Hum. Reprod. Advance Access originally published online on December 3, 2008
Human Reproduction 2009 24(2):315-319; doi:10.1093/humrep/den426
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mifepristone-induced abortion and placental complications in subsequent pregnancy
1 Department of Reproductive Epidemiology and Social Science, National Population and Family Planning Key Laboratory of Contraceptive Drugs and Devices, Shanghai Institute of Planned Parenthood Research, Shanghai 200032, China 2 Department of Preventive Medicine and Public Health, Creighton University School of Medicine, Nebraska 68178 USA 3 Department of Reproductive Epidemiology, Chengdu Donghua Reproductive Health Research Institute, Sichuan 610041, China 4 Department of Reproductive Epidemiology and Social Science, National Research Institute for Family Planning, Beijing 100081, China
5 Correspondence address. 2140 Xietu Road, Shanghai 200032, China. Tel: +86-21-64034614; Fax: +86-21-64043701; E-mail: yuanwei{at}sippr.stc.sh.cn
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
BACKGROUND: The aim of the study was to explore the effect of first-trimester mifepristone-induced abortion (MA) on placental complications in subsequent pregnancy.
METHODS: Two cohorts of nulliparous pregnant women were recruited in China during early pregnancy, one with a history of one MA and the other with no abortion (NA). Women were followed up until delivery.
RESULTS: The incidence proportions of abruptio placenta, placenta previa, placenta accreta and retained placenta in the MA group (4673) and NA group (4690) were, respectively, 0.5 and 0.3, 0.8 and 0.9, 0.5 and 0.5, and 0.7 and 0.8% (all differences non-significant). After adjustment for center, age, education, occupation, residence, income, BMI and type of delivery, the incidence rates of placenta previa, accreta and retained placenta in the MA and NA groups showed no significant differences. The risk of abruptio placenta in women with a MA was nearly double that of women with no abortion, although this apparent increased risk was not statistically significant. Furthermore, this increased risk of abruptio placenta was found only in those with a gestational age >6 weeks at abortion (aOR: 2.46; 95% CI: 1.00–6.04), a curettage after abortion (aOR: 3.00; 95% CI: 1.25–7.20) or a longer inter-pregnancy interval (P-value for trend: 0.022).
CONCLUSIONS: Mifepristone-induced abortion itself is not associated with placental complications in subsequent pregnancy, but other factors related to medical abortion—such as a gestational age >6 weeks at abortion, a curettage after abortion, and a longer interpregnancy interval—may increase the risk of abruptio placenta.
Key words: mifepristone/placental complication/subsequent pregnancy
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Ever since 1988, when mifepristone (RU486) was first registered for use in first-trimester medical abortion in France and China (Creinin, 2000
; Wu, 2000), it has been introduced and used widely to terminate unwanted pregnancies all over the world, including in numerous European countries, the USA and China. When used alone, mifepristone has a 65–95% effectiveness rate, which improves to 95% when used in combination with a prostaglandin (Healy, 1990). Mifepristone has provided a safe alternative to surgical termination of pregnancy and the immediate side effects of mifepristone-induced abortion have been well-documented (Avrech et al., 1991; Donaldson et al., 1994; Say et al., 2002; Rørbye et al., 2004; Hamoda and Flett, 2005). However, only limited information has been available on the long-term effects of early pregnancy abortion induced by mifepristone. In China, past studies indicated that 65.1% of women experiencing a mifepristone-induced abortion were nulliparous, most of whom would undergo pregnancy and delivery later in life (Yan et al., 2004). Although recent studies in China and Denmark have not found increased risks of preterm delivery, low birthweight, stillbirth, neonatal death or malformation in pregnancy subsequent to a previous mifepristone-induced abortion (Chen et al., 2004; Virk et al., 2007), other effects of mifepristone-induced abortion on the subsequent pregnancy have remained an important public health concern.
