- Email: [email protected]
Mifepristone vs. osmotic dilator insertion for cervical preparation prior to surgical abortion at 14–16 weeks: a randomized trial
Contraception 86 (2012) 567 – 571
Original research article
Mifepristone vs. osmotic dilator insertion for cervical preparation prior to surgical abortion at 14–16 weeks: a randomized trial☆,☆☆,★ Lynn Borgatta a,⁎, Danielle Roncari b , Sarita Sonalkar a , Alice Mark c , Melody Y. Hou a , Molly Finneseth a , Olivera Vragovic a b
a Department of Obstetrics and Gynecology, Boston University School of Medicine, Boston, MA 021128, USA Department of Obstetrics and Gynecology, Tufts Medical Center, 800 Washington St, Boston, MA 02111, USA c 5 Perrin Road, Brookline, MA 02445, USA Received 23 March 2012; revised 27 April 2012; accepted 2 May 2012
Abstract Background: Cervical preparation is recommended before second-trimester abortion. We investigated the use of a pharmacologic method of preparation, mifepristone, as compared to osmotic dilators for surgical abortions at 14–16 weeks. Study Design: This was a randomized, parallel-group study with concealed allocation. Women were allocated to receive osmotic dilators or mifepristone 200 mg orally 24 h prior to abortion. The study population was 50 women seeking surgical abortion at 14–16 menstrual weeks in a hospital-based abortion service. The primary outcome was the length of time to perform the procedure; the study had 80% power to detect a difference of more than 3 min in procedure time. Secondary outcomes included cervical dilation, side effects and acceptability. Results: The mean abortion time for the osmotic dilator group was 8.00 min [95% confidence interval (CI) 6.75–11.47], and that for the mifepristone group was 9.87 min (95% CI 8.93–11.36). Side effects of pain were more common in the osmotic dilator group. Conclusion: Mifepristone did not increase the time for abortion by more than the prespecified margin (3 min). Women preferred mifepristone to osmotic dilators. © 2012 Elsevier Inc. All rights reserved. Keywords: Abortion; Second trimester; Mifepristone; Osmotic dilators; Laminaria; Cervical priming
1. Introduction Cervical preparation is recommended before secondtrimester surgical abortion [1,2]. The purpose of the preparation is to soften and open the cervix so that instruments can be safely introduced. Osmotic dilators are one of the most widely used methods of cervical preparation. Dilator insertion can be uncomfortable, and many women experience cramping and pain while the dilators are in place. Dilator insertion also requires a trained provider. To avoid dilator insertion, pharmaceutical alternatives have been studied. One alternative is misoprostol, a prostaglandin that ☆ This study was supported by a grant from the Society of Family Planning Research Fund. ☆☆ The authors have no conflicts of interest. ★ This study was registered with clinicaltrials.gov #NCT00986921. ⁎ Corresponding author. Tel.: +1 617 414 3440; fax: +1 617 414 7300. E-mail address: [email protected] (L. Borgatta).
0010-7824/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.contraception.2012.05.002
a woman takes several hours prior to a planned surgical procedure. However, misoprostol can have side effects including cramping, nausea and pain, and the efficacy for cervical preparation in the second trimester is not well established [2]. Mifepristone, a progesterone receptor modulator, is used in the first trimester and second trimester as part of medical abortion procedures. Compared to regimens that use misoprostol alone, mifepristone increases the efficacy of medical abortion in the first trimester [3] and decreases the induction-to-abortion interval in the second trimester [4]. Mifepristone alone has also been shown to be effective for cervical preparation before first-trimester surgical abortion [5–7]. The purpose of this study is to determine whether mifepristone taken the day before abortion is comparable to osmotic dilators for patients undergoing a surgical termination of pregnancy between 14 and 16 weeks. If mifepristone proves to be effective, it will have the
568
L. Borgatta et al. / Contraception 86 (2012) 567–571
advantages of allowing women to avoid dilator insertion the day prior to abortion, or a wait of several hours for the misoprostol to take effect, while still gaining adequate cervical preparation. The primary hypothesis is that mifepristone is not significantly inferior to osmotic dilators using the measure of total abortion time. We chose to measure procedure time as the primary outcome, as we felt that procedure time was an important facet of procedure acceptability, both to clinicians and to patients. However, recognizing that the time required to perform an abortion is only one aspect of acceptability, we also assessed patient discomfort before, during and after the abortion, and patient and provider opinions about the process.
