- Email: [email protected]
Ovulation resumption after medical abortion with mifepristone and misoprostol
Contraception 84 (2011) 230 – 233
Original research article
Ovulation resumption after medical abortion with mifepristone and misoprostol☆ Courtney A. Schreibera,⁎, Stephanie Sobera , Sarah Ratcliffea , Mitchell D. Creininb a
University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA b University of Pittsburgh, Pittsburgh, PA, USA Received 15 October 2010; revised 14 January 2011; accepted 15 January 2011
Abstract Background: As an antiprogestin, mifepristone may have an impact on the return to ovulation in a manner that is not only attributable to its abortifacient activity. Our aim was to measure the time-to-ovulation in women who received mifepristone 200 mg orally and misoprostol 800 mcg vaginally for abortion up to 63 days of gestation. Study Design: This planned substudy was part of a multicenter randomized trial of mifepristone 200 mg followed immediately or 24 h later by misoprostol 800 mcg vaginally. Women who had successful expulsion of the gestational sac based on ultrasound examination 1 week after mifepristone treatment were enrolled. All subjects used nonhormonal contraception until study completion. Baseline serum progesterone (P) levels were drawn on day 8±1 after mifepristone administration and then twice weekly until the P level was N3 ng/mL, consistent with ovulation. The mean time-to-ovulation was calculated using interval censored regression to address the censoring due to participant discontinuation. Results: Fourteen (52%) of 27 enrolled women completed the substudy. The longest period of time that a subject who did not complete the study was followed was 29 days. Ovulation occurred 20.6±5.1 (range 8–36) days after mifepristone administration. Time-to-ovulation was not affected by participant age, gestational age, study arm, body mass index or presence or absence of human chorionic gonadotropin. Conclusions: Return to ovulation following medical abortion with mifepristone and misoprostol occurs on average 3 weeks postabortion. Mifepristone 200 mg does not appear to have a lasting effect on ovarian function. Our results should be contextualized by the small sample size, although this is one of the larger studies on return to ovulation after abortion. © 2011 Elsevier Inc. All rights reserved. Keywords: Ovulation; Abortion; Ovarian function; Pregnancy termination; Progesterone
1. Introduction The timing of fertility return after induced abortion is important because women should be aware of when to initiate contraception or anticipate conception. It is also of biologic interest and provides us with insight into the restoration of the
☆ This study was conducted with the support of an anonymous foundation. Authors' disclosures: Dr. Schreiber currently receives research funding from an anonymous foundation, the Society of Family Planning, Medicines360 and the National Institutes of Child Health & Human Development Contraception & Reproductive Health (CRH) Branch and has served as a consultant for Bayer Pharmaceuticals. ⁎ Corresponding author. Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA. Tel.: +1 215 615 6531; fax: +1 215 615 6853. E-mail address: [email protected] (C.A. Schreiber).
0010-7824/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2011.01.013
nonpregnant physiology after the ending of a pregnancy. A surgical abortion under 12 weeks of gestation seems to have little or no lasting effect on ovarian function; ovulation occurs in 6–21 days [1–4]. In studies that have evaluated postabortal ovulatory function over 4 weeks' time, 83% showed hormonal evidence of ovulation [5], and 91% showed a luteal rise in either serum progesterone (P) or urinary pregnanediol compatible with postovulatory values [6]. Mifepristone is an antiprogestin that binds competitively to the P receptor and blocks progesterone's biologic activity. This effect is present in pregnant and nonpregnant women. Because there are P receptors in the ovary, mifepristone has direct effects on follicular development and ovulation [7–9]. Accordingly, mifepristone may have an impact on the return to ovulation independent of its abortifacient action. We aimed to establish the timing of ovulation resumption after medical abortion with mifepristone and misoprostol.
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
2. Materials and methods We hypothesized that women who undergo medical abortion with mifepristone and misoprostol will resume ovulation within 6 weeks after the abortion as measured by serum P levels. To test our hypothesis, we conducted an observational trial as a planned substudy of a prospective, multicenter, randomized clinical trial that evaluated the impact of changing the timing of misoprostol administration after mifepristone use [10]. The substudy protocol was performed only at the University of Pittsburgh site and was approved by the University of Pittsburgh Institutional Review Board. In the main study protocol, women used mifepristone 200 mg orally and misoprostol 800 mcg vaginally. Subjects were randomized to use misoprostol within 15 min of swallowing mifepristone (“same time”) or 23–25 h later (“24 hours”). Subjects were scheduled to return for a follow-up examination between 7 and 9 days after using mifepristone. Women 18–35 years old who returned within that time frame and had expelled the gestational sac based on transvaginal sonography were offered enrollment into the substudy. All participants underwent informed consent specifically for the substudy and agreed to use nonhormonal contraception
231
until study completion. Serum P (Siemens ACS chemiluminescence assay) levels were performed at this first postabortal visit and then twice weekly until the P level was ≥3 ng/mL, consistent with ovulation. High-sensitivity urinary human chorionic gonadotropin (hCG) (SureVue 25 mIU hCG; Fisher Healthcare, Houston, TX, USA) was performed weekly at the research visits. Women were notified by telephone after each laboratory value was obtained. Once the P level was ≥3 ng/mL, the subject was informed of her postovulatory status, and complete contraceptive counseling and method initiation were conducted. If the P value was b3 ng/mL, she was asked to return in 2–4 days for another blood draw. Women who could not be contacted via three phone attempts and one certified letter were considered lost to follow-up. We planned to enroll a convenience sample of 40 women, with 20 women from the group who received the randomized trial study medications at the same time and 20 from the group who took the medications 24 h apart. We calculated the mean, maximum, minimum and standard deviation of P levels over the duration of the study. The mean time-toovulation was calculated using interval censored regression to address the censoring due to participant discontinuation.
Table 1 Baseline demographic and clinical characteristics of the sample; mean ± SD or n (%) reported Timing of misoprostol administration after mifepristone
Age, years Race White Black Asian Married Years of school completed Some high school High school Some college College Graduate or professional school Other BMI Gravidity Parity Prior abortion Prior miscarriage Prior cesarean section Gestational age at time of abortion, days Primary contraceptive used during 3 months prior to enrollment Male condom Vaginal ring Contraceptive patch OCP Other None
Total (N=27)
Same time (n=12)
24 h (n=15)
25.0 ± 5.4
26.2 ± 6.3
25.7 ± 5.8
8 (66.7) 4 (33.3) 0 (0.0) 2 (16.7)
6 (40.0) 7 (46.7) 2 (13.3) 3 (20.0)
14 (51.9) 11 (40.7) 2 (7.4) 5 (18.5)
0 (0.0) 6 (50.0) 4 (33.3) 1 (8.3) 1 (8.3) 0 (0.0) 28.9 ± 7.8 3±3 1±1 5 (41.7) 3 (25.0) 0 (0.0) 50.8 ± 5.6
1 (6.7) 6 (40.0) 2 (13.3) 2 (13.3) 2 (13.3) 2 (13.3) 28.8 ± 6.9 3±2 1±1 7 (46.7) 3 (20.0) 3 (20.0) 52.7 ± 7.9
1 (3.7) 12 (44.4) 6 (22.2) 3 (11.1) 3 (11.1) 2 (7.4) 28.9 ± 7.2 3±2 1±1 12 (44.4) 6 (22.2) 3 (11.1) 51.9 ± 6.9
p valuea
.75 .43
N.99 .75
.96 .86 .86 N.99 N.99 .23 .51 N.99
8 (66.7) 0 (0.0) 1 (8.3) 1 (8.3) 0 (0.0) 2 (16.7)
9 (60.0) 1 (6.7) 0 (0.0) 2 (13.3) 1 (6.7) 2 (13.3)
17 (63.0) 1 (3.7) 1 (3.7) 3 (11.1) 1 (3.7) 4 (14.8)
OCP, oral contraceptive pill. a Comparison of groups based on timing of misoprostol administration using Mann–Whitney U test or Fisher's Exact Test.
232
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
Comparison of our findings to the time-to-ovulation following prostaglandin-induced abortion (24 days) [4] and surgical abortion (29 days) [4] was performed using the Student's t test.
