Contraception 92 (2015) 301 – 307
Original research article
Levonorgestrel-releasing intrauterine system versus a low-dose combined oral contraceptive for treatment of adenomyotic uteri: a randomized clinical trial☆,☆☆,★ Omar M. Shaaban⁎, Mohammed K. Ali, Ali Mohamed A. Sabra, Diaa Eldeen M. Abd El Aal Department of Obstetrics and Gynecology, Faculty of Medicine, Assiut University, Egypt Received 25 February 2015; revised 26 May 2015; accepted 30 May 2015
Abstract Introduction: This study compares the efficacy of a levonorgestrel-releasing intrauterine system (LNG-IUS) and a low-dose combined oral contraceptive (COC) in reducing adenomyosis-related pain and bleeding. Materials and methods: A randomized clinical trial included 62 participants complaining of pain and bleeding that was associated with adenomyosis. Participants were randomly assigned to either LNG-IUS or COC treatment. The outcomes included the improvement of pain using a visual analogue scale, menstrual blood loss using a menstrual diary and estimated uterine volume by ultrasound for 6 months of treatment. We also compared uterine arteries and intramyometrial Doppler indices before and 6 months after treatment with both LNG-IUS and COCs. Results: Both treatments significantly reduced pain after 6 months of use; however, the reduction was greater in the LNG-IUS group (from 6.23±0.67 to 1.68±1.25) compared with the COCs group (from 6.55±0.68 to 3.90±0.54). Both treatment arms significantly decreased the number of bleeding days, uterine volume and Doppler blood flow in the uterus from before to after treatment. These effects were more significant in the LNG-IUS arm compared with the COC arm. Conclusion: Both LNG-IUS and COCs decreased the pain and menstrual bleeding that is associated with adenomyosis. However, LNG-IUS is more effective than the COCs in reducing pain and menstrual blood loss. This effect may be secondary to the decrease in uterine volume and the increase in blood flow resistance. © 2015 Elsevier Inc. All rights reserved. Keywords: Adenomyosis; COCs; LNG-IUS; Mirena; Dysmenorrhea
1. Introduction Adenomyosis is a common disease in women aged 40–50 years and is considered a significant cause of dysmenorrhea and menorrhagia [1]. Until a few years ago, a hysterectomy was considered the main treatment that could definitively cure this disease [2]. Other treatment options are increasingly offered,
☆
Funding: The authors acknowledge a research grant obtained from the Assiut Medical School Grant's Office. ☆☆ Conflict of interest: Omar Shaaban, Mohammed Ali, Ali Sabra and Diaa Abd El Aal declare no conflict of interest. ★ Clinical trial registration number: NCT01601366. ⁎ Corresponding author. Faculty of Medicine, Assiut University, Women's Health Hospital, Assiut University Campus, 7111, Assiut Egypt, Egypt. Tel.: +20-882412616, +20-1223971457 (mobile). E-mail address: [email protected] (O.M. Shaaban). http://dx.doi.org/10.1016/j.contraception.2015.05.015 0010-7824/© 2015 Elsevier Inc. All rights reserved.
including hormonal suppression with GnRH agonists or danazol and endometrial ablation [3]. However, deep adenomyosis responds weakly to the above treatment options, which are commonly not considered for long-term management because of the associated side effects [4]. The levonorgestrel-releasing intrauterine system (LNG-IUS) is a reversible method of contraception that is effective in treating dysmenorrhea and menorrhagia [5]. LNG-IUS has been suggested as a treatment option for adenomyosis, secondary to its effect in down-regulation of estrogen receptors (ER) in both glandular and stromal endometrial tissues, its effect on decidualization and subsequent noticeable atrophy of the endometrium [6]. In addition, the LNG-IUS may offer pain relief because of the reduction of prostaglandin production within the endometrium as well as atrophy of the adenomyosis foci [7]. Low-dose combined oral contraceptive (COC) pills have been widely
302
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
used as the primary treatment for menorrhagia [8]. COCs can also be used to induce endometrial atrophy and to decrease endometrial prostaglandin production, which can consequently improve menorrhagia and dysmenorrhea that are associated with adenomyosis [9]. The current study compares the efficacy of LNG-IUS with that of low-dose COCs in the treatment of adenomyosisassociated pain with or without abnormal uterine bleeding. To our knowledge, no randomized clinical trial (RCT) has been conducted to compare the above two modalities in the management of adenomyosis. In addition, in this study, we are also testing the hypothesis that adenomyosis increases uterine volume and blood flow to feed the ectopic endometrial foci, thus causing the pain and bleeding that are associated with adenomyosis. Color Doppler ultrasound has been used previously for diagnostic purposes [10,11]. However, the effects of these treatments on the uterine artery or intramyometrial blood flow have not been studied before. 2. Materials and methods The current study is a clinically registered open, parallel, RCT (NCT01601366) comparing the effect of the LNG-IUS (Mirena) with that of low-dose COCs in treating adenomyosisrelated pain with or without uterine bleeding. The ethical review board of the Faculty of Medicine of the Assiut University approved the study. The participants were recruited from the Outpatient Gynecology Clinic of the Women's Health Hospital. It was carried out in the period between the first of August 2013 and the first of November 2014. This trial was designed and reported according to the revised recommendations of ClinicalTrials.gov for improving the quality of reporting RCTs. 2.1. Eligible participants All participants who presented to the above clinic with complaints of pelvic pain (dysmenorrhea and or chronic pelvic pain) with or without uterine bleeding were entered in a screening phase. This phase included history taking (including basal evaluation of the degree of pain and bleeding), clinical examination, two-dimensional transvaginal ultrasound (2D TV/US) and color Doppler ultrasound. Ultrasound scanning was performed using a Sonoline G60S ultrasound imaging system (Siemens, Germany) using a 4- to 7-MHz endovaginal probe. The following ultrasound criteria were used to diagnose adenomyosis: globular uterus, myometrial linear striation, myometrial cysts and uterine wall asymmetry [9]. Myometrial cysts were defined as rounded anechoic areas of 2–6 mm in diameter [10]. Color Doppler was used to differentiate the myometrial cysts from myometrial blood vessels [11]. Other inclusion criteria included the participant's request for contraception for at least 6 months, an age between 20 and 45 years, a resident in the nearby vicinity to make the follow-up easy and feasible
and acceptance of use of either type of management. Exclusion criteria included history of ectopic pregnancy, puerperal sepsis, pelvic inflammatory disease, evidence of coagulopathy and/or abnormalities of the uterine cavity such as submucous fibroids distorting the cavity. Women were also excluded if they had a history of malignancy or histological evidence of endometrial hyperplasia, any adnexal abnormality on ultrasound, undiagnosed vaginal bleeding or any other contraindication to receive COCs. 2.2. Randomization Randomization was done using a computer-generated random table. Eligible patients who consented were randomly assigned to receive either LNG-IUS or COCs. Allocation concealment was done using serially numbered closed opaque envelopes. Each envelope was labeled with a serial number and had a card noting the intervention type inside. Allocation was never changed after opening the envelopes. Preparation and sorting of the serially numbered envelopes was done by an investigator who did not participate in evaluating patients either before recruitment or in the follow-up stage. 2.3. Intervention Eligible participants were allocated to one of two groups. The LNG-IUS group received LNG-IUS (Mirena; Bayer Schering Healthcare AG, Bayer HealthCare, Berlin, Germany), inserted according to the manufacturer's instructions [12]. Transvaginal sonography ascertained proper insertion of the device immediately after insertion. The second group received COCs (Gynera; Bayer Schering Healthcare AG, Bayer HealthCare, Berlin, Germany), which included 30 mcg of ethinyl estradiol and 75 mcg of gestodene. Participants were instructed to use the COCs as prescribed (one pill every day for 21 days followed by a 1-week pill-free interval). Both treatment groups were provided the treatment for free. Baseline assessment of the degree of pelvic pain as measured by visual analogue scale (VAS), the menstrual blood loss by menstrual diary, uterine volume and Doppler blood flow in the uterine and intramyometrial blood flow were evaluated (as described in the study outcomes section). Ultrasound evaluation and clinical visits were free of charge, and the participants were requested to come to a monthly follow-up with our clinic for at least 6 months. Each participant had a special follow-up card for the study that included the study serial number, the study group and the required follow-up schedule. The card also included contact details for reaching a trained nurse for any required help or advice between visits. The contact details of each participant, including cell phone number (if provided), were also recorded on the data collection sheet. 2.4. Study outcomes The primary outcome of this study was the improvement of pelvic pain (dysmenorrhea and or chronic pelvic pain) as