2. Materials and methods Women aged 18–45 years requesting an induced abortion between 14 and 16 menstrual weeks were eligible for the study. All women had ultrasound dating of the pregnancy prior to study entry. Women were excluded from the study if they had fetal demise, ruptured membranes, spontaneous abortion, active substance abuse or did not speak English or Spanish. The study was approved by the Boston University Medical Center Institutional Review Board. Informed consent was obtained from all participants prior to the start of study procedures by one of the investigators. Study procedures began on the day prior to the abortion. After eligibility was confirmed and informed consent was given, women were allocated to one of two groups: the osmotic dilator group or the mifepristone group. The 1:1 randomization was computer generated, using block sizes varying between 6 and 10. Allocation was carried out using sequentially numbered opaque vials; the vials concealed the group assignment until they were opened. The vials were prepared by the research pharmacy. Once allocation had occurred, treatment was not blinded. Women in the osmotic dilator group had osmotic dilators placed 20–24 h prior to scheduled abortion. The procedure was as follows: women were first given ketorolac 60 mg im or ibuprofen 800 mg orally. The cervix was cleansed with povidone–iodine solution, and the cervix was infiltrated with 10 mL 1% lidocaine. Between three and six dilators were placed in the cervix based on clinician preference. Both laminaria and Dilapan (GEL-MED International spol s.r.o.) dilators were used. Doxycycline 200 mg orally was given after the dilators were inserted. Women were asked to rate the pain of insertion on an 11-point scale. Women in the mifepristone group received mifepristone 200 mg orally with ingestion observed. No antibiotics or other medications were used. Women in both groups were scheduled to return at 20–24 h after treatment for their abortion procedure. They answered a short questionnaire about overnight and morning symptoms prior to the abortion. All procedures were performed by
experienced attending physicians or family planning fellows at Boston Medical Center. For women in the dilator group, the dilators were removed immediately before the procedure or after speculum placement, according to operator preference. For all women, the time of speculum placement was recorded. A cervical block of 20 mL 1% buffered lidocaine with 4 U vasopressin was used for all women. A 14-mm suction cannula was used first. If the 14-mm cannula passed, the abortion was completed with suction and forceps. If the 14-mm cannula did not pass, cervical dilation was measured starting with a 39Fr Pratt dilator, followed by sequentially smaller dilators until a dilator passed easily. The operator could then dilate mechanically as much as deemed appropriate. The time that suction began was recorded. The time of completion of the procedure was defined as the time of speculum removal, which was also recorded. Suction was always used as the first intrauterine procedure; forceps were used at any time after the initial use of suction as needed. After the procedure was completed, women were asked about the amount of discomfort during the procedure and their preferences for cervical preparation. The primary outcome was the time from speculum placement to speculum removal (“procedure time”). Secondary outcomes included the length of time from the initial use of suction to speculum removal (“operative time”) and cervical dilation at the beginning of the procedure. Secondary outcomes of discomfort were assessed with pain scores or by collapsing descriptive outcomes into binary outcomes. An 11-point scale (values from 0 to 10) was used to assess pain at a particular point in time, while descriptive categories were used to assess symptoms during a prolonged period, e.g., “overnight.” Opinions about the method of cervical preparation were assessed with 5-point Likert scales. Physicians were asked to rate the ease of procedure on a 5point scale. The study was structured with a noninferiority design. The primary outcome, procedure time measured in minutes, was defined as inferior if the mean time for the mifepristone group was more than 3 min longer than the time for the osmotic dilator group, using a Student's t test. The 3-min figure was chosen because we felt that 3 min would be noticeable to the operator. The mean procedure time was expected to be 10 min. A one-sided t test with a 95% confidence interval (CI) and 80% power yielded a sample size of 24 women in each group. We planned to enroll 25 women in each group in case there were any withdrawals. Data were analyzed as intent-to-treat using SAS (Version 9, SAS Institute, Inc., Cary, NC, USA). Outcomes related to time were expressed as means and 95% CIs. A multiple regression analysis was performed to evaluate the relationship between the primary outcome and covariates, i.e., parity, gravidity and gestational age at the time of the procedure. Secondary outcomes were evaluated with χ 2 tests. A value of pb.05 was considered to be significant.
L. Borgatta et al. / Contraception 86 (2012) 567–571 Table 1 Characteristics of study participants
3. Results Fifty women were enrolled from October 2009 to March of 2011 (Fig. 1). Characteristics of enrolled women are presented in Table 1. The groups had similar baseline characteristics. All women were randomly allocated and received their assigned treatment. One woman in the osmotic dilator group aborted spontaneously shortly before the planned uterine aspiration. Although we do not have outcome data related to the procedure, her information about symptoms and opinions are included. All other women were analyzed as assigned. All procedures were completed successfully; there were no incomplete procedures or complications. For the women in the osmotic dilator group, between three and six dilators were placed in all women (mean, 4.9). Three women had laminaria only; 22 women had a combination of laminaria and Dilapan. Women who received osmotic dilators reported mean pain scores of 5.2 (95% CI 4.0–6.4) right after insertion. When asked about the evening and night after insertion, 20 (80%) of the women in the osmotic dilator group had moderate or severe pain compared to 9 (36%) women in the mifepristone group (p=.002) (Table 2). Women in the osmotic dilator group were more likely to take pain medication overnight [14 (56%) vs. 3 (15%), pb.001]. Two women in the osmotic dilator group made visits to the emergency department to obtain narcotic pain medication. The incidence of nausea and vomiting overnight was similar: 60% vs. 56%. Four women (16%) in the osmotic dilator group had diarrhea overnight compared to none in the mifepristone group. Overnight bleeding was Screened for eligibility 107 Not eligible – 21 Declined to participate – 24 Other reasons or not asked --12 Enrolled and randomized 50
Allocated to receive mifepristone 25
Allocated to receive osmotic dilators 25
Received mifepristone 25
Received osmotic dilators 25
Analyzed 25
569
Analyzed 24 1 woman aborted before analysis
Fig. 1. Flowchart of participants in trial.
