Medical Abortion Deaths Update

An update on the reports of deaths following medical abortion with misoprostol and mifepristone (RU-486, Mifeprex). From the Centers for Disease Control and Prevention meeting on clostridial infection:

At the meeting, researchers discussed the latest information in order to draft a research agenda for C. sordellii and C. difficile, another emerging infection associated with toxin-mediated sepsis that has also affected pregnant women. About half of the day was spent discussing each organism.

Dr. Marc Fischer, a medical epidemiologist at the CDC, summarized the published literature to date on infections involving C. sordellii, a gram-positive anaerobic bacillus that resides in soil and colonizes the gastrointestinal and/or genital tracts of healthy humans. In various case reports and series, the organism has been identified in cases of pneumonia, endocarditis, arthritis, peritonitis, corneal ulcer, and bacteremia, as well as in wound infections among patients with necrotizing fasciitis, tissue allograft infections, neonatal omphalitis, postpartum endometritis, and episiotomy infections.

Some interesting information on genital tract infection (and toxic shock-like syndrome) cases from a review of the literature (1977 to present):

Childbirth -- 8 infections (C. sordellii)

Medical abortion -- 5 cases (C. sordellii)

Between 1977 and 2001, C. sordellii genital tract infections and toxic shock-like syndrome were reported in 10 women, among whom the preceding events were childbirth (8) and medical abortion (1), reported from Canada in 2001. Another four cases were identified between 2003 and 2005, all involving women who had undergone medical abortions using the common "off-label" regimen of 200 mg oral mifepristone followed by 800 mcg vaginal misoprostol....

The four recent cases were all previously healthy women from California who developed symptoms including tachycardia, hypotension, vomiting or diarrhea, and abdominal pain within 5 days of taking mifepristone. Clinical laboratory findings in three of the patients included leukemoid reaction in all three, hemoconcentration in two, and thrombocytopenia in two. All died within a day of hospitalization. The clinical and pathologic findings in these cases were similar to those of the 10 previously reported cases, Dr. Fischer commented.

[More on medical abortion drug regimens here.]

Still under investigation, 3 additional cases of fatal toxic shock-like syndrome following medical abortion:

Each of these differs in various ways from the previous five: One, in a woman who had taken oral mifepristone followed by vaginal misoprostol, was associated with C. perfringens, not C. sordellii. A second case, also of C. perfringens, involved the use of misoprostol with the cervical dilator Laminaria, not mifepristone. The third, although initially reported as being associated with a medical abortion, could not be confirmed as such. Moreover, investigation has shown pathologic findings consistent with appendicitis, serositis, and pneumonia....

Spontaneous abortion -- 3 cases (C. sordellii/C. perfringens)

[T]here have been three reported cases of toxic shock-like syndrome following spontaneous abortion, all involving C. sordellii. One of these patients was coinfected with C. perfringens. Another patient, in whom the C. sordellii did not possess the genes encoding the lethal toxin, was the only one who survived.

So, what does all this mean? Is there a connection between RU-486/misoprostol medical abortions and clostridial infection? Based on the available data, I don't see any evidence for it [ACOG hasn't issued an opinion yet; I'll update when it does]. Then what about pregnancy? Does being pregnant--irrespective of outcome--predispose one to developing clostridial infections? Recall that pregnancy is an immunodepressed state, and, along the same lines,

...mifepristone is a potent inhibitor of both progesterone and glucocorticoid receptors. As such, mifepristone may impair host immune responses and predispose women to lethal infections caused by toxigenic C. sordellii and other pathogens that exist normally in low numbers in the reproductive tracts of many women....

Again, the evidence doesn't support this hypotheses. Of the tens of millions of pregnancies over the past ~30 years, we have ~24 cases of clostridial infection. Moreover, as Dr. Toub notes in response to this post:

Vaginal misoprostol had also been used very successfully in conjunction with methotrexate for medical abortion before the approval of mifepristone, and I'm not aware of any increase in these infections with the methotrexate regimen. Yes, there is a possibility that mifepristone could be affecting the immune system. But methotrexate definitely affects the immune system, as a systemic chemotherapeutic agent.