August 26, 2026

expert reaction to recombinant shingles vaccination and the risk of cardiovascular events

A study published in Nature Medicine looks at the recombinant shingles vaccine (Shingrix) and the risk of cardiovascular events.

Prof Ian Jones, Professor of Virology, University of Reading, said:

“This natural experiment study took advantage of the changeover in the United States from a live to recombinant shingles vaccine to ask if cardiovascular disease rates in the vaccinated population were similar prior to and after vaccination. As the type of person taking the vaccines does not change, behavioural differences are avoided and any differences seen should be solely the result of the vaccine change, aside from potential unknown confounders. The outcome is a clear association for cardiovascular events similar to that reported earlier for the same recombinant vaccine and protection against dementia onset. But how could a vaccine to a virus infection potentially benefit age related non-infectious disease? The vaccine works by generating antibodies to a single virus protein that block or reduce virus entry, giving protection against shingles, the re-awakened form of chickenpox. While the antibodies are specific to the virus, the vaccine also includes chemicals to kick start the immune system and it is possible that this immune jolt also clears up low level persistent inflammation which would otherwise build to later clinical disease, including neuronal and cardiovascular disease. The data amplify the obvious benefit of the shingles vaccination, which is free for older individuals, but it also suggests there may be an immune dimension to a number of age-related conditions and that focusing on boosting immunity could be generally beneficial.“

Dr Mark Russell, Clinical Senior Lecturer and Consultant Rheumatologist, King’s College London, said:

“This is a well-conducted study that uses a clever natural experiment to try to address one of the big problems with research into the benefits and risks of vaccinations. Namely, people who choose to be vaccinated are often different from those who do not, making it difficult to know whether differences in outcomes are due to the vaccine itself or other factors.

“In this study, the researchers took advantage of a relatively sudden switch in the type of shingles vaccine being used in the US. Similar changes in shingles vaccines have also occurred in the UK. The authors compared cardiovascular events (e.g. heart attacks) between people vaccinated with the previously used live shingles vaccine and those vaccinated a year later, when the recombinant (non-live) vaccine was predominantly used.

“Their finding of a 9% reduction in cardiovascular disease burden among people who received the recombinant vaccine could have important implications at a population level if the benefits are confirmed to be true. Although the benefits are relatively small on an individual basis, millions of people receive shingles vaccination, meaning that even a small cardiovascular benefit could translate into a substantial number of cardiovascular events prevented at a population level.

“The authors used several different approaches to test whether something other than the vaccine could have explained their findings. They matched the groups across lots of different characteristics that put people at increased risk of cardiovascular events (e.g. other health conditions). They showed that cardiovascular disease rates were similar between the groups before they received their vaccination. They also showed that events unrelated to shingles vaccination (known as a negative control) did not differ between the vaccinated groups.

“Together, these findings strengthen the possibility that at least some of the observed difference in cardiovascular events could be due to the recombinant vaccine. Interestingly, the benefits were observed relatively early on after receiving the vaccine and were not entirely explained by better prevention against shingles infection. This raises the possibility that the recombinant vaccine itself (or components of the vaccine) might provide some protection against cardiovascular events.

“However, it is important to note that this is an observational study, not a randomised trial, and it cannot prove definitively that the recombinant vaccine caused the reduction in cardiovascular disease. The groups in this study were vaccinated a year apart, during which time other changes could have occurred. For example, the introduction of a new vaccine is often accompanied by changes in vaccination recommendations, which can influence which people come forward for vaccination or not. Similarly, early adopters of a new type of vaccine might differ from those receiving established versions of the vaccine. Additionally, when using real-world data, there is always the potential for unmeasured differences (known as confounding factors) that could explain some or all of the observed differences between groups. For example, the authors noted that they were unable to fully account for factors such as socioeconomic status and some aspects of lifestyle.

“One of the key ways to overcome this problem and establish whether an association is causal is through a randomised trial, where participants are randomly allocated to receive a vaccine or placebo. Interestingly, cardiovascular benefits were not clearly observed in the large, randomised trials (ZOE-50 and ZOE-70) that evaluated the effectiveness of the recombinant shingles vaccine. Serious cardiovascular adverse events (e.g. heart attacks) occurred at similar rates in the vaccine and placebo groups in these trials. However, these trials were not designed to detect the relatively small reduction in cardiovascular risk reported in the current study, and so this does not rule out a genuine protective effect.

“Overall, this is an important piece of evidence highlighting a potential link between receiving a recombinant shingles vaccine and a reduced burden of cardiovascular disease. It provides an intriguing signal that deserves further evaluation.”

