Waning Of Vaccine-induced Immunity Research Articles

Chickenpox Outbreak in a Highly Vaccinated School Population

To the Editor.—We commend Tugwell et al1 for reporting on the important issue of chickenpox outbreaks in a highly vaccinated school population. However, the authors present conclusions on the time since vaccination as a risk factor with a measure of certainty not supported by the data.Varicella is a highly contagious disease, and varicella cases can be infectious before rash onset. Exposure may occur on school buses, in lunchrooms, in hallways, and on playgrounds; therefore, it is difficult to ascertain the likelihood of exposure. The authors presume that there was zero exposure for children in the unaffected classrooms and that exposure was maximal in each of the affected classrooms. These assumptions do not reflect the reality of exposure through population mixing.It is important to know the degree of exposure in each class before drawing conclusions regarding duration of vaccine protection. Because the authors stratified classes based on the risk of exposure by excluding unaffected classes, that method needs to be applied consistently. We would suggest that classes be analyzed by the number of index cases and by the vaccination status of the index cases (because unvaccinated cases are more contagious than vaccinated cases).2 Moreover, we would have excluded the 4 cases with household exposures, because the risk of transmission is higher in households.Age at varicella disease is highly correlated with risk of exposure. If multiple or unvaccinated index cases occurred predominantly among children in grades 1 to 3 versus those in kindergarten, more secondary cases would be expected in these classes. It then would seem that increased time since vaccination rather than increased risk of exposure was the risk factor for vaccine failure. This association between time since vaccination and risk of exposure was evident in follow-up studies in which the risk of breakthrough was highest 2 to 5 years postvaccination (years of higher likelihood of varicella zoster virus exposure) and declined thereafter.3 If waning immunity was a key factor, breakthrough rates would have increased over time.The abrupt increase in attack rates is not consistent with waning of vaccine-induced immunity. Because time since vaccination is confounded by age at exposure and correlated with age at vaccination, not taking that into consideration when interpreting findings would lead to policy change that is not necessary. Given the above biases, the definitiveness of the conclusions regarding the waning of immunity is not warranted.