Antibiotic Exposure and Inactivated Influenza Vaccine Effectiveness

Background: Gut microbiota catalyze immune responses to inactivated influenza vaccines (IIV). Antibiotics have been shown to impair anti-influenza immune responses in animal models and, experimentally, in certain healthy adults, but real-world evidence on this topic is lacking.

Objectives: To evaluate the risks of severe influenza in children and adults following antibiotic exposure around the time of IIV administration.

Methods: We conducted a retrospective cohort study of commercially insured children and adults (ages 1-64) using claims data from one US state (2015-2018). Eligible subjects had ≥6 months of baseline data prior to receipt of IIV. Antibiotic exposure was defined by pharmacy claims and classified by timing and type. The primary exposure window was 29 days before to 2 days after vaccination (peri-vaccine); secondary exposure windows extended up to 365 days prior to vaccination. Severe influenza was defined by ICD-9/10-CM codes from primary discharge diagnoses of hospital or emergency room (ER) encounters, excluding events within 14 days after vaccination. Odds ratios with 95% confidence intervals (CIs) were produced from generalized estimating equation models adjusted for age, sex, year, comorbidities, prior influenza diagnosis, baseline ER or hospital encounter, and number of baseline outpatient encounters. Subgroup analyses were performed by age group and antibiotic type.

Results: A total of 118,292 subjects (median age 40, 58% female) received IIV. Crude rates of severe influenza were higher after peri-vaccine antibiotic exposure (0.23% antibiotic exposed, 0.09% unexposed). Peri-vaccine antibiotic exposure was associated with greater risk of severe influenza (aOR 1.71, 95% CI 1.04, 2.82). Greater risks of severe influenza were observed after more recent vs. more remote exposure (0-3 month prior: aOR 1.6; 3-12 month prior: aOR 1.0-1.1). Greater risks of severe influenza were also observed after peri-vaccine exposure to broad-spectrum penicillins (aOR 3.19) vs. narrow-spectrum penicillins (aOR 2.66) or cephalosporins (aOR 1.32). Peri-vaccine antibiotic-associated risks of severe influenza were higher among subjects ages 9-17 and 18-49 compared to younger and older age groups.

Conclusions: Antibiotic exposure around the receipt of IIV was associated with increased risks of severe influenza, with stronger effects after more recent exposures and exposure to broader-spectrum agents. These results suggest that antibiotics may reduce IIV effectiveness via a microbiota-dependent mechanism, although residual confounding is also possible. While these findings warrant replication, they could have implications for the optimal timing of IIV administration after antibiotic exposure.