IUS) in contraception and reproductive health. J Fam Plann Reprod Health Care 30, 99–108.[ISI][Medline]

Fedele L, Bianchi S, Zanconato G, Portuese A and Raffaelli R (2001) Use of levonorgestrel-releasing intrauterine system in the treatment of recto-vaginal endometriosis. Fertil Steril 75, 485–488.[CrossRef][ISI][Medline]

Haney AF and Weinberg JB (1988) Reduction in the intraperitoneal inflammation associated with endometriosis by treatment with medroxyprogesterone acetate. Am J Obstet Gynecol 159, 450–454.[ISI][Medline]

Higham J, O'Brien PM and Shaw RW (1990) Assessment of menstrual loss using a pictorial chart. Br J Obstet Gynaecol 97, 734–739.[ISI][Medline]

Hurskainen R, Salmi A, Paavonen J, Teperi J and Rutanen E (2000) Expression of sex steroid receptors and Ki-67 in the endometrium of menorrhagic women: effects of intrauterine levonorgestrel. Mol Hum Reprod 6, 1013–1018.[Abstract/Free Full Text]

Hurskainen R, Teperi J, Rissan P, Aalto A, Grenman S, Kivela A, Kujansuu E, Vuorma S, Yliskoki M and Paavpnen J (2004) Clinical outcomes and costs with levonorgestrel-releasing intrauterine system or hysterectomy for treatment of menorrhagia. JAMA 291, 1456–1463.[Abstract/Free Full Text]

Jones RL and Critchley HO (2000) Morphological and functional changes in human endometrium following intrauterine levonorgestrel delivery. Hum Reprod 15 (suppl 3), 162–172.[Medline]

Kupker W, Schultze-Mosgau A and Diedritch K (1998) Paracrine changes in the peritoneal environment of women with endometriosis. Hum Reprod Update 4, 719–723.[Abstract/Free Full Text]

Lahteenmaki P, Haukkamaa M, Puolakka J, Riikonen U, Sainio S, Suvisaari J and Nilsson CG (1998) Open randomised study of the use of levonorgestrel releasing intrauterine system as alternative to hysterectomy. BMJ 316, 1122–1126.[Abstract/Free Full Text]

Lahteenmaki P, Rauramo I and Backman T (2000) The levonorgestrel intrauterine system in contraception. Steroids 65, 693–697.[CrossRef][ISI][Medline]

Lockhat FB, Emembolu JO and Konje JC (2004a) The evaluation of the effectiveness of an intrauterine administered progestogen (levonorgestrel) in the symptomatic treatment of endometriosis and on the staging of the disease. Hum Reprod 19, 179–184.[Abstract/Free Full Text]

Lockhat FB, Emembolu JO and Konje JC (2005) Serum and peritoneal fluid levels of levonorgestrel in women with endometriosis treated with an intrauterine contraceptive device containing levonorgestrel (Mirena). Fertil Steril, in press.

Luciano AA, Turksoy RN and Carleo J (1988) Evaluation of oral medroxyprogesterone acetate in the treatment of endometriosis. Obstet Gynecol 72, 323–327.[Abstract]

Luukkainen T (2000) The levonorgestrel intrauterine system: therapeutic aspects. Steroids 65, 699–702.[CrossRef][ISI][Medline]

Monghissi K (1999) Medical treatment of endometriosis. Clin Obstet Gynaecol 42, 620–632.[CrossRef]

Nilsson CG and Luukkainen T (1977) Improvement of a D-norgestrel-releasing IUD. Contraception 15, 295–306.[CrossRef][ISI][Medline]

Nilsson CG, Lahteenmaki P, Robertson DN and Luukkainen P (1980) Plasma concentrations of levonorgestrel as a function of the release rate levonorgestrel from medicated intrauterine devices. Acta Endocrinol 93, 380–384.[ISI][Medline]

Pakarinen P, Luukkainen T, Laine H and Lahteenmaki P (1995) The effect of local intrauterine levonorgestrel administration on endometrial thickness and uterine circulation. Hum Reprod 10, 2390–2394.[Abstract]

Ramey JW and Archer DF (1993) Peritoneal fluid: its relevance to the development of endometriosis. Fertil Steril 60, 1–14.[ISI][Medline]

Stewart A, Cummins C, Jordan R and Phillips W (2001) The effectiveness of the levonorgestrel-releasing intrauterine system in menorrhagia: a systematic review. Br J Obstet Gynaecol 108, 74–86.[CrossRef][ISI]

Surrey ES and Hornstein MD (2002) Prolonged GnRH agonist and add-back therapy for symptomatic endometriosis: long-term follow-up. Obstet Gynecol 99, 709–719.[Abstract/Free Full Text]

Takayama K, Zeitoun K, Gumby RT, Sasano H, Carr BR and Bulun SE (1998) Treatemnt of severe postmenopausal endometriosis with an aromatase inhibitor. Fertil Steril 69, 709–713.[CrossRef][ISI][Medline]

Van de Hurk PJ and O'Brien PMS (1999) Non-contraceptive use of levonorgestrel-releasing intrauterine system. Obstet Gynecol 1, 13–19.[CrossRef]

Vercellini P, Aimi G, Panazza S, De Giorgi O, Pesole A and Crosignani PG (1999) A levonorgestrel releasing intrauterine system for the treatment of dysmenorrhoea associated with endometriosis—a pilot study. Fertil Steril 72, 505–508.[CrossRef][ISI][Medline]

Vercellini P, Frontino G, De Giorgi O, Aimi G, Zaina B and Crosignani PG (2003) Comparison of a levonorgestrel-releasing intrauterine device versus expectant management after concervative surgery for symptomatic endometriosis: a pilot study. Fertil Steril 80, 305–309.[ISI][Medline]

Winkel CA (2003) Evaluation and management of women with endometriosis. Obstet Gynecol 102, 397–408.[Abstract/Free Full Text]

Winkel CA and Scialli AR (2001) Medical and surgical therapies for pain associated with endometriosis. J Womens Health Gend Based Med 10, 137–162.[CrossRef][ISI][Medline]

Submitted on May 20, 2004; resubmitted on July 9, 2004; accepted on October 20, 2004.