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
measured using a VAS. Operationally, a VAS is usually a horizontal line that is 10 cm in length and is anchored by number descriptors from 0 to 10 at either end. Each participant was advised to consider the 10-end of the line as representing the maximum pain she had ever felt and the 0-end as representing the status of no pain. The patient marked the line according to her maximum perception of pain at the last month prior to assessment of the degree of pain [13]. To increase the reliability of the measurement, the participating women were instructed on how to fill out the menstrual diary and had the chance to complete one cycle during the screening phase of the study. We evaluated the perception of pain at 4 points: baseline and after the first, third and sixth months of treatment. At each visit, the women were asked how they had evaluated their pain and/or dysmenorrhea in their last cycle. Secondary outcomes included menstrual blood loss as measured using the menstrual diary. Uterine arteries and intramyometrial blood flow were measured by color Doppler before the start of treatment. Each uterine vessel was observed with a color Doppler technique with real-time spectral analysis using a convex 3.5-MHz probe with a high pass filter, which was set at 125 Hz; the uterine artery was obtained immediately after crossing the hypogastric artery. The systolic/diastolic (S/D) ratio, resistance index (RI) and the pulsatility index (PI) of uterine arteries and intramyometrial blood vessels were calculated when three similar consecutive waves were obtained. The patients' overall satisfaction was evaluated by three simple questions about the effect of her pain and/or bleeding on her physical health, sexual health and religious duties. Participants were considered to be satisfied if she denied that pain and/or bleeding affected any of the above three aspects of her life and were considered to be unsatisfied if any of them had been affected. Patient responses were assessed after the first, third and sixth months of treatment and were compared with the basal assessment before the intervention. Treatment failure was defined as shifting the patient to another treatment for the same symptoms, opting to remove the uterus by hysterectomy or confirming expulsion or removal of a LNG-IUS. 2.5. Follow-up schedule All study participants were followed at the end of the first, third and sixth months from the start of treatment. During each visit, we evaluated the patients clinically, we asked them to report their perception of pain by VAS in the last month prior to the scheduled flow up visit, their records of menstrual diary were also revised and 2D transvaginal sonography and Doppler indices were done to evaluate uterine volume and blood flow (details above). Finally, the overall satisfaction was assessed only in the final visit. 2.6. Sample size The sample size calculation was based on the primary outcome (improvement of pain as indicated by VAS before
303
and after treatment). Previous nonrandomized studies reported that the LNG-IUS improved dysmenorrhea and chronic pelvic pain from 77.9% to 11.8% (approximately 66%) before and after LNG-IUS use, respectively [14]. Using a two-sided chi-square (χ 2) test with an α of 0.05, a total sample size was calculated to be at least 62 patients in the 2 groups (31 in each arm) with 80% power to detect a 36% difference in the VAS between LNG-IUS and COCs {i.e., 77.9% in the LNG-IUS group vs. 40% in the COCs [odds ratio of 5.8] assuming a rate of loss to follow-up of 10% (Epi-info™, Centers for Disease Control and Prevention, USA). 2.7. Statistical analysis The data were collected and entered into a Microsoft Access database and were analyzed using the Statistical Package for Social Science (SPSS Inc., Chicago, version 16). The demographic characteristics and baseline data were compared between the groups. The outcome variables were calculated using a paired t test to compare continuous variables before and after treatment and using an unpaired t test between groups. For dichotomous variables, chi-square was used to estimate the significance value. For analysis, pb.05 was considered to be significant.
3. Results Out of 142 patients with evidence that was suggestive of adenomyosis, 62 consented to participate. The main reasons for not consenting included preferring hysterectomy as a final option (22 patients) or planning a pregnancy (18 patients). They were either not willing to share in an RCT or had no reason for declining to participate (40 patients). Consenting women were randomized into two groups: LNG-IUS and COCs. At the end of the study, two patients (6.45%) were not included in the analysis in the LNG-IUS group (one for spontaneous expulsion and one for removal because no improvement was observed). In the second group, three patients (9.67%) were not analyzed; two of them were lost for follow-up and one stopped using the pills because of side effects a few days after recruitment (study flow chart, Fig. 1). Histopathological examination of hysterectomized uteri of those who opted to have a hysterectomy (22 women) revealed 18 uteri (81.8%) with typical adenomyosis, 2 uteri (9.1%) with minimal myometrial hyperplasia and 2 uteri (9.1%) with nonspecific infections. Basal characteristics of the study participants are given in Table 1 and showed that the two study groups were similar regarding the mean age, parity, number of previous cesarean section (CS) and the duration that had passed since the last pregnancy. The mean age of the study participants was approximately 39.0 years with a mean body mass index (BMI) of approximately 26.0. Moreover, the mean parity of the participants in the study was approximately 5.0.
304
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Fig. 1. The study flow chart.
Regarding pelvic pain (Table 2), both groups reported a significant decrease in this complaint from before the treatment to the sixth month of treatment (pb.001). However, the LNG-IUS group reported a larger reduction in the VAS in comparison with the COCs group, from 6.23±0.67 to 1.68± 1.25 in the LNG-IUS group and from 6.55±0.68 to 3.90±0.54 in the COCs group. The bleeding pattern was improved in both Table 1 Baseline characteristics of the 62 participants with adenomyosis-related pain and bleeding
Age (years) 30–40 years N40 years Mean±S.D. BMI Number of abortion Number of normal delivery Number of CS Number of living children Duration from the last pregnancy (months) S.D., standard deviation.
Group I LNG-IUS (n= 31)
Group II COCs (n=31)
No.
No.
%
17 54.8 14 45.2 39.39±4.43 25.9±6.3 1.98±1.26 5.64±1.26 0.42±0.67 4.84±1.29 89.03±37.79
p Value
%
19 61.3 12 38.7 39.16±3.21 26.1±4.4 2.0±1.39 5.03±1.35 0.55±0.92 5.55±0.85 87.74±16.84
.607 .819 .985 .170 .150 .532 .013 .126
treatment arms; in the LNG-IUS, the mean number of bleeding days per month decreased from 9.81±1.82 days prior to recruitment to 2.63±2.13 days after the sixth month of insertion. In the COCs group, the number of bleeding days per month decreased from 9.97±1.52 days to 5.52±1.00 days. Regarding the sanitary pads used per day, in the LNG-IUS group, the mean number of pads decreased from 6.29±0.69 to 2.0±1.44. In the COCs group, the number of pads per day decreased from 6.13±0.85 to 3.58±0.67 after 6 months of treatment. The number of bleeding-free days increased from before to 6 months after treatment (from 19.32±2.65 to 25.39± 1.11 days) in the LNG-IUS group and from 19.10±1.98 days to 23.65±2.57 days in the COCs group (Table 2). Uterine volume decreased in both treatment arms after 6 months of treatment; the decrease was more significant in the LNG-IUS arm (from 10.23±1.06 mL to 7.63±0.49 mL) than in the COCs arm (from 10.42±0.99 mL to 8.32±0.91 mL) (pb.001). Collectively, the mean overall reductions in the VAS, the number of bleeding days per month, the number of sanitary pads per day and the uterine volume in the LNG-IUS group were significantly more than that observed in the COCs group (pb.001) (Table 2). From the Doppler indices, S/D, RI and PI of the uterine arteries and intramyometrial blood vessels increased significantly from baseline values to 6 month values; the increases in the LNG-IUS group were more than that of the COCs group (pb.000) (Table 3).