Age⁎ (years) Gravidity N1 1 Parity ≥1 0 Previous abortions ≥1 0 Gestational age⁎ (days) Ethnic group Caucasian African American Hispanic Other Type of provider Attending Fellows
Mifepristone group N=25
Osmotic dilator group N=25
24±5
25±6
17 (68%) 8 (32%)
22 (88%) 3 (12%)
13 (52%) 12 (48%)
19 (76%) 6 (24%)
15 (60%) 10 (40%) 105±4
16 (64%) 9 (36%) 104±4
3 (12%) 15 (60%) 4 (16%) 3 (12%)
7 (28%) 11 (44%) 6 (24%) 1 (1%)
13 (54%) 12 (46%)
13 (54%) 11 (46%)
⁎ Mean and standard deviation.
more common in the osmotic dilator group [16 (64%)] compared to the mifepristone group [1 (4%)]. In the morning before the abortion procedure, women in the osmotic dilator group were more likely to report pain and severe pain than women in the mifepristone group (Table 2). The pattern was similar to that of overnight symptoms. The incidence of nausea was similar between the two groups. The abortion procedure began at a median of 24.0 h (range 21.6–25.9) after osmotic dilator placement and 23.7 h (range 19.2–27.9) after mifepristone ingestion. Three women, all in the mifepristone group, had their procedure with local anesthesia only; the remaining 46 women had intravenous sedation. The procedure times are displayed in Fig. 2. The times are not normally distributed as evidenced by a Kolmogorov– Smirnov value of 0.01. We did a log transformation of the data, and after the log transformation, the Kolmogorov– Smirnov value was 0.45, which is very close to normal distribution. The means, computed with logarithms, were 8.00 min (95% CI 6.75–11.47) for the osmotic dilator group and 9.87 min (95% CI 8.93, 11.36) for the mifepristone group. The corresponding medians were 7 and 10 min, respectively. The means do not vary by more than 3 min, and the CIs for the means overlap. In the osmotic dilator group, after the osmotic dilators were removed, a 14-mm cannula could be passed in 18/24 women. After additional mechanical dilation, if necessary, a 14-mm cannula was used for all women. In the mifepristone group, a 14-mm cannula could be passed in one woman. The median dilation at the start of the procedure was 35Fr in the mifepristone group, which differs from the median dilation of over 39Fr in the osmotic dilator group (pb.0001).
570
L. Borgatta et al. / Contraception 86 (2012) 567–571
Table 2 Secondary outcomes Night prior to abortion Mifepristone N=25 Pain None Mild Moderate Severe Nausea None Mild Moderate Severe and/or vomiting a b
Morning of abortion procedure Osmotic dilator N=25
16 7 2 0
5 7 6 7
10 4 4 6a
12 3 1 8a
p value .001
.49
Mifepristone N=25
Osmotic dilator N=25
16 6 2 1
6 8 6 5
19 2 2 0b
20 3 2 0
p value .02
.93
One value missing. Two values missing.
Mechanical dilation was done to 39Fr or above in 15 women, and a 14 cannula was ultimately used in 17/25 women. Several of the physicians volunteered that, after mifepristone use, mechanical dilation was easy in most women as the cervix was soft. Once dilation, if necessary, had been completed, the operative time from the start of vacuum aspiration to speculum removal was similar. In the osmotic dilator group, the logarithmic mean was 4.90 min (95% CI 3.93–6.11), and in the mifepristone group, it was 5.10 min (95% CI 4.06–6.42). The multivariable analysis suggested that increasing gestational age (p=.007) and nulliparity (p=.05) were associated with increased procedure time. An association was not seen for prior abortion, patient age, operator (attending or fellow) or ethnic group. Physicians rated each procedure after operating. No procedure in either group was rated very difficult or unable to be completed. Eleven (46%) of the procedures in the osmotic dilator group and nine (36%) of the mifepristone
procedures were rated easy or very easy. Two (8%) of the procedures after osmotic dilators were rated difficult compared to six (24%) after mifepristone. All of the physicians felt that both methods of cervical dilation should be available. Women were asked about acceptability of mifepristone compared to osmotic dilators, and opinions about effectiveness. Few women in either group (seven in the osmotic dilator group and four in the mifepristone group) thought that osmotic dilators were more effective. Based on their experience with one method, women were asked, if they had another procedure, whether they would use the same method again or try another method. In the mifepristone group, 24 women (96%) said they would use mifepristone again, and 1 woman (4%) was uncertain. In the osmotic dilator group, 14 women (56%) said they would prefer to try mifepristone instead, 4 (16%) were uncertain, and 7 women (28%) said they would prefer osmotic dilators again.
4. Discussion
Fig. 2. Outcomes of time measurements of abortion procedures. The time required for the abortion procedure, measured from the time the speculum was inserted until it was removed, is shown on the left. The mean times were 8.0 min for the osmotic dilator group and 9.9 min for the mifepristone group. When the time interval from the start of suction (after dilation was completed) to speculum removal is considered (right), the times are 4.9 min and 5.1 min, respectively.
