Mechanistic Evidence

The committee identified three publications reporting encephalitis or encephalopathy after administration of a hepatitis B vaccine. The publications did not provide evidence beyond temporality (Deisenhammer et al., 1994; Manna et al., 1996; Yang et al., 2006). The publications did not contribute to the weight of mechanistic of evidence.

Weight of Mechanistic Evidence

The committee assesses the mechanistic evidence regarding an association between hepatitis B vaccine and encephalitis or encephalopathy as lacking.

Causality Conclusion

Conclusion 8.1: The evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and encephalitis.

Conclusion 8.2: The evidence is inadequate to accept or reject a causal relationship between hepatitis B vaccine and encephalopathy.

SEIZURES

Epidemiologic Evidence

The committee reviewed four studies to evaluate the risk of seizures after the administration of hepatitis B vaccine. Two studies (Dobson et al., 1995; Niu et al., 1996) were not considered in the weight of epidemiologic evidence because they provided data from passive surveillance systems and lacked unvaccinated comparison populations. One controlled study (Zipp et al., 1999) had very serious methodological limitations that precluded its inclusion in this assessment. The study by Zipp et al. (1999) was a letter to the editor judged by the committee to have insufficient methodological detail.

The one remaining controlled study (Lewis et al., 2001) contributed to the weight of epidemiologic evidence and is described below.

Lewis et al. (2001) conducted a cohort study in children enrolled in the San Francisco Medical Center of Northern California Kaiser Permanente. Patients born from November 1991 through April 1994 were included in the study; premature and low-birth-weight infants, and infants with diagnoses (e.g., sepsis, congenital infection, hematologic disorder, cardiac disease, neurologic disease, and lung disease) that made vaccination less likely in the opinion of the authors were excluded. A total of 5,655 patients met the in-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement