Low-dose oral contraceptive pill for dysmenorrhea associated with endometriosis: a placebo-controlled, double-blind, randomized trial

Transcription

1 ENDOMETRIOSIS Low-dose oral contraceptive pill for dysmenorrhea associated with endometriosis: a placebo-controlled, double-blind, randomized trial Tasuku Harada, M.D., a Mikio Momoeda, M.D., b Yuji Taketani, MD., b Hiroshi Hoshiai, M.D., c and Naoki Terakawa, M.D. a a Department of Obstetrics and Gynecology, Tottori University School of Medicine, Yonago, Japan; b Department of Obstetrics and Gynecology, Tokyo University, Tokyo, Japan; and c Department of Obstetrics and Gynecology, Kinki University, Osaka, Japan Objective: To evaluate the efficacy of a low-dose oral contraceptive pill (OCP) for patients with dysmenorrhea associated with endometriosis. Design: A double-blind, randomized, placebo-controlled trial. Settings: Clinical trial sites in Japan. Patient(s): One hundred patients with dysmenorrhea associated with endometriosis. Most enrolled patients had radiologic evidence of endometriosis rather than surgical diagnosis. Intervention(s): Patients were randomly assigned to receive either monophasic OCP (ethinylestradiol plus norethisterone) or placebo. Participants used their usual pain medications as needed during the trial. Main Outcome Measure(s): After four cyclic treatments, we used a zero- to three-point verbal rating scale and a visual analogue scale to measure the severity of disability because of dysmenorrhea in daily life, and the patients use of analgesics. Result(s): Total dysmenorrhea scores assessed by the verbal rating scale were significantly decreased at the end of treatment in both groups. From the first cycle through the end of treatment, dysmenorrhea in the OCP group was significantly milder than in the placebo group. Nonmenstrual pelvic pain was present at baseline in 24.5% (12 of 49) of the OCP group and 34.0% (16 of 47) of the placebo group. The volume of endometrioma (larger than 3 cm in diameter) was significantly decreased in the OCP group, but not in the placebo group. No serious adverse events related to using OCPs occurred. Conclusion(s): The present study clearly demonstrated for the first time that OCPs could be used to effectively and safely treat pain associated with endometriosis. (Fertil Steril Ò 2008;90: Ó2008 by American Society for Reproductive Medicine.) Key Words: Oral contraceptive (OC), ethinylestradiol, norethisterone, randomized clinical trial Endometriosis, a chronic gynecologic disease, is characterized by the presence and growth of endometrial-like glands and stromas outside the uterine cavity. Endometriosis causes dysmenorrhea, dyspareunia, pelvic pain, and infertility in reproductive-age women. Dysmenorrhea is the most common symptom in patients with endometriosis. Ovulation inhibition is known to relieve dysmenorrhea, and oral contraceptive pills (OCPs) that contain synthetic estrogen and progestin can be used for this purpose (1). Oral contraceptive pills suppress ovulation and reduce the growth of endometrial tissue, thus reducing both menstrual flow and prostaglandins production (2). Recently, OCPs were also shown to down-regulate cell Received May 16, 2007; revised and accepted August 21, All authors have received consulting fee from Nobelpharma Co., Ltd. Tokyo, Japan. Reprint requests to: Tasuku Harada, M.D., Department of Obstetrics and Gynecology, Tottori University School of Medicine, Yonago , Japan (FAX: ; tasuku@grape.med.tottori-u.ac.jp). proliferation and increase apoptosis in the eutopic endometrium of women with endometriosis (3). Oral contraceptive pills, which have been used empirically to alleviate dysmenorrhea for many years, are generally well tolerated, with fewer metabolic and hormonal side effects than danazol or gonadotropin-releasing hormone agonist (GnRHa). Open clinical trials have shown that OCPs relieve dysmenorrhea. The Cochrane Database describes a few randomized clinical trials conducted from 1960 to Proctor et al. (4) revealed that OCPs with medium-dose estrogen and first- and second-generation progestogens were more effective than placebos for dysmenorrhea, but