Cochrane for Clinicians

Putting Evidence into Practice

Medical Methods for First-Trimester Abortion

MELISSA NOTHNAGLE, M.D., and JULIE SCOTT TAYLOR, M.D., M.SC., Brown Medical School, Providence, Rhode Island

This clinical content conforms to AAFP criteria for evidence-based continuing medical education (EB CME). EB CME is clinical content presented with practice recommendations supported by evidence that has been systematically reviewed by an AAFP-approved source. The practice recommen-dations in this activity are available online at http://www.cochrane.org/cochrane/revabstr/AB002855.htm. The Cochrane Abstract below is a summary of a review from the Cochrane Library. It is accompanied by an interpretation that will help clinicians put evidence into practice. Melissa Nothnagle, M.D., and Julie Scott Taylor, M.D., M.Sc., present a clinical scenario and question based on the Cochrane Abstract, along with the evidence-based answer and a full critique of the abstract.

Clinical Scenario

A 34-year-old patient, gravida 3, para 2, has an unplanned pregnancy and requests referral for abortion. Her last menstrual period began 39 days ago. She asks what you know about "the abortion pill."

Cochrane Abstract Background. Surgical abortion by vacuum aspiration or dilatation and curettage up to 63 days' gestation has been the method of choice since the 1960s. Medical abortion became an alternative method of first-trimester pregnancy termination with the availability of prostaglandins in the early 1970s and anti-progesterones in the 1980s. The most widely researched drugs are prostaglandins alone, mifepristone alone, methotrexate alone, mifepristone with prostaglandins, and methotrexate with prostaglandins. Objectives. To compare different medical methods for first-trimester abortion. Search Strategy. The authors1 searched the Cochrane Controlled Trials Register, MEDLINE, and Popline. Reference lists of retrieved papers also were searched. Experts in the World Health Organization Human Reproduction Programme were contacted. Selection Criteria. Randomized controlled trials comparing different methods (e.g., single-drug and combination regimens), methods of application, or dosage regimens for medical abortion were considered. After assessment, trials were included if they had adequate randomization procedure, concealment of allocation, and follow-up. Women in the first trimester of pregnancy who were undergoing medical abortion were the participants. Different medical methods used for first-trimester abortion, compared with each other or placebo, were included. Outcomes included mortality, failure to achieve complete abortion, surgical evacuation (as emergency procedure, nonemergency procedure, or undefined), ongoing pregnancy at follow-up, time until passing of conceptus (more than three to six hours), blood transfusion, blood loss (measured or clinically relevant drop in hemoglobin), days of bleeding, pain resulting from the procedure (reported by the women or measured by use of analgesics), additional uterotonics used, women's dissatisfaction with the procedure, nausea, vomiting, and diarrhea. Data Collection and Analysis. Two reviewers independently selected trials for inclusion from the results of the search strategy described previously. Trials under consideration were evaluated for appropriateness for inclusion and methodologic quality without consideration of their results. A form was designed to facilitate the data extraction. Data were processed using Revman software. Primary Results. Thirty-nine trials were included in the review. Unless otherwise stated, the effectiveness outcomes refer to failure to achieve complete abortion with the intended method. Mifepristone in a dosage of 600 mg compared with a dosage of 200 mg has similar effectiveness in achieving complete abortion (four trials; relative risk [RR], 1.07; 95 percent confidence interval [CI], 0.87 to 1.32). Misoprostol administered orally is less effective (more failures) than misoprostol administered vaginally (RR, 3.00; 95 percent CI, 1.44 to 6.24) and may be associated with more frequent side effects, such as nausea and diarrhea. Mifepristone alone is less effective than a combination of mifepristone and prostaglandin (RR, 3.76; 95 percent CI, 2.30 to 6.15). Similarly, all but one of the five trials comparing prostaglandin with the combined regimen reported higher effectiveness with the combined regimen. The results of these studies were not pooled, but the RR of failure with prostaglandin alone is between 1.4 and 3.75, and the 95 percent CI indicates statistical significance. In one trial comparing gemeprost in a dosage of 0.5 mg with misoprostol in a dosage of 800 mcg, misoprostol was more effective (failure with gemeprost: RR, 2.86; 95 percent CI, 1.14 to 7.18). [Editor's note: Gemeprost is a prostaglandin available in the United Kingdom but not in the United States.] There was no difference between using a split dose or a single dose of prostaglandin. There was no statistically significant difference in failure to achieve complete abortion when intramuscular methotrexate was compared with oral methotrexate (RR, 2.04; 95 percent CI, 0.51 to 8.07). Similarly, early (day 3) versus late (day 5) administration of prostaglandin showed no significant difference (RR, 0.72; 95 percent CI, 0.36 to 1.43). One trial compared the effects of tamoxifen versus methotrexate, and no statistically significant differences were observed between the groups. Reviewers' Conclusions. Safe and effective medical abortion methods are available. Combined regimens are more effective than use of single agents. In the combined regimen, the dosage of mifepristone can be lowered to 200 mg without significantly decreasing effectiveness. Vaginal misoprostol is more effective than oral misoprostol. Some results are based on small studies only and therefore carry some uncertainty. Almost all trials were conducted in hospital settings with good access to support and emergency services. It is therefore not clear if the results are readily applicable to under-resourced settings where such services are lacking, even if the agents used are available.

These summaries have been derived from Cochrane reviews published in the Cochrane Database of Systematic Reviews in The Cochrane Library. Their content has, as far as possible, been checked with the authors of the original reviews, but the summaries should not be regarded as an official product of the Cochrane Collaboration; minor editing changes have been made to the text (www.cochrane.org).

Clinical Question

What are the most effective regimens for first-trimester medical abortion?

Evidence-Based Answer

A regimen that includes mifepristone in a dosage of 200 mg administered orally, followed by misoprostol in a dosage of 800 mcg administered vaginally, is highly effective for medical abortion up to 63 days' gestation. Side effects include bleeding, pain, nausea, vomiting, and diarrhea. Serious complications are rare. In settings where mifepristone is unavailable, methotrexate, followed by misoprostol, is effective up to 49 days' gestation.

Practice Pointers

In the United States, 49 percent of all pregnancies are unintended, and about one half of women with an unintended pregnancy choose abortion.2 The U.S. Food and Drug Administration (FDA) has approved mifepristone (also known as RU-486) for use with the prostaglandin misoprostol to induce medical abortion up to 49 days after the last menstrual period.

The original FDA-approved protocol is based on early clinical trials in the United States and France. On the first visit, after appropriate counseling and dating ultrasonography, the patient takes 600 mg of mifepristone orally. She returns on day 3 to take 400 mcg of misoprostol orally, and again on day 14 for a follow-up examination and ultrasonography to ensure that the abortion is complete.

Based on evidence from recent trials, including those in the Cochrane Review, current practice patterns include several variations of this protocol, incorporating the following changes: using a lower dosage (200 mg) of oral mifepristone; administering 800 mcg of misoprostol vaginally instead of orally; allowing the patient to use misoprostol at home on day 2, 3, or 4; and following up sooner than 14 days. This protocol is used until 63 days' gestation. These variations are equally effective and allow more flexibility for patients. Vaginal misoprostol causes fewer gastrointestinal side effects than oral misoprostol.

The rate of complete abortion with mifepristone and misoprostol was 92 to 95 percent using the FDA-approved regimen.3,4 Data in this review were insufficient to describe differences in efficacy by gestational age, but a recent study using the evidence-based regimen reported completion rates of 96 to 97 percent, including pregnancies up to 63 days' gestation.5 Patients choosing medical abortion must consent to undergo vacuum aspiration in cases of incomplete abortion. Misoprostol and methotrexate can cause fetal anomalies.

Prior to mifepristone's approval in the United States, medical abortion regimens were developed that used methotrexate in combination with prostaglandins. None of the reviewed studies compared methotrexate with mifepristone. One high-quality, unblinded, randomized controlled trial comparing methotrexate (50 mg per m2 intramuscularly) with mifepristone (200 mg orally), each followed by 800 mcg of vaginal misoprostol, reported similar success rates, side effects, and complications up to 49 days' gestation.6 Time to completion of the abortion was shorter with the mifepristone-based regimen.

Common side effects of medical abortion include pain and bleeding from the abortion itself; these effects may be comparable to those of an early miscarriage, except that the timing of bleeding is more predictable. Other side effects of prostaglandin use may include nausea, vomiting, diarrhea, low-grade fever, and chills. Most side effects can be managed with oral analgesics and careful counseling. Complications of medical abortion, such as hemorrhage and infection, are rare. Bleeding that requires transfusion occurs in approximately 0.2 percent of cases.1 Infection is less frequent.7 The current review did not address differences in side effects, bleeding patterns, and acceptability of the various regimens.

Almost all of the studies in the current review were conducted in hospital-based clinics with access to ultrasonography and surgical abortion services. Caution should be used when applying this research to settings with fewer resources. In summary, medical abortion is a safe, effective, noninvasive alternative to early surgical abortion.

The information and opinions presented in AFP reflect the views of the authors, not those of the journal or the American Academy of Family Physicians, unless so stated. The Academy's policy on reproductive decisions, including medical abortion, is available online at: http://www.aafp.org/x7053.xml.

References

Cochrane Briefs

Gowning in Newborn and Special-Care Nurseries

Clinical Question

Does gowning attendants and visitors in newborn and special-care nurseries improve outcomes?

Evidence-Based Answer

There is no evidence that gowning in newborn nurseries and neonatal intensive care units (NICUs) improves clinically important outcomes.

