A medical abortion regimen designed to reduce the rate and severity of side effects by using a lower-dose drug combination than is typically used in Europe is highly effective, especially for nulliparous women and women who seek abortion during the first seven or eight weeks of pregnancy, according to a retrospective study conducted in Edinburgh, Scotland.1The researchers, working in a routine clinic setting, found that 96% of women given 200 mg of oral mifepristone followed by 0.5 mg of vaginal gemeprost had a complete abortion without surgical intervention. The likelihood of complete abortion was significantly higher among women who had never given birth than for their parous counterparts, and it declined as the length of gestation increased.
The study was based on data for women who sought abortions at an Edinburgh clinic in 1995-1999. Women who were interested in medical abortion were first counseled and given a medical examination. Gestational age was based on a clinical assessment, with ultrasound used for confirmation in cases of uncertainty. Subsequently, women were given 200 mg of oral mifepristone and asked to return to the clinic in 48 hours. At that time, 0.5 mg of gemeprost was inserted into the vagina. (The standard regimen uses 600 mg of mifepristone and 1 mg of gemeprost.) Nursing staff monitored the women for up to six hours and recorded any passage of fetal tissue. Analgesia was given as requested.
Women returned for a follow-up visit 10-14 days later. A diagnosis of complete abortion was made at follow-up on the basis of the amount of bleeding and pain since treatment, a physical examination of the uterus and, for women who had not passed fetal tissue at the previous visit, a vaginal ultrasound examination. Abortion-related complications, such as bleeding or pelvic infection, were also assessed and treated at this time.
Initially, 3,161 women who obtained medical abortions were studied. Twelve women were excluded because their complete records were not available, and another 310 were excluded because they did not return for follow-up, leaving a total of 2,839 women included in the analyses. On average, the women were 26 years old; 48% had given birth. Most sought their abortion within the first eight weeks of pregnancy: 51% before 49 days' gestation and 32% at 50-56 days. Sixteen percent were 57-63 days pregnant, and 1% were at longer gestations. One-fifth had had a previous abortion.
Overall, 96% of women had a complete abortion after treatment with the drugs and did not require any surgical procedure; in these instances, the researchers considered the method a success. Abortion failures fell into two categories: For 2% of women, the abortion was incomplete (i.e., fetal tissue was still detected in the uterus), while for 1%, the pregnancy was ongoing (i.e., a fetal heartbeat was still detectable) at follow-up. In cases of failure, women underwent surgical evacuation of the uterus.
In bivariate analyses, women who obtained their abortion during the first 49 days of gestation had a higher rate of complete abortion (97%) than those whose abortion occurred later (95%); the results were similar when the researchers compared women who had an abortion within 56 days with those who had the procedure later. Additionally, women who had never given birth were more likely to have a complete abortion (98%) than were parous women (95%). Complete abortion also was more likely among younger women and women who had not had a previous pregnancy termination than among their respective counterparts.
In multivariate analyses using logistic regression, parity and gestation remained significant as predictors of completed abortion after medical therapy. Nulliparity was more strongly associated with the probability of successful medical pregnancy termination (p<.001) than was short gestation (p=.01). These analyses also showed that women with no children were less likely than parous women to have an incomplete abortion (p=.04) or an ongoing pregnancy (p<.001), and that the likelihood of ongoing pregnancy rose as gestation increased (p<.001).
Seven in 10 women required analgesics, and the proportion was significantly higher among nulliparous women (81%) than among parous women (59%). Gestational age was not associated with analgesic use. Abortion-related complications were rare: Fewer than 1% of women suffered severe hemorrhage, and 6% were given antibiotics during the two weeks following the abortion. The occurrence of complications was not associated with any characteristics examined.
The researchers point out that the proportion of women who failed to attend follow-up visits (10%) was higher than the proportion found in other prospective studies; but they add that this proportion is more likely reflective of what happens in a routine clinical setting. Given the increased fetal and maternal risks from an ongoing pregnancy and incomplete abortion, respectively, the researchers mention that this lack of follow-up after medical pregnancy termination "demands persistent attention."
The researchers conclude that the reduced-dose regimen used in the study is effective and safe. They observe that the difference in medical abortion success between nulliparous and parous women was unexpected, and suggest that it reflects differences in early pregnancy between these two groups. Moreover, they note that clinicians can use parity, along with gestation, as criteria when trying to predict the likelihood of a successful medical abortion.--A. Brochert
REFERENCE
1. Bartley J et al., Parity is a major determinant of success rate in medical abortion: a retrospective analysis of 3,161 consecutive cases of early medical abortion treated with reduced doses of mifepristone and vaginal gemeprost, Contraception, 2000, 62(6):297-303.