Abortion Reversal Study Stopped Early for Severe Bleeding

A trial to test whether a medical abortion might be halted, allowing a pregnancy to continue after the patient takes the first drug (mifepristone) of a two-drug regimen but before taking the second drug (misoprostol), was ended early for safety concerns. One quarter of the patients experienced severe hemorrhage, according to findings published December 5 in Obstetrics & Gynecology.

"Although the study sample size was powered to demonstrate a difference in continuing pregnancy rates between progesterone and placebo treatment after mifepristone ingestion, we could not evaluate this outcome owing to stopping enrollment for safety reasons," report Mitchell D. Creinin, MD, of the University of California Davis in Sacramento, and colleagues. "Because of the potential dangers for patients who opt not to use misoprostol after mifepristone ingestion, any mifepristone antagonization treatment must be considered experimental."

In an interview with Medscape Medical News, Creinin emphasized that the study is "not about the danger of progesterone" but about the "safety of not using the combination treatment as planned by not using misoprostol."

The researchers had planned to enroll 40 patients with pregnancies between 44 to 63 days gestation in a double-blind randomized trial to determine whether pregnancy continuation was possible after patients took mifepristone, the first of a two-drug regimen for medical abortion. The trial intended to test the medical "abortion reversal" protocol promoted by a handful of physicians on the basis of two incomplete methodologically flawed case series. Those previous studies have led to multiple state laws in recent years that require abortion providers to inform women of the existence of the protocol, over the objections of the American College of Obstetricians and Gynecologists (ACOG).

Study enrollees were women who had planned a surgical abortion but agreed to delay termination for 2 weeks to participate in the trial, which involved administering the first half of a medical abortion regimen. The original study protocol called for half the participants to take 400 mg of progesterone twice daily for 3 days starting 24 hours after taking 200 mg of mifepristone. The other half took a placebo twice daily for 3 days. After 3 days, the groups would take the progesterone or placebo once daily until their planned surgical abortion.

Twelve participants had enrolled between February to July 2019 before the researchers halted the study after three patients required ambulance transport to the hospital for severe hemorrhage. One of these patients was in the progesterone group and two were in the placebo group; the latter both required suction aspiration and one required a transfusion.

Two of the 12 patients left the study as a result of nausea, vomiting, and bleeding. Of those who remained, four of the five women in the progesterone group and two of the women in the placebo group had confirmed fetal cardiac activity 2 weeks after taking mifepristone.

Creinin told Medscape Medical News he notified the institutional review board that he had to stop the study after the third hemorrhage, a very rare occurrence with medical abortions.

"When I'm doing research, I care about every single person who enrolls in my studies. I always remind women that the study is about them, not the drug," Creinin said.

He expressed concerns about the previously published case series, which claimed high rates of continuing pregnancies when women took large doses of oral progesterone to allegedly cancel the effects of mifepristone. Those studies, however, did not operate under an institutional review board and did not report the outcomes of all women who took progesterone after mifepristone instead of misoprostol. Those aspects and other flaws led the ACOG to issue a statement that "claims regarding abortion 'reversal' treatment are not based on science and do not meet clinical standards."

Until this study, progesterone as a mifepristone antagonist had not been tested in a randomized controlled trial. A systematic review of 13 studies in 11 publications in which women did not take misoprostol showed that 8% to 46% of pregnancies continued 1 to 2 weeks after mifepristone. Prior to this new study, no evidence existed to suggest serious safety concerns.

"As physicians within the US, we have the ability to prescribe any medication off-label according to [US Food and Drug Administration] guidance that we feel is appropriate as long as we tell the patient it's off-label and there's data to support the safety and efficacy," Creinin said. "There is no data to support that this is an effective treatment. We do have some data to support it may not be safe."

The high proportion of hemorrhages, despite the small number of participants, also surprised Daniel Grossman, MD, a professor of obstetrics and gynecology at the University of California, San Francisco, and Lisa Harris, MD, PhD, a professor of obstetrics and gynecology at the University of Michigan in Ann Arbor, two of the authors of the systematic review.

"I didn't anticipate that there was a serious safety issue," Harris told Medscape Medical News. The new data "really magnifies all the concerns I already had," she said. "Not only is that other treatment unproven but it may be unsafe."

Since this study was so small and stopped early, we still lack evidence on the efficacy of using progesterone to continue a pregnancy after mifepristone, Grossman told Medscape Medical News.

"Anybody who is offering this treatment should be doing it under an experimental protocol that is under an IRB or ethical review panel, particularly now that these safety concerns have been raised," Grossman said. He added that it is extremely rare for patients to change their mind after starting a medical abortion, even rarer than for many other medical decisions.

Yet that remote possibility has inspired a spate of state laws. Starting with an Arkansas law passed in 2015, state legislators across the United States have introduced bills or passed laws that require abortion providers to tell women prior to a medical abortion that the abortion might be able to be "reversed" if the woman changes her mind after taking the first drug. The Arkansas law includes a script providers must read that directs women to search online for the term "abortion pill reversal."

States with similar laws currently include South Dakota, Utah, Idaho, North Dakota, Kentucky, Nebraska, and Oklahoma, though the North Dakota law was blocked in September by a federal judge who ruled it violated providers' right to free speech.

"What these laws do is interfere with that relationship between the provider and the patient, and when you have to give patients medically inaccurate information, it gives them doubt about their own decision," Creinin said. "I know of no providers who would provide an abortion if the patients weren't sure."

Harris also expressed concern about the message sent by these laws.

"The question remains why are we moving the certainty of the decision to a point after the treatment starts rather than before," Harris said. "If there's really a problem with women not being certain about their decision, then we need to focus our efforts on achieving decisional certainty rather than on starting treatment and then changing course." She, too, however, expressed doubt that women changing their minds is a serious problem.

The passage of the laws raises other ethical concerns as well, according to Bonnie Steinbock, PhD, a medical ethicist and professor emerita of the University at Albany/State University of New York.

"The global point is that you're not supposed to practice medicine without a license, and all these laws are practicing medicine without a license," Steinbock told Medscape Medical News. Physicians "are absolutely put in a bind," she said, if they are expected "to follow the new laws, which conflict with the entire body of medical ethics and what the law has to say about informed consent." As written, the current laws do not address potential risks of "abortion reversal" procedures.

This issue also raises the question of whether patients should receive any procedure they request, "and I think that has to be 'no,'" Steinbock said. "There are always ethical limits on what doctors can provide or should provide. What a doctor is obligated to do is follow best practices," not do a procedure "just because a patient wants it."

The research was funded by the Society of Family Planning Research Fund. Dr Creinin consults for Danco Laboratories, and coauthor Laura Dalton is an employee of Planned Parenthood.

Obstet Gynecol. Published online December 5, 2019. Abstract

Follow Medscape on Facebook, Twitter, Instagram, and YouTube.