Printer friendly

Medical methods for first trimester abortion have been demonstrated to be both safe and effective. Regimens that combine mifepristone or methotrexate with a prostaglandin such as misoprostol are more efficacous than a prostaglandin alone. In the case of regimens that combine mifepristone with a prostaglandin, the dose of mifepristone may be reduced from 600 mg to 200 mg without affecting efficacy; in 4 trials that compared the efficacy of the 200-mg dose with the 600-mg dose of mifepristone the relative risk (RR) of failure was 1.07 with the 95% confidence interval (CI) being 0.87–1.32. In combination with mifepristone, vaginally administered misoprostol in a 800-µg dose , appears to be more efficacious compared with a prostaglandin E1 analogue 0.5 mg. When mifepristone is used with misoprostol to terminate pregnancies of up to 63 days’ gestation, misoprostol administered vaginally is more efficacious than when administered orally (2 trials, RR 4.41 of failure with oral misoprostol compared with vaginal adminstration, 95% CI 2.32–8.38). Data were insufficient in the meta-analysis to determine the effect of gestational age on the efficacy of the various regimens. Oral administration of misoprostol was found to be associated more frequently with nausea and diarrhoea compared with vaginal administration (2 trials, RR 1.13, 95% CI 1.0–1.25; RR 1.80, 95% CI 1.49–2.18, respectively). With regard to the timing of administration of the prostaglandin following the administration of mifepristone on day 0, one trial indicated that misoprostol administered on day 3 was less effective compared with its administration on day 1 (RR of failure 1.94, 95% CI 1.05–3.58). However, in the three trials included in the review (but not pooled), there was no difference in efficacy when comparing prostaglandin administration on day 3 with day 2, day 2 with day 1, and day 2 with day 0.

With the combined methotrexate-misoprostol regimen, one study compared intramuscular administration with oral administration of methotrexate and found no statistically significant difference in efficacy . The review also found no statistically significant difference in efficacy between the administration of misoprostol on day 3 compared with day 5, or on day 4 compared with day 5, after methotrexate.

The review does not include two recent trials as they were published after the review was submitted for publication. The first is a WHO multinational study (1), of mifepristone combined with three different misoprostol regimens-(a) 800 µg vaginally on day 3 only; (b) 800 µg vaginally on day 3 followed by 400 µg orally twice daily for 7 days; or (c) 800 µg orally on day 3 followed by 400 µg orally twice daily for 7 days; n=2219. This study found that in women with gestation periods ≥57 days, the risk of failure was higher in group c compared with group b (RR 2.8, 95% CI 1.3-5.8), but there were no significant differences in efficacy in women with gestation periods <57 days.

The second study not included in the current version of the review is a randomized controlled trial of mifepristone followed by either sublingual or vaginal misoprostol (2). The dose of mifepristone was 200 mg, and that of misoprostol was 800 µg, administered 48 hours after mifepristone. The study included women with gestation periods up to 63 days (n=224). The authors found no significant difference in efficacy between the two regimens (sublingual: 98.2%; vaginal: 93.8%), although women in the sublingual group experienced significantly more nausea, vomiting, diarrhoea, fever and chills. While more research is needed to identify the optimal dosage of sublingual misoprostol, it appears that this route is another effective option that may be offered to women, especially given the evidence that women often prefer oral administration over vaginal (3).

One of the outcomes this review aimed to address was “women’s dissatisfaction with the procedure,” yet in the discussion the authors state that acceptability has not been sufficiently assessed in the identified trials.

The WHO multinational trial referred to above reported data on side-effects and acceptability and found that 85% of participants with a successful outcome for the procedure would choose medical abortion again should the need arise (4). Significantly higher acceptability was found among parous women (compared to nulliparous women) and those whose procedures were successful (compared with those in whom the procedure failured).

Two reviewers selected trials for inclusion from the results of the search and performed data extraction independently. The quality of the trials included was assessed appropriately, and there was adequate justification to exclude the studies that were left out. Study location is not always specified for the included trials, which would have been useful. Data were appropriately analysed and presented clearly. It is especially useful that data were presented for trials even when their methodologies were not similar enough to allow for combined analysis using meta-analytic techniques. The presentation of the data is user-friendly for such a review that includes many trials with dissimilar interventions and methods.

The full RHL commentary also includes sections on: