• abortion, induced
• Reproductive Rights
• Patient Safety

The increasing literature on early pregnancy mifepristone antagonisation by progesterone is encouraging, as we seek to better serve women in this difficult situation. Nevertheless, there are ongoing examples of misleading information regarding scientific and clinical aspects of ‘abortion reversal’. The recent review by Stifani and Lavelanet claimed that clinicians “should be cautious about providing high-dose progesterone off-label given the lack of evidence of benefit and the potential safety concerns”.1 They cite the only randomised controlled trial (RCT) in the field, which also claimed that their small study provided important insights into the safety of mifepristone antagonisation with progesterone in early pregnancy.2

In the case of where a woman has taken mifepristone but then chooses not to take misoprostol, we concur that there is a lack of quality evidence concerning any benefit of progesterone treatment. However, we refute that there are safety concerns for prescribing progesterone compared with the prevailing recommendation of conservative management.

In figure 1, the woman has enacted her reproductive autonomy at I, II and III. At III(b), if she chooses not to take misoprostol she will maintain the adverse bleeding risk attributable to taking mifepristone alone. At III(c), if she chooses not to take misoprostol and instead commences abortion reversal, she does so with the pre-existing risk attributable to taking mifepristone alone being present before she commences taking progesterone.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Relative bleeding risk comparisons in pregnancy and medical abortion. RCT, randomised controlled trial.

As acknowledged by Stifani and Lavelanet,1 in the only abortion reversal RCT2 there were two serious bleeding events in the placebo arm, and one bleeding event in the progesterone arm, which resolved without any treatment required (figure 1, IV). Clearly, however, the small numbers in this trial precluded drawing any statistically robust conclusions. Furthermore, Stifani and Lavelanet concluded that ongoing pregnancy rates were not significantly different between groups of women managed conservatively or with progesterone partly on the basis that there were two of four ongoing pregnancies (50%, 95% CI 15% to 85%) in the placebo group of the RCT for gestation 7–8 weeks.2 Making a statistical claim based on such low numbers is misleading.

In another example of misleading (or outdated) information being promulgated in the literature, the cited figure of women changing their mind after taking mifepristone of 0.004%, which was based on a personal communication with the drug distributor, has been noted to be inaccurate.3 Elsewhere, Heartbeat International have published on their website unverified figures of women who have called their 24-hour service seeking abortion reversal information. This includes a reported 2614 contacts from the USA in 2022, which is higher than the 0.004% figure by two orders of magnitude. One UK study found that of 2743 scheduled medical abortion cases, 21 women (0.77%) used mifepristone alone but did not abort, although it is not documented what their subsequent pregnancy outcome was.4 Irrespective of the actual numbers, these data do demonstrate that for some women there are no guidelines for providing active treatment options to maintain viability of their pregnancy after taking mifepristone.5

A recent law-related article claimed that all published abortion reversal case reports demonstrated a failure to obtain consent from the women involved.6 However, one of these reports clearly documented that women were consented, and institutional research ethics approval was obtained. Ironically, this inaccurate law-related article expounded abortion policy as being formed by unsound science, yet partly based this claim on their unsound statement discussed above. It should further be noted that attempts by this author to correct the public record in this regard were rejected by the journal concerned.

At the core, good medicine and collaborative science need to serve these vulnerable women. This includes provision of the full range of treatment options by clinicians, which should be underpinned by quality evidence from rigorous and statistically validated clinical trials.