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
305
Table 2 Effect of LNG-IUS and COCs on pain, uterine bleeding and uterine volume before and 6 months after treatment Outcomes (mean± S.D.)
VAS score (mean±S.D.) Number of bleeding days per month Number of sanitary pads per day Number of bleeding-free days
Group I LNG-IUS group
Group II COC group
Intergroup comparisons p value
Baseline
6 months a
p Value
Baseline
6 months a
p Value
Baseline
6 months a
6.23±0.67 9.81±1.82 6.29± 0.69 19.32±2.65
1.68±1.25 2.63± 2.13 2±1.44 25.39±1.11
b .001⁎ b .001⁎ b .001⁎ .001⁎
6.55±0.68 9.97±1.52 6.13±0.85 19.10±1.98
3.90±0.54 5.25 ± 1 3.58±0.67 23.65±2.57
b .001⁎ b .001⁎ b .001⁎ .000⁎
.064 .441 .415 .792
b.001⁎ b.001⁎ b.001⁎ .000⁎
S.D., standard deviation. ⁎ Statistically significant difference (pb.05). a The data presented at 6 months follow-up are only for those women who completed the follow-up period and did not include those who had treatment failure prior to the end of follow-up or those who were lost during follow-up.
At the start of the study, 7 out of 31 women in the LNG-IUS group and 5 out of 31 women in the COCs group expressed their satisfaction with the different aspects of their lives. After 6 months of treatment, 25 participants in the LNG-IUS arm and 18 in the COCs arm expressed their overall satisfaction.
4. Discussion The present work demonstrates that both COCs and LNG-IUS were effective in the treatment of adenomyosis. The LNG-IUS demonstrated superiority to COCs regarding the outcomes of pain and uterine bleeding profile. Furthermore, we demonstrated that both treatment arms reduced uterine volume and blood flow inside the uterus, a change that could have been the cause of the effect of the treatment. There is great variation in the ultrasound diagnostic accuracy of adenomyosis; we depended on this accuracy for the diagnosis of eligible participants [15]. However, both LNG-IUS and COCs are beneficial in treating pain and bleeding of adenomyotic and nonadenomyotic patients. The suggestion of using LNG-IUS was based on its ability to down-regulate ERs in endometrial tissues and its decidualization with subsequent atrophy of the endometrium [12]. In addition, the LNG-IUS has a pain-relieving effect through the reduction of prostaglandin production within the endometrium and induction of adenomyosis foci atrophy [13]. However, the suggestion of using COCs in the treatment of adenomyosis was secondary to its inhibition of prostaglandins and gonadotropin release, which decrease pain and bleeding [16]. The associated pain (dysmenorrhea and/or chronic pelvic pain) was the primary outcome measure of the present study. The patients in the two groups reported an improvement in pain. The mean overall reduction in the VAS in the LNG-IUS group was significantly more than that observed in the COCs group (pb.001). These findings are in agreement with that from the study of Deng et al. in 2006 [17] who reported that the baseline VAS decreased from 8.4±1.5 to 3.99±3.87 after 1 year of insertion of the LNG-IUS.
However, we achieved a more marked reduction in the VAS after 6 months, most likely because the baseline VAS in our study was lower (6.23) than that in the Deng et al. study (8.41). In 2007, Bragheto and colleagues followed up 29 women after 6 months insertion of the LNG-IUS, and they reported a similar reduction in the VAS pain score between baseline and 6 months after insertion (pb.001). However, six women in their study continued to report a VAS of more than 3 after 6 months of treatment [12]. In 2008, Cho and colleagues reported the effect of LNG-IUS in 47 patients with adenomyosis, and they found that the VAS decreased significantly after 6 months. More importantly, they noted a significant rebound increase in the pain scores at 36 months [13]. In 2010, Zheng and colleagues reported results similar to ours after concurrent treatment with LNG-IUS and GnRH agonist. The VAS was significantly lowered from baseline (6.83±9.3) to 12 months (0.68±0.42) after placement of the LNG-IUS [14]. In 2011, a randomized trial was reported on the treatment of adenomyosis conducted by Bayoglu Tekin and colleagues. They compared a depot GnRH analogue with the LNG-IUS in the treatment of adenomyosis-related chronic pelvic pain. However, the GnRH-a group showed a significant decrease in the VAS compared with the LNG-IUS group at the end of 1 year. However, the GnRH-a group showed menopausal-like symptoms such as hot flashes and vaginal dryness [18]. These symptoms limit the use of this analogue for a long time and decrease patient satisfaction. Menorrhagia associated with adenomyosis was a secondary outcome measure in our RCT. LNG-IUS reduced menstrual blood more effectively than COCs. The first study that reported a reduction in menstrual blood loss after insertion of the LNG-IUS in patients with adenomyosis was by Fedele and colleagues in 1997. They studied the effect of LNG-IUS in 25 women with adenomyosis-associated menorrhagia. They found that all studied women experienced an improvement of bleeding symptoms, with a majority experiencing regular scanty flow. They also reported significant increases in hemoglobin, hematocrit and serum ferritin levels [19].
306
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Table 3 Uterine arteries and intramyometrial blood vessel Doppler indices before and 6 months after treatment with LNG-IUS and COCs Group I LNG-IUS group End point Uterine artery S/D RI PI Intramyometrial S/D RI PI
Group II COCs group
Intergroup comparisons p value
Baseline
6 months a
p Value
Baseline
6 months a
p Value
Baseline
6 months a
4.25±1.59 0.75±0.07 1.71±0.35
12.07±2.78 0.94±0.05 2.76±0.44
b.001⁎ b.001⁎ b.001⁎
3.96±0.76 0.72±0.05 1.62±0.39
6.92±1.27 0.85±0.04 2.31±0.47
b.001⁎ b.001⁎ b.001⁎
.015⁎ .678 .041⁎
b.001⁎ b.001⁎ b.001⁎
3.08±0.83 0.66±0.16 1.02±0.36
8.86±4.72 0.87±0.07 2.33±0.44
b.001⁎ b.001⁎ b.001⁎
3.19±0.84 0.76±0.25 1.12±0.41
3.92±0.97 0.75±0.09 1.62±0.48
.058⁎ .018⁎ .004⁎
.955 .084 .263
b.001⁎ b.001⁎ b.001⁎
⁎ Statistically significant difference (pb.05). a The data presented at 6 months follow-up are only for those women who completed the follow-up period and did not include those who had treatment failure prior to the end of follow-up or those who were lost during follow-up.
In 2003, Maia and colleagues [20] evaluated the effect of LNG-IUS versus no treatment as an adjuvant to endometrial ablation in 95 adenomyotic patients suffering from heavy menstrual bleeding. Our results are in accord with their results. In 2008, Cho et al. [13] evaluated the effect of LNG-IUS in 47 patients with adenomyosis-associated abnormal uterine bleeding. Their results were in agreement with ours. The bleeding dropped dramatically after 6 months of LNG-IUS insertion. In our study, LNG-IUS was effective in decreasing uterine volume more so when compared with COCs after 6 months of treatment. There were 2 previous studies addressed this point; Cho et al. [13] reported a significant decrease in mean uterine volume after 6 months of LNG-IUS insertion (156.85±49.79 mL to 118.64±41.36 mL; pb.001). The reduction in uterine volume was more marked after 24 months (128.84 ± 48.70 mL; pb.001). However, Bragheto et al. observed no significant change in uterine volume (142.6 mL vs. 136.4 mL; p=.2077) between baseline and the 6-month evaluation [12]. The difference may be attributed to interobserver variability. The Doppler study of uterine arteries and intramyometrial blood vessels before and after LNG-IUS and COCs treatment in adenomyosis revealed a significant decrease in uterine arteries and intramyometrial blood flow in adenomyotic patients in both groups. The reduction was more significant in the LNG-IUS group than in the COCs group after 6 months, and these changes were positively correlated with the uterine volume changes. Cho et al. has been the only study that reported an effect of LNG-IUS on uterine artery Doppler of adenomyotic uteri. They reported that the mean pulsatility indices of both uterine arteries increased significantly 12 months after insertion (p=.002 for right; p=.011 for left) [13]. Reduction in uterine volume and the decrease in blood flow to and inside the uterus may be responsible at least partially for the clinical improvement observed as a result of the treatment in our study. The present work had some limitations. First, our diagnostic criteria of 2D TV/US and color Doppler ultrasound for adenomyosis had a reported sensitivity and
specificity of 86.0% and 86.67%, respectively; this was because noninvasive confirmations of adenomyosis were not possible. Moreover, neither treatment option would harm nonadenomyotic patients as long as they were suffering from pain and bleeding and requiring contraception. Second, it was infeasible to blind study participants. Third, the small sample size that was available for the final analysis at 6 months is 62 patients. We preferred to use cyclic rather than a continuous COC regimen because developing amenorrhea (ultimate result of continuous COC use) is not always an acceptable option to women in our culture. Especially in the first few months of use of any contraceptive method, they may be worried about pregnancy and or retention of toxic materials inside their body. Continuous use of COCs may have a beneficial effect over cyclic use with regard to a reduction in pain associated with primary dysmenorrhea [21]. However, this has not been proved in previous trials regarding congestive dysmenorrhea associated with adenomyosis. Moreover, continuous COC use may help more with pain relief; however, that should make comparison with LNG-IUS with regard to menstrual pattern very difficult. Amenorrhea results from continuous COC use, while LNG-IUS users continue to menstruate and or spot within the first 6 months of use.
5. Conclusion Both LNG-IUS and COCs are effective in reducing pain and bleeding associated with adenomyosis after 6 months of use. However, LNG-IUS is superior to COCs for relief of both symptoms. The decrease in uterine volume and the increase in blood flow resistance in both the uterine arteries and intramyometrial blood vessels may be the cause of the treatment effect, which is more pronounced with the use of LNG-IUS.
Conflict of Interest The authors declare that there was no conflict of interest.
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Acknowledgments This research was funded by a research grant obtained from the Assiut Faculty of Medicine Grant's Office. References [1] Jason M, Dechenne V, Chantraine F, Nisolle M. Adenomyosis. Rev Med Liege 2013;68:160–2. [2] Uysal A, Taner CE, Mun S, Uysal F, Celimli FH. Use of a levonorgestrel-releasing intrauterine device in the treatment of adenomyosis associated heavy menstrual bleeding. J Pak Med Assoc 2013;63:1349–52. [3] Huchon C, Fauconnier A. Adenomyosis. Rev Prat 2014;64:551–2. [4] Shrestha A, Sedai LB. Understanding clinical features of adenomyosis: a case control study. Nepal Med Coll J 2012;14:176–9. [5] Sanghera S, Roberts TE, Barton P, Frew E, Daniels J, Middleton L, et al. Levonorgestrel-releasing intrauterine system vs. usual medical treatment for menorrhagia: an economic evaluation alongside a randomised controlled trial. PLoS One 2014;9:e91891. [6] Beatty MN, Blumenthal PD. The levonorgestrel-releasing intrauterine system: safety, efficacy, and patient acceptability. Ther Clin Risk Manag 2009;5:561–74. [7] Guo SW, Mao X, Ma Q, Liu X. Dysmenorrhea and its severity are associated with increased uterine contractility and overexpression of oxytocin receptor (OTR) in women with symptomatic adenomyosis. Fertil Steril 2013;99:231–40. [8] Shaaban MM, Zakherah MS, El-Nashar SA, Sayed GH. Levonorgestrel-releasing intrauterine system compared to low dose combined oral contraceptive pills for idiopathic menorrhagia: a randomized clinical trial. Contraception 2011;83:48–54. [9] Sakhel K, Abuhamad A. Sonography of adenomyosis. J Ultrasound Med 2012;31:805–8. [10] Champaneria R, Abedin P, Daniels J, Balogun M, Khan KS. Ultrasound scan and magnetic resonance imaging for the diagnosis of adenomyosis: systematic review comparing test accuracy. Acta Obstet Gynecol Scand 2010;89:1374–84.
307
[11] Hanafi M. Ultrasound diagnosis of adenomyosis, leiomyoma, or combined with histopathological correlation. J Hum Reprod Sci 2013;6:189–93. [12] Bragheto AM, Caserta N, Bahamondes L, Petta CA. Effectiveness of the levonorgestrel-releasing intrauterine system in the treatment of adenomyosis diagnosed and monitored by magnetic resonance imaging. Contraception 2007;76:195–9. [13] Cho S, Nam A, Kim H, Chay D, Park K, Cho DJ, et al. Clinical effects of the levonorgestrel-releasing intrauterine device in patients with adenomyosis. Am J Obstet Gynecol 2008;198, http://dx.doi.org/ 10.1016/j.ajog.2007.10.798. [14] Zheng Z, Wang NN, Wang JH, Gan XQ, Zheng QQ, Ke PQ. Effect of levonorgestrel-releasing intrauterine system combined with GnRH analogue for treatment of large adenomyosis. Nan Fang Yi Ke Da Xue Xue Bao 2010;30:541–3. [15] Dueholm M, Lundorf E, Sorensen JS, Ledertoug S, Olesen F, Laursen H. Reproducibility of evaluation of the uterus by transvaginal sonography, hysterosonographic examination, hysteroscopy and magnetic resonance imaging. Hum Reprod 2002;17:195–200. [16] Streuli I, Dubuisson J, Santulli P, de Ziegler D, Batteux F, Chapron C. An update on the pharmacological management of adenomyosis. Expert Opin Pharmacother 2014;15:2347–60. [17] Deng S, Lang JH, Leng JH, Liu ZF, Sun DW, Zhu L. Effects of levonorgestrel-releasing intrauterine system on pain and recurrence associated with endometriosis and adenomyosis. Zhonghua Fu Chan Ke Za Zhi 2006;41:664–8. [18] Bayoglu Tekin Y, Dilbaz B, Altinbas SK, Dilbaz S. Postoperative medical treatment of chronic pelvic pain related to severe endometriosis: levonorgestrel-releasing intrauterine system versus gonadotropinreleasing hormone analogue. Fertil Steril 2011;95:492–6. [19] Fedele L, Bianchi S, Raffaelli R, Portuese A, Dorta M. Treatment of adenomyosis-associated menorrhagia with a levonorgestrel-releasing intrauterine device. Fertil Steril 1997;68:426–9. [20] Maia H, Maltez A, Coelho G, Athayde C, Coutinho EM. Insertion of Mirena after endometrial resection in patients with adenomyosis. J Am Assoc Gynecol Laparosc 2003;10:512–6. [21] Dmitrovic R, Kunselman AR, Legro RS. Continuous compared with cyclic oral contraceptives for the treatment of primary dysmenorrhea: a randomized controlled trial. Obstet Gynecol 2012;119:1143–50.
Original research article
Levonorgestrel-releasing intrauterine system versus a low-dose combined oral contraceptive for treatment of adenomyotic uteri: a randomized clinical trial☆,☆☆,★ Omar M. Shaaban⁎, Mohammed K. Ali, Ali Mohamed A. Sabra, Diaa Eldeen M. Abd El Aal Department of Obstetrics and Gynecology, Faculty of Medicine, Assiut University, Egypt Received 25 February 2015; revised 26 May 2015; accepted 30 May 2015
Abstract Introduction: This study compares the efficacy of a levonorgestrel-releasing intrauterine system (LNG-IUS) and a low-dose combined oral contraceptive (COC) in reducing adenomyosis-related pain and bleeding. Materials and methods: A randomized clinical trial included 62 participants complaining of pain and bleeding that was associated with adenomyosis. Participants were randomly assigned to either LNG-IUS or COC treatment. The outcomes included the improvement of pain using a visual analogue scale, menstrual blood loss using a menstrual diary and estimated uterine volume by ultrasound for 6 months of treatment. We also compared uterine arteries and intramyometrial Doppler indices before and 6 months after treatment with both LNG-IUS and COCs. Results: Both treatments significantly reduced pain after 6 months of use; however, the reduction was greater in the LNG-IUS group (from 6.23±0.67 to 1.68±1.25) compared with the COCs group (from 6.55±0.68 to 3.90±0.54). Both treatment arms significantly decreased the number of bleeding days, uterine volume and Doppler blood flow in the uterus from before to after treatment. These effects were more significant in the LNG-IUS arm compared with the COC arm. Conclusion: Both LNG-IUS and COCs decreased the pain and menstrual bleeding that is associated with adenomyosis. However, LNG-IUS is more effective than the COCs in reducing pain and menstrual blood loss. This effect may be secondary to the decrease in uterine volume and the increase in blood flow resistance. © 2015 Elsevier Inc. All rights reserved. Keywords: Adenomyosis; COCs; LNG-IUS; Mirena; Dysmenorrhea
1. Introduction Adenomyosis is a common disease in women aged 40–50 years and is considered a significant cause of dysmenorrhea and menorrhagia [1]. Until a few years ago, a hysterectomy was considered the main treatment that could definitively cure this disease [2]. Other treatment options are increasingly offered,
☆
Funding: The authors acknowledge a research grant obtained from the Assiut Medical School Grant's Office. ☆☆ Conflict of interest: Omar Shaaban, Mohammed Ali, Ali Sabra and Diaa Abd El Aal declare no conflict of interest. ★ Clinical trial registration number: NCT01601366. ⁎ Corresponding author. Faculty of Medicine, Assiut University, Women's Health Hospital, Assiut University Campus, 7111, Assiut Egypt, Egypt. Tel.: +20-882412616, +20-1223971457 (mobile). E-mail address: [email protected] (O.M. Shaaban). http://dx.doi.org/10.1016/j.contraception.2015.05.015 0010-7824/© 2015 Elsevier Inc. All rights reserved.
including hormonal suppression with GnRH agonists or danazol and endometrial ablation [3]. However, deep adenomyosis responds weakly to the above treatment options, which are commonly not considered for long-term management because of the associated side effects [4]. The levonorgestrel-releasing intrauterine system (LNG-IUS) is a reversible method of contraception that is effective in treating dysmenorrhea and menorrhagia [5]. LNG-IUS has been suggested as a treatment option for adenomyosis, secondary to its effect in down-regulation of estrogen receptors (ER) in both glandular and stromal endometrial tissues, its effect on decidualization and subsequent noticeable atrophy of the endometrium [6]. In addition, the LNG-IUS may offer pain relief because of the reduction of prostaglandin production within the endometrium as well as atrophy of the adenomyosis foci [7]. Low-dose combined oral contraceptive (COC) pills have been widely
302
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
used as the primary treatment for menorrhagia [8]. COCs can also be used to induce endometrial atrophy and to decrease endometrial prostaglandin production, which can consequently improve menorrhagia and dysmenorrhea that are associated with adenomyosis [9]. The current study compares the efficacy of LNG-IUS with that of low-dose COCs in the treatment of adenomyosisassociated pain with or without abnormal uterine bleeding. To our knowledge, no randomized clinical trial (RCT) has been conducted to compare the above two modalities in the management of adenomyosis. In addition, in this study, we are also testing the hypothesis that adenomyosis increases uterine volume and blood flow to feed the ectopic endometrial foci, thus causing the pain and bleeding that are associated with adenomyosis. Color Doppler ultrasound has been used previously for diagnostic purposes [10,11]. However, the effects of these treatments on the uterine artery or intramyometrial blood flow have not been studied before. 2. Materials and methods The current study is a clinically registered open, parallel, RCT (NCT01601366) comparing the effect of the LNG-IUS (Mirena) with that of low-dose COCs in treating adenomyosisrelated pain with or without uterine bleeding. The ethical review board of the Faculty of Medicine of the Assiut University approved the study. The participants were recruited from the Outpatient Gynecology Clinic of the Women's Health Hospital. It was carried out in the period between the first of August 2013 and the first of November 2014. This trial was designed and reported according to the revised recommendations of ClinicalTrials.gov for improving the quality of reporting RCTs. 2.1. Eligible participants All participants who presented to the above clinic with complaints of pelvic pain (dysmenorrhea and or chronic pelvic pain) with or without uterine bleeding were entered in a screening phase. This phase included history taking (including basal evaluation of the degree of pain and bleeding), clinical examination, two-dimensional transvaginal ultrasound (2D TV/US) and color Doppler ultrasound. Ultrasound scanning was performed using a Sonoline G60S ultrasound imaging system (Siemens, Germany) using a 4- to 7-MHz endovaginal probe. The following ultrasound criteria were used to diagnose adenomyosis: globular uterus, myometrial linear striation, myometrial cysts and uterine wall asymmetry [9]. Myometrial cysts were defined as rounded anechoic areas of 2–6 mm in diameter [10]. Color Doppler was used to differentiate the myometrial cysts from myometrial blood vessels [11]. Other inclusion criteria included the participant's request for contraception for at least 6 months, an age between 20 and 45 years, a resident in the nearby vicinity to make the follow-up easy and feasible
and acceptance of use of either type of management. Exclusion criteria included history of ectopic pregnancy, puerperal sepsis, pelvic inflammatory disease, evidence of coagulopathy and/or abnormalities of the uterine cavity such as submucous fibroids distorting the cavity. Women were also excluded if they had a history of malignancy or histological evidence of endometrial hyperplasia, any adnexal abnormality on ultrasound, undiagnosed vaginal bleeding or any other contraindication to receive COCs. 2.2. Randomization Randomization was done using a computer-generated random table. Eligible patients who consented were randomly assigned to receive either LNG-IUS or COCs. Allocation concealment was done using serially numbered closed opaque envelopes. Each envelope was labeled with a serial number and had a card noting the intervention type inside. Allocation was never changed after opening the envelopes. Preparation and sorting of the serially numbered envelopes was done by an investigator who did not participate in evaluating patients either before recruitment or in the follow-up stage. 2.3. Intervention Eligible participants were allocated to one of two groups. The LNG-IUS group received LNG-IUS (Mirena; Bayer Schering Healthcare AG, Bayer HealthCare, Berlin, Germany), inserted according to the manufacturer's instructions [12]. Transvaginal sonography ascertained proper insertion of the device immediately after insertion. The second group received COCs (Gynera; Bayer Schering Healthcare AG, Bayer HealthCare, Berlin, Germany), which included 30 mcg of ethinyl estradiol and 75 mcg of gestodene. Participants were instructed to use the COCs as prescribed (one pill every day for 21 days followed by a 1-week pill-free interval). Both treatment groups were provided the treatment for free. Baseline assessment of the degree of pelvic pain as measured by visual analogue scale (VAS), the menstrual blood loss by menstrual diary, uterine volume and Doppler blood flow in the uterine and intramyometrial blood flow were evaluated (as described in the study outcomes section). Ultrasound evaluation and clinical visits were free of charge, and the participants were requested to come to a monthly follow-up with our clinic for at least 6 months. Each participant had a special follow-up card for the study that included the study serial number, the study group and the required follow-up schedule. The card also included contact details for reaching a trained nurse for any required help or advice between visits. The contact details of each participant, including cell phone number (if provided), were also recorded on the data collection sheet. 2.4. Study outcomes The primary outcome of this study was the improvement of pelvic pain (dysmenorrhea and or chronic pelvic pain) as
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
measured using a VAS. Operationally, a VAS is usually a horizontal line that is 10 cm in length and is anchored by number descriptors from 0 to 10 at either end. Each participant was advised to consider the 10-end of the line as representing the maximum pain she had ever felt and the 0-end as representing the status of no pain. The patient marked the line according to her maximum perception of pain at the last month prior to assessment of the degree of pain [13]. To increase the reliability of the measurement, the participating women were instructed on how to fill out the menstrual diary and had the chance to complete one cycle during the screening phase of the study. We evaluated the perception of pain at 4 points: baseline and after the first, third and sixth months of treatment. At each visit, the women were asked how they had evaluated their pain and/or dysmenorrhea in their last cycle. Secondary outcomes included menstrual blood loss as measured using the menstrual diary. Uterine arteries and intramyometrial blood flow were measured by color Doppler before the start of treatment. Each uterine vessel was observed with a color Doppler technique with real-time spectral analysis using a convex 3.5-MHz probe with a high pass filter, which was set at 125 Hz; the uterine artery was obtained immediately after crossing the hypogastric artery. The systolic/diastolic (S/D) ratio, resistance index (RI) and the pulsatility index (PI) of uterine arteries and intramyometrial blood vessels were calculated when three similar consecutive waves were obtained. The patients' overall satisfaction was evaluated by three simple questions about the effect of her pain and/or bleeding on her physical health, sexual health and religious duties. Participants were considered to be satisfied if she denied that pain and/or bleeding affected any of the above three aspects of her life and were considered to be unsatisfied if any of them had been affected. Patient responses were assessed after the first, third and sixth months of treatment and were compared with the basal assessment before the intervention. Treatment failure was defined as shifting the patient to another treatment for the same symptoms, opting to remove the uterus by hysterectomy or confirming expulsion or removal of a LNG-IUS. 2.5. Follow-up schedule All study participants were followed at the end of the first, third and sixth months from the start of treatment. During each visit, we evaluated the patients clinically, we asked them to report their perception of pain by VAS in the last month prior to the scheduled flow up visit, their records of menstrual diary were also revised and 2D transvaginal sonography and Doppler indices were done to evaluate uterine volume and blood flow (details above). Finally, the overall satisfaction was assessed only in the final visit. 2.6. Sample size The sample size calculation was based on the primary outcome (improvement of pain as indicated by VAS before
303
and after treatment). Previous nonrandomized studies reported that the LNG-IUS improved dysmenorrhea and chronic pelvic pain from 77.9% to 11.8% (approximately 66%) before and after LNG-IUS use, respectively [14]. Using a two-sided chi-square (χ 2) test with an α of 0.05, a total sample size was calculated to be at least 62 patients in the 2 groups (31 in each arm) with 80% power to detect a 36% difference in the VAS between LNG-IUS and COCs {i.e., 77.9% in the LNG-IUS group vs. 40% in the COCs [odds ratio of 5.8] assuming a rate of loss to follow-up of 10% (Epi-info™, Centers for Disease Control and Prevention, USA). 2.7. Statistical analysis The data were collected and entered into a Microsoft Access database and were analyzed using the Statistical Package for Social Science (SPSS Inc., Chicago, version 16). The demographic characteristics and baseline data were compared between the groups. The outcome variables were calculated using a paired t test to compare continuous variables before and after treatment and using an unpaired t test between groups. For dichotomous variables, chi-square was used to estimate the significance value. For analysis, pb.05 was considered to be significant.
3. Results Out of 142 patients with evidence that was suggestive of adenomyosis, 62 consented to participate. The main reasons for not consenting included preferring hysterectomy as a final option (22 patients) or planning a pregnancy (18 patients). They were either not willing to share in an RCT or had no reason for declining to participate (40 patients). Consenting women were randomized into two groups: LNG-IUS and COCs. At the end of the study, two patients (6.45%) were not included in the analysis in the LNG-IUS group (one for spontaneous expulsion and one for removal because no improvement was observed). In the second group, three patients (9.67%) were not analyzed; two of them were lost for follow-up and one stopped using the pills because of side effects a few days after recruitment (study flow chart, Fig. 1). Histopathological examination of hysterectomized uteri of those who opted to have a hysterectomy (22 women) revealed 18 uteri (81.8%) with typical adenomyosis, 2 uteri (9.1%) with minimal myometrial hyperplasia and 2 uteri (9.1%) with nonspecific infections. Basal characteristics of the study participants are given in Table 1 and showed that the two study groups were similar regarding the mean age, parity, number of previous cesarean section (CS) and the duration that had passed since the last pregnancy. The mean age of the study participants was approximately 39.0 years with a mean body mass index (BMI) of approximately 26.0. Moreover, the mean parity of the participants in the study was approximately 5.0.
304
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Fig. 1. The study flow chart.
Regarding pelvic pain (Table 2), both groups reported a significant decrease in this complaint from before the treatment to the sixth month of treatment (pb.001). However, the LNG-IUS group reported a larger reduction in the VAS in comparison with the COCs group, from 6.23±0.67 to 1.68± 1.25 in the LNG-IUS group and from 6.55±0.68 to 3.90±0.54 in the COCs group. The bleeding pattern was improved in both Table 1 Baseline characteristics of the 62 participants with adenomyosis-related pain and bleeding
Age (years) 30–40 years N40 years Mean±S.D. BMI Number of abortion Number of normal delivery Number of CS Number of living children Duration from the last pregnancy (months) S.D., standard deviation.
Group I LNG-IUS (n= 31)
Group II COCs (n=31)
No.
No.
%
17 54.8 14 45.2 39.39±4.43 25.9±6.3 1.98±1.26 5.64±1.26 0.42±0.67 4.84±1.29 89.03±37.79
p Value
%
19 61.3 12 38.7 39.16±3.21 26.1±4.4 2.0±1.39 5.03±1.35 0.55±0.92 5.55±0.85 87.74±16.84
.607 .819 .985 .170 .150 .532 .013 .126
treatment arms; in the LNG-IUS, the mean number of bleeding days per month decreased from 9.81±1.82 days prior to recruitment to 2.63±2.13 days after the sixth month of insertion. In the COCs group, the number of bleeding days per month decreased from 9.97±1.52 days to 5.52±1.00 days. Regarding the sanitary pads used per day, in the LNG-IUS group, the mean number of pads decreased from 6.29±0.69 to 2.0±1.44. In the COCs group, the number of pads per day decreased from 6.13±0.85 to 3.58±0.67 after 6 months of treatment. The number of bleeding-free days increased from before to 6 months after treatment (from 19.32±2.65 to 25.39± 1.11 days) in the LNG-IUS group and from 19.10±1.98 days to 23.65±2.57 days in the COCs group (Table 2). Uterine volume decreased in both treatment arms after 6 months of treatment; the decrease was more significant in the LNG-IUS arm (from 10.23±1.06 mL to 7.63±0.49 mL) than in the COCs arm (from 10.42±0.99 mL to 8.32±0.91 mL) (pb.001). Collectively, the mean overall reductions in the VAS, the number of bleeding days per month, the number of sanitary pads per day and the uterine volume in the LNG-IUS group were significantly more than that observed in the COCs group (pb.001) (Table 2). From the Doppler indices, S/D, RI and PI of the uterine arteries and intramyometrial blood vessels increased significantly from baseline values to 6 month values; the increases in the LNG-IUS group were more than that of the COCs group (pb.000) (Table 3).