Literature has indicated that placental complications are associated with a prolonged third stage of labor, post-partum hemorrhage and uterus injury (Thomas et al., 1994; Zhou et al., 1999). The effects of surgical abortion on placental complications of subsequent pregnancies have occasionally been reported (Lopes et al., 1991; Taylor et al., 1993; Zhou et al., 2001; Thorp et al., 2003), and most likely a similar process of endocrine interruptions (reproductive hormones) and endometrium changes occurs in both surgical and mifepristone-induced abortions. We therefore investigated, in two cohorts of nulliparous pregnant women with histories of one mifepristone-induced abortion and no abortion, whether a mifepristone-induced early abortion in nulliparous women affects the placental complications of subsequent wanted pregnancies.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Subjects and data collection
With the approval of the institutional Ethical Review Committee, this study was conducted in a number of clinics in Beijing, Shanghai and Chengdu between June 1998 and March 2001. Nulliparas who were 20–34 years old and who visited clinics for antenatal care before 16 completed weeks' gestation were eligible for the study if they had a history either of one first-trimester mifepristone-induced abortion or no abortion, and were willing and able to give informed consent. Women with a history of surgical or spontaneous abortion, severe heart, liver, kidney or lung disease were excluded. Women with no previous abortion (NA) were frequency-matched proportionately, in 5-year bands of age (20–24, 25–29 and 30–34 years), to women with one previous mifepristone abortion (MA).
At enrollment, information was collected by trained interviewers through a structured questionnaire that included demographic characteristics, lifestyle, reproductive history, medical history, current pregnancy status and, for women who had undergone one abortion, details on the previous induced abortion. Physical and other necessary examinations were conducted and recorded by obstetricians and/or midwives.
Women were followed up at 28–30 weeks of gestation, at delivery and 4–6 weeks after delivery, respectively. At the second follow-up, information was collected on the course and type of delivery, the duration of labor, and especially on placental complications.
Abortion regimens
In China, the commonly used regimens for mifepristone-induced abortion for the period in which the women's abortions had occurred were: (i) 200 mg administered orally in a single dose; (ii) 50 mg taken twice daily for 2 days; (iii) 150 mg administered in one dose; (iv) one 50-mg dose and one 25-mg dose taken daily for 2 days and (v) three 25-mg doses taken daily for 2 days (Wu, 2000). Forty-eight hours after the first dose of mifepristone, either prostaglandin F2
(1 mg), was given, usually vaginally, or misoprostol (0.6 mg), was given, usually orally.
Statistical analysis
The main outcome variables included the following four placental complications: abruptio placenta was defined as the detachment of the placenta from its decidual seat before delivery of the fetus; placenta previa was defined as the location of the placenta over or near the internal os of the cervix, but was not excluded from the group of abruptio placenta; placenta accreta was defined as the abnormal invasion of the placenta in the musculature of the lower uterine; retained placenta was defined as the failure to expel all fetal membranes within 3 h of delivery (Benirschke and Kaufmann, 1990).
Supervisors checked questionnaires for completeness and consistency, and then data were coded and double-entered with Epi info 6.04 at each center. Finally, data were integrated, logically checked and analyzed with SAS 9.1 at the Shanghai center.
We compared placental complications in labor in the MA group with those in the NA group. Study center and age were included in the adjusted Multivariate Logistic regression model. If data on any of the covariates differed between the MA and NA groups, the covariate was considered a potential confounder and was adjusted in the model. Subjects were further divided into subgroups, according to certain factors related to the previous induced abortion, and the incidences of placental complications of different subgroups were compared through stratified chi-square analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
A total of 9856 pregnant women were enrolled in the study, corresponding to a participation rate of 95.6%. In this analysis, 493 records were excluded due to loss to follow-up (51), spontaneous abortion (251), surgical abortion (50), medical abortion (14), fetal death (28), stillbirth (48) and twins pregnancy (51). The results were, therefore, based on 9363 singleton live birth records, including 4673 in the MA group and 4690 in the NA group.