Twenty-seven of a possible 321 women agreed to participate. Twelve women had received the medication for medical abortion at the same time, and 15 had received the medications 24 h apart. Baseline demographics and contraceptive histories are shown in Table 1. Fourteen (52%) women completed the substudy. Ovulation occurred on average 20.6±5.1 (range 8–36) days after mifepristone administration and did not vary by timing of misoprostol administration (Table 2). Of those 13 women who started but did not complete the study, the length of participation ranged from 8 to 29 days. Thus, the longest follow-up for a subject in whom we did not witness ovulation was 29 days. Time-to-ovulation was not affected by participant age (p=.48), race (p=.61), gestational age (p=.37), study arm (p=.55), body mass index (BMI) (p=.93) or gravidity (p=.22). The time-to-ovulation and time-to-resolution of hCG for all participants are shown in Fig. 1. The mean timeto-ovulation was not statistically significantly different from that reported following prostaglandin-induced abortion (24 days, p=.19). However, our patients ovulated approximately 8 days earlier than the mean 29 days reported after surgical abortion (p=.001).
4. Discussion Mifepristone is an antiprogestin that binds competitively to the P receptor and blocks progesterone's biologic activity. Because there are P receptors in the ovary, mifepristone has direct effects on follicular development and ovulation [7–9]. Studies in nonpregnant women have shown that repeated daily doses of mifepristone delays ovulation regardless of the time in Table 2 Comparison of outcomes by timing of misoprostol administration Timing of misoprostol Entire administration after mifepristone sample (N=27) Same time (n=12) 24 h (n=15) Ovulated, n (%) Days to ovulationb Negative hCG obtained, n (%) Days to negative hCGb
p valuea
7 (58.3) 18.9 ± 3.7 5 (41.7)
7 (46.7) 21.9 ± 3.5 6 (40.0)
14 (51.9) .70 20.6 ± 2.6 .55 11 (40.7) N.99
11.2 ± 1.0
13.4 ± 1.0
12.5 ± 0.8
.12
a Comparison of groups via Fisher's Exact Test or interval censored regression. b Mean values ± SD estimated using interval censored regression.
0.5
Proportion
3. Results
0.6
0.4 0.3 0.2 0.1 0 0
5
10
15
20
25
30
35
40
Days Ovulated
Negative Urine hCG
Fig. 1. Proportion of women with ovulation and resolution of hCG measurements after medical abortion with mifepristone and misoprostol in all women enrolled. The longest follow-up for a subject in whom we did not witness a negative hCG was 29 days.
the cycle at which it is administered, and that resumption of follicle recruitment and ovulation occur predictably (10–15 days) after treatment withdrawal [8,11]. Because mifepristone affects both the pregnancy and ovary, we sought to determine if ovulation resumption is different for women who undergo abortion with mifepristone than women who undergo an early termination of pregnancy by other means. In this study, the first to measure time-to-ovulation after use of mifepristone and misoprostol to induce abortion, ovulation occurred on average 3 weeks postabortion. The earliest that ovulation occurred in our study was as soon as 8 days after the abortion was induced. Cameron and Baird [4] compared time-to-ovulation in women undergoing surgical abortion to those having an abortion induced with a prostaglandin analogue (gemeprost) alone. The mean return to ovulation was 29 days (16–37) in the surgical cohort and 24 days (16–32) in the gemeprost cohort (p=not significant). Time-to-ovulation in our study was shorter than that in either of these groups, and the 8-day difference between the mean time of ovulation resumption in patients whose abortion was induced with mifepristone and misoprostol and surgical abortion was statistically significant (p=.001). A single study has evaluated the trend of P levels following the administration of mifepristone and misoprostol 36–48 h apart for medical abortion, noting that serum P levels fall rapidly after misoprostol administration [6]. However, P levels initially increased after the administration of mifepristone. By 14 days postabortion, all 34 subjects had near-zero P levels. That study [6] did not follow women beyond 14 days, which, it appears, is required to observe the initiation of the next ovulatory cycle in most women. The difference in ovulation resumption that we observed between our patients and the earlier cohort of surgical abortion patients [4] may simply reflect the large range in ovulation resumption among women after the end of an early
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