four classic randomized clinical trials used a small number of patients and are of poor quality. In those studies, most of the study subjects were women with primary dysmenorrhea, but no endometriosis. Although OCPs for dysmenorrhea associated with endometriosis have been the first choice of treatments in the /08/$34.00 Fertility and Sterility â Vol. 90, No. 5, November doi: /j.fertnstert Copyright ª2008 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 current medical management of endometriosis (1, 5), no high-quality clinical evidence is available regarding the efficacy of low-dose OCP. Therefore, we undertook a randomized, placebo-controlled, clinical trial to examine the efficacy of low-dose OCP on dysmenorrhea associated with endometriosis. METHODS Patient Selection Patient inclusion criteria included women of 18 years and older; regular menstrual cycles (28 2 days); symptomatic endometriosis (diagnosed by laparoscopy or laparotomy) or ovarian endometrioma (diagnosed by ultrasound or magnetic resonance imaging); normal cervical and endometrial smear cytology; moderate or severe dysmenorrhea (evaluated by a modified pain scale) and no medical or surgical treatment for endometriosis within 8 weeks before entry into the study, including hormonal agents, such as OCP, GnRHa, and danazol. The study patients must have had moderate or severe dysmenorrhea, scoring higher than three points at the admission visit on a modified pain scale originally developed by Biberoglu et al. (6) and Andersch et al. (7). This verbal rating scale (VRS) defines pain according to a limited ability to work and the need for analgesics, categorized by the number of days analgesics were used (Table 1). The degree of dysmenorrhea and nonmenstrual pelvic pain was also evaluated according to a visual analogue scale (VAS). Study Design This was a phase III, randomized, double-blind, placebocontrolled, multicenter trial of low-dose OCP versus placebo in 100 patients with endometriosis performed in 18 centers (13 clinics, 5 hospitals) in Japan. Subjects were randomly TABLE 1 Grading and scoring of symptoms and requirement of analgesics. Grade Score Dysmenorrhea (or nonmenstrual pelvic pain) None 0 none Mild 1 some loss of work (or study) efficiency Moderate 2 want to take some rest in bed, loss of work Severe 3 in bed more than 1 day Use of analgesics (previous or present period) None 0 none Mild 1 take analgesics for 1 day Moderate 2 take analgesics for 2 days Severe 3 take analgesics for >3 days Note: Grading scales developed by Biberoglu et al. and Andersch et al. were modified. assigned in a ratio of 1:1 to receive monophasic OCP (ethinylestradiol mg plus norethisterone 1 mg) for 21 days, plus 7 days of placebo or identical placebo for 28 days. The OCP and the placebo were prepared by the manufacturer in 28-day blister packs and appeared identical. The use of analgesic agents was allowed, but other hormonal treatments for pain or vaginal bleeding were prohibited. Randomization was done by the pharmaceutical company (Nobelpharma Co., Ltd. Tokyo, Japan), using the permuted block method. One block, consisting of six sets of drugs (three active drug, three placebo), was allocated to each of the 18 centers. Allocation concealment was accomplished centrally by the company, not broken until after all data were collected. Both the patients and the doctors were blinded regarding the medication. Treatment began on the third day (2 days) of the menstrual cycle and continued for four cycles. This study was conducted in accordance with the Declaration of Helsinki and the Japanese Pharmaceutical Affairs Law. The investigators obtained approval for the protocol from their institutional review boards, and each patient provided written informed consent. Patient Monitoring During the two menstrual cycles before starting the study protocol, each patient underwent a prerecruitment evaluation, consisting of a general medical and gynecologic history, physical and pelvic examination, clinical evaluation of signs and symptoms, cervical and endometrial smear, patient evaluation of pain, and review of the menstrual record, that is, scoring