Our experience with mifepristone for cervical preparation was satisfactory. Procedures after mifepristone lasted about 2 min longer than procedures after osmotic dilators, although the CIs overlapped, and the difference did not exceed the prespecified difference of 3 min. The difference in time was due to the frequent need for additional mechanical dilation after mifepristone, as cervical dilation was less after mifepristone. However, the amount of time necessary was short, as the starting dilation was usually 35 to 39Fr, and a limited amount of dilation was used. Although we did not attempt to quantify resistance to dilation, several operators noted that, after mifepristone, the cervix was soft and dilated easily to the desired diameter. Study physicians felt that both methods of cervical preparation should be available. We found that women frequently reported pain with insertion of osmotic dilators. Discomfort continued during the night and the next morning, resulting in increased medication use. The amount of discomfort for women taking
L. Borgatta et al. / Contraception 86 (2012) 567–571
mifepristone was much less. Although nausea may be a side effect of mifepristone, the incidence of nausea and vomiting was the same in both groups. In contrast to using same-day misoprostol, women using mifepristone did have a visit the day before the abortion. However, they did not need to have a pelvic examination for the purpose of inserting dilators, and there was no waiting period on the day of the abortion. There is limited published experience comparing pharmacologic and mechanical cervical preparation. Goldberg et al. [8] compared women at 13–16 weeks of gestation who were randomly assigned to receive laminaria the day before abortion or misoprostol 3 h before abortion. They found that cervical dilation was greater in the osmotic dilator group, although operative time was similar. However, women preferred misoprostol. Carbonell et al. [9] randomly assigned women with pregnancies 14–18 weeks to receive mifepristone or not to receive mifepristone 48 h before abortion. Women were also randomized to receive misoprostol either vaginally or sublingually 1.5–2.5 h before abortion; all women received misoprostol. They found that the women who received mifepristone had greater mean cervical dilation at the start of the procedure compared to women who received misoprostol alone. For women who received mifepristone, they found that 2 of 450 women (0.2%) aborted after mifepristone alone and 3.3% aborted after misoprostol was given but before the surgical abortion was started. For women who did not receive mifepristone, 0.6% aborted after misoprostol. In our small study, none of the women aborted after mifepristone; in fact, the only woman to abort spontaneously was a woman who received osmotic dilators. We used mifepristone for 24 h and did not add misoprostol; both these practices may decrease the rate of expulsion of the pregnancy prior to surgical abortion. A limitation of our study was the lack of blinding, which could have led to bias by both patients and physicians. However, blinding was not possible as the physician could easily tell from the appearance of the cervix whether dilators had been used and the woman was obviously aware as to whether dilators had been placed. Pharmacologic cervical preparation is attractive to women and is less uncomfortable than osmotic dilators. Even when used without misoprostol, at the gestational ages of 14–16 weeks, mifepristone allowed completion of the abortion without undue difficulty. Mifepristone resulted in the need for additional mechanical dilation for many women; however, the procedure times for mifepristone were not
571
inferior to those of osmotic dilators. Using mifepristone requires a 2-day procedure and does not present an advantage over osmotic dilators in this regard. However, women were more comfortable after mifepristone. As women in our study had a marked preference for mifepristone, we believe that this option would be desirable to other women. Acknowledgments We would like to acknowledge Dr. Howard Cabral for his assistance with the statistical analysis. This study was supported by a grant from the Society of Family Planning Research Fund. The Society of Family Planning is a nonprofit organization and is not involved in the actual conduct of the study. References [1] Newmann SJ, Dalve-Endres A, Diedrich JT, Steinauer JE, Meckstroth K, Drey EA. Cervical preparation for second trimester dilation and evacuation. Cochrane Database Systematic Review 2010 Aug 4;8: CD007310. [2] Fox MC, Hayes JL. Society of Family Planning. Cervical preparation for second-trimester surgical abortion prior to 20 weeks of gestation. Contraception 2007;76:486–95. [3] Ngoc NT, Shochet T, Raghavan S, et al. Mifepristone and misoprostol compared with misoprostol alone for second-trimester abortion: a randomized controlled trial. Obstet Gynecol 2011;118:601–8. [4] Borgatta L, Kapp N, Society of Family Planning. Clinical guidelines. Labor induction abortion in the second trimester. Contraception 2011; 84:4–18. [5] Kapp N, Lohr PA, Ngo TD, Hayes JL. Cervical preparation for first trimester surgical abortion. Cochrane Database Syst Rev 2010 Feb 17;2: CD007207. [6] Bokstrom H, Atterfelt P, Alexandersson M, Brannstrom M, Norstrom A. Preoperative cervical softening before first trimester legal abortion by mifepristone and misoprostol: a double-blind, randomized, clinical, biochemical and immunohistochemical study. Contraception 1998;58: 157–63. [7] World Health Organization, Task Force on Post-ovulatory Methods for Fertility Regulation Special Programme of Research, Development and Research Training in Human Reproduction. The use of mifepristone (RU486) for cervical preparation in first trimester pregnancy termination by vacuum aspiration. BJOG 1990;97:260–6. [8] Goldberg AB, Drey EA, Whitaker AK, Kang MS, Meckstroth KR, Darney PD. Misoprostol compared with laminaria before early secondtrimester surgical abortion: a randomized trial. Am J Obstet Gynecol 2005;106:234–41. [9] Carbonell JL, Gallego FG, Llorente MP, et al. Vaginal vs. sublingual misoprostol with mifepristone for cervical priming in second-trimester abortion by dilation and evacuation: a randomized trial. Contraception 2007;75:230–7.