Background characteristics were comparable between the MA and NA groups except for education, residence and income. Mean ages (SD) of the MA and NA groups were 25.58 (2.39) and 25.51 (2.50) years, respectively. Women with no abortion history had lower incomes than women with a previous MA (P < 0.001), and more likely resided in rural areas (P = 0.015) (Table I). Tobacco smoking, alcohol drinking, occupational exposure to toxicants, histories of severe diseases, the use of hormonal contraceptives and reproductive tract infections during pregnancy were rare and were not significantly different between the MA and NA groups.
|
The overall incidence rates of placental complications were 2.0% (92/4673) and 2.0% (92/4690), respectively, in the MA and NA groups (P = 0.98). The risks of sub-types of placental complications are presented in Table II. After adjustment for center, age, education, occupation, residence, income, BMI and type of delivery, the incidence rates of placenta previa, accreta and retained placenta in the MA and NA groups showed no significant differences. However, the risk of abruptio placenta in women with a MA (22/4651) was nearly double that of women with no abortion (12/4678) (aOR: 1.88, 95% CI: 0.93–3.82), although this increased risk was not statistically significant.
|
Factors related to previous MA were examined to explore whether they were associated with extra risks of placental complications, as compared with the NA group (Table III). As with the overall analysis, the risks of placenta previa, placenta accreta and retained placenta were not different from those in the NA group, regardless of the women's age at abortion, hospital where abortion took place, inter-pregnancy interval, gestational age at previous abortion and curettage after abortion (data not shown). It was interesting, however, that, as compared with the NA group, increased risks of abruptio placenta were found in those MA subgroups with a gestational age >6 weeks at abortion (aOR: 2.46, 95% CI: 1.00–6.04), a subsequent curettage (aOR: 3.00, 95% CI: 1.25–7.20) or a longer inter-pregnancy interval (P-value for trend: 0.022).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Studies on the short-term side effects of mifepristone have addressed its safety (Donaldson et al., 1994
; Westfall et al., 1998). Common side-effects of mifepristone are shivering, vomitting, diarrhea, abdominal pain, headache and/or dizziness (Mahajan and London, 1997); allergic or hemorrhagic shock, arrhythmia and convulsion occasionally occur (Zou et al., 2004). However, few studies have investigated the long-term side effects of mifepristone-induced abortion. Virk et al. (2007) recently reported that medical abortion did not increase the risk of spontaneous abortion, ectopic pregnancy, preterm birth or low birthweight in their study. These results are similar to our previously published data (Chen et al., 2004). In this present study, however, while risks of placental complications in the MA group were similar to those in the NA group, we found that a gestational age >6 weeks at previous MA, a curettage after MA and a longer inter-pregnancy interval significantly increased the risk of abruptio placenta in the subsequent pregnancy.
Possible effects of surgical abortion on the placenta have been previously reported. Taylor et al. (1993) found a strong association between a previous spontaneous or induced abortion and the subsequent development of placenta previa. Zhou et al. (2001) found a higher risk of retained placenta in women who had had one abortion, but only when followed by a short pregnancy interval, and this association was weak. Lopes et al. (1991) has recently found an increased incidence of retained placenta only in a subgroup of women with three or more previous induced abortions. These ambiguous results may be due to different control groups, selection bias and differential or non-differential misclassification of the exposure. However, few studies on mifepristone are available to corroborate our results.
Mifepristone is an 11 beta-dimethyl-amino-phenyl derivative of norethindrone with a high affinity for progesterone and glucocorticoid receptors. Mifepristone can effectively block the binding of progesterone to the corresponding receptor in the placenta, resulting in the termination of pregnancy (Mahajan and London, 1997). The use of mifepristone as an inducer of abortion is associated with both endometrial hemorrhage and extracellular matrix (ECM) degradation. Such processes reflect reduced perivascular decidual cell hemostatic and increased ECM-degrading protease activity (Papp et al., 2000). Injury to the endometrium caused by mifepristone might induce the onset of the self-repair process. However, the use of mifepristone at a gestational age of >6 weeks or curettage after medical abortion might cause more serious injury to the endometrium. If the degree of injury exceeded the capacity for self-repair, there may be an irreparable long-term effect. Abruptio placenta may be one manifestation of such injury.
Our results appear to contradict the usual expectation that a shorter interval between abortion and subsequent pregnancy increases the risk of abruptio placenta (Zhou et al., 2001). One plausible explanation is a prolonged inter-pregnancy interval may result from sexually transmitted infections (STIs) or other negative health conditions, which may also give rise to placental complications. However, our current data did not permit the examination of this hypothesis.
