A new study (Mifepristone Antagonization with Progesterone to Prevent Medical Abortion: A Randomised Controlled Trial ) conducted in January 2020 is often cited as evidence that the Mifeprex  (abortion pill) is safe while the Abortion Pill Reversal (progesterone) is dangerous. When in fact, quite the opposite is true.

Dermot Kearney offers a critique of the study:

On the Creinin study, and particularly in relation to using the findings in the study to attempt to demonstrate that APR is useless or ineffective, there are many shortcomings that need to be highlighted. 

  • The stated aim of the study was allegedly to estimate the efficacy and safety of Mifepristone antagonisation with high-dose oral Progesterone. In designing the study, it was estimated that the numbers that would be needed to demonstrate a significant difference between the treatment and placebo groups would be 40 patients with 20 in each group. 
  • This is obviously a very small cohort of study participants and it doesn’t leave any room for a dropout rate which is inevitable in this type of study.
  • That might raise questions as to how serious the investigators really were in attempting to answer the questions relating to efficacy and safety.
  • Two participants, one in each group, voluntarily withdrew from the study within a few days, and the study was discontinued after only twelve participants had been enrolled, allegedly because of safety concerns. 
  •  Three participants, two in the placebo group and one in the Progesterone group, attended emergency departments due to concerns over bleeding
  •  It is important to highlight that bleeding did not occur in these ladies because of placebo or because of Progesterone. If anything, Progesterone might prevent or at least limit the degree of haemorrhage. 
  • Mifepristone caused the bleeding in all cases. That is the nature of the drug and that is what it does. It interferes with the normal process of decidualisation (development of the endometrium in preparation for implantation and for maintaining pregnancy) by blocking Progesterone receptors, thereby preventing Progesterone from having its normal, intended, physiological effect.
  • Of the three who attended for emergency assessment, only one was deemed to be haemodynamically compromised so as to require a blood transfusion. She was in the placebo group and did not receive Progesterone. 
  • Because of these alleged safety concerns, the study was discontinued prematurely after only twelve participants had been enrolled. 
  • There were six in each group. When two withdrew early on, it left only five study participants in each group.
  • It is worth noting that, at the planned time of study end for each participant (before they were subsequently scheduled to proceed with their planned surgical abortion) at fifteen days, four subjects in the group treated with Progesterone still had ultrasonic evidence of continuing viable pregnancy (80%). Two subjects in the placebo group had evidence of continuing pregnancy (40%).
  • While the numbers are too small to reach any firm conclusion, it suggests that there may already have been an early trend towards a significantly improved foetal survival rate in the Progesterone treatment group. It is possible that this trend may have been noticed by the investigators who may have then decided to terminate the study because of a possible “unwelcome” outcome if it was allowed to continue.
  • The main investigator was previously an outspoken critic of the notion that Progesterone administration after Mifepristone had already been administered could be effective in some cases in preventing abortion. He may have had a biased mindset in the designing of the study (flawed with too low intended enrollment numbers) and in the decision to terminate the study (for fear that a previously held opinion might be undermined by study results). This, of course, is pure speculation, but may be worth considering.  
  • The main conclusion that the authors attempt to derive from this very limited and inconclusive study is that attempted Mifepristone antagonisation with Progesterone or expectant management (as per use of placebo in this study) after Mifepristone ingestion without subsequent Misoprostol administration in early pregnancy may be dangerous and may result in an increased risk of major haemorrhage. 

This conclusion is quite remarkable

  1.  First of all, as mentioned above, the bleeding was caused in all cases by Mifepristone.
  2. Haemorrhage, sometimes very heavy haemorrhage, is very common after its administration.
  3. In first trimester pregnancy, our experience in the UK is that some degree of haemorrhage is almost inevitable after Mifepristone and we have to warn and also reassure our patients who seek rescue treatment that they are very likely to experience some bleeding.
  4. Sometimes the haemorrhage starts within a matter of hours after Mifepristone, even before Progesterone can be administered.
  5. Sometimes it occurs several days later.
  6. Sometimes bleeding can continue for several days after Mifepristone.
  7. Sometimes it is mild and short-lived.
  8. Some women need to attend hospital for resuscitation measures, including blood transfusion.
  9. Most women don’t require emergency resuscitation.
  • It is interesting and noteworthy that the Creinin conclusion regarding the potential danger of taking Mifepristone and not following it with Misoprostol is based on the results from this very small and incomplete study.
  • In fact, the conclusion was based on the experience of three participants only (25% of the study co-hort). 
  • The authors ignored the fact that the improved foetal survival rate in the treatment group, although not statistically significant due to the small numbers, was based on the experience of six participants (50% of the study cohort). 
  • It is also worth noting that the author’s conclusion that Mifepristone administration without subsequent Misoprostol may increase the risk of significant haemorrhage is contradictory to the advice generally given to women who change their minds about proceeding with abortion after they have already taken Mifepristone. 

Those who do not accept the validity of abortion pill rescue tend to advise “expectant” management by doing nothing but watching and waiting rather than attempting to block the effects of Mifepristone with Progesterone. 

A leading example of someone who advises “expectant” management is Daniel Grossman, another outspoken opponent of abortion pill rescue programmes (Continuing pregnancy after Mifepristone and “reversal” of first-trimester medical abortion: a systematic review. Grossman et al, Contraception 92 (2015):206-211).

 Similarly, in the UK, before we established our rescue service, we tried to obtain support for the programme of offering reversal to mothers seeking help, by writing to the Royal College of Obstetricians and Gynaecologists, the Royal College of General Practitioners and NHS England. Our request for support was rejected by all bodies. 

One of the reasons given for not supporting our proposal was that “expectant” management was the recommended strategy in the highly unlikely event that a woman might change her mind about proceeding with abortion after taking Mifepristone. If the Creinin study conclusion is correct, although certainly not conclusive, then this would be the worst possible advice to offer mothers who change their minds after taking Mifepristone.

 Critique of the Mitchell Creinin Study by Dermot Kearney, Cardiologist and General Internal Physician working in the North-East of England. Current President of the Catholic Medical Association (UK), along with one other colleague, Dr Eileen Reilly in Glasgow. They have initiated and continued an Abortion Pill Rescue Programme in the UK since May 2020. (Written in an email to AFLO, January 19, 2021).