There are only a few published peer-reviewed studies examining the safety of the anthrax vaccine in humans. The studies discussed below, with the exception of the Ft. Detrick studies, administered only the anthrax vaccine and were not intended to examine the effects of multiple vaccinations. The committee notes a recent literature review (Demicheli et al., 1998) on anthrax vaccine studies conducted according to the Cochrane Collaboration guidelines for systematic reviews of health care interventions. Only the Brachman study (described below) met the Cochrane criteria for prospective randomized or quasi-randomized studies of a protective antigen anthrax vaccine.
During the development of the anthrax vaccine, several studies examined adverse reactions in humans. These studies used early versions of the culture filtrate (protective-antigen) vaccine. Wright and colleagues (1954) described the reactions of 660 persons who received a total of 1,936 injections. They found that 0.7% of the vaccinated subjects reported systemic reactions—typically consisting of mild muscle aches, headaches, and mild-to-moderate malaise lasting 1 to 2 days. Significant local reactions—typically swelling (5–10 cm in diameter) and local pruritus (itching)—were reported for 2.4% of the injections. The incidence of local reactions increased with the number of previous injections. Two additional early studies also showed low rates of mild, brief local reactions (Darlow et al., 1956; Puziss and Wright, 1963). There is no long-term follow-up reported on the subjects in these studies.
Brachman and colleagues (1962) conducted the only randomized clinical trial of vaccination with a protective-antigen anthrax vaccine. Although the vaccine used in this study was similar to the vaccine currently available in the United States in that it was a protective-antigen vaccine, the manufacturing process has since changed and a different strain of anthrax bacillus is now used (GAO, 1999a).
The clinical trial was conducted among 1,249 eligible workers1 at four goat hair processing mills in which some raw materials were contaminated by the anthrax bacillus. After the initial series of three injections, the study had to be terminated at the largest mill, which employed nearly half of the subjects, because of an outbreak of inhalation anthrax that required the immunization of all employees. At the remaining mills, 480 participants completed the series of injections (230 of whom were randomized to active vaccination and 250 of whom were randomized to receive placebo injections) and 81 participants did not complete the series of injections.2 The study subjects did not know