ACOG PRACTICE BULLETIN CLINICAL MANAGEMENT GUIDELINES FOR OBSTETRICIAN–GYNECOLOGISTS NUMBER 67, OCTOBER 2005 (Replaces Practice Bulletin Number 26, April 2001) Medical Management of Abortion This Practice Bulletin was developed by the ACOG Com- Over the past two decades,medical methods of abortion have been developed mittee on Practice Bulletins— throughout the world and are now used in the United States. Medical abortion, Gynecology with the assistance which involves the use of medications to induce an abortion rather than a sur- of Mitchell D. Creinin, MD. gical abortion, is an option for women who wish to terminate a pregnancy up The information is designed to to 63 days of gestation (calculated from the first day of the last menstrual peri- aid practitioners in making od). Medical abortions currently account for 6% of all abortions in the United decisions about appropriate obstetric and gynecologic care. States (1). The purpose of this document is to present evidence of the effective- These guidelines should not be ness, benefits, and risks of medical abortion and provide a framework for the construed as dictating an exclu- evaluation and counseling of women who are considering medical abortion. sive course of treatment or pro- cedure. Variations in practice may be warranted based on the Background needs of the individual patient, Medications Currently Used in Medical Abortion resources, and limitations unique to the institution or type Mifepristone of practice. Mifepristone (RU-486),a derivative of norethindrone,binds to the progesterone receptor with an affinity greater than progesterone but does not activate the receptor, thereby acting as an antiprogestin (2). Mifepristone’s known actions on a pregnant uterus include necrotizing the decidua,softening the cervix,and increasing both uterine contractility and prostaglandin sensitivity (3,4). Human studies have suggested that uterine contractility does not increase until 24–36 hours after mifepristone administration (4). At this point,the myometrium is five times more sensitive to the stimulatory effects of exogenous prostaglandins (4). Administration of mifepristone followed by a prostaglandin analogue,usu- ally misoprostol, is the most commonly used medical abortion regimen throughout the world. As a progesterone receptor antagonist,mifepristone also has several other potential medical applications, including emergency contra- ception, cervical ripening for labor induction, and treatment of conditions VOL. 106, NO. 4, OCTOBER 2005 OBSTETRICS & GYNECOLOGY 871 such as symptomatic leiomyomata uteri, endometriosis, nation do not confirm expulsion,ultrasonography is per- Cushing’s syndrome,breast cancer,and glaucoma. formed. If a gestational sac is seen,aspiration is typical- ly performed. Misoprostol Efficacy with this regimen is approximately 92% in Misoprostol is an inexpensive prostaglandin analogue in women with pregnancies up to 49 days of gestation (11, a tablet form that is stable at room temperature. 12). Complete abortion rates are higher with earlier ges- Misoprostol is used clinically for prevention of gastric tations:approximately 96–98% for pregnancies up to 42 ulcers in individuals taking antiinflammatory drugs on a days of gestation (13,14),91–95% from 43 to 49 days of long-term basis, for abortion, and for labor induction. gestation (13,14),and less than 85% beyond 49 days of Pharmacokinetic evaluation of oral and vaginal adminis- gestation (11,13,14). tration of misoprostol demonstrates that oral misoprostol Alternative Regimens is absorbed more rapidly, resulting in a higher peak serum level (5, 6), but vaginal administration results in Other evidence-based medical abortion regimens have greater uterine contractility. Recent evaluations of sublin- been developed in an effort to reduce side effects and to gual administration show higher peak serum concentra- make medical abortion less expensive, safer, and more tions (7, 8), which may result in more unnecessary side rapid. Regimens using 200-mg doses of mifepristone effects (7). Further study of buccal administration may be orally have efficacy rates comparable to the FDA- warranted because its pharmacokinetic profile appears to approved regimen (12, 15) at one third of the cost. be similar to vaginal administration (7). Additionally, increasing the misoprostol dose to 800 µg and administering the medication vaginally decreases the Other Agents time to expulsion (16), results in fewer side effects (16, Methotrexate is used less often today for medical abor- 17), and improves complete abortion rates when com- tion because of the greater availability of mifepristone. pared with oral administration of a 400-µg dose of miso- Methotrexate blocks dihydrofolate reductase,an enzyme prostol (16, 18–20). Multiple large studies in the United involved in producing thymidine during DNA synthesis. States have demonstrated that a patient can safely and Methotrexate exerts its action primarily on the cytotro- effectively self-administer the misoprostol (orally or phoblast rather than the developing embryo. Methotrex- vaginally) in her home (18,19,21–27). ate has been used for more than 40 years to treat Investigations also have demonstrated the flexibility neoplastic diseases, rheumatoid arthritis, and psoriasis; in timing between the two medications. One study other medical applications include treatment of systemic demonstrated that 800 µg of misoprostol may be admin- lupus erythematosus, dermatomyositis, severe asthma, istered either 24,48,or 72 hours after 200 mg of mifepri- Crohn’s disease,and extrauterine pregnancy. stone with equal efficacy in pregnancies up to 56 days of Tamoxifen has been used in combination with miso- gestation (23); a follow-up study using a regimen with a prostol in some studies of early abortion. However, ran- 24-hour interval between medications yielded similar domized trials have demonstrated no benefit of using a results in pregnancies up to 63 days of gestation (24). tamoxifen–misoprostol regimen compared with a metho- Moreover, the results