Placental complications have been reported as associated with a prolonged third stage of labor, post-partum hemorrhage and uterus injury (Thomas et al., 1994; Zhou et al., 1999). If curettage after MA increases the risk of placental complications, gynecologists should, based on individual conditions, seriously assess the necessity of curettage after MA. This is an especially important and practical issue in China because our data shows that curettage after medical abortion has been as high as 25.4%, much higher than that in other countries (Borgatta et al., 2001). Because of varying protocols and different waiting periods for medical abortions, a uniform scheme for defining failure and analyzing efficacy has never been fully articulated or adopted (Trussell et al., 1999). This is an important detail, for it suggests that some patients may be undergoing an unnecessary curettage even after a successful expulsion.
The study attained an almost complete follow-up. The independence of interviewers from investigators has in part reduced differential misclassification. In China, induced abortion is a legal and acceptable medical practice and has no conflict with the traditional culture. Therefore, it is less likely that women would under- or over-report their abortion history. Although most variables in the questionnaire were explicitly defined, non-differential misclassification may exist. However, we would not expect this to magnify the odds ratios presented. Although most potential confounders were collected and adjusted for by multivariate analyses, residual confounding could not be totally ruled out.
The subjects of this study were all nulliparas, and women in the MA group had undergone only one abortion, so that it is not possible to comment on the influence of two or more mifepristone-induced abortions. The current comparison of women with a previous mifepristone-induced abortion and women with no previous abortion may at least partly imply a comparison of women of gravidities 2 and 1, since gravidity itself can to some extent influence outcome of pregnancy. However, the potential confounding effects of gravidity on outcome of pregnancy may bias the results toward a decreased risk in pregnancy outcome in women with a previous mifepristone-induced abortion, given that the influence of gravidity on pregnancy is similar to that of parity, and that primigravidae are at higher risk during pregnancy (Bai et al., 2002). Although with the use of a surgical abortion group would enable comparisons of equivalent gravidity between groups of surgical and medical abortions, the potential adverse effects of surgical abortion on placenta complications make the comparison not a flawless one.
Our data did not provide evidence that a previous mifepristone-induced abortion was itself associated with placental complications in subsequent pregnancy. However, other factors related to medical abortion—such as a gestational age >6 weeks at abortion, a curettage after abortion and a longer interpregnancy interval—may increase the risk of abruptio placenta. Still, the limited capabilities of sub-analysis with regard to testing statistical significance clearly require that these results be corroborated in further studies.
|
Authors' contribution |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Q.-X.Z.: data analysis and interpretation, drafting manuscript, final approval. E.-S.G.: study design, data analysis, final approval.A.-M.C.H, L.L. and Y.-M.C.: study design, data collection, final approval. W.Y.: data analysis and interpretation, revising manuscript, final approval.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Financial and technical support for this study was provided by the Task Force on Surveillance and Evaluation in Reproductive Health of the Special Programme of Research, Development, and Research Training in Human Reproduction, World Health Organization (grant WHO97823).
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAuthors' contributionFundingReferences |
|---|
Avrech OM, Golan A, Weinraub Z, Bukovsky I, Caspi E. Mifepristone (RU486) alone or in combination with a prostaglandin analogue for termination of early pregnancy: a review. Fertil Steril (1991) 56:385–393.[Web of Science][Medline]
Bai J, Wong FW, Bauman A, Mohsin M. Parity and pregnancy outcomes. Am J Obstet Gynecol (2002) 186:274–278.[CrossRef][Web of Science][Medline]
Benirschke K, Kaufmann P. Pathology of the Human Placenta (1990) 4th edn. Berlin: Springer-Verlag.
Borgatta L, Burnhill MS, Tyson J, Leonhardt KK, Hausknecht RU, Haskell S. Early medical abortion with methotrexate and misoprostol. Obstet Gynecol (2001) 97:11–16.[Medline]
Chen A, Yuan W, Meirik O, Wang X, Wu S, Zhou L, Luo L, Gao E, Cheng Y. Mifepristone-induced early abortion and outcome of subsequent wanted pregnancy. Am J Epidemiol (2004) 160:110–117.
