A large observational study exploiting a natural experiment in US vaccine policy found that recipients of the recombinant herpes zoster vaccine Shingrix had a 9% lower cardiovascular disease burden over seven years compared with recipients of the older live attenuated vaccine Zostavax. The study was published August 26, 2026 in Nature Medicine by researchers led by Paul J. Harrison and Maxime Taquet at the University of Oxford.
Using electronic health records from the TriNetX US Collaborative Network, researchers exploited the rapid US transition from Zostavax to Shingrix after October 2017. They compared adults aged 60 and older vaccinated in April–September 2018, when Shingrix accounted for 93.5% of vaccinations, with those vaccinated during the same period in 2017, when Zostavax accounted for 98.6%. After propensity score matching on 83 covariates, each cohort comprised 36,460 participants.
Over seven years, the predominantly Shingrix cohort had a 9% lower burden of the primary composite of ischemic heart disease, ischemic stroke, and heart failure (restricted mean time lost ratio 0.91; 95% CI 0.88–0.95; p<0.001). Ischemic heart disease burden was 10% lower and heart failure burden 12% lower in both sexes, while ischemic stroke burden was reduced by 12% in men but not significantly across the overall population.
The cardiovascular associations were strongest during the first 3.5 years and attenuated later in follow-up. Sensitivity and active-comparator analyses supported the main finding, although the authors cautioned that residual confounding cannot be excluded.
One possible mechanism involves the AS01 adjuvant used in Shingrix but not Zostavax. Previous research suggests AS01-adjuvanted vaccination can induce sustained epigenetic changes in circulating monocytes, including reduced interleukin-6 (IL-6) responses. The authors argued that differences in shingles prevention alone were unlikely to account for the cardiovascular association, although the proposed immune mechanism remains experimentally unconfirmed.