of a randomized multicenter study trexate–misoprostol regimen (9) or misoprostol alone indicated that 800 µg of misoprostol administered vagi- (10). nally 6–8 hours after 200 mg of mifepristone resulted in significantly fewer side effects (and no decrease in effi- Mifepristone Regimens cacy) than regimens using a 24-hour interval (27). Compared with the FDA-approved regimen, mife- Protocol Approved by the U.S. Food and pristone–misoprostol regimens using mifepristone, 200 Drug Administration mg orally,and misoprostol,800 µg vaginally,are associ- Mifepristone regimens vary according to dosage,timing, ated with a decreased rate of continuing pregnancies, and route of administration (see Table 1). The U.S. Food decreased time to expulsion,fewer side effects,improved and Drug Administration (FDA)approved the protocol of complete abortion rates, and lower cost for women with mifepristone, 600 mg orally, followed approximately 48 pregnancies up to 63 days of gestation based on LMP. hours later by misoprostol,400 µg orally. This is safe and effective for medical abortion through 49 days of gesta- Nonmifepristone Regimens tion (calculated from the first day of the last menstrual Methotrexate and Misoprostol period [LMP]). A follow-up evaluation is scheduled approximately 14 days after administration of mifepris- The combination of methotrexate and misoprostol is an tone. At that time,if clinical history and physical exami- alternative early medical abortion regimen. Among 872 ACOG Practice Bulletin Medical Management of Abortion OBSTETRICS & GYNECOLOGY Table 1.Comparison of Common Medical Abortion Regimens Overall Success Common Regimens Rate (%) Advantages and Disadvantages Gestational Age Mifepristone, 600 mg orally + misoprostol, 92a Must remain in office or clinic 4 hours Up to 49 days 400 µg orally (FDA-approved regimen) after administration Mifepristone, 200 mg orally + misoprostol, 95–99b–f Compared with FDA-approved regimen: Up to 63 days 800 µg vaginally (alternative evidence- • More effective based regimen) • Less time to expulsion • Fewer side effects • Requires vaginal administration of a medication Methotrexate, 50 mg/m2IM or 50 mg 92–96g–i Compared with mifepristone–misoprostol Up to 49 days vaginally, + misoprostol, 800µg vaginally regimen: 3–7 days later • Takes longer for expulsion in 20–30% of women • Readily available medications • Low drug cost Misoprostol only, 800µg vaginally 88j • Requires complicated dosing regimens Up to 56 days repeated for up to three doses • Significantly higher incidence of side effects than other regimens • Low drug cost Abbreviations: FDA, U.S. Food and Drug Administration; IM, intramuscularly aSpitz IM, Bardin CW, Benton L, Robbins A. Early pregnancy termination with mifepristone and misoprostol in the United States. N Engl J Med 1998;338:1241–7. bSchaff EA, Eisinger SH, Stadalius LS, Franks P, Gore BZ, Poppema S. Low-dose mifepristone 200 mg and vaginal misoprostol for abortion. Contraception 1999;59:1–6. cSchaff EA, Fielding SL, Westhoff C. Randomized trial of oral versus vaginal misoprostol at one day after mifepristone for early medical abortion. Contraception 2001;64:81–5. del-Refaey H, Rajasekar D, Abdalla M, Calder L, Templeton A. Induction of abortion with mifepristone (RU 486) and oral or vaginal misoprostol. N Engl J Med 1995;332:983–7. evon Hertzen H, Honkanen H, Piaggio G, Bartfai G, Erdenetungalag R, Gemzell-Danielsson K, et al. WHO multinational study of three misoprostol regimens after mifepris- tone for early medical abortion. I: Efficacy. WHO Research Group on Post-Ovulatory Methods for Fertility Regulation. BJOG 2003;110:808–18. fCreinin MD, Fox MC, Teal S, Chen A, Schaff EA, Meyn LA. A randomized comparison of misoprostol 6 to 8 hours versus 24 hours after mifepristone for abortion. MOD Study Trial Group. Obstet Gynecol 2004;103:851–9. gCreinin MD, Vittinghoff E, Schaff E, Klaisle C, Darney PD, Dean C. Medical abortion with oral methotrexate and vaginal misoprostol. Obstet Gynecol 1997;90:611–6. hCreinin MD, Carbonell JL, Schwartz JL, Varela L, Tanda R. A randomized trial of the effect of moistening misoprostol before vaginal administration when used with methotrexate for abortion. Contraception 1999;59:11–6. iWiebe E, Dunn S, Guilbert E, Jacot F, Lugtig L. Comparison of abortions induced by methotrexate or mifepristone followed by misoprostol. Obstet Gynecol 2002;99:813–9. jJain JK, Dutton C, Harwood B, Meckstroth KR, Mishell DR Jr. A prospective randomized, double-blinded, placebo-controlled trial comparing mifepristone and vaginal misoprostol to vaginal misoprostol alone for elective termination of early pregnancy. Hum Reprod 2002;17:1477–82. women with pregnancies up to 49 days of gestation,this formed approximately 1 week after methotrexate admin- regimen results in a complete abortion rate of 92–96%. istration; a vaginal ultrasound examination is performed Between 50 days and 56 days of gestation,however,effi- to confirm passage of the gestational sac. If abortion has cacy decreases to 82% (28). Although overall efficacy is not occurred, the misoprostol dose is repeated. Further equal to the standard regimen of mifepristone–miso- follow-up for women requiring a second dose of miso- prostol,approximately 15–25% of women using metho- prostol is performed in 4 weeks unless embryonic car- trexate regimens may wait up to 4 weeks for complete diac activity is still visible on ultrasound examination,in abortion to occur (25,29,30). which case patients return in 1 week. If gestational car- Methotrexate is most commonly administered diac activity is present 2 weeks after initiating treatment intramuscularly at a dose based on body surface area or expulsion has not occurred by the 4-week follow-up (50 mg/m2), the same dose used for the management of visit,aspiration is performed. ectopic pregnancy (31). However,regimens using 50 mg Misoprostol Alone of methotrexate orally appear to be as effective as those using methotrexate, 50 mg/m2 intramuscularly (29, 32, Misoprostol, 800 µg vaginally, when moistened with 