Creinin MD. Medical abortion regimens: historical context and overview. Am J Obstet Gynecol (2000) 183:S3–S9.[CrossRef][Medline]
Donaldson K, Briggs J, McMaster D. RU 486: an alternative to surgical abortion. J Obstet Gynecol Neonatal Nurs (1994) 23:555–559.[Medline]
Hamoda H, Flett GM. Medical termination of pregnancy in the early first trimester. J Fam Plann Reprod Health Care (2005) 31:10–14.[Medline]
Healy DL. Progesterone receptor antagonists and prostaglandins in human fertility regulation: a clinical review. Reprod Fertil Dev (1990) 2:477–490.[Medline]
Lopes A, King PA, Duthie SJ, To WK, Ma HK. The impact of multiple induced abortions on the outcome of subsequent pregnancy. Aust N Z J Obstet Gynaecol (1991) 31:41–43.[Medline]
Mahajan DK, London SN. Mifepristone (RU486): a review. Fertil Steril (1997) 68:967–976.[CrossRef][Web of Science][Medline]
Papp C, Schatz F, Krikun G, Hausknecht V, Lockwood CJ. Biological mechanisms underlying the clinical effects of mifepristone (RU 486) on the endometrium. Early Pregnancy (2000) 4:230–239.[Medline]
Rørbye C, Nørgaard M, Nilas L. Medical versus surgical abortion efficacy, complications and leave of absence compared in a partly randomized study. Contraception (2004) 70:393–399.[Medline]
Say L, Kulier R, Gülmezoglu M, Campana A. Medical versus surgical methods for first trimester termination of pregnancy. Cochrane Database Syst Rev (2002) CD003037.
Taylor VM, Kramer MD, Vaughan TL, Peacock S. Placental previa in relation to induced and spontaneous abortion: a population-based study. Obstet Gynecol (1993) 82:88–91.[Medline]
Thomas AG, Alvarez M, Friedman F Jr, Brodman ML, Kim J, Lockwood C. The effect of placenta previa on blood loss in second-trimester pregnancy termination. Obstet Gynecol (1994) 84:58–60.[Medline]
Thorp JM Jr, Hartmann KE, Shadigian E. Long-term physical and psychological health consequences of induced abortion: review of the evidence. Obstet Gynecol Surv (2003) 58:67–79.[CrossRef][Web of Science][Medline]
Trussell J, Ellertson C. Estimating the efficacy of medical abortion. Contraception (1999) 60:119–135.[Medline]
Virk J, Zhang J, Olsen J. Medical abortion and the risk of subsequent adverse pregnancy outcomes. N Engl J Med (2007) 357:648–653.
Westfall JM, O'Brien-Gonzales A, Barley G. Update on early medical and surgical abortion. J Womens Health (1998) 7:991–995.[Medline]
Wu S. Medical abortion in China. J Am Med Womens Assoc (2000) 55:S197–S199. 204.
Yan M, Shen J, Su SQ, Wang YY. The correlation between pregnancy and delivery history and abortion effect in women with medical abortion. Zhongguo Fu You Bao Jian (2004) 19:70–71. (Chinese).
Zhou W, Gao E, Che Y, Olsen J. Induced abortion and duration of third stage labour in a subsequent pregnancy. J Obstet Gynaecol (1999) 19:349–354.[Medline]
Zhou W, Nielsen GL, Larsen H, Olsen J. Induced abortion and placenta complications in the subsequent pregnancy. Acta Obstet Gynecol Scand (2001) 80:1115–1120.[CrossRef][Medline]
Zou Y, Li YP, Lei ZW, Lu L, Jiang S, Li Q. Side effect of mifepristone in combination with misoprostol for medical abortion. Zhonghua Fu Chan Ke Za Zhi (2004) 39:39–42. (Chinese).[Medline]
Submitted on July 10, 2008; resubmitted on October 17, 2008; accepted on October 29, 2008.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||