pregnancy, the fact that our participants were of an earlier gestation than the surgical abortion patients or the biological effects of mifepristone on the ovary. Alternatively, these differences may be present because the two studies were performed at different times with different patient populations. Further study in this area is required to better understand the reasons for this observed difference. Our study has a few weaknesses. Because we only measured serum progesterone every 2–4 days, the accrued data are limited to defining when ovulation had not occurred and when it had already occurred: we did not directly observe ovulation. It is also possible that the earliest ovulation occurred before 8 days but was missed because our first measurement took place at approximately 1 week after mifepristone use. The use of interval censored regression enabled us to make use of all the available information without assuming that the first day that we knew ovulation had occurred was the actual day of ovulation, and helped to reduce bias due to measurement error. Another limitation was our inability to meet our sample size. Our requirement that women refrain from hormonal contraceptive use during the study period was an unavoidable barrier to recruitment. Recruitment was also likely hampered by competing substudies and the discomfort of venipuncture. Nonetheless, this is one of the largest studies to examine ovulation resumption after abortion, with most prior studies including under 20 women [1–3,5] and the largest study reporting outcomes on 29 women [4]. The fact that our findings are similar to those seen following medical abortion with prostaglandin alone lends credibility to our results. Ultimately, larger trials would be needed to provide a broader understanding of the potentially wide variability in return to ovulation following medical abortion, and the differences between ovulation resumption after medical and surgical abortion. In our study, we found that mifepristone 200 mg to induce abortion does not result in delayed ovulation in the following cycle. Therefore, prompt contraceptive administration is warranted for these patients. We also found that, among women with a completed medical abortion, the resolution of urinary hCG is highly variable. Other studies have shown that the majority of high-sensitivity urine pregnancy tests
233
remain positive 2 weeks after medical abortion [12] and that, at 30 days after treatment, 25% of women will still have detectable hCG in the urine [13]. Our data support that the fall of hCG and the beginning of the next ovulatory cycle are independent occurrences. Thus, women and clinicians need to be advised that fertility may return before biochemical resolution of the prior pregnancy. References [1] Marrs RP, Kletzky OA, Howard WF, Mishell DR. Disappearance of human chorionic gonadotropin and resumption of ovulation following abortion. Am J Obstet Gynecol 1979;135:731–6. [2] Lahteenmaki P. The disappearance of hCG and return of pituitary function after abortion. Clin Endocrinol 1978;9:101–12. [3] Lahteenmaki P, Luukkainen T. Return of ovarian function after abortion. Clin Endocrinol 1978;8:123–32. [4] Cameron IT, Baird DT. The return to ovulation following early abortion: a comparison between vacuum aspiration and prostaglandin. Acta Endocrinol-Cop 1988;118:161–7. [5] Donnet ML, Howie M, Cooper W, Lewis M. Return of ovarian function following spontaneous abortion. Clin Endocrinol 1990;33:13–20. [6] Honkanen H, Ranta S, Ylikorkala O, Heikinheimo O. The kinetics of serum hCG and progesterone in response to oral and vaginal administration of misoprostol during medical termination of pregnancy. Hum Reprod 2002;17:2315–9. [7] Curry T, Nothnick W. Mifepristone and ovarian function. Clin Obstet Gynecol 1996;39:486–97. [8] Spitz IM, Croxatto HB, Lahteenmaki P, Heikinheimo O, Bardin CW. Effect of mifepristone on inhibition of ovulation and induction of luteolysis. Hum Reprod 1994;1:69–76. [9] Batista MC, Nieman LK, Cartledge TP, et al. The antiprogestin RU486 delays the midcycle gonadotropin surge and ovulation in gonadatropinreleasing hormone-induced cycles. Fertil Steril 1994;1:28–34. [10] Creinin MD, Schreiber CA, Bednarek P, Lintu H, Wagner MS, Meyn LA. Medical Abortion at the Same Time (MAST) Study Trial Group. Mifepristone and misoprostol administered simultaneously versus 24 hours apart for abortion: a randomized controlled trial. Obstet Gynecol 2007;109:885–94. [11] Luukkainen T, Heikinheimo O, Haukkammaa M, Lahteenmaki P. Inhibition of folliculogenesis and ovulation by the antiprogesterone RU 486. Fertil Steril 1988;49:961–3. [12] Godfrey EM, Anderson A, Fielding SL, Meyn L, Creinin MD. Clinical utility of urine pregnancy assays to determine medical abortion outcome is limited. Contraception 2007;75:378–82. [13] Perriera LK, Reeves MF, Chen BA, Hohmann HL, Hayes J, Creinin MD. Feasibility of telephone follow-up after medical abortion. Contraception 2010;81:143–9.
Original research article
Ovulation resumption after medical abortion with mifepristone and misoprostol☆ Courtney A. Schreibera,⁎, Stephanie Sobera , Sarah Ratcliffea , Mitchell D. Creininb a
University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA b University of Pittsburgh, Pittsburgh, PA, USA Received 15 October 2010; revised 14 January 2011; accepted 15 January 2011
Abstract Background: As an antiprogestin, mifepristone may have an impact on the return to ovulation in a manner that is not only attributable to its abortifacient activity. Our aim was to measure the time-to-ovulation in women who received mifepristone 200 mg orally and misoprostol 800 mcg vaginally for abortion up to 63 days of gestation. Study Design: This planned substudy was part of a multicenter randomized trial of mifepristone 200 mg followed immediately or 24 h later by misoprostol 800 mcg vaginally. Women who had successful expulsion of the gestational sac based on ultrasound examination 1 week after mifepristone treatment were enrolled. All subjects used nonhormonal contraception until study completion. Baseline serum progesterone (P) levels were drawn on day 8±1 after mifepristone administration and then twice weekly until the P level was N3 ng/mL, consistent with ovulation. The mean time-to-ovulation was calculated using interval censored regression to address the censoring due to participant discontinuation. Results: Fourteen (52%) of 27 enrolled women completed the substudy. The longest period of time that a subject who did not complete the study was followed was 29 days. Ovulation occurred 20.6±5.1 (range 8–36) days after mifepristone administration. Time-to-ovulation was not affected by participant age, gestational age, study arm, body mass index or presence or absence of human chorionic gonadotropin. Conclusions: Return to ovulation following medical abortion with mifepristone and misoprostol occurs on average 3 weeks postabortion. Mifepristone 200 mg does not appear to have a lasting effect on ovarian function. Our results should be contextualized by the small sample size, although this is one of the larger studies on return to ovulation after abortion. © 2011 Elsevier Inc. All rights reserved. Keywords: Ovulation; Abortion; Ovarian function; Pregnancy termination; Progesterone
1. Introduction The timing of fertility return after induced abortion is important because women should be aware of when to initiate contraception or anticipate conception. It is also of biologic interest and provides us with insight into the restoration of the
☆ This study was conducted with the support of an anonymous foundation. Authors' disclosures: Dr. Schreiber currently receives research funding from an anonymous foundation, the Society of Family Planning, Medicines360 and the National Institutes of Child Health & Human Development Contraception & Reproductive Health (CRH) Branch and has served as a consultant for Bayer Pharmaceuticals. ⁎ Corresponding author. Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, PA 19104, USA. Tel.: +1 215 615 6531; fax: +1 215 615 6853. E-mail address: [email protected] (C.A. Schreiber).
0010-7824/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2011.01.013
nonpregnant physiology after the ending of a pregnancy. A surgical abortion under 12 weeks of gestation seems to have little or no lasting effect on ovarian function; ovulation occurs in 6–21 days [1–4]. In studies that have evaluated postabortal ovulatory function over 4 weeks' time, 83% showed hormonal evidence of ovulation [5], and 91% showed a luteal rise in either serum progesterone (P) or urinary pregnanediol compatible with postovulatory values [6]. Mifepristone is an antiprogestin that binds competitively to the P receptor and blocks progesterone's biologic activity. This effect is present in pregnant and nonpregnant women. Because there are P receptors in the ovary, mifepristone has direct effects on follicular development and ovulation [7–9]. Accordingly, mifepristone may have an impact on the return to ovulation independent of its abortifacient action. We aimed to establish the timing of ovulation resumption after medical abortion with mifepristone and misoprostol.