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
305
Table 2 Effect of LNG-IUS and COCs on pain, uterine bleeding and uterine volume before and 6 months after treatment Outcomes (mean± S.D.)
VAS score (mean±S.D.) Number of bleeding days per month Number of sanitary pads per day Number of bleeding-free days
Group I LNG-IUS group
Group II COC group
Intergroup comparisons p value
Baseline
6 months a
p Value
Baseline
6 months a
p Value
Baseline
6 months a
6.23±0.67 9.81±1.82 6.29± 0.69 19.32±2.65
1.68±1.25 2.63± 2.13 2±1.44 25.39±1.11
b .001⁎ b .001⁎ b .001⁎ .001⁎
6.55±0.68 9.97±1.52 6.13±0.85 19.10±1.98
3.90±0.54 5.25 ± 1 3.58±0.67 23.65±2.57
b .001⁎ b .001⁎ b .001⁎ .000⁎
.064 .441 .415 .792
b.001⁎ b.001⁎ b.001⁎ .000⁎
S.D., standard deviation. ⁎ Statistically significant difference (pb.05). a The data presented at 6 months follow-up are only for those women who completed the follow-up period and did not include those who had treatment failure prior to the end of follow-up or those who were lost during follow-up.
At the start of the study, 7 out of 31 women in the LNG-IUS group and 5 out of 31 women in the COCs group expressed their satisfaction with the different aspects of their lives. After 6 months of treatment, 25 participants in the LNG-IUS arm and 18 in the COCs arm expressed their overall satisfaction.
4. Discussion The present work demonstrates that both COCs and LNG-IUS were effective in the treatment of adenomyosis. The LNG-IUS demonstrated superiority to COCs regarding the outcomes of pain and uterine bleeding profile. Furthermore, we demonstrated that both treatment arms reduced uterine volume and blood flow inside the uterus, a change that could have been the cause of the effect of the treatment. There is great variation in the ultrasound diagnostic accuracy of adenomyosis; we depended on this accuracy for the diagnosis of eligible participants [15]. However, both LNG-IUS and COCs are beneficial in treating pain and bleeding of adenomyotic and nonadenomyotic patients. The suggestion of using LNG-IUS was based on its ability to down-regulate ERs in endometrial tissues and its decidualization with subsequent atrophy of the endometrium [12]. In addition, the LNG-IUS has a pain-relieving effect through the reduction of prostaglandin production within the endometrium and induction of adenomyosis foci atrophy [13]. However, the suggestion of using COCs in the treatment of adenomyosis was secondary to its inhibition of prostaglandins and gonadotropin release, which decrease pain and bleeding [16]. The associated pain (dysmenorrhea and/or chronic pelvic pain) was the primary outcome measure of the present study. The patients in the two groups reported an improvement in pain. The mean overall reduction in the VAS in the LNG-IUS group was significantly more than that observed in the COCs group (pb.001). These findings are in agreement with that from the study of Deng et al. in 2006 [17] who reported that the baseline VAS decreased from 8.4±1.5 to 3.99±3.87 after 1 year of insertion of the LNG-IUS.
However, we achieved a more marked reduction in the VAS after 6 months, most likely because the baseline VAS in our study was lower (6.23) than that in the Deng et al. study (8.41). In 2007, Bragheto and colleagues followed up 29 women after 6 months insertion of the LNG-IUS, and they reported a similar reduction in the VAS pain score between baseline and 6 months after insertion (pb.001). However, six women in their study continued to report a VAS of more than 3 after 6 months of treatment [12]. In 2008, Cho and colleagues reported the effect of LNG-IUS in 47 patients with adenomyosis, and they found that the VAS decreased significantly after 6 months. More importantly, they noted a significant rebound increase in the pain scores at 36 months [13]. In 2010, Zheng and colleagues reported results similar to ours after concurrent treatment with LNG-IUS and GnRH agonist. The VAS was significantly lowered from baseline (6.83±9.3) to 12 months (0.68±0.42) after placement of the LNG-IUS [14]. In 2011, a randomized trial was reported on the treatment of adenomyosis conducted by Bayoglu Tekin and colleagues. They compared a depot GnRH analogue with the LNG-IUS in the treatment of adenomyosis-related chronic pelvic pain. However, the GnRH-a group showed a significant decrease in the VAS compared with the LNG-IUS group at the end of 1 year. However, the GnRH-a group showed menopausal-like symptoms such as hot flashes and vaginal dryness [18]. These symptoms limit the use of this analogue for a long time and decrease patient satisfaction. Menorrhagia associated with adenomyosis was a secondary outcome measure in our RCT. LNG-IUS reduced menstrual blood more effectively than COCs. The first study that reported a reduction in menstrual blood loss after insertion of the LNG-IUS in patients with adenomyosis was by Fedele and colleagues in 1997. They studied the effect of LNG-IUS in 25 women with adenomyosis-associated menorrhagia. They found that all studied women experienced an improvement of bleeding symptoms, with a majority experiencing regular scanty flow. They also reported significant increases in hemoglobin, hematocrit and serum ferritin levels [19].
306
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Table 3 Uterine arteries and intramyometrial blood vessel Doppler indices before and 6 months after treatment with LNG-IUS and COCs Group I LNG-IUS group End point Uterine artery S/D RI PI Intramyometrial S/D RI PI
Group II COCs group
Intergroup comparisons p value
Baseline
6 months a
p Value
Baseline
6 months a
p Value
Baseline
6 months a
4.25±1.59 0.75±0.07 1.71±0.35
12.07±2.78 0.94±0.05 2.76±0.44
b.001⁎ b.001⁎ b.001⁎
3.96±0.76 0.72±0.05 1.62±0.39
6.92±1.27 0.85±0.04 2.31±0.47
b.001⁎ b.001⁎ b.001⁎
.015⁎ .678 .041⁎
b.001⁎ b.001⁎ b.001⁎
3.08±0.83 0.66±0.16 1.02±0.36
8.86±4.72 0.87±0.07 2.33±0.44
b.001⁎ b.001⁎ b.001⁎
3.19±0.84 0.76±0.25 1.12±0.41
3.92±0.97 0.75±0.09 1.62±0.48
.058⁎ .018⁎ .004⁎
.955 .084 .263
b.001⁎ b.001⁎ b.001⁎
⁎ Statistically significant difference (pb.05). a The data presented at 6 months follow-up are only for those women who completed the follow-up period and did not include those who had treatment failure prior to the end of follow-up or those who were lost during follow-up.
In 2003, Maia and colleagues [20] evaluated the effect of LNG-IUS versus no treatment as an adjuvant to endometrial ablation in 95 adenomyotic patients suffering from heavy menstrual bleeding. Our results are in accord with their results. In 2008, Cho et al. [13] evaluated the effect of LNG-IUS in 47 patients with adenomyosis-associated abnormal uterine bleeding. Their results were in agreement with ours. The bleeding dropped dramatically after 6 months of LNG-IUS insertion. In our study, LNG-IUS was effective in decreasing uterine volume more so when compared with COCs after 6 months of treatment. There were 2 previous studies addressed this point; Cho et al. [13] reported a significant decrease in mean uterine volume after 6 months of LNG-IUS insertion (156.85±49.79 mL to 118.64±41.36 mL; pb.001). The reduction in uterine volume was more marked after 24 months (128.84 ± 48.70 mL; pb.001). However, Bragheto et al. observed no significant change in uterine volume (142.6 mL vs. 136.4 mL; p=.2077) between baseline and the 6-month evaluation [12]. The difference may be attributed to interobserver variability. The Doppler study of uterine arteries and intramyometrial blood vessels before and after LNG-IUS and COCs treatment in adenomyosis revealed a significant decrease in uterine arteries and intramyometrial blood flow in adenomyotic patients in both groups. The reduction was more significant in the LNG-IUS group than in the COCs group after 6 months, and these changes were positively correlated with the uterine volume changes. Cho et al. has been the only study that reported an effect of LNG-IUS on uterine artery Doppler of adenomyotic uteri. They reported that the mean pulsatility indices of both uterine arteries increased significantly 12 months after insertion (p=.002 for right; p=.011 for left) [13]. Reduction in uterine volume and the decrease in blood flow to and inside the uterus may be responsible at least partially for the clinical improvement observed as a result of the treatment in our study. The present work had some limitations. First, our diagnostic criteria of 2D TV/US and color Doppler ultrasound for adenomyosis had a reported sensitivity and
specificity of 86.0% and 86.67%, respectively; this was because noninvasive confirmations of adenomyosis were not possible. Moreover, neither treatment option would harm nonadenomyotic patients as long as they were suffering from pain and bleeding and requiring contraception. Second, it was infeasible to blind study participants. Third, the small sample size that was available for the final analysis at 6 months is 62 patients. We preferred to use cyclic rather than a continuous COC regimen because developing amenorrhea (ultimate result of continuous COC use) is not always an acceptable option to women in our culture. Especially in the first few months of use of any contraceptive method, they may be worried about pregnancy and or retention of toxic materials inside their body. Continuous use of COCs may have a beneficial effect over cyclic use with regard to a reduction in pain associated with primary dysmenorrhea [21]. However, this has not been proved in previous trials regarding congestive dysmenorrhea associated with adenomyosis. Moreover, continuous COC use may help more with pain relief; however, that should make comparison with LNG-IUS with regard to menstrual pattern very difficult. Amenorrhea results from continuous COC use, while LNG-IUS users continue to menstruate and or spot within the first 6 months of use.
5. Conclusion Both LNG-IUS and COCs are effective in reducing pain and bleeding associated with adenomyosis after 6 months of use. However, LNG-IUS is superior to COCs for relief of both symptoms. The decrease in uterine volume and the increase in blood flow resistance in both the uterine arteries and intramyometrial blood vessels may be the cause of the treatment effect, which is more pronounced with the use of LNG-IUS.
Conflict of Interest The authors declare that there was no conflict of interest.
O.M. Shaaban et al. / Contraception 92 (2015) 301–307
Acknowledgments This research was funded by a research grant obtained from the Assiut Faculty of Medicine Grant's Office. References [1] Jason M, Dechenne V, Chantraine F, Nisolle M. Adenomyosis. Rev Med Liege 2013;68:160–2. [2] Uysal A, Taner CE, Mun S, Uysal F, Celimli FH. Use of a levonorgestrel-releasing intrauterine device in the treatment of adenomyosis associated heavy menstrual bleeding. J Pak Med Assoc 2013;63:1349–52. [3] Huchon C, Fauconnier A. Adenomyosis. Rev Prat 2014;64:551–2. [4] Shrestha A, Sedai LB. Understanding clinical features of adenomyosis: a case control study. Nepal Med Coll J 2012;14:176–9. [5] Sanghera S, Roberts TE, Barton P, Frew E, Daniels J, Middleton L, et al. Levonorgestrel-releasing intrauterine system vs. usual medical treatment for menorrhagia: an economic evaluation alongside a randomised controlled trial. PLoS One 2014;9:e91891. [6] Beatty MN, Blumenthal PD. The levonorgestrel-releasing intrauterine system: safety, efficacy, and patient acceptability. Ther Clin Risk Manag 2009;5:561–74. [7] Guo SW, Mao X, Ma Q, Liu X. Dysmenorrhea and its severity are associated with increased uterine contractility and overexpression of oxytocin receptor (OTR) in women with symptomatic adenomyosis. Fertil Steril 2013;99:231–40. [8] Shaaban MM, Zakherah MS, El-Nashar SA, Sayed GH. Levonorgestrel-releasing intrauterine system compared to low dose combined oral contraceptive pills for idiopathic menorrhagia: a randomized clinical trial. Contraception 2011;83:48–54. [9] Sakhel K, Abuhamad A. Sonography of adenomyosis. J Ultrasound Med 2012;31:805–8. [10] Champaneria R, Abedin P, Daniels J, Balogun M, Khan KS. Ultrasound scan and magnetic resonance imaging for the diagnosis of adenomyosis: systematic review comparing test accuracy. Acta Obstet Gynecol Scand 2010;89:1374–84.
307
[11] Hanafi M. Ultrasound diagnosis of adenomyosis, leiomyoma, or combined with histopathological correlation. J Hum Reprod Sci 2013;6:189–93. [12] Bragheto AM, Caserta N, Bahamondes L, Petta CA. Effectiveness of the levonorgestrel-releasing intrauterine system in the treatment of adenomyosis diagnosed and monitored by magnetic resonance imaging. Contraception 2007;76:195–9. [13] Cho S, Nam A, Kim H, Chay D, Park K, Cho DJ, et al. Clinical effects of the levonorgestrel-releasing intrauterine device in patients with adenomyosis. Am J Obstet Gynecol 2008;198, http://dx.doi.org/ 10.1016/j.ajog.2007.10.798. [14] Zheng Z, Wang NN, Wang JH, Gan XQ, Zheng QQ, Ke PQ. Effect of levonorgestrel-releasing intrauterine system combined with GnRH analogue for treatment of large adenomyosis. Nan Fang Yi Ke Da Xue Xue Bao 2010;30:541–3. [15] Dueholm M, Lundorf E, Sorensen JS, Ledertoug S, Olesen F, Laursen H. Reproducibility of evaluation of the uterus by transvaginal sonography, hysterosonographic examination, hysteroscopy and magnetic resonance imaging. Hum Reprod 2002;17:195–200. [16] Streuli I, Dubuisson J, Santulli P, de Ziegler D, Batteux F, Chapron C. An update on the pharmacological management of adenomyosis. Expert Opin Pharmacother 2014;15:2347–60. [17] Deng S, Lang JH, Leng JH, Liu ZF, Sun DW, Zhu L. Effects of levonorgestrel-releasing intrauterine system on pain and recurrence associated with endometriosis and adenomyosis. Zhonghua Fu Chan Ke Za Zhi 2006;41:664–8. [18] Bayoglu Tekin Y, Dilbaz B, Altinbas SK, Dilbaz S. Postoperative medical treatment of chronic pelvic pain related to severe endometriosis: levonorgestrel-releasing intrauterine system versus gonadotropinreleasing hormone analogue. Fertil Steril 2011;95:492–6. [19] Fedele L, Bianchi S, Raffaelli R, Portuese A, Dorta M. Treatment of adenomyosis-associated menorrhagia with a levonorgestrel-releasing intrauterine device. Fertil Steril 1997;68:426–9. [20] Maia H, Maltez A, Coelho G, Athayde C, Coutinho EM. Insertion of Mirena after endometrial resection in patients with adenomyosis. J Am Assoc Gynecol Laparosc 2003;10:512–6. [21] Dmitrovic R, Kunselman AR, Legro RS. Continuous compared with cyclic oral contraceptives for the treatment of primary dysmenorrhea: a randomized controlled trial. Obstet Gynecol 2012;119:1143–50.