dysmenorrhea and nonmenstrual pain and recording details about the use of analgesics. Blood was drawn for pretreatment clinical laboratory determinations, including hematology (white blood cells, platelet, red blood cells, hemoglobin, and hematocrit), coagulation tests (fibrinogen, plasminogen, PT, APTT, and AT-III), biochemical test (total protein, albumin, GOT, GPT, r-gtp, Al-P, LDH, total bilirubin, total cholesterol, high-density lipoprotein-cholesterol, TG, Creatinine, Ca, Fe, P, Na, K, and Cl), antiphospholipid antibodies, CA125, and serum IL-6. At this point, eligibility for entry into the study was determined, pain assessed, and the drug provided. Patients were followed every 4 weeks and their diaries reviewed to evaluate pain and bleeding, side effects, and other health concerns. Efficacy was evaluated at the final visit after completing treatment. Symptoms were assessed using the modified pain scale (Table 1). The primary efficacy endpoint was patient response to treatment for dysmenorrhea associated with endometriosis, as evaluated by VRS. Secondary efficacy endpoints included changes in the VAS of dysmenorrhea, VRS, and VAS in nonmenstrual pain, clinical evaluation of pelvic induration, and size of ovarian endometrioma. Statistical Analysis A minimum of 40 patients per treatment arm was required for the trial to ensure that this trial would have a power of 80% to 1584 Harada et al. OCP for dysmenorrhea with endometriosis Vol. 90, No. 5, November 2008

3 detect 1.2-point differences in mean change in the dysmenorrhea score between the OCP group and the placebo group. The estimated difference between the groups was calculated from our pilot study and from reported data on placebo (8). Assuming a dropout rate during treatment of 10%, we tried to recruit 50 patients per group. The mean changes of the dysmenorrhea score from baseline at the final evaluation were tested, using the Student t test or Wilcoxon-2 sample test. Repeated-measured analysis using a mixed model was performed to evaluate the effect of treatment. Statistical significance was accepted at P<.05. The intent-to-treat population included patients who received at least one cycle of study medication. RESULTS Disposition of Patients Of 107 patients entered in the study, 7 were excluded before randomization because they had abnormal smear cytology (n ¼ 3), exclusion criteria (n ¼ 3), or positive antiphospholipid antibodies (n ¼ 1). One hundred patients were randomized to receive either OCP (n ¼ 51) or placebo (n ¼ 49). One patient in the OCP group did not take OCPs because she became pregnant after randomization. One patient in the OCP and two in the placebo group were lost to follow-up. Therefore, 96 patients were included in at least one of the efficacy analyses. A patient flow chart is presented in Figure 1. Fourteen patients (seven OCP, seven placebo) discontinued the study. Four of the OCP patients were discontinued because of adverse effects (one, rupture of ovarian cyst; one, nausea and headache; one, ovarian hemorrhagic cyst; one, edema), two patients were lost to follow-up, and one took a prohibited drug. Seven of the placebo patients terminated: three had adverse effects (one, edema and headache; one, ovarian hemorrhagic cyst; one, worsened dysmenorrhea), FIGURE 1 A flow-chart of participants through the trial. OCP: oral contraceptive pill. three were lost to follow-up, and one used a prohibited drug. Continuation rates were similar between the treatment groups, with 88% of patients receiving OCPs and 86% receiving placebo continuing in the study. Patient Characteristics The two treatment groups were similar with respect to pretreatment and demographic characteristics (Table 2). Most patients (47 of 49 in the OCP group and 44 of 47 patients in the placebo group) had endometrioma. Changes in Pain Score The mean scores before and after treatment in the dysmenorrhea and nonmenstrual pain evaluation are shown in Table 2. Treatment groups did not differ significantly for any of the primary variables (dysmenorrhea and nonmenstrual pain scores) at the start of treatment. Total dysmenorrhea scores assessed by VRS (dysmenorrhea