33). Misoprostol (800 µg) is administered by the woman water, can result in complete abortion rates of 90% in 3–7 days later at home. A follow-up examination is per- women with pregnancies up to 56 days of gestation VOL. 106, NO. 4, OCTOBER 2005 ACOG Practice Bulletin Medical Management of Abortion 873 (10, 34–38). Studies with nonmoistened vaginal miso- Table 2.Features of Medical and Surgical Abortion prostol demonstrated lower rates of 50–67% (39–41). Studies showing that misoprostol alone is effective for Medical Abortion Surgical Abortion abortion often involve complex dosing regimens or •Usually avoids invasive procedure • Involves invasive procedure require clinician application of the tablets. Additionally, • Usually avoids anesthesia • Allows use of sedation if desired this treatment results in significantly higher rates of side • Requires two or more visits • Usually requires one visit effects (nausea,vomiting,diarrhea,and fever and chills) • Days to weeks to complete • Complete in a predictable than those using misoprostol after pretreatment with period of time either mifepristone or methotrexate (10, 34–37). Recent • Available during early pregnancy • Available during early pregnancy trials with sublingual misoprostol in repeated doses do • High success rate (~95%) • High success rate (99%) not appear to improve outcome over vaginal misoprostol and may cause even more side effects (38,42). • Bleeding moderate to heavy • Bleeding commonly perceived for short time as light A recent randomized, double-blind trial in women with pregnancies up to 56 days of gestation compared a • Requires follow-up to ensure • Does not require follow-up in all misoprostol-only regimen (800 µg vaginally) with a reg- completion of abortion cases imen of 200 mg of mifepristone orally followed 48 hours • Patient participation throughout • Patient participation in a single- later by 800 µg of misoprostol vaginally (43). In both a multiple-step process step process groups, misoprostol was repeated every 24 hours for up Adapted from Breitbart V. Counseling for medical abortion. Am J Obstet to three doses. Complete abortion rates for each regimen Gynecol 2000;183:S26–33. were 88.0% and 95.7%, respectively (P<.05). The women who received mifepristone aborted much more quickly and required fewer doses of misoprostol com- Counseling and Symptom Management pared with women who received misoprostol alone. Mifepristone–misoprostol regimens using 200 mg of Some degree of bleeding and uterine cramping are nec- mifepristone orally and 800 µg of misoprostol vaginally essary for the medical abortion process to occur. Other generally are preferred to regimens using methotrexate potential side effects of medical abortion include nausea, and misoprostol or misoprostol only for medical abortion. vomiting,diarrhea,warmth or chills,headache,dizziness, and fatigue (Table 3). Counseling should emphasize that Counseling Patients bleeding may be much heavier than menses, potentially with severe cramping. The woman should understand how Medical Versus Surgical Abortion much bleeding is considered too much. An easy reference for the patient to use is soaking of two pads per hour for Patient counseling must first include discussion of preg- 2 hours in a row (46). This is not necessarily a point at nancy options to be sure that a woman is certain about which intervention is needed but a time when the woman her decision to have an abortion. If she is uncertain, the should call the health care provider. Whether or not it is decision about abortion technique must be delayed until imperative for the patient to seek emergency care depends she has reached a firm decision,even if the delay means on how she is feeling, her baseline hemoglobin, whether that she will be unable to choose a medical option. It is the bleeding seems to be slowing,and how far she is from important to respect the patient’s autonomy and to sepa- an emergency treatment facility. rate the decision to terminate the pregnancy from the Pain management is especially important for the decision about the method to be used. woman aborting at home. She should be sent home with After a woman has considered her options and has appropriate instructions for analgesia with over-the- decided to have an abortion, the method must be select- counter medications, as well as with prescriptions for ed. Most women seeking early abortion will be eligible for both medical and surgical methods. Medical abortion oral narcotics to use if needed. should be considered a medically acceptable alternative The incidence of each symptom will depend on the to surgical abortion in selected,carefully counseled,and regimen used,the dose and route of administration of the informed women. The general advantages and disadvan- prostaglandin analogue, and gestational age. Gastro- tages of each approach (Table 2) should be explained intestinal side effects are less common when dry miso- early in the counseling process because most women will prostol is administered vaginally compared with have a clear preference (44, 45). Even among women regimens that use oral misoprostol or moistened vaginal who think they are unsure, most will have some prefer- misoprostol. Oral ulcers with methotrexate use,although ence after counseling (44). rare,have been reported in the literature. 