VZV-containing vaccines and hospitalization for herpes zoster: careful optimism

Abstract

The introduction of a universal varicella vaccine program for children in the USA in 1996 sparked a concern for an increase in the incidence of herpes zoster (HZ) in older adults. Some experts posited that adults exposed to children with naturally acquired active varicella receive an immune boost that could help protect them from HZ, the “exogenous boosting hypothesis” [1–5]. Ogunjimi et al. found a very modest magnitude of boosting of zoster-specific T cells in only 25% of grandparents exposed to grandchildren with chickenpox that was diminished at 1 year [6]. It is not clear that loss of this small level of boosting would have clinically significant effect on immunity to prevent reactivation of zoster. Also, it is not known how actively or frequently a person’s immune system is endogenously boosted by failed attempts at zoster reactivation that do not result in HZ. However, this theory remains controversial, and attempts to evaluate the impact of chickenpox vaccination on HZ incidence have led to mixed results at best, as has been illustrated in a systematic review and meta-analysis by Harder et al. [7]. In the current study, Pham et al. evaluate the impact of two policies relating to VZV vaccination on the rate of HZ in different age groups: the recommendation of a booster dose of VZV vaccine at age 4–6 years in 2006, and the recommendation of a live, attenuated varicella vaccine at age 60 in 2008 [8]. Using Nationwide Inpatient Sample discharge data, the authors compared the average annual rates of hospitalization for HZ across age groups between a period prior to the above policy changes (2001–2005) and a period afterwards (2012–2015). They describe an overall increase in the hospitalization rate for HZ through 2008, followed by a decline between 2009 and 2015 which is the period following the implementation of the two-dose varicella vaccine schedule. Conceptually, a decrease in herpes zoster hospitalization rates may be caused by a (1) decrease in the incidence of HZ, (2) reduced severity of HZ whether due to greater underlying immunity or better outpatient treatment, (3) a higher threshold disease severity before triggering admission, or (4) a reduction of coding for the HZ diagnosis in hospital claims. HZ incidence had been on the rise in prior decades. A retrospective study of Medicare claims for close to 3 million beneficiaries older than 65 reported an ageand sex-adjusted increase of 40% in the incidence of HZ between 1992 and 2010; however, there was no observed change in the rate of increase coinciding with the introduction of the varicella vaccine [9]. Similarly, a population-based cohort study in Olmsted County, MN, using medical records data from 1945–1960 and 1980–2007 found an ageand sex-adjusted yearly increase of HZ incidence of fourfold across the six decades, at a rate of 2.5% per year, unaffected by the introduction of the varicella vaccination program [10] Therefore, more plausibly, a decrease in the proportion of persons with HZ being hospitalized would explain the observed decrease in hospitalization for HZ. Barring some indications, the vast majority of cases of HZ in the US are treated without hospital admission, as has been reported by multiple studies examining the cost of care and resource utilization associated with cases of HZ; some reported a hospitalization rate as low as 1% [11–15]. Yawn et al. 2007 found a rate of 12.1 cases/1000 person-years in the Olmstead cohort in the 80 + year population. In this study, the baseline period (2001) rate in the 80 + population was 115 cases/100 000 person-years * Elie A. Saade [email protected]