Can women accurately assess the outcome of medical abortion based on symptoms alone?

Abstract

Background

The primary purpose of this study was to evaluate whether women undergoing medical abortion can accurately assess abortion outcome based on symptoms alone. Our secondary aim was to identify predictors of medical abortion failure.

Study Design

We conducted a case–control study of women undergoing medical abortion from January 1, 2004, to December 31, 2005, who were 63 days' gestation or less and received 200 mg mifepristone followed by 800 mcg of vaginal misoprostol 6–72 h later. Cases were defined as women who required uterine evacuation for a retained gestational sac or ongoing pregnancy. Separate analyses were conducted for the subset of cases with ongoing pregnancies. Controls were defined as women who successfully expelled the pregnancy without uterine evacuation.

Results

During the study period, 53 women had a retained gestational sac (N=26) or ongoing pregnancy (N=27), and a total of 53 controls were selected, matched by site and date of procedure. Case subjects were more likely than controls to report minimal vaginal bleeding and ongoing pregnancy symptoms and to express doubt that they expelled the pregnancy. When predictive modeling was performed, ongoing pregnancy symptoms, minimal bleeding and gestational age as determined by ultrasound measurement of gestational sac or crown-rump length accurately identified only 68% of medical abortion failures. We also found that the odds of medical abortion failure decreased progressively from approximately 4 to 7 weeks' gestational age, was lowest at approximately 7 weeks and increased from 7 to 9 weeks' gestation.

Conclusion

Patient symptomatology and self-assessment of complete abortion alone are moderately useful in identifying medical abortion failure. An objective measure of complete abortion, such as a pregnancy test, is still required.

Introduction

Medical abortion regimens currently utilized in the United States require at least two clinical visits. At the first visit, gestational age is determined, and the patient is evaluated for contraindications to medical abortion. If the patient is a candidate for medical abortion, and there is no state regulation requiring an additional 24-h waiting period, mifepristone is administered and misoprostol is dispensed for home use. At the second visit, completion of the abortion is objectively confirmed either via ultrasonography or serum human chorionic gonadotropin (hCG) measurement [1]. It is important to identify ongoing pregnancies in a timely fashion, so that women who fail a medical abortion still have the option of a surgical abortion and also because misoprostol has teratogenic potential [2], [3]. However, women often find follow-up visits cumbersome. In one medical abortion study, 22% of subjects reported problems scheduling multiple visits and noted difficulties with work, transportation, school and childcare [4]. Additionally, clinicians have cited the need for multiple visits as being one of the top obstacles to both start-up and continuation of medical abortion services [5].

The majority of clinicians in the United States confirm completion of medical abortion with ultrasonography [5]. Simplifying the medical abortion protocol by eliminating the second visit could lead to increased access and cost savings. Previous studies have examined the ability of clinicians and subjects to assess the success of medical abortion based on symptoms and history. However, given the low failure rate of medical abortion [1], [5], most published studies to date have had few ongoing pregnancies and therefore are limited in their ability to assess how well women can identify a medical abortion failure.

Published data support that when medical abortion is successful, which it is 95%–98% of the time, women and physicians can reliably make that assessment based on history alone. However, the outcome of greater concern is medical abortion failure, in particular, ongoing pregnancy. We do not have adequate evidence thus far that women who have a medical abortion failure can accurately determine that they have ongoing pregnancies in a timely fashion based on symptoms alone.

Given that medical abortion failure is a rare adverse event, we conducted a case–control study, identifying a cohort of women with medical abortion failures and comparing them to controls from the same population in order to evaluate whether women can accurately identify medical abortion failure.