乌鸦传媒

No association was detected between COVID-19 vaccination and herpes zoster infection in a database study involving more than 2 million individuals in the U.S.

Using a self-controlled risk interval (SCRI) analysis, the risk of herpes zoster in the month after receiving a dose of COVID-19 vaccine was no different when compared with a control period 60 to 90 days following an individuals' last dose (incidence rate ratio [IRR] 0.91, 95% CI 0.82-1.01, P=0.08), reported Nisha Acharya, MD, MS, of the University of California San Francisco, and colleagues.

"There was no increase in risk of herpes zoster after COVID-19 vaccination when individuals were stratified by age, immunocompromised status, or type of vaccine administered," the group wrote in .

And when compared with influenza vaccination during a period before the pandemic, COVID-19 vaccination was associated with a slightly lower shingles risk (P<0.001), while no difference between the two vaccines were seen early on in the pandemic, according to the findings.

"Compared with previous work, our study had a larger sample, which enabled us to control for confounders that were not included in other studies, such as measures of healthcare use, history of zoster vaccination, and comorbidity-related risk factors for herpes zoster rather than weighted indices," Acharya and co-authors wrote.

Shingles following COVID-19 vaccination became a worry after some previous studies reported an elevated risk.

For example, an international involving more than 1 million vaccinated individuals detected a nearly twofold higher risk of shingles within 60 days of COVID-19 vaccination. And a large case-control study from Israel that involved nearly 900,000 vaccinated individuals and an equal number of comparators reported a 43% higher risk of herpes zoster after vaccination.

"Our study included a racially and ethnically heterogenous cohort compared with the studies conducted in Israel and may therefore better represent post-vaccination herpes zoster event risk in diverse populations," according to the researchers.

They noted that "several mechanisms have been suggested by which latent varicella zoster virus could be reactivated after vaccination," and added that "mRNA vaccines stimulate toll-like receptor signaling, a pathway involved in the latency and reactivation of varicella zoster virus, so there is a plausible mechanism by which the mRNA-based COVID-19 vaccines could lead to herpes zoster infection."

For their study, the researchers looked at the health records of 2,039,854 U.S. individuals in the Optum Labs Data Warehouse -- a national deidentified claims-based database -- who had received at least one dose of COVID-19 vaccination.

Individuals in the cohort had a mean age of 43 years, and were evenly divided by sex, while 66% were white, 10% were Hispanic, 7% Asian, and 7% Black. The vast majority received either the Pfizer (61%) or Moderna (35%) mRNA vaccines, while the remaining received the Johnson & Johnson vaccine.

In total, 1,451 had a herpes zoster infection in the defined risk period (30 days following a first or second dose) or control period (60-90 days after the last dose in the medical record).

Both a diagnosis and a prescription for a systemic antiviral medication (acyclovir, valacyclovir, or famciclovir) were required for a herpes zoster event. If there was no medication at the time of diagnosis, an antiviral prescription within 5 days after the first herpes zoster diagnosis was needed to be included. If they were already receiving antiviral medications at the time of diagnosis, either a dose increase or an oral or ophthalmic steroid medication within 5 days after the first herpes zoster diagnosis had to be recorded.

The researchers noted that the SCRI design "inherently controlled for time-invariant factors, such as age, demographic characteristics, chronic diseases, and long-term medication use, and has been used extensively to study vaccine safety."

In the SCRI analysis, no difference in risk for herpes zoster was seen for first or second dose, or for the type of vaccine. Furthermore, the researchers saw no increased risk for immunocompromised subgroups (IRR 0.73, 95% CI 0.53-1.02, P=0.06), nor for people age 50 and over (IRR 0.90, 95% CI 0.78-1.04, P=0.14).

For the influenza vaccine comparison, Acharya's group included 5,058,394 individuals who received the flu shot in a pre-pandemic period (January 2018-December 2019), along with 4,029,131 individuals who received the flu vaccine early on in the pandemic (March-November 2020).

Hazard ratios (HRs) here suggested a lower risk with COVID-19 vaccination versus influenza vaccination during the pre-pandemic period (HRs of 0.78 and 0.79 for dose one and two, respectively; P<0.001) and no difference compared with the flu vaccine during the early pandemic period (HRs of 0.89 and 0.91 for dose one and two; P≥0.05).

Limitations cited by the authors included the lack of granularity in claims data as well as the change in healthcare use during the pandemic. They also noted that the database used does not include patients on basic Medicare or Medicaid plans, nor the uninsured. Furthermore, some groups, such as Black individuals, were underrepresented in the COVID-vaccination cohort.

Comment