874 ACOG Practice Bulletin Medical Management of Abortion OBSTETRICS & GYNECOLOGY Table 3.Incidence of Side Effects in Selected North American Trials of Medical Abortion Regimens* Incidence of Side Effects (%) Nausea Vomiting Diarrhea Headache Dizziness Thermoregulatory† Trial Mife Miso Mife Miso Mife Miso Mife Miso Mife Miso Mife Miso Schaff et al (1997)‡ 36 36 14 14 8 22 18 19 22 37 20 37 Schaff et al (1999)§ 45 43 13 26 11 23 14 13 15 28 14 32 Wiebe et al (2002)|| 45 39 13 15 5 16 19 29 NR NR NR 23 Creinin (2004)¶ 20 44 5 23 1 27 10 37 12 35 9 56 Creinin (2004)# 39 52 14 30 7 25 20 37 20 37 19 53 Abbreviations: Mife, mifepristone; Miso, misoprostol; NR, not reported *Studies are included only if the incidences of side effects were differentiated between the medications. †Fever, warmth, hot flashes, or chills ‡Mifepristone, 600 mg, followed by misoprostol, 800 µg vaginally, 36–48 hours later. (Schaff EA, Stadalius LS, Eisinger SH, Franks P. Vaginal miso- prostol administered at home after mifepristone (RU486) for abortion. J Fam Pract 1997;44:353–60.) §Mifepristone, 200 mg, followed by misoprostol, 800 µg vaginally, 48 hours later. (Schaff EA, Eisinger SH, Stadalius LS, Franks P, Gore BZ, Poppema S. Low-dose mifepristone 200 mg and vaginal misoprostol for abortion. Contraception 1999;59:1–6.) ||Mifepristone, 600 mg, followed by misoprostol, 400 µg orally, 36–48 hours later. (Wiebe E, Dunn S, Guilbert E, Jacot F, Lugtig L. Comparison of abortions induced by methotrexate or mifepristone followed by misoprostol. Obstet Gynecol 2002;99:813–9.) ¶Mifepristone, 200 mg, followed by misoprostol, 800 µg vaginally, 6–8 hours later (first row). (Creinin MD, Fox MC, Teal S, Chen A, Schaff EA, Meyn LA. A randomized comparison of misoprostol 6 to 8 hours versus 24 hours after mifepristone for abortion. Obstet Gynecol 2004;103:851–9.) #Mifepristone, 200 mg, followed by misoprostol, 800 µg vaginally, 23–25 hours later (second row). (Creinin MD, Fox MC, Teal S, Chen A, Schaff EA, Meyn LA. A randomized comparison of misoprostol 6 to 8 hours versus 24 hours after mifepristone for abortion. Obstet Gynecol 2004;103: 851–9.) Need for Follow-up Dilation and who has not aborted and is awaiting delayed expulsion Curettage will no longer feel pregnant or have medication-induced symptoms; the patient will be waiting for the onset of A failed medical abortion is defined as the presence of bleeding or cramping similar to anticipating the start of gestational cardiac activity on vaginal ultrasonography menses (28). Providers must differentiate this scenario 2 weeks after the initiation of treatment. No studies have from women who have incomplete expulsion of the preg- assessed the efficacy of additional doses of mifepristone, nancy tissue,for whom symptoms can include prolonged methotrexate, or misoprostol after a medical abortion and irregular bleeding episodes. Early trials of failure. Continuing pregnancies, which should be termi- methotrexate and misoprostol showed that serial β-hCG nated by surgical evacuation,are typically reported in less evaluations did not aid in the diagnosis of incomplete than 1% of women who begin treatment at 49 days of ges- abortion. All women with an incomplete abortion pre- tation or less regardless of regimen. sented clinically, and the incomplete abortion was not Intervention guidelines vary for women who have a diagnosed by increasing or plateaued β-hCG levels (40, persistent gestational sac seen on ultrasonography with- 49,50). out evidence of embryonic cardiac activity or continuing Understanding the difference between incomplete development. Typically, protocols used in mifepristone abortion and the normal course of medical abortion is studies define a retained sac 2 weeks after the admin- important. The sole purpose of ultrasound examination istration of mifepristone as an indication for suction after misoprostol administration is to determine whether evacuation. However, medical abortion studies using the gestational sac is present. After expulsion,the uterus methotrexate and misoprostol demonstrate that interven- will normally contain ultrasonographically hyperechoic tion for a nonviable pregnancy is unnecessary and that tissue consisting of blood, blood clots, and decidua. expulsion will occur,on average,between 22 and 29 days Rarely does this finding during medical abortion indicate after the methotrexate is administered (28, 29, 40, 47, a need for intervention. In the absence of excessive 48). With this understanding, the mifepristone studies bleeding, providers can follow such patients conserva- performed in the United States over the past 6 years have tively (51). allowed approximately 36 days to elapse after mifepris- Overall, large studies demonstrate that less than tone administration before recommending surgical inter- 1% of women undergoing medical abortion will need vention (18, 19, 22–25, 27). Most commonly, a woman emergent curettage because of excessive bleeding VOL. 106, NO. 4, OCTOBER 2005 ACOG Practice Bulletin Medical Management of Abortion 875 (13, 22, 52–54). Moreover, the risk of clinically signifi- Although medical contraindications are infrequent, cant bleeding and transfusion may be lower in women social or psychologic contraindications to medical abortion with pregnancies up to 49 days of gestation compared are more common. Women are not good candidates for with those beyond 49 days (11); this relative risk will medical abortion if they do not wish to take responsibility vary depending on the regimen used. Still, just as for for their care,are anxious to have the abortion over quick- women undergoing surgical abortion, surgical curettage ly,cannot return for follow-up visits,or cannot understand must be available on a 24-hour basis for cases of hemor- the instructions because of language or comprehension rhage. Clinicians who wish to provide medical abortion barriers. Other nonmedical criteria to be considered are services either should be trained in surgical abortion or access to a phone in case of an emergency and access to should work in conjunction with a clinician who is 24-hour emergency medical treatment (eg, surgical curet- trained in surgical abortion. tage for hemorrhage). Counseling should include a descrip- tion of cramping and bleeding and should indicate that, rarely,the process may not be completed for several weeks. Clinical Considerations and ▲ Which pretreatment laboratory tests are Recommendations needed? ▲ What factors determine whether a woman is a No special pretreatment laboratory tests are necessary for candidate for medical abortion? medical abortion beyond those for surgical abortion. Con- firmation of pregnancy by ultrasonography or pregnancy Gestational Age testing is necessary before attempting abortion regardless The upper limit of