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
2. Materials and methods We hypothesized that women who undergo medical abortion with mifepristone and misoprostol will resume ovulation within 6 weeks after the abortion as measured by serum P levels. To test our hypothesis, we conducted an observational trial as a planned substudy of a prospective, multicenter, randomized clinical trial that evaluated the impact of changing the timing of misoprostol administration after mifepristone use [10]. The substudy protocol was performed only at the University of Pittsburgh site and was approved by the University of Pittsburgh Institutional Review Board. In the main study protocol, women used mifepristone 200 mg orally and misoprostol 800 mcg vaginally. Subjects were randomized to use misoprostol within 15 min of swallowing mifepristone (“same time”) or 23–25 h later (“24 hours”). Subjects were scheduled to return for a follow-up examination between 7 and 9 days after using mifepristone. Women 18–35 years old who returned within that time frame and had expelled the gestational sac based on transvaginal sonography were offered enrollment into the substudy. All participants underwent informed consent specifically for the substudy and agreed to use nonhormonal contraception
231
until study completion. Serum P (Siemens ACS chemiluminescence assay) levels were performed at this first postabortal visit and then twice weekly until the P level was ≥3 ng/mL, consistent with ovulation. High-sensitivity urinary human chorionic gonadotropin (hCG) (SureVue 25 mIU hCG; Fisher Healthcare, Houston, TX, USA) was performed weekly at the research visits. Women were notified by telephone after each laboratory value was obtained. Once the P level was ≥3 ng/mL, the subject was informed of her postovulatory status, and complete contraceptive counseling and method initiation were conducted. If the P value was b3 ng/mL, she was asked to return in 2–4 days for another blood draw. Women who could not be contacted via three phone attempts and one certified letter were considered lost to follow-up. We planned to enroll a convenience sample of 40 women, with 20 women from the group who received the randomized trial study medications at the same time and 20 from the group who took the medications 24 h apart. We calculated the mean, maximum, minimum and standard deviation of P levels over the duration of the study. The mean time-toovulation was calculated using interval censored regression to address the censoring due to participant discontinuation.
Table 1 Baseline demographic and clinical characteristics of the sample; mean ± SD or n (%) reported Timing of misoprostol administration after mifepristone
Age, years Race White Black Asian Married Years of school completed Some high school High school Some college College Graduate or professional school Other BMI Gravidity Parity Prior abortion Prior miscarriage Prior cesarean section Gestational age at time of abortion, days Primary contraceptive used during 3 months prior to enrollment Male condom Vaginal ring Contraceptive patch OCP Other None
Total (N=27)
Same time (n=12)
24 h (n=15)
25.0 ± 5.4
26.2 ± 6.3
25.7 ± 5.8
8 (66.7) 4 (33.3) 0 (0.0) 2 (16.7)
6 (40.0) 7 (46.7) 2 (13.3) 3 (20.0)
14 (51.9) 11 (40.7) 2 (7.4) 5 (18.5)
0 (0.0) 6 (50.0) 4 (33.3) 1 (8.3) 1 (8.3) 0 (0.0) 28.9 ± 7.8 3±3 1±1 5 (41.7) 3 (25.0) 0 (0.0) 50.8 ± 5.6
1 (6.7) 6 (40.0) 2 (13.3) 2 (13.3) 2 (13.3) 2 (13.3) 28.8 ± 6.9 3±2 1±1 7 (46.7) 3 (20.0) 3 (20.0) 52.7 ± 7.9
1 (3.7) 12 (44.4) 6 (22.2) 3 (11.1) 3 (11.1) 2 (7.4) 28.9 ± 7.2 3±2 1±1 12 (44.4) 6 (22.2) 3 (11.1) 51.9 ± 6.9
p valuea
.75 .43
N.99 .75
.96 .86 .86 N.99 N.99 .23 .51 N.99
8 (66.7) 0 (0.0) 1 (8.3) 1 (8.3) 0 (0.0) 2 (16.7)
9 (60.0) 1 (6.7) 0 (0.0) 2 (13.3) 1 (6.7) 2 (13.3)
17 (63.0) 1 (3.7) 1 (3.7) 3 (11.1) 1 (3.7) 4 (14.8)
OCP, oral contraceptive pill. a Comparison of groups based on timing of misoprostol administration using Mann–Whitney U test or Fisher's Exact Test.
232
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
Comparison of our findings to the time-to-ovulation following prostaglandin-induced abortion (24 days) [4] and surgical abortion (29 days) [4] was performed using the Student's t test.