pain plus drug score) significantly decreased at the end of treatment in both the OCP and placebo groups. However, the reduction in pain score was significantly higher in the OCP group ( 2.0) compared with the placebo group ( 0.6) (P<.0001). The mean dysmenorrhea VAS scores were also significantly lower in both groups after treatment. Again, reduction of the dysmenorrhea score in the OCP group was significantly greater than in the placebo group. The total dysmenorrhea score was significantly reduced when evaluating the first treatment cycle. Dysmenorrhea (total score) in the OCP group became significantly milder than that of the placebo group from the first visit to the end of treatment (Fig. 2). The dysmenorrhea pain score was also reduced at the first visit through the end of treatment. The dysmenorrhea drug score decreased at the end of second cycle through final evaluation (data not shown). Nonmenstrual pelvic pain was present at baseline in 24.5% (12 of 49) of the OCP group and 34.0% (16 of 47) of the placebo group. Total nonmenstrual pelvic pain scores by VRS were not significantly reduced after treatment in either the OCP or placebo groups. The nonmenstrual pelvic pain VAS score was significantly reduced only in the OCP group (Table 2). Clinical Evaluation of Pelvic Induration Pelvic induration was identified at the baseline examination in 32 of 49 OCP patients and 33 of 47 placebo patients. At the final visit, OCP patients showed less induration (21 of 49) than the placebo group (14 of 47). The difference was not significant. Ovarian Endometrioma Ovarian endometrioma had a high prevalence rate at baseline (Table 2). The volume of the endometrioma was significantly reduced at the end of treatment in both the OCP and placebo groups (Table 2). When ovarian endometriomas larger than 3 cm in diameter were analyzed, the volume of the Fertility and Sterility â 1585

4 TABLE 2 Demographic data of patients and effects of treatment on pain score and endometrioma size. Oral contraceptive (n [ 49) Placebo (n [ 47) P value Age (y) Age of menarche (y) Cycle length (days) Previous pregnancy Median parity (range) 0 (0 4) 0 (0 3) Surgical diagnosis 10 7 Presence of Endometrioma a Adenomyosis 7 7 Dysmenorrhea score VRS Pretreatment End of treatment P< P¼.0047 <.0001 VAS Pretreatment End of treatment P< P¼.0064 <.0001 Nonmenstrual pelvic pain score VRS Pretreatment End of treatment P¼ P¼ VAS Pretreatment End of treatment P¼ P¼ Volume of endometrioma (median, ml) Pretreatment End of treatment 7.6 P< P¼ Average diameter of endometrioma (mean, mm) Pretreatment End of treatment P< P¼ Note: Values are mean SD. a Endometrioma was diagnosed by ultrasound examination or magnetic resonance imaging. endometrioma was significantly reduced only in the OCP group (P<.01, data not shown). Side Effects No serious adverse events related to OCP use occurred. However, the incidence of irregular uterine bleeding and nausea was significantly higher in the OCP group than in the placebo group (60.0% vs. 26.5%, 24.0% vs. 0.0%, respectively). No clinically relevant changes in hematology and chemistry assay were noted. DISCUSSION Oral contraceptive pills have been used as a first-choice drug for primary dysmenorrhea and dysmenorrhea associated with endometriosis (1, 5). However, clinical evidence supporting the efficacy of low-dose OCP is lacking. This multicenter, double-blind, placebo-controlled, clinical trial evaluated the efficacy of low-dose OCP for dysmenorrhea associated with endometriosis. Our results demonstrated for the first time that low-dose OCPs effectively reduced dysmenorrhea associated with endometriosis compared with placebo. OCPs lessened dysmenorrhea and reduced the number of days analgesics were used, suggesting that OCPs improved quality of life in endometriosis patients. No serious adverse events related to OCPs use were observed. Our results support the common clinical practice of treating dysmenorrhea resulting from endometriosis with low-dose OCP. A meta-analysis by The Cochrane Collaboration suggested that OCPs may be more effective than placebo, but the investigators pointed out that the studies included in the meta-analysis were conducted >20 years ago using high-dose OCPs. Recent studies have demonstrated the efficacy of various 1586 Harada et al. OCP for dysmenorrhea with endometriosis Vol. 90, No. 5, November 2008