gestational age at which a medical of method. Pretreatment assessment of hemoglobin or abortion regimen is still an option varies depending on hematocrit and blood typing are imperative, and anti-D the types, dosages, and routes of administration of the immune globulin should be administered if indicated. medications. Outpatient treatment with mifepristone– misoprostol regimens up to 63 days of gestation and for ▲ What is the risk of infection with medical methotrexate–misoprostol regimens up to 49 days of ges- abortion? tation are highly effective. Complete abortion rates Endometritis is a rare complication of medical abortion. among all regimens are highest for earlier gestations and In trials involving more than 500 participants, infection are clinically similar in women with pregnancies up to 49 rates typically vary from 0.09% to 0.6% (11,17,18,22, days of gestation. Between 50 and 63 days of gestation, 23,27,56,57). No data exist to support the universal use the use of vaginal misoprostol in regimens with mifepris- of prophylactic antibiotics for medical abortion. Five tone results in complete abortion in 96–99% of women cases of death have been reported in women using mife- (18, 22–24, 26, 27, 43, 53, 55), whereas regimens using pristone,200 mg,followed by misoprostol,800 µg vagi- oral misoprostol demonstrate significantly lower success nally, in North America since 2001; all appear to be rates for these gestational ages. infectious, with Clostridium sordellii identified in three Contraindications of the cases (58). The cause of these infections and the relationship of the deaths to mifepristone and misopros- Medical contraindications to abortion with mifepristone tol are still under investigation. Even if related,the death regimens include confirmed or suspected ectopic preg- rate would be less than 1 per 100,000 mifepristone pro- nancy or undiagnosed adnexal mass, intrauterine device cedures,a rate comparable to that for early surgical abor- in place, current long-term systemic corticosteroid ther- tion and miscarriage (59). apy,chronic adrenal failure,severe anemia,known coag- ulopathy or anticoagulant therapy, and mifepristone ▲ Is ultrasonography useful in the medical intolerance or allergy. Most clinical trials also exclude management of abortion before treatment? women with severe liver, renal, or respiratory disease, uncontrolled hypertension, cardiovascular disease (an- Gestational age should be confirmed by clinical evalua- gina,valvular disease,arrhythmia,or cardiac failure),or tion or ultrasonography. Only 85% of U.S. women are severe anemia. Misoprostol should not be used in women able to predict gestational age within 2 weeks of the ges- with an uncontrolled seizure disorder or those who have tational age assigned by the clinician using ultrasound an allergy or intolerance to misoprostol or other examination (60). Additionally,medical abortion studies prostaglandins. Asthma is not a contraindication because in U.S. women have found that the gestational age deter- misoprostol is a weak bronchodilator. mined by LMP was confirmed for only 40–60% of study 876 ACOG Practice Bulletin Medical Management of Abortion OBSTETRICS & GYNECOLOGY participants (14,27,61). Because efficacy for some reg- comfortable with their assessment without ultrasonogra- imens decreases significantly with increasing gestational phy or if they would feel better in that situation with an age, the clinical relevance of erroneous gestational age ultrasound examination to confirm their impression based assignment will vary according to the regimen used. on the patient’s history and physical examination. Although not required, all major U.S. trials of Physicians thought an ultrasound examination was not mifepristone or methotrexate have relied on transvaginal needed in 60% of the women who were ultimately found ultrasonography for dating and follow-up. In France, to have expelled the gestational sac. However,the gesta- however, clinicians use ultrasonography for preabortion tional sac was still present in 29% of women for whom screening only when they find a discrepancy between physicians believed ultrasonography was not indicated. uterine size and dating by LMP and when patients pres- Methods to verify abortion include reports of bleed- ent with bleeding or symptoms suggestive of ectopic ing combined with evidence of uterine involution on pregnancy. Pregnancy termination services in France are pelvic examination or hCG testing. When misoprostol is offered only by authorized abortion clinics staffed by administered 2–5 days after methotrexate or mifepristone, β-hCG concentrations should decrease by at least 50% highly experienced providers. The high efficacy and within 1 week of initiating the medication regimen. safety results in the French trials suggest that this selec- However,performing sensitive serum or urine hCG assays tive use of ultrasonography suffices when medical abor- (detection threshold, 25–50 mIU/mL) too soon after the tion is provided by experienced clinicians. termination of a pregnancy may result in an erroneous A concern when providing early abortion services is diagnosis of failed medical abortion. Two trials using the possibility of an undiagnosed extrauterine gestation. methotrexate and misoprostol found that the average time Although the ectopic pregnancy rate in the general popu- to disappearance of β-hCG is 33–34 days and may take as lation is currently around 19–21 per 1,000 pregnancies long as 90 days (40,67). The utility of nonsensitive urine (62, 63), ectopic pregnancy rates in studies of women hCG assays in follow-up after mifepristone and misopros- seeking abortion are consistently lower. A study of surgi- tol administration warrants investigation. cal abortion in women with pregnancies less than 42 days In clinical trials with methotrexate and misoprostol, of gestation in the United States found the ectopic preg- only about half of the women who thought they had nancy rate to be 5.9 per 1,000 pregnancies (64). Similarly, aborted actually had done so (28). Moreover, women the largest published study of medical abortion involved may experience symptom resolution consistent with a 16,369 women with pregnancies up to 49 days of gesta- complete medical abortion and still have a persistent tion,21 of whom were excluded from the analysis because gestational sac (28), or even an ectopic pregnancy (22). of an ectopic pregnancy, yielding an ectopic pregnancy The importance of patient follow-up must be empha- rate of 1.3 per 1,000 pregnancies (57). Although ectopic sized because failure rates for medical abortion are high- pregnancy in a population of women seeking early abor- er than those for surgical techniques. tion is rare,women with significant medical risk factors or However,recent data suggest that for most women history (eg, unilateral pain and vaginal bleeding) should having an abortion with mifepristone and misoprostol, have pretreatment ultrasonography. no follow-up may be needed other than a telephone conversation. One report compared clinicians’ and ▲ What methods can be used to confirm patients’ impressions of whether or not expulsion complete abortion? occurred based solely on the patient’s history with the results of vaginal ultrasonography performed approxi- Transvaginal ultrasonography offers an efficient means of mately 1 week after initiating treatment (68). When the assessing outcome in patients who undergo medical abor- clinician and the patient both thought that expulsion had tion. Its primary objective is to determine if the gestation- occurred,they were correct 99% of the time. Additional al sac is absent (with or without the presence of other studies are needed to validate the premise that such ultrasonographically hyperechoic tissue). However, women need only a home pregnancy test for follow-up French clinicians, who have extensive experience with and that office evaluation should be required only if medical abortion, use ultrasonography significantly less either the clinician or the patient is not certain that than American clinicians (65). One explanation for this expulsion has occurred. difference may be less familiarity with the process by both American clinicians and patients; another reason ▲ Do nonsteroidal antiinflammatory drugs could be liability concerns in the United States. A study of affect the success rates for medical abortion? U.S. providers indicated that ultrasonography is perceived to be unnecessary to assess abortion outcome for most Cramping pain for patients who are not undergoing abor- women (66). Researchers asked physicians if they felt tion usually is treated with ibuprofen or other non- VOL. 106, NO. 4, OCTOBER 2005 ACOG Practice Bulletin Medical Management of Abortion 877 steroidal antiinflammatory drugs (NSAIDs). Although Summary of NSAIDs inhibit the synthesis of new prostaglandins, Recommendations and they do not block the action of prostaglandin receptors; therefore,such agents should not inhibit the action of a Conclusions prostaglandin used for medical abortion. The only report to evaluate the effects of analgesics on abortion The following recommendations are based prima- outcome was a retrospective analysis of NSAIDs and rily on good and consistent scientific evidence complete abortion in 416 women who received miso- (Level A): prostol following methotrexate for medical abortion of pregnancies up to 56 days of gestation (69). The use of ▲ Medical abortion should be considered a medically acceptable alternative to surgical abortion in select- ibuprofen did not seem to interfere with the action of ed,carefully counseled,and informed women. misoprostol to induce uterine contractions and preg- nancy expulsion. Therefore,NSAIDs such as ibuprofen ▲ The FDA-approved protocol of 600 mg of mifepris- are not contraindicated for women undergoing a med- tone orally followed approximately 48 hours later by ical abortion. 400 µg of misoprostol orally is safe and effective for medical abortion through 49 days of gestation (cal- culated from the first day of the LMP). ▲ How should a patient be counseled about ▲ Compared with the FDA-approved regimen, potential teratogenicity if a medical method mifepristone–misoprostol regimens using 200 mg of fails to lead to abortion? mifepristone orally and 800 µg of misoprostol vagi- nally are associated with a decreased rate of contin- Because teratogenicity of medical abortifacients uing pregnancies, decreased time to expulsion, becomes an important issue if the pregnancy continues, fewer side effects, improved complete abortion patients must be informed of the need for a surgical abor- rates,and lower cost for women with pregnancies up tion in the event of a continuing pregnancy. There is no to 63 days of gestation based on LMP. evidence to date of a teratogenic effect of mifepristone. ▲ A methotrexate–misoprostol regimen is appropriate However,methotrexate is an antimetabolite that can cause for medical abortion only in pregnancies up to 49 fetal anomalies (70, 71). Evidence suggests that miso- days of gestation. Women using this regimen may prostol also can result in congenital anomalies when used wait up to 4 weeks for complete abortion to occur. during the first trimester,possibly due to mild uterine con- ▲ Mifepristone–misoprostol regimens using 200 mg of tractions resulting in decreased blood flow during organo- mifepristone orally and 800 µg of misoprostol vagi- genesis (72). Anomalies associated with misoprostol use nally are generally preferred to regimens using that have been described in the literature include defects methotrexate and misoprostol or misoprostol only in the frontal or temporal bones (73) and limb abnormal- for medical abortion. ities with or without Möbius sequence (masklike facies ▲ A patient can administer misoprostol safely and with bilateral sixth and seventh nerve palsy and frequently effectively,orally or vaginally,in her home as part of coincident micrognathia) (74–77). No conclusions a medical abortion regimen. regarding teratogenicity can be drawn from these reports because of the extremely small sample sizes. The