Practice Pointers

Gowns are an uncomfortable and not very fashionable fact of life at many newborn nurseries. It is thought that they reduce transmission of infection from clinicians to infants and limit the introduction of infectious agents by visitors from outside the nursery. Webster and Pritchard reviewed the literature and identified 12 relevant studies, four of which were excluded because they were not randomized or used historical controls (i.e., outcomes before and after a gowning requirement was begun or stopped were compared).

Three studies randomly assigned staff and visitors to a gown or no gown and observed their handwashing behavior (n = 2,285 infants). Five studies alternated periods when gowning was or was not required for all staff and visitors; all of these studies examined infants (n = 3,979) in special-care nurseries or NICUs. Not wearing a gown was associated with a lower death rate (relative risk, 0.84; 95 percent CI, 0.70 to 1.02) in the four NICU studies that studied this outcome.

The five NICU studies did not show any effect on the incidence of nosocomial infections such as septicemia, meningitis, necrotizing enterocolitis, or pneumonia. The relative risk of infection ranged from 0.62 (less infection with gowning) to 2.52 (more infection with gowning), but none of the differences between groups was significant, and the overall relative risk of 0.95 was not statistically significant. Four studies of localized nosocomial infection also found no benefit from gowning. Secondary outcomes such as length of hospital stay, likelihood that patients or clinicians would wash hands, or colonization rates did not differ between groups.

MARK H. EBELL, M.D., M.S.

Webster J, Pritchard MA. Gowning by attendants and visitors in newborn nurseries for prevention of neonatal morbidity and mortality. Cochrane Database Syst Rev 2003;3:CD003670.

Intensive Management of Gestational Diabetes

Clinical Question

Does intensive management of gestational diabetes improve outcomes?

Evidence-Based Answer

There is not enough evidence to support dietary or drug treatment in patients with gestational diabetes.

Practice Pointers

Gestational diabetes and impaired glucose tolerance are associated with macrosomia and may be associated with increased risk for cesarean delivery, shoulder dystocia, and birth trauma. Although preexisting diabetes has been shown to increase the risk of poor perinatal outcomes, it is not clear that data relating to preexisting diabetes can be extrapolated to patients with gestational diabetes.

Tuffnell and colleagues searched the Cochrane Pregnancy and Childbirth Group trials register, the Cochrane Central Register of Controlled Trials, and bibliographies of relevant articles. They identified three studies of 223 women with impaired glucose tolerance; none of these studies was a randomized controlled trial comparing management strategies. Treatment of women with impaired glucose tolerance did not offer a statistically significant benefit over nontreatment in terms of abdominal operative delivery rates, neonatal intensive care admissions, or reduction in birth weight. Treatment may be associated with a reduced incidence of neonatal hypoglycemia. The trials had wide confidence intervals and methodologic shortcomings. The small number of patients studied means that a small but clinically meaningful benefit may have been missed.

In the face of limited and inconsistent research, the American College of Obstetricians and Gynecologists (ACOG) continues to recommend universal screening for gestational diabetes.1 It recommends that insulin therapy be considered in patients for whom nutritional therapy does not result in a fasting glucose level of less than 95 mg per dL (5.3 mmol per L), a one-hour postprandial glucose level of less than 130 to 140 mg per dL (7.2 to 7.8 mmol per L), or a two-hour postprandial glucose level of less than 120 mg per dL (6.7 mmol per L). ACOG also recommends that physicians consider elective cesarean delivery for women with gestational diabetes and an estimated fetal weight greater than 4,500 g (9 lb, 15 oz). ACOG does not make a recommendation for or against calorie restriction in obese women with gestational diabetes.

Intensive management of gestational diabetes is time-consuming and resource-intensive. Overall, evidence is insufficient to support therapy for gestational diabetes. However, universal screening is the standard of care in most communities. When faced with abnormal results, most family physicians will opt to follow the consensus opinion of our specialist colleagues.

CLARISSA KRIPKE, M.D.

Tuffnell DJ, et al. Treatments for gestational diabetes and impaired glucose tolerance in pregnancy. Cochrane Database Syst Rev 2003;3:CD003395.

---

The Authors

Melissa Nothnagle, M.D., is clinical assistant professor of family medicine and assistant residency director in family medicine at Brown Medical School in Providence, R.I.

Julie Scott Taylor, M.D., M.Sc., is assistant professor of family medicine and director of predoctoral education in family medicine at Brown Medical School.

Address correspondence to Melissa Nothnagle, M.D., Department of Family Medicine, Memorial Hospital of Rhode Island, 111 Brewster St., Providence, RI 02860 (e-mail: Melissa_nothnagle@mhri.org). Reprints are not available from the authors.

Reference

Copyright © 2004 by the American Academy of Family Physicians. This content is owned by the AAFP. A person viewing it online may make one printout of the material and may use that printout only for his or her personal, non-commercial reference. This material may not otherwise be downloaded, copied, printed, stored, transmitted or reproduced in any medium, whether now known or later invented, except as authorized in writing by the AAFP. Contact afpserv@aafp.org for copyright questions and/or permission requests.

---

---