Dr Anne Suffel, Research Fellow, London School of Hygiene and Tropical Medicine (LSHTM), said:

“Corsi-Zuelli et al. present an interesting and rigorously designed natural experiment comparing the incidence of cardiovascular events in individuals predominantly vaccinated against shingles with the live attenuated vaccine versus individuals predominantly vaccinated with a recombinant vaccine. This design compared people vaccinated from April 2017-September 2017 when 98.6% of people received the live vaccine to people vaccinated from April 2018-September 2018 when 93.5% of people received the recombinant vaccine.

“Through their inclusion of a completely vaccinated population alone, the authors account for healthy vaccinee bias, i.e., that healthier people are more likely to take up a vaccine. However, individuals who receive the recombinant vaccine may still systematically differ from those who were eligible for the live attenuated vaccine. This may be due to changes in recommendations from the Advisory Committee on Immunization Practices. The authors accounted for these potential differences in lifestyle factors and comorbidities in their methods, however it may not have eliminated the impacts of unmeasured confounding.

“The authors conclude that the reduction in cardiovascular events is mainly due to immune and endothelial changes induced by the recombinant vaccine, citing a single study based on a natural experiment design which suggested the live-attenuated vaccine did not reduce risk of heart disease or stroke. However, a recent systematic review [1] showed a reduction in cardiovascular events for both the recombinant and live attenuated vaccine based on several studies using both randomised and robust observational designs.

“The authors utilize shingles cases as a positive control outcome to demonstrate the differences in protective efficacy against shingles between the two vaccines, however they dismiss the finding of greater protective efficacy as insufficient to explain the difference in cardiovascular events. If both vaccines are protective against cardiovascular outcomes, and the recombinant vaccine more so in line with its greater efficacy, this provides a simpler explanation for the study findings.

“A mechanism based on a greater reduction of shingles cases also aligns with evidence indicating that herpes zoster may trigger stroke and myocardial infarction [2]. The authors argue that further research on the mechanism of protection for cardiovascular outcomes is needed, which may still be the case, but this should be informed fully by the existing evidence base.“

1-    https://academic.oup.com/eurheartj/article/46/Supplement_1/ehaf784.3633/8312335

2-    https://academic.oup.com/jid/article/218/suppl_2/S102/5105944

Dr Hugo Pedder, Research Fellow in Statistical Modelling at the Bristol Medical School, University of Bristol, said:

“In terms of methodology this study does a good job of trying to analyse observational data in a way that tries to account for many of the limitations that observational analyses usually encounter. These are primarily addressed via the natural experiment design.

“The authors also calculate an “E-value”, a method used to explore how strong an imbalance in baseline characteristics would have to be to cause the effect observed. The value found by the authors (E=1.42) is lower than the potential effect of some strong confounders that were not collected or adjusted for within the study. If these variables were not well-balanced by the natural experiment design (i.e. if they differed between the recombinant and live vaccine groups) then they could plausibly explain the relationships observed.

“My main concern is regarding the reduction in hazard ratios over time (first 3.5 vs last 3.5 years). The paper (and press release) interpret this as a time-dependent effect of the recombinant vaccine, and state that the effect was “strongest in the years soon after vaccination”, but this may simply be the statistical artifact of more frail patients experiencing early events in the live vaccine group.
Residual confounding also produces this exact same observed statistical effect in the HR, though it can be somewhat challenging to explain. If an unmeasured characteristic that increases the risk of CHD is more common in the live vaccine group at baseline, those patients would experience CHD events early and so lead to a larger HR in early years, that then returns towards 1 once those more susceptible patients have experienced an event (and are no longer included in the last 3.5 years HR calculation).

“Fundamentally the data are still observational and there may still be confounding remaining between the live and recombinant groups that explains the treatment effect seen in the study, even though the ways in which that could occur are less likely than in more traditional observational studies that do not rely on a natural experiment.

“A randomised trial would be the best way to confidently answer whether recombinant vaccines do provide the benefits seen here versus live vaccines, but until such a trial is done this study provides some of the best quality observational evidence supporting an effect of recombinant vs live vaccines for cardiovascular outcomes.”

* ‘Recombinant shingles vaccination and the risk of cardiovascular events’ by Fabiana Corsi-Zuelli et al. will be published in Nature Medicine at 10:00AM UK time on Wednesday 26 August 2026, which is when the embargo will lift.

DOI: 10.1038/s41591-026-04606-0

Declared interests

Prof Ian Jones: “I state no conflicts of interest. “

Dr Mark Russell: “I have no conflicts relevant to this.”

Dr Anne Suffel: “I have no conflicts of interests to report.”

Dr Hugo Pedder: “For my previous employer (a Health Economics and Outcomes Research consultancy) I worked on a project for GSK looking at dementia and the Shingrix vaccine (recombinant vaccine). I did not receive payment from GSK directly.”

This Roundup was accompanied by an SMC Briefing.