Twenty-seven of a possible 321 women agreed to participate. Twelve women had received the medication for medical abortion at the same time, and 15 had received the medications 24 h apart. Baseline demographics and contraceptive histories are shown in Table 1. Fourteen (52%) women completed the substudy. Ovulation occurred on average 20.6±5.1 (range 8–36) days after mifepristone administration and did not vary by timing of misoprostol administration (Table 2). Of those 13 women who started but did not complete the study, the length of participation ranged from 8 to 29 days. Thus, the longest follow-up for a subject in whom we did not witness ovulation was 29 days. Time-to-ovulation was not affected by participant age (p=.48), race (p=.61), gestational age (p=.37), study arm (p=.55), body mass index (BMI) (p=.93) or gravidity (p=.22). The time-to-ovulation and time-to-resolution of hCG for all participants are shown in Fig. 1. The mean timeto-ovulation was not statistically significantly different from that reported following prostaglandin-induced abortion (24 days, p=.19). However, our patients ovulated approximately 8 days earlier than the mean 29 days reported after surgical abortion (p=.001).
4. Discussion Mifepristone is an antiprogestin that binds competitively to the P receptor and blocks progesterone's biologic activity. Because there are P receptors in the ovary, mifepristone has direct effects on follicular development and ovulation [7–9]. Studies in nonpregnant women have shown that repeated daily doses of mifepristone delays ovulation regardless of the time in Table 2 Comparison of outcomes by timing of misoprostol administration Timing of misoprostol Entire administration after mifepristone sample (N=27) Same time (n=12) 24 h (n=15) Ovulated, n (%) Days to ovulationb Negative hCG obtained, n (%) Days to negative hCGb
p valuea
7 (58.3) 18.9 ± 3.7 5 (41.7)
7 (46.7) 21.9 ± 3.5 6 (40.0)
14 (51.9) .70 20.6 ± 2.6 .55 11 (40.7) N.99
11.2 ± 1.0
13.4 ± 1.0
12.5 ± 0.8
.12
a Comparison of groups via Fisher's Exact Test or interval censored regression. b Mean values ± SD estimated using interval censored regression.
0.5
Proportion
3. Results
0.6
0.4 0.3 0.2 0.1 0 0
5
10
15
20
25
30
35
40
Days Ovulated
Negative Urine hCG
Fig. 1. Proportion of women with ovulation and resolution of hCG measurements after medical abortion with mifepristone and misoprostol in all women enrolled. The longest follow-up for a subject in whom we did not witness a negative hCG was 29 days.
the cycle at which it is administered, and that resumption of follicle recruitment and ovulation occur predictably (10–15 days) after treatment withdrawal [8,11]. Because mifepristone affects both the pregnancy and ovary, we sought to determine if ovulation resumption is different for women who undergo abortion with mifepristone than women who undergo an early termination of pregnancy by other means. In this study, the first to measure time-to-ovulation after use of mifepristone and misoprostol to induce abortion, ovulation occurred on average 3 weeks postabortion. The earliest that ovulation occurred in our study was as soon as 8 days after the abortion was induced. Cameron and Baird [4] compared time-to-ovulation in women undergoing surgical abortion to those having an abortion induced with a prostaglandin analogue (gemeprost) alone. The mean return to ovulation was 29 days (16–37) in the surgical cohort and 24 days (16–32) in the gemeprost cohort (p=not significant). Time-to-ovulation in our study was shorter than that in either of these groups, and the 8-day difference between the mean time of ovulation resumption in patients whose abortion was induced with mifepristone and misoprostol and surgical abortion was statistically significant (p=.001). A single study has evaluated the trend of P levels following the administration of mifepristone and misoprostol 36–48 h apart for medical abortion, noting that serum P levels fall rapidly after misoprostol administration [6]. However, P levels initially increased after the administration of mifepristone. By 14 days postabortion, all 34 subjects had near-zero P levels. That study [6] did not follow women beyond 14 days, which, it appears, is required to observe the initiation of the next ovulatory cycle in most women. The difference in ovulation resumption that we observed between our patients and the earlier cohort of surgical abortion patients [4] may simply reflect the large range in ovulation resumption among women after the end of an early
C.A. Schreiber et al. / Contraception 84 (2011) 230–233
