To the Editor: Victor et al. (Oct. 25 issue)1 report on theirstudy of hepatitis A vaccine as compared with immune globulinfor postexposure prophylaxis. I applaud the recommendation forthe postexposure use of hepatitis A vaccine to provide long-termprotection. However, I am concerned that the statistical criteriaused for determining substantial inferiority detract from thestudy findings, which indicate a 30% higher rate of hepatitisA after exposure in the group receiving vaccine. The findingssupport the biologically plausible position that the administrationof immediately available antibodies through immune globulinhas a time-dependent advantage over the delayed developmentof vaccine-generated antibodies. Although the statistical findingsallow the rejection of the null hypothesis, they cannot be usedto conclude that the difference demonstrated is not real. Dataprovided in the text and tables indicate that in the vaccinegroup there were more cases of hepatitis A, more severe symptoms,higher alanine aminotransferase levels, more cases with viralRNA, and more cases occurring in the second week after exposure.These findings are consistent with current knowledge of passiveversus active immunity. An alternative recommendation shouldbe considered: the use of both immune globulin and vaccine forpostexposure situations.
Thomas G. Betz, M.D., M.P.H. Texas Department of State Health Services Austin, TX 78756 tom.betz{at}dshs.state.tx.us
References
Victor JC, Monto AS, Surdina TY, et al. Hepatitis A vaccine versus immune globulin for postexposure prophylaxis. N Engl J Med 2007;357:1685-1694. [Free Full Text]
To the Editor: Victor et al. conclude that immune globulin isnot inferior to vaccination for postexposure prophylaxis againsthepatitis A. The study was conducted in a developing country,and the conclusion may not be applicable to developed countries.For instance, the seroprevalence rate for hepatitis A is nearlyzero in children in Taiwan.1 An outbreak of hepatitis A in sucha population may affect older persons, who are at risk for severeand even fatal infections.2 The study by Victor et al. showedthat recipients of immune globulin had lower average peak alanineaminotransferase levels than did vaccine recipients. This protectiveeffect should be taken into consideration in developed countries.Furthermore, the study excluded 3110 study subjects becausethey were immune or already IgM-positive at enrollment. In thereal world, we cannot check for hepatitis A antibody in allexposed persons. It would be interesting to know whether immuneglobulin can modify the disease severity in these seropositivepersons.
Ping-Ing Lee, M.D. National Taiwan University Hospital Taipei 10002, Taiwan pinging{at}ntu.edu.tw
References
Tzen KT, Chang MH, Tsen YJ, Lee CY, Chen DS. Hepatitis A virus infection in Taipei City in 1989. J Formos Med Assoc 1991;90:138-140. [Medline]
Kyrlagkitsis I, Cramp ME, Smith H, Portmann B, O'Grady J. Acute hepatitis A virus infection: a review of prognostic factors from 25 years experience in a tertiary referral center. Hepatogastroenterology 2002;49:524-528. [Medline]
The author replies: In response to Betz: although the relativerisk indicated an approximately 30% higher risk among recipientsof vaccine than among recipients of immune globulin, absoluterisks were low, and absolute risk differences were never greaterthan 1.5%. Still, in our report, my colleagues and I draw attentionto the potential relevance of these differences for personswho are likely to have severe illness if infected with the hepatitisA virus. Indeed, the Advisory Committee on Immunization Practicesconsidered these small differences when updating recommendationsfor postexposure prophylaxis.1
Several of Betz's characterizations of our findings merit clarification.It is arguable whether the proportions of vaccine recipientsand immune-globulin recipients with detectable hepatitis A virusRNA (62% and 56%, respectively; P=0.761) are different; theymight equally well be considered equivalent. As we noted, thestudy was not designed to measure disease severity. Furthermore,alanine aminotransferase levels should be interpreted with caution,since measurements were performed once at nonstandardized timepoints during the course of the illness; in addition, differenceswere limited to levels among children. Finally, although inthe modified intention-to-treat population, the relative riskwith vaccine as compared with immune globulin was lower in thefirst week after exposure than in the second week, in the per-protocolpopulation, it was higher in the first week after exposure thanin the second week (1.73 vs. 1.30).
Betz's statement that "immune globulin has a time-dependentadvantage over the delayed development of vaccine-generatedantibodies" appears to be based on the assumption that hepatitisA vaccine used after exposure operates solely through antibody-dependentmechanisms. However, antibody-independent T-cell effects maybe important, and vaccination after exposure might modulatean already initiated immune response in participants with incubatinghepatitis A virus — something immune globulin is unlikelyto do.
Unlike Lee, we believe that our study results are applicableto persons in developed countries. Our analyses of relativeefficacy were conducted among only those persons found to havebeen susceptible at the time of receipt of vaccine or immuneglobulin, precisely so that conclusions would be applicableto susceptible persons in any population. Moreover, the generalizabilityof the results is supported by studies of the transmission dynamicsof hepatitis A virus in the population, conducted before implementationof the trial.2,3
John C. Victor, Ph.D., M.P.H. Program for Appropriate Technology in Health Seattle, WA 98107 cvictor{at}path.org
References
Update: prevention of hepatitis A after exposure to hepatitis A virus and in international travelers: updated recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 2007;56:1080-1084. [Medline]
Victor JC, Surdina TY, Suleimenova SZ, Favorov MO, Bell BP, Monto AS. Person-to-person transmission of hepatitis A virus in an urban area of intermediate endemicity: implications for vaccination strategies. Am J Epidemiol 2006;163:204-210. [Free Full Text]
Victor JC, Surdina TY, Suleimenova SZ, Favorov MO, Bell BP, Monto AS. The increasing prominence of household transmission of hepatitis A in an area undergoing a shift in endemicity. Epidemiol Infect 2006;134:492-497. [CrossRef][Medline]
Previous
