Leads from the MMWR. Suboptimal response to hepatitis B vaccine given by injection into the buttock.

Hepatitis B (HB) vaccine was licensed in November 1981 as a highly immunogenic and ef­ fective vaccine against hepatitis B virus (HBV) infection. Large studies before licensure demonstrated, with one exception, that the vaccine induced antibody* in over 90% of healthy adult recipients of the three-dose series ( 1-3). The one exception, in which only 85% of recip­ ients responded to vaccination, was later shown to be caused by partial freezing of the vac­ cine during shipment (4). Since vaccine licensure, however, the vaccine manufacturer (Merck, Sharp & Dohme) and CDC have received reports of suboptimal response to vaccine in the health-care personnel of a number of hospitals and other vaccine users. Two such examples, in which only 82% and 68% of normal adults responded to vaccination, have recently been published (5-6). Initial in­ vestigations of these and other reports by the manufacturer and by CDC included site visits, repeat serologic testing of vaccine recipients to confirm poor response, assays of residual vaccine for evidence of freezing and for retention of potency, and review of vaccine lots used. These investigations generally confirmed suboptimal vaccine response but failed to identify any specific cause. The investigations did indicate that, in many such instances, vaccine had been given by buttock (gluteal) injection, in contrast to the arm (deltoid) injection used in all prelicensure vaccine studies. Two recent investigations, one by the vaccine manufacturer and the other by CDC, indicate that site of vaccine injection is important in explaining suboptimal response to vaccine in many vaccine programs. Both studies were retrospective telephone surveys of hospitals or hemodialysis units that had vaccinated and then serotested significant numbers of persons after vaccination. Vaccine manufacturer's study: In December 1984, the vaccine manufacturer surveyed two groups of vaccine users: over 90 hospitals that had contacted the manufacturer reporting suboptimal vaccine response and an additional 1 2 hospitals known to have conducted large vaccination programs and to have done postvaccination testing. The telephone survey verified the exact number of persons completing vaccination and the number failing to respond to vaccine and determined the vaccine injection site. Injection site for the hospital was classified as arm if over 90% of persons received vaccine in the arm; buttock if over 90% received vac­ cine in the buttock; and mixed for all others. In both surveys, vaccine response rate was significantly higher in hospitals using arm injec­ tion than in those using buttock injection (Table 1). Among hospitals that reported suboptimal vaccine response, the pooled response rate for vaccinees was 88% in hospitals using arm in­ jection and 73% in those using buttock injection (p < 0.01). Among the 1 2 other hospitals, re-