Original research article
Mifepristone vs. osmotic dilator insertion for cervical preparation prior to surgical abortion at 14–16 weeks: a randomized trial☆,☆☆,★ Lynn Borgatta a,⁎, Danielle Roncari b , Sarita Sonalkar a , Alice Mark c , Melody Y. Hou a , Molly Finneseth a , Olivera Vragovic a b
a Department of Obstetrics and Gynecology, Boston University School of Medicine, Boston, MA 021128, USA Department of Obstetrics and Gynecology, Tufts Medical Center, 800 Washington St, Boston, MA 02111, USA c 5 Perrin Road, Brookline, MA 02445, USA Received 23 March 2012; revised 27 April 2012; accepted 2 May 2012
Abstract Background: Cervical preparation is recommended before second-trimester abortion. We investigated the use of a pharmacologic method of preparation, mifepristone, as compared to osmotic dilators for surgical abortions at 14–16 weeks. Study Design: This was a randomized, parallel-group study with concealed allocation. Women were allocated to receive osmotic dilators or mifepristone 200 mg orally 24 h prior to abortion. The study population was 50 women seeking surgical abortion at 14–16 menstrual weeks in a hospital-based abortion service. The primary outcome was the length of time to perform the procedure; the study had 80% power to detect a difference of more than 3 min in procedure time. Secondary outcomes included cervical dilation, side effects and acceptability. Results: The mean abortion time for the osmotic dilator group was 8.00 min [95% confidence interval (CI) 6.75–11.47], and that for the mifepristone group was 9.87 min (95% CI 8.93–11.36). Side effects of pain were more common in the osmotic dilator group. Conclusion: Mifepristone did not increase the time for abortion by more than the prespecified margin (3 min). Women preferred mifepristone to osmotic dilators. © 2012 Elsevier Inc. All rights reserved. Keywords: Abortion; Second trimester; Mifepristone; Osmotic dilators; Laminaria; Cervical priming
1. Introduction Cervical preparation is recommended before secondtrimester surgical abortion [1,2]. The purpose of the preparation is to soften and open the cervix so that instruments can be safely introduced. Osmotic dilators are one of the most widely used methods of cervical preparation. Dilator insertion can be uncomfortable, and many women experience cramping and pain while the dilators are in place. Dilator insertion also requires a trained provider. To avoid dilator insertion, pharmaceutical alternatives have been studied. One alternative is misoprostol, a prostaglandin that ☆ This study was supported by a grant from the Society of Family Planning Research Fund. ☆☆ The authors have no conflicts of interest. ★ This study was registered with clinicaltrials.gov #NCT00986921. ⁎ Corresponding author. Tel.: +1 617 414 3440; fax: +1 617 414 7300. E-mail address: [email protected] (L. Borgatta).
0010-7824/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.contraception.2012.05.002
a woman takes several hours prior to a planned surgical procedure. However, misoprostol can have side effects including cramping, nausea and pain, and the efficacy for cervical preparation in the second trimester is not well established [2]. Mifepristone, a progesterone receptor modulator, is used in the first trimester and second trimester as part of medical abortion procedures. Compared to regimens that use misoprostol alone, mifepristone increases the efficacy of medical abortion in the first trimester [3] and decreases the induction-to-abortion interval in the second trimester [4]. Mifepristone alone has also been shown to be effective for cervical preparation before first-trimester surgical abortion [5–7]. The purpose of this study is to determine whether mifepristone taken the day before abortion is comparable to osmotic dilators for patients undergoing a surgical termination of pregnancy between 14 and 16 weeks. If mifepristone proves to be effective, it will have the
568
L. Borgatta et al. / Contraception 86 (2012) 567–571
advantages of allowing women to avoid dilator insertion the day prior to abortion, or a wait of several hours for the misoprostol to take effect, while still gaining adequate cervical preparation. The primary hypothesis is that mifepristone is not significantly inferior to osmotic dilators using the measure of total abortion time. We chose to measure procedure time as the primary outcome, as we felt that procedure time was an important facet of procedure acceptability, both to clinicians and to patients. However, recognizing that the time required to perform an abortion is only one aspect of acceptability, we also assessed patient discomfort before, during and after the abortion, and patient and provider opinions about the process.