following recommendations are based prima- rily on limited scientific evidence (Level B): ▲ Does medical abortion affect future fertility? ▲ Because teratogenicity of medical abortifacients becomes an important issue if the pregnancy contin- Future fertility following medical abortion has been ues,patients must be informed of the need for a sur- evaluated only within a 1-year period after medical gical abortion in the event of a failed abortion. abortion in a group of 93 women who received meth- otrexate and misoprostol for abortion (78). Although ▲ Gestational age should be confirmed by clinical evaluation or ultrasonography. none of the women were actively attempting to achieve pregnancy, 25% became pregnant, a rate higher than The following recommendations are based prima- would be expected for a group of women using contra- rily on consensus and expert opinion (Level C): ception. By comparison, another report indicated a pregnancy rate of 13% within 1 year after a first surgi- ▲ Surgical curettage must be available on a 24-hour cal abortion (79). basis for cases of hemorrhage,even though less than 878 ACOG Practice Bulletin Medical Management of Abortion OBSTETRICS & GYNECOLOGY 1% of women having a medical abortion will need a 13. Aubény E, Peyron R,Turpin CL, Renault M,Targosz V, curettage because of excessive bleeding. Silvestre L,et al. Termination of early pregnancy (up to 63 days of amenorrhea) with mifepristone and increasing ▲ Pretreatment anti-D immune globulin should be doses of misoprostol [published erratum appears in Int J administered if indicated. Fertil Menopausal Stud 1996;41:56]. Int J Fertil Menopausal Stud 1995;40 (suppl 2):85–91. (Level II-3) ▲ No data exist to support the universal use of pro- phylactic antibiotics for medical abortion. 14. Creinin MD,Spitz IM. Use of various ultrasound criteria to evaluate the efficacy of mifepristone and misoprostol for medical abortion. Am J Obstet Gynecol 1999; References 181:1419–24. (Level II-3) 15. McKinley C,Thong KJ,Baird DT. The effect of dose of 1. Finer LB,Henshaw SK. Abortion incidence and services mifepristone and gestation on the efficacy of medical in the United States in 2000. Perspect Sex Reprod Health abortion with mifepristone and misoprostol. Hum Reprod 2003;35:6–15. (Level II-3) 1993;8:1502–5. (Level I) 2. Gravanis A, Schaison G, George M, de Brux J, 16. el-Refaey H, Rajasekar D, Abdalla M, Calder L, Satyaswaroop PG, Baulieu EE, et al. Endometrial and Templeton A. Induction of abortion with mifepristone pituitary responses to the steroidal antiprogestin RU 486 (RU 486) and oral or vaginal misoprostol. N Engl J Med in postmenopausal women. J Clin Endocrinol Metab 1995;332:983–7. (Level I) 1985;60:156–63. (Level II-3) 17. Honkanen H, Piaggio G, Hertzen H, Bartfai G, 3. Johannisson E,Oberholzer M,Swahn ML,Bygdeman M. Erdenetungalag R, Gemzell-Danielsson K, et al. WHO Vascular changes in the human endometrium following multinational study of three misoprostol regimens after the administration of the progesterone antagonist RU 486. mifepristone for early medical abortion. WHO Research Contraception 1989:39;103–17. (Level II-2) Group on Post-Ovulatory Methods for Fertility Regulation. 4. Swahn ML,Bygdeman M. The effect of the antiprogestin BJOG 2004;111:715–25. (Level I) RU 486 on uterine contractility and sensitivity to prostaglandin and oxytocin. Br J Obstet Gynaecol 18. Schaff EA,Fielding SL,Eisinger SH,Stadalius LS,Fuller 1988:95;126–34. (Level II-2) L. Low-dose mifepristone followed by vaginal misopros- tol at 48 hours for abortion up to 63 days. Contraception 5. Danielsson KG,Marions L,Rodriguez A,Spur BW,Wong 2000;61:41–6. (Level II-3) PY,Bygdeman M. Comparison between oral and vaginal administration of misoprostol on uterine contractility. 19. Schaff EA,Fielding SL,Westhoff C. Randomized trial of Obstet Gynecol 1999;93:275–80. (Level II-2) oral versus vaginal misoprostol 2 days after mifepristone 200 mg for abortion up to 63 days of pregnancy [pub- 6. Zieman M,Fong SK,Benowitz NL,Banskter D,Darney lished erratum appears in Contraception 2002;66:481]. PD. Absorption kinetics of misoprostol with oral or vagi- Contraception 2002;66:247–50. (Level I) nal administration. Obstet Gynecol 1997;90:88–92. (Level II-1) 20. von Hertzen H, Honkanen H, Piaggio G, Bartfai G, Erdenetungalag R, Gemzell-Danielsson K, et al. WHO 7. Schaff EA,DiCenzo R,Fielding SL. Comparison of miso- prostol plasma concentrations following buccal and sub- multinational study of three misoprostol regimens after lingual administration. Contraception 2005;71:22–5. mifepristone for early medical abortion. I:Efficacy. WHO (Level II-3) Research Group on Post-Ovulatory Methods for Fertility Regulation. BJOG 2003;110:808–18. (Level I) 8. Tang OS, Schweer H, Seyberth HW, Lee SW, Ho PC. Pharmacokinetics of different routes of administration of 21. Schaff EA,Stadalius LS,Eisinger SH,Franks P. Vaginal misoprostol. Hum Reprod 2002;17:332–6. (Level I) misoprostol administered at home after mifepristone (RU486) for abortion. J Fam Pract 1997;44:353–60. 9. Wiebe ER. Tamoxifen compared to methotrexate when (Level II-3) used with misoprostol for abortion. Contraception 1999;59:265–70. (Level I) 22. Schaff EA,Eisinger SH,Stadalius LS,Franks P,Gore BZ, Poppema S. Low-dose mifepristone 200 mg and vaginal 10. Jain JK,Meckstroth KR,Park M,Mishell DR Jr. A com- misoprostol for abortion. Contraception 1999;59:1–6. parison of tamoxifen and misoprostol to misoprostol (Level II-3) alone for early pregnancy termination. Contraception 2000;60:353–6. (Level I) 23. Schaff EA,Fielding SL,Westhoff C,Ellertson C,Eisinger 11. Spitz IM,Bardin CW,Benton L,Robbins A. Early preg- SH,Stadalius LS,et al. Vaginal misoprostol administered nancy termination with mifepristone and misoprostol in 1,2,or 3 days after mifepristone for early medical abor- the United States. N Engl J Med 1998;338:1241–7. (Level tion: a randomized trial [published erratum appears in II-3) JAMA 2000;284:2597]. JAMA 2000;284:1948–53. (Level I) 12. Comparison of two doses of mifepristone in combination