5 FIGURE 2 Changes in mean dysmenorrhea score during the trial. OCP: oral contraceptive pill. combinations of low-dose OCPs as a treatment choice for primary dysmenorrhea (9, 10). Endometriosis is often manifested by pain symptoms, such as dysmenorrhea, pelvic pain, and dyspareunia. Current therapeutic options include conservative surgery and medical management with GnRH agonist, danazol, and progestins. Medical treatment of endometriosis-associated pain is generally effective, with little difference in efficacy observed among the different types of agents used; however, the adverse-event profiles of the various drug regimens can differ markedly. GnRH agonists are the most frequently used current medical treatment for endometriosis. Although the effectiveness of GnRH agonists is well established, the side effects include hot flushes, vaginal dryness, emotional lability, and loss of libido. Administration of GnRH agonist also results in a 3% to 5% loss in bone mineral density, and this decrease may not be reversed until a few years after completion of treatments (1, 2). The OCP is the most widely used hormonal drug in the world. Although some minor adverse events have been reported, all symptoms are well tolerated and much safer than other hormonal agents. Vercellini et al. (8) evaluated the use of a combination OCP (ethinylestradiol and desogestrel) compared with monthly subcutaneous injections of GnRH agonist in reducing dysmenorrhea, dyspareunia, and nonmenstrual pain in the presence of a laparoscopically confirmed diagnosis of endometriosis. Of the 57 patients evaluated, no significant differences between the two treatments for any of the three symptoms evaluated were found at the end of the 6-month follow-up. Another small, prospective, randomized study reported on the postoperative administration of 6 months of cyclic low-dose OCP containing ethinylestradiol and gestodene compared with no treatment. The results indicated that OCP therapy after operative laparoscopy did not significantly influence long-term recurrence rates for symptoms or endometrioma, although it showed a significant trend toward a delay in disease recurrence (11). Recently, Zupi et al. (12) conducted a controlled study to assess whether GnRH agonist plus add-back therapy allows longer treatment and is associated with better pain control and quality of life than GnRH analogue alone or OCP. The results showed that 12 months of GnRH agonist plus add-back, GnRH analogue alone, and OCP significantly reduced pelvic pain, dysmenorrhea, and the dyspareunia score. They concluded that add-back therapy allows for safe and comfortable treatment over a long period for women with relapsed endometriosis-associated pain. A major challenge in managing endometriosis is that of chronic or recurrent symptoms that require long-term or repeated courses of medication. Treatment with GnRH analogues, such as leuprolide, is limited to only 6 months because these agents induce a hypoestrogenic state that substantially decreases bone mineral density. The OCP as a first choice is a simple and effective way to manage endometriosis. Vercellini et al. (13) reported that avoidance or delaying of menstruation by continuous OCP administration for 2 years significantly reduced the severity of dysmenorrhea in women who experienced recurrent dysmenorrhea despite cyclic OCP use. By preventing ovulation, OCPs suppress the progesteronedriven proliferation of the secretory endometrium during the luteal phase, thus resulting in a decrease in prostaglandin synthesis and the volume of menstrual fluid (2). The most widely accepted theory of the pathogenesis of endometriosis proposes that endometrial tissue reaches the pelvic cavity through retrograde menstruation. Epidemiologic data suggest that potential risk factors for developing endometriosis include shorter menstrual cycles, longer duration of menstrual flow, outflow obstruction, and