2. Materials and methods Women aged 18–45 years requesting an induced abortion between 14 and 16 menstrual weeks were eligible for the study. All women had ultrasound dating of the pregnancy prior to study entry. Women were excluded from the study if they had fetal demise, ruptured membranes, spontaneous abortion, active substance abuse or did not speak English or Spanish. The study was approved by the Boston University Medical Center Institutional Review Board. Informed consent was obtained from all participants prior to the start of study procedures by one of the investigators. Study procedures began on the day prior to the abortion. After eligibility was confirmed and informed consent was given, women were allocated to one of two groups: the osmotic dilator group or the mifepristone group. The 1:1 randomization was computer generated, using block sizes varying between 6 and 10. Allocation was carried out using sequentially numbered opaque vials; the vials concealed the group assignment until they were opened. The vials were prepared by the research pharmacy. Once allocation had occurred, treatment was not blinded. Women in the osmotic dilator group had osmotic dilators placed 20–24 h prior to scheduled abortion. The procedure was as follows: women were first given ketorolac 60 mg im or ibuprofen 800 mg orally. The cervix was cleansed with povidone–iodine solution, and the cervix was infiltrated with 10 mL 1% lidocaine. Between three and six dilators were placed in the cervix based on clinician preference. Both laminaria and Dilapan (GEL-MED International spol s.r.o.) dilators were used. Doxycycline 200 mg orally was given after the dilators were inserted. Women were asked to rate the pain of insertion on an 11-point scale. Women in the mifepristone group received mifepristone 200 mg orally with ingestion observed. No antibiotics or other medications were used. Women in both groups were scheduled to return at 20–24 h after treatment for their abortion procedure. They answered a short questionnaire about overnight and morning symptoms prior to the abortion. All procedures were performed by
experienced attending physicians or family planning fellows at Boston Medical Center. For women in the dilator group, the dilators were removed immediately before the procedure or after speculum placement, according to operator preference. For all women, the time of speculum placement was recorded. A cervical block of 20 mL 1% buffered lidocaine with 4 U vasopressin was used for all women. A 14-mm suction cannula was used first. If the 14-mm cannula passed, the abortion was completed with suction and forceps. If the 14-mm cannula did not pass, cervical dilation was measured starting with a 39Fr Pratt dilator, followed by sequentially smaller dilators until a dilator passed easily. The operator could then dilate mechanically as much as deemed appropriate. The time that suction began was recorded. The time of completion of the procedure was defined as the time of speculum removal, which was also recorded. Suction was always used as the first intrauterine procedure; forceps were used at any time after the initial use of suction as needed. After the procedure was completed, women were asked about the amount of discomfort during the procedure and their preferences for cervical preparation. The primary outcome was the time from speculum placement to speculum removal (“procedure time”). Secondary outcomes included the length of time from the initial use of suction to speculum removal (“operative time”) and cervical dilation at the beginning of the procedure. Secondary outcomes of discomfort were assessed with pain scores or by collapsing descriptive outcomes into binary outcomes. An 11-point scale (values from 0 to 10) was used to assess pain at a particular point in time, while descriptive categories were used to assess symptoms during a prolonged period, e.g., “overnight.” Opinions about the method of cervical preparation were assessed with 5-point Likert scales. Physicians were asked to rate the ease of procedure on a 5point scale. The study was structured with a noninferiority design. The primary outcome, procedure time measured in minutes, was defined as inferior if the mean time for the mifepristone group was more than 3 min longer than the time for the osmotic dilator group, using a Student's t test. The 3-min figure was chosen because we felt that 3 min would be noticeable to the operator. The mean procedure time was expected to be 10 min. A one-sided t test with a 95% confidence interval (CI) and 80% power yielded a sample size of 24 women in each group. We planned to enroll 25 women in each group in case there were any withdrawals. Data were analyzed as intent-to-treat using SAS (Version 9, SAS Institute, Inc., Cary, NC, USA). Outcomes related to time were expressed as means and 95% CIs. A multiple regression analysis was performed to evaluate the relationship between the primary outcome and covariates, i.e., parity, gravidity and gestational age at the time of the procedure. Secondary outcomes were evaluated with χ 2 tests. A value of pb.05 was considered to be significant.
L. Borgatta et al. / Contraception 86 (2012) 567–571 Table 1 Characteristics of study participants
3. Results Fifty women were enrolled from October 2009 to March of 2011 (Fig. 1). Characteristics of enrolled women are presented in Table 1. The groups had similar baseline characteristics. All women were randomly allocated and received their assigned treatment. One woman in the osmotic dilator group aborted spontaneously shortly before the planned uterine aspiration. Although we do not have outcome data related to the procedure, her information about symptoms and opinions are included. All other women were analyzed as assigned. All procedures were completed successfully; there were no incomplete procedures or complications. For the women in the osmotic dilator group, between three and six dilators were placed in all women (mean, 4.9). Three women had laminaria only; 22 women had a combination of laminaria and Dilapan. Women who received osmotic dilators reported mean pain scores of 5.2 (95% CI 4.0–6.4) right after insertion. When asked about the evening and night after insertion, 20 (80%) of the women in the osmotic dilator group had moderate or severe pain compared to 9 (36%) women in the mifepristone group (p=.002) (Table 2). Women in the osmotic dilator group were more likely to take pain medication overnight [14 (56%) vs. 3 (15%), pb.001]. Two women in the osmotic dilator group made visits to the emergency department to obtain narcotic pain medication. The incidence of nausea and vomiting overnight was similar: 60% vs. 56%. Four women (16%) in the osmotic dilator group had diarrhea overnight compared to none in the mifepristone group. Overnight bleeding was Screened for eligibility 107 Not eligible – 21 Declined to participate – 24 Other reasons or not asked --12 Enrolled and randomized 50
Allocated to receive mifepristone 25
Allocated to receive osmotic dilators 25
Received mifepristone 25
Received osmotic dilators 25
Analyzed 25
569
Analyzed 24 1 woman aborted before analysis
Fig. 1. Flowchart of participants in trial.