with misoprostol for early medical abortion:a randomised 24. Schaff EA,Fielding SL,Westhoff C. Randomized trial of trial. World Health Organisation Task Force on Post-ovu- oral versus vaginal misoprostol at one day after mifepris- latory Methods of Fertility Regulation. BJOG 2000; tone for early medical abortion. Contraception 2001;64: 107:524–30. (Level I) 81–5. (Level I) VOL. 106, NO. 4, OCTOBER 2005 ACOG Practice Bulletin Medical Management of Abortion 879 25. Wiebe E, Dunn S, Guilbert E, Jacot F, Lugtig L. 39. Bugalho A, Faundes A, Jamisse L, Usfa M, Maria E, Comparison of abortions induced by methotrexate or Bique C. Evaluation of the effectiveness of vaginal miso- mifepristone followed by misoprostol. Obstet Gynecol prostol to induce first trimester abortion. Contraception 2002;99:813–9. (Level I) 1996;53:244–6. (Level II-1) 26. Creinin MD,Potter C,Holovanisin M,Janczukiewicz L, 40. Creinin MD,Vittinghoff E. Methotrexate and misoprostol Pymar HC, Schwartz JL, et al. Mifepristone and miso- vs misoprostol alone for early abortion. A randomized prostol and methotrexate/misoprostol in clinical practice controlled trial. JAMA 1994;272:1190–5. (Level I) for abortion. Am J Obstet Gynecol 2003;188:664–9. 41. Koopersmith TB,Mishell DR Jr. The use of misoprostol (Level II-3) for termination of early pregnancy. Contraception 27. Creinin MD,Fox MC,Teal S,Chen A,Schaff EA,Meyn 1996;53:238–42. (Level I) LA. A randomized comparison of misoprostol 6 to 8 hours 42. Tang OS, Miao BY, Lee SW, Ho PC. Pilot study on the versus 24 hours after mifepristone for abortion. MOD use of repeated doses of sublingual misoprostol in termi- Study Trial Group. Obstet Gynecol 2004;103:851–9. nation of pregnancy up to 12 weeks gestation: efficacy (Level I) and acceptability. Hum Reprod 2002;17:654–8. (Level II-3) 28. Creinin MD,Vittinghoff E,Keder L,Darney PD,Tiller G. 43. Jain JK,Dutton C,Harwood B,Meckstroth KR,Mishell Methotrexate and misoprostol for early abortion:a multi- DR Jr. A prospective randomized,double-blinded,place- center trial. I. Safety and efficacy. Contraception bo-controlled trial comparing mifepristone and vaginal 1996;53:321–7. (Level II-3) misoprostol to vaginal misoprostol alone for elective ter- 29. Creinin MD,Vittinghoff E, Schaff E, Klaisle C, Darney mination of early pregnancy. Hum Reprod 2002;17: PD,Dean C. Medical abortion with oral methotrexate and 1477–82. (Level I) vaginal misoprostol. Obstet Gynecol 1997;90:611–6. 44. Creinin MD. Randomized comparison of efficacy,accept- (Level II-3) ability and cost of medical versus surgical abortion. 30. Creinin MD,Carbonell JL,Schwartz JL,Varela L,Tanda Contraception 2000;62:117–24. (Level I) R. A randomized trial of the effect of moistening miso- 45. Henshaw RC, Naji SA, Russell IT, Templeton AA. prostol before vaginal administration when used with Comparison of medical abortion with surgical vacuum methotrexate for abortion. Contraception 1999;59:11–6. aspiration: women’s preferences and acceptability of (Level I) treatment. BMJ 1993;307:714–7. (Level II-1) 31. Lipscomb GH, Bran D, McCord ML, Portera JC, Ling 46. Creinin MD,Aubény E. Medical abortion in early preg- FW. Analysis of three hundred fifteen ectopic pregnancies nancy. In: Paul M, Lichtenberg ES, Borgatta L, Grimes treated with single-dose methotrexate. Am J Obstet DA,Stubblefield PG,editors. A clinician’s guide to med- Gynecol 1998;178:1354–8. (Level II-3) ical and surgical abortion. New York (NY): Churchill 32. Carbonell JL,Varela L,Velazco A,Cabezas E,Fernandez Livingstone; 1999. p. 91–106. (Level III) C,Sanchez C. Oral methotrexate and vaginal misoprostol 47. Creinin MD,Vittinghoff E,Galbraith S,Klaisle C. A ran- for early abortion. Contraception 1998;57:83–8. (Level I) domized trial comparing misoprostol three and seven days 33. Creinin MD. Oral methotrexate and vaginal misoprostol after methotrexate for early abortion. Am J Obstet for early abortion. Contraception 1996;54:15–8. (Level I) Gynecol 1995;173:1578–84. (Level I) 34. Carbonell JL,Varela L,Velazco A,Fernandez C. The use 48. Creinin MD. Medical abortion with methotrexate 75 mg of misoprostol for termination of early pregnancy. intramuscularly and vaginal misoprostol. Contraception Contraception 1997;55:165–8. (Level II-3) 1997;56:367–71. (Level II-3) 35. Carbonell JL,Varela L,Velazco A,Fernandez C,Sanchez 49. Creinin MD, Darney PD. Methotrexate and misoprostol C. The use of misoprostol for abortion at ≤9 weeks’ges- for early abortion [published erratum appears in Contra- tation. Eur J Contracept Reprod Health Care 1997;2: ception 1994;49:99]. Contraception 1993;48:339–48. 181–5. (Level II-3) (Level II-3) 36. Esteve JL, Varela L, Velazco A, Tanda R, Cabezas E, 50. Creinin MD. Methotrexate for abortion at ≤42 days gesta- Sanchez C. Early abortion with 800 micrograms of miso- tion. Contraception 1993;48:519–25. (Level II-3) prostol by the vaginal route. Contraception 1999; 59:219–25. (Level II-3) 51. Harwood B,Meckstroth KR,Mishell DR,Jain JK. Serum beta-human chorionic gonadotropin levels and endometri- 37. Jain JK,Meckstroth KR,Mishell DR Jr. Early pregnancy al thickness after medical abortion. Contra- termination with intravaginally administered sodium ception 2001;63:255–6. (Level III) chloride solution-moistened misoprostol tablets: histori- cal comparison with mifepristone and oral misoprostol. 52. Allen RH,Westhoff C,De Nonno L,Fielding SL,Schaff Am J Obstet Gynecol 1999;181:1386–91. (Level II-2) EA. Curettage after mifepristone-induced abortion: fre- quency, timing, and indications. Obstet Gynecol 2001; 38. Singh K,Fong YF,Dong F. A viable alternative to surgi- 98:101–6. (Level II-3) cal vacuum aspiration: repeated doses of intravaginal misoprostol over 9 hours for medical termination of preg- 53. Ashok PW,Penney GC,Flett GM,Templeton A. An effec- nancies up to eight weeks. BJOG 2003;110:175–80. tive regimen for early medical abortion:a report of 2000 (Level II-3) consecutive cases. Hum Reprod 1998;13:2962–5. (Level II-3) o880 ACOG Practice Bulletin Medical Management of Abortion OBSTETRICS & GYNECOLOGY