pregnancy, the fact that our participants were of an earlier gestation than the surgical abortion patients or the biological effects of mifepristone on the ovary. Alternatively, these differences may be present because the two studies were performed at different times with different patient populations. Further study in this area is required to better understand the reasons for this observed difference. Our study has a few weaknesses. Because we only measured serum progesterone every 2–4 days, the accrued data are limited to defining when ovulation had not occurred and when it had already occurred: we did not directly observe ovulation. It is also possible that the earliest ovulation occurred before 8 days but was missed because our first measurement took place at approximately 1 week after mifepristone use. The use of interval censored regression enabled us to make use of all the available information without assuming that the first day that we knew ovulation had occurred was the actual day of ovulation, and helped to reduce bias due to measurement error. Another limitation was our inability to meet our sample size. Our requirement that women refrain from hormonal contraceptive use during the study period was an unavoidable barrier to recruitment. Recruitment was also likely hampered by competing substudies and the discomfort of venipuncture. Nonetheless, this is one of the largest studies to examine ovulation resumption after abortion, with most prior studies including under 20 women [1–3,5] and the largest study reporting outcomes on 29 women [4]. The fact that our findings are similar to those seen following medical abortion with prostaglandin alone lends credibility to our results. Ultimately, larger trials would be needed to provide a broader understanding of the potentially wide variability in return to ovulation following medical abortion, and the differences between ovulation resumption after medical and surgical abortion. In our study, we found that mifepristone 200 mg to induce abortion does not result in delayed ovulation in the following cycle. Therefore, prompt contraceptive administration is warranted for these patients. We also found that, among women with a completed medical abortion, the resolution of urinary hCG is highly variable. Other studies have shown that the majority of high-sensitivity urine pregnancy tests
233
remain positive 2 weeks after medical abortion [12] and that, at 30 days after treatment, 25% of women will still have detectable hCG in the urine [13]. Our data support that the fall of hCG and the beginning of the next ovulatory cycle are independent occurrences. Thus, women and clinicians need to be advised that fertility may return before biochemical resolution of the prior pregnancy. References [1] Marrs RP, Kletzky OA, Howard WF, Mishell DR. Disappearance of human chorionic gonadotropin and resumption of ovulation following abortion. Am J Obstet Gynecol 1979;135:731–6. [2] Lahteenmaki P. The disappearance of hCG and return of pituitary function after abortion. Clin Endocrinol 1978;9:101–12. [3] Lahteenmaki P, Luukkainen T. Return of ovarian function after abortion. Clin Endocrinol 1978;8:123–32. [4] Cameron IT, Baird DT. The return to ovulation following early abortion: a comparison between vacuum aspiration and prostaglandin. Acta Endocrinol-Cop 1988;118:161–7. [5] Donnet ML, Howie M, Cooper W, Lewis M. Return of ovarian function following spontaneous abortion. Clin Endocrinol 1990;33:13–20. [6] Honkanen H, Ranta S, Ylikorkala O, Heikinheimo O. The kinetics of serum hCG and progesterone in response to oral and vaginal administration of misoprostol during medical termination of pregnancy. Hum Reprod 2002;17:2315–9. [7] Curry T, Nothnick W. Mifepristone and ovarian function. Clin Obstet Gynecol 1996;39:486–97. [8] Spitz IM, Croxatto HB, Lahteenmaki P, Heikinheimo O, Bardin CW. Effect of mifepristone on inhibition of ovulation and induction of luteolysis. Hum Reprod 1994;1:69–76. [9] Batista MC, Nieman LK, Cartledge TP, et al. The antiprogestin RU486 delays the midcycle gonadotropin surge and ovulation in gonadatropinreleasing hormone-induced cycles. Fertil Steril 1994;1:28–34. [10] Creinin MD, Schreiber CA, Bednarek P, Lintu H, Wagner MS, Meyn LA. Medical Abortion at the Same Time (MAST) Study Trial Group. Mifepristone and misoprostol administered simultaneously versus 24 hours apart for abortion: a randomized controlled trial. Obstet Gynecol 2007;109:885–94. [11] Luukkainen T, Heikinheimo O, Haukkammaa M, Lahteenmaki P. Inhibition of folliculogenesis and ovulation by the antiprogesterone RU 486. Fertil Steril 1988;49:961–3. [12] Godfrey EM, Anderson A, Fielding SL, Meyn L, Creinin MD. Clinical utility of urine pregnancy assays to determine medical abortion outcome is limited. Contraception 2007;75:378–82. [13] Perriera LK, Reeves MF, Chen BA, Hohmann HL, Hayes J, Creinin MD. Feasibility of telephone follow-up after medical abortion. Contraception 2010;81:143–9.