Age⁎ (years) Gravidity N1 1 Parity ≥1 0 Previous abortions ≥1 0 Gestational age⁎ (days) Ethnic group Caucasian African American Hispanic Other Type of provider Attending Fellows
Mifepristone group N=25
Osmotic dilator group N=25
24±5
25±6
17 (68%) 8 (32%)
22 (88%) 3 (12%)
13 (52%) 12 (48%)
19 (76%) 6 (24%)
15 (60%) 10 (40%) 105±4
16 (64%) 9 (36%) 104±4
3 (12%) 15 (60%) 4 (16%) 3 (12%)
7 (28%) 11 (44%) 6 (24%) 1 (1%)
13 (54%) 12 (46%)
13 (54%) 11 (46%)
⁎ Mean and standard deviation.
more common in the osmotic dilator group [16 (64%)] compared to the mifepristone group [1 (4%)]. In the morning before the abortion procedure, women in the osmotic dilator group were more likely to report pain and severe pain than women in the mifepristone group (Table 2). The pattern was similar to that of overnight symptoms. The incidence of nausea was similar between the two groups. The abortion procedure began at a median of 24.0 h (range 21.6–25.9) after osmotic dilator placement and 23.7 h (range 19.2–27.9) after mifepristone ingestion. Three women, all in the mifepristone group, had their procedure with local anesthesia only; the remaining 46 women had intravenous sedation. The procedure times are displayed in Fig. 2. The times are not normally distributed as evidenced by a Kolmogorov– Smirnov value of 0.01. We did a log transformation of the data, and after the log transformation, the Kolmogorov– Smirnov value was 0.45, which is very close to normal distribution. The means, computed with logarithms, were 8.00 min (95% CI 6.75–11.47) for the osmotic dilator group and 9.87 min (95% CI 8.93, 11.36) for the mifepristone group. The corresponding medians were 7 and 10 min, respectively. The means do not vary by more than 3 min, and the CIs for the means overlap. In the osmotic dilator group, after the osmotic dilators were removed, a 14-mm cannula could be passed in 18/24 women. After additional mechanical dilation, if necessary, a 14-mm cannula was used for all women. In the mifepristone group, a 14-mm cannula could be passed in one woman. The median dilation at the start of the procedure was 35Fr in the mifepristone group, which differs from the median dilation of over 39Fr in the osmotic dilator group (pb.0001).
570
L. Borgatta et al. / Contraception 86 (2012) 567–571
Table 2 Secondary outcomes Night prior to abortion Mifepristone N=25 Pain None Mild Moderate Severe Nausea None Mild Moderate Severe and/or vomiting a b
Morning of abortion procedure Osmotic dilator N=25
16 7 2 0
5 7 6 7
10 4 4 6a
12 3 1 8a
p value .001
.49
Mifepristone N=25
Osmotic dilator N=25
16 6 2 1
6 8 6 5
19 2 2 0b
20 3 2 0
p value .02
.93
One value missing. Two values missing.
Mechanical dilation was done to 39Fr or above in 15 women, and a 14 cannula was ultimately used in 17/25 women. Several of the physicians volunteered that, after mifepristone use, mechanical dilation was easy in most women as the cervix was soft. Once dilation, if necessary, had been completed, the operative time from the start of vacuum aspiration to speculum removal was similar. In the osmotic dilator group, the logarithmic mean was 4.90 min (95% CI 3.93–6.11), and in the mifepristone group, it was 5.10 min (95% CI 4.06–6.42). The multivariable analysis suggested that increasing gestational age (p=.007) and nulliparity (p=.05) were associated with increased procedure time. An association was not seen for prior abortion, patient age, operator (attending or fellow) or ethnic group. Physicians rated each procedure after operating. No procedure in either group was rated very difficult or unable to be completed. Eleven (46%) of the procedures in the osmotic dilator group and nine (36%) of the mifepristone
procedures were rated easy or very easy. Two (8%) of the procedures after osmotic dilators were rated difficult compared to six (24%) after mifepristone. All of the physicians felt that both methods of cervical dilation should be available. Women were asked about acceptability of mifepristone compared to osmotic dilators, and opinions about effectiveness. Few women in either group (seven in the osmotic dilator group and four in the mifepristone group) thought that osmotic dilators were more effective. Based on their experience with one method, women were asked, if they had another procedure, whether they would use the same method again or try another method. In the mifepristone group, 24 women (96%) said they would use mifepristone again, and 1 woman (4%) was uncertain. In the osmotic dilator group, 14 women (56%) said they would prefer to try mifepristone instead, 4 (16%) were uncertain, and 7 women (28%) said they would prefer osmotic dilators again.
4. Discussion
Fig. 2. Outcomes of time measurements of abortion procedures. The time required for the abortion procedure, measured from the time the speculum was inserted until it was removed, is shown on the left. The mean times were 8.0 min for the osmotic dilator group and 9.9 min for the mifepristone group. When the time interval from the start of suction (after dilation was completed) to speculum removal is considered (right), the times are 4.9 min and 5.1 min, respectively.
Our experience with mifepristone for cervical preparation was satisfactory. Procedures after mifepristone lasted about 2 min longer than procedures after osmotic dilators, although the CIs overlapped, and the difference did not exceed the prespecified difference of 3 min. The difference in time was due to the frequent need for additional mechanical dilation after mifepristone, as cervical dilation was less after mifepristone. However, the amount of time necessary was short, as the starting dilation was usually 35 to 39Fr, and a limited amount of dilation was used. Although we did not attempt to quantify resistance to dilation, several operators noted that, after mifepristone, the cervix was soft and dilated easily to the desired diameter. Study physicians felt that both methods of cervical preparation should be available. We found that women frequently reported pain with insertion of osmotic dilators. Discomfort continued during the night and the next morning, resulting in increased medication use. The amount of discomfort for women taking
L. Borgatta et al. / Contraception 86 (2012) 567–571
mifepristone was much less. Although nausea may be a side effect of mifepristone, the incidence of nausea and vomiting was the same in both groups. In contrast to using same-day misoprostol, women using mifepristone did have a visit the day before the abortion. However, they did not need to have a pelvic examination for the purpose of inserting dilators, and there was no waiting period on the day of the abortion. There is limited published experience comparing pharmacologic and mechanical cervical preparation. Goldberg et al. [8] compared women at 13–16 weeks of gestation who were randomly assigned to receive laminaria the day before abortion or misoprostol 3 h before abortion. They found that cervical dilation was greater in the osmotic dilator group, although operative time was similar. However, women preferred misoprostol. Carbonell et al. [9] randomly assigned women with pregnancies 14–18 weeks to receive mifepristone or not to receive mifepristone 48 h before abortion. Women were also randomized to receive misoprostol either vaginally or sublingually 1.5–2.5 h before abortion; all women received misoprostol. They found that the women who received mifepristone had greater mean cervical dilation at the start of the procedure compared to women who received misoprostol alone. For women who received mifepristone, they found that 2 of 450 women (0.2%) aborted after mifepristone alone and 3.3% aborted after misoprostol was given but before the surgical abortion was started. For women who did not receive mifepristone, 0.6% aborted after misoprostol. In our small study, none of the women aborted after mifepristone; in fact, the only woman to abort spontaneously was a woman who received osmotic dilators. We used mifepristone for 24 h and did not add misoprostol; both these practices may decrease the rate of expulsion of the pregnancy prior to surgical abortion. A limitation of our study was the lack of blinding, which could have led to bias by both patients and physicians. However, blinding was not possible as the physician could easily tell from the appearance of the cervix whether dilators had been used and the woman was obviously aware as to whether dilators had been placed. Pharmacologic cervical preparation is attractive to women and is less uncomfortable than osmotic dilators. Even when used without misoprostol, at the gestational ages of 14–16 weeks, mifepristone allowed completion of the abortion without undue difficulty. Mifepristone resulted in the need for additional mechanical dilation for many women; however, the procedure times for mifepristone were not
571
inferior to those of osmotic dilators. Using mifepristone requires a 2-day procedure and does not present an advantage over osmotic dilators in this regard. However, women were more comfortable after mifepristone. As women in our study had a marked preference for mifepristone, we believe that this option would be desirable to other women. Acknowledgments We would like to acknowledge Dr. Howard Cabral for his assistance with the statistical analysis. This study was supported by a grant from the Society of Family Planning Research Fund. The Society of Family Planning is a nonprofit organization and is not involved in the actual conduct of the study. References [1] Newmann SJ, Dalve-Endres A, Diedrich JT, Steinauer JE, Meckstroth K, Drey EA. Cervical preparation for second trimester dilation and evacuation. Cochrane Database Systematic Review 2010 Aug 4;8: CD007310. [2] Fox MC, Hayes JL. Society of Family Planning. Cervical preparation for second-trimester surgical abortion prior to 20 weeks of gestation. Contraception 2007;76:486–95. [3] Ngoc NT, Shochet T, Raghavan S, et al. Mifepristone and misoprostol compared with misoprostol alone for second-trimester abortion: a randomized controlled trial. Obstet Gynecol 2011;118:601–8. [4] Borgatta L, Kapp N, Society of Family Planning. Clinical guidelines. Labor induction abortion in the second trimester. Contraception 2011; 84:4–18. [5] Kapp N, Lohr PA, Ngo TD, Hayes JL. Cervical preparation for first trimester surgical abortion. Cochrane Database Syst Rev 2010 Feb 17;2: CD007207. [6] Bokstrom H, Atterfelt P, Alexandersson M, Brannstrom M, Norstrom A. Preoperative cervical softening before first trimester legal abortion by mifepristone and misoprostol: a double-blind, randomized, clinical, biochemical and immunohistochemical study. Contraception 1998;58: 157–63. [7] World Health Organization, Task Force on Post-ovulatory Methods for Fertility Regulation Special Programme of Research, Development and Research Training in Human Reproduction. The use of mifepristone (RU486) for cervical preparation in first trimester pregnancy termination by vacuum aspiration. BJOG 1990;97:260–6. [8] Goldberg AB, Drey EA, Whitaker AK, Kang MS, Meckstroth KR, Darney PD. Misoprostol compared with laminaria before early secondtrimester surgical abortion: a randomized trial. Am J Obstet Gynecol 2005;106:234–41. [9] Carbonell JL, Gallego FG, Llorente MP, et al. Vaginal vs. sublingual misoprostol with mifepristone for cervical priming in second-trimester abortion by dilation and evacuation: a randomized trial. Contraception 2007;75:230–7.