younger age at menarche (14, 15). Collectively, OCPs that decrease menstrual flow may influence the development of endometriosis. Although the literature is in conflict as to how OCPs affect the risk of endometriosis (16), the currently used low-dose OCPs, although not curative, appear beneficial because they may suppress endometriosis. Recent experimental data show that OCPs also down-regulate cell proliferation and increase apoptosis in the eutopic endometrium of women with endometriosis (3). In conclusion, the present study demonstrated that lowdose OCPs are an effective treatment with few adverse effects for dysmenorrhea associated with endometriosis. Acknowledgments: The authors thank the following doctors for participation of the study: Dr. Kazue Yoshino (Yoshino Women s Clinic), Dr. Toyohiko Miyazaki (Akasakamitsuke Miyazaki Clinic), Dr. Takayuki Kamiya (Shinagawa Ladies Clinic Kamiya), Dr. Tsuneo Yokokura (Yokokura Clinic), Dr. Shigeaki Kurasawa (Ginza Women s Clinic), Dr. Masaru Sakamoto (Sasaki Foundation Kyoundo Hospital), Dr. Ken Yoshikawa (Kichijyoji Ladies Clinic), Dr. Hiroshi Sakuma (Kumagaya General Hospital), Dr. Fumio Sone (Takatsu Central Hospital), Dr. Kiyoko Iesaka (Iesaka Obstetrics and Gynecological Clinic), Dr. Kazunori Kinoshita (Seijo Kinoshita Hospital), Dr. Mitsuma Shimadu (Medical Corporation Daiyukai Daiyukai Daiichi Hospital), Dr. Susumu Namihira (Namihira Ladies Clinic), Dr. Hitoshi Ohkubo (Sapporo Maternity Women s Hospital), Dr. Minoru Yaegashi (Sapporo Maternity Women s S-1 Clinic), Dr. Masaki Hashimoto (Hashimoto Clinic), Fertility and Sterility â 1587

6 Dr. Teruko Yasuda (Yoshio Clinic), and Dr. Shinichi Tanaka (Primo Women s Clinic). REFERENCES 1. Olive DL, Pritts EA. Treatment of endometriosis. N Engl J Med 2001;345: Crosignani P, Olive D, Bergvist A, Luciano A. Advances in the management of endometriosis: an update for clinicians. Hum Reprod Update 2006;12: Meresman GF, Auge L, Baranao RI, Lombardi E, Tesone M, Sueldo C. Oral contraceptives suppress cell proliferation and enhance apoptosis of eutopic endometrial tissue from patients with endometriosis. Fertil Steril 2002;77: Proctor ML, Roberts H, Farquhar CM. Combined oral contraceptive pill (OCP) as treatment for primary dysmenorrhea. Cochrane Database Syst Rev 2001;CD Cecil Text Book of Medicine. 22nd ed. Philadelphia, PA: Elsevier, 2004: Biberoglu KO, Behrman SJ. Dosage aspects of danazol therapy in endometriosis: short-term and long-term effectiveness. Am J Obstet Gynecol 1981;139: Andersch B, Milson I. An epidemiologic study of young women with dysmenorrhea. Am J Obstet Gynecol 1982;144: Vercellini P, Trespide L, Colombo A, Ventola N, Marchini M, Crosignani PG. A gonadotropin-releasing hormone agonist versus a low-dose oral contraceptive for pelvic pain associated with endometriosis. Fertil Steril 1993;60: Hendlix SL, Alexander NJ. Primary dysmenorrhea treatment with a desogestrel-containing low-dose oral contraceptive. Contraception 2002; 66: Davis AR, Westhoff C, O Connell K, Gallagher N. Oral contraceptive for dysmenorrhea in adolescent girls. Obstet Gynecol 2005;106: Muzzi L, Marana R, Caruana P, Catalano GF, Marugutti F, Panici PB. Postoperative administration of monphasic combined oral contraceptives after laparoscopic treatment of ovarian endometriomas: a prospective randomized trial. Am J Obstet Gynecol 2000;183: Zupi E, Marconi D, Sbracia M, Zullo F, Vivo BD, Exacustos C, Sorrenti G. Add-back therapy in the treatment of endometriosis-associated pain. Fertil Steril 2004;82: Vercellini P, Frontino G, Giorgi OD, Pietropaolo G, Pasin R, Crosignani PG. Continuous use of oral contraceptive for endometriosis-associated recurrent dysmenorrheal that does not respond to a cyclic pill regimen. Fertil Steril 2003;80: Apgar BS. Endometriosis. Diagnostic clues and new treatment options. Postgrad Med 1992;92: Barbieri RL. Etiology and epidemiology of endometriosis. Am J Obstet Gynecol 1990;162: Vercellini P, Rangni G, Trespidi L, Oldani S, Crosignani PG. Does contraception modify the risk of endometriosis? Hum Reprod 1993;8: Harada et al. OCP for dysmenorrhea with endometriosis Vol. 90, No. 5, November 2008