Sunday, January 19, 2014

Detouring around evidence: CDC's new recommendations on treatment of anthrax exposures in pregnant and post-partum women

CDC just published new guidelines for treatment of anthrax exposure and anthrax disease in pregnancy and the postpartum period.  It is said the guidelines are based on meetings  "of national subject matter experts to review key clinical elements of anthrax prevention and treatment for pregnant, postpartum, and lactating (P/PP/L) women."

A lot of effort was made to involve 77 experts, convene 5 workgroups, and undertake a series of conference calls as well as a 2.5 day meeting. Then CDC created the new guidelines.

Sounds pretty impressive.  The only problem is that they had very little solid information to talk about, since the data on treatment of anthrax exposure is limited to 1-2 published studies of the 10-30,000 people who received antibiotics after the anthrax letters.  One hundred ninety-eight of them also received anthrax vaccine, most receiving 3 doses on an every two week schedule while taking antibiotics.  I have seen nothing published about any pregnant women who participated in the CDC post-exposure treatment trial of anthrax vaccine.

The only data on anthrax vaccine in pregnancy comes from a) a large epidemiological study done by Ryan et al. of female soldiers vaccinated during pregnancy, and b) a poorer quality study by Weisen et al, which looked at pregnancy outcomes after anthrax vaccinations, but had a very high loss to followup, including all pregnancies referred outside the military base when they became complicated.  Weisen's abstract admits, "this study did not have sufficient power to detect adverse birth outcomes," yet the supposed reason for the study was to detect them. However, his paper ends with a misleading conclusion, "Anthrax vaccination had no effect on pregnancy and birth rates or adverse birth outcomes."

Ryan's study found that for women vaccinated for anthrax during the first trimester, birth defects were increased by 18%.  Her full-text report can be found here.  I wrote a detailed analysis of her paper, and the context in which her study was done.  Her early data led to major policy changes regarding careful avoidance of anthrax vaccine during potential pregnancy. They furthermore led to a change in the anthrax vaccine label, changing the vaccine from Category C for pregnancy (Animal reproduction studies have shown an adverse effect on the fetus and there are no adequate and well-controlled studies in humans, to Category D (positive evidence of human fetal risk), where it remains.  Could the new guidelines be part of an attempt by CDC to remove the Category D designation?

I mention this because the latest anthrax vaccine label change downplayed the existing adverse event data on anthrax vaccine.  In one case, the label reduced the rate of serious adverse events in CDC's pivotal trial of anthrax vaccine by a factor of ten, from 15% of subjects to 1.5%. Although analysis of this trial was completed in 2009, the completed analysis has never been published, a hint that the results may be highly unfavorable for anthrax vaccine.

Of 14 known first-trimester pregnancies that received anthrax vaccine in this CDC trial, there were 2 miscarriages, a fetal death and an infant born with a club foot:  first trimester anthrax vaccination was associated with a 29% rate of serious adverse outcomes.

To return to the new guidelines, doctors can earn CME credit when they learn that women exposed to anthrax spores, but not sick, (where "high risk of exposure" is poorly defined) should be treated with both antibiotics and anthrax vaccine, whether pregnant, lactating or not. 

However, not a shred of evidence exists that vaccine adds to the 100% protection offered by antibiotics for exposure, if there are no signs of anthrax disease, since the only evidence was obtained after the anthrax letters.  Sure, next time it could be different.  

I would suggest that if the anthrax next time is resistant to available antibiotics that have good CNS penetration, then probably vaccine or (preferably) antitoxin/monoclonal antibodies should be used. Antitoxin works right away, and in some cases has also functioned to provide long-lasting antibodies against anthrax, like a vaccine. The existing vaccine, on the other hand, takes weeks to generate high antibody levels. This could be the basis for simple, straightforward guidelines that appropriately respond to the specifics of an anthrax spore exposure.

Neither Dr. Ryan nor Dr. Weisen were included in forming these guidelines. It took a lot of people who lacked experience regarding anthrax and pregnancy, a lot of meetings, a lot of money and nearly two years to create guidelines that don't make good medical sense.

In fact, the guidelines fail to offer any scientific justification for using the anthrax vaccine. What did the 77 experts and all those meetings actually conclude?  All their guidelines say is that a 2008 ACIP report authored by several of the same 77 experts from CDC said the vaccine was okay in pregnancy. The 2008 guideline authors blew off the admitted increased risk in pregnancy with the following fluff:
"... In addition, late recognition of pregnancy, a moderate risk factor for many birth defects, including ASD, might explain the number of women vaccinated during their first trimester. After review of these data and discussions with the authors of this study, ACIP concluded that AVA is safe to administer during pregnancy but recommended that pregnant women defer vaccination unless exposure to anthrax poses an immediate risk for disease."  
This is a "blame the victim" strategy.  If you got vaccinated while you were pregnant, the assumption is that you did not know you were pregnant, and therefore you were in a group with a higher than normal risk of adverse birth outcomes. And that was the problem, not anthrax vaccine!  What a convenient, non-data-driven assumption.

However, it was military policy to vaccinate women who thought they might be pregnant but had not proven it, before Ryan's research came out in 2001.  

CDC, what happened to "First, Do No Harm" and "Second, use your head"? 

No comments: