Antibiotic Use and Vaccine Antibody Levels

Abstract

Background: The majority of children are prescribed antibiotics in the first 2 years of life while vaccine-induced immunity develops. Researchers have suggested a negative association of antibiotic use with vaccine-induced immunity in adults, but data are lacking in children.

Methods: From 2006 to 2016, children aged 6 to 24 months were observed in a cohort study. A retrospective, unplanned secondary analysis of the medical record regarding antibiotic prescriptions and vaccine antibody measurements was undertaken concurrently. Antibody measurements relative to diphtheria-tetanus-acellular pertussis (DTaP), inactivated polio (IPV), Haemophilus influenzae type b (Hib), and pneumococcal conjugate (PCV) vaccines were made.

Results: In total, 560 children were compared (342 with and 218 without antibiotic prescriptions). Vaccine-induced antibody levels to several DTaP and PCV antigens were lower (P < .05) in children given antibiotics. A higher frequency of vaccine-induced antibodies below protective levels in children given antibiotics occurred at 9 and 12 months of age (P < .05). Antibiotic courses over time was negatively associated with vaccine-induced antibody levels. For each antibiotic course the child received, prebooster antibody levels to DTaP antigens were reduced by 5.8%, Hib by 6.8%, IPV by 11.3%, and PCV by 10.4% (all P ≤ .05), and postbooster antibody levels to DTaP antigens were reduced by 18.1%, Hib by 21.3%, IPV by 18.9%, and PCV by 12.2% (all P < .05).

Conclusions: Antibiotic use in children <2 years of age is associated with lower vaccine-induced antibody levels to several vaccines.

FIGURE 1

Frequency of subprotective vaccine antibody levels from 9 to 24 months of age. A, The frequency of subprotective vaccine levels were determined for children with antibiotic prescriptions and those with no antibiotic prescriptions for each vaccine antigen from 9 to 24 months of age combined. Streptococcus pneumoniae (Spn) and the specific serotype in the PCV tested are shown as numbered. B, Data for all vaccine antigens compared by child age. C, Time points where a single antibiotic prescription was completed within 30 days of measurement of vaccine-induced antibody levels were compared with those of children with no antibiotic prescriptions (control). Dosing: amoxicillin, 80 to 100 mg/kg/day divided twice daily for 10 days; cefdinir, 14 mg/kg/day once daily for 10 days; amoxicillin/clavulanate, 80 to 100 mg/kg/day on the basis of the amoxicillin component divided twice daily for 5 or 10 days; and ceftriaxone, 50 mg/kg intramuscular injection per day for 3 sequential days or every other day. Amox/clav-5 and Amox/clav-10 groups represent children who received 5- and 10-day courses of amoxicillin/clavulanate, respectively. *P < .05 by Fisher’s exact test. Abx, antibiotics.

FIGURE 2

Association of antibiotic courses with pre- and postbooster vaccine antibody levels. A and B, Regression analyses of log-transformed vaccine-induced antibody levels across groups were performed for DTaP, Hib, IPV, and PCV as linear functions of the number of antibiotic prescriptions from 6 to 12 months (A) or from 6 to 15 months (B) of age. To show the impact on the nontransformed antibody level, the exponential relationship is shown. Prebooster antibody levels to DTaP antigens were reduced by 5.8% (P = .01), Hib by 6.8% (P = .05), IPV by 11.3% (P = .04), and PCV by 10.4% (P = .01) for each antibiotic course the child received. Postbooster antibody levels to DTaP antigens were reduced by 18.1% (P = .001), Hib by 21.3% (P = .02), IPV by 18.9% (P = .02), and PCV by 12.2% (P = .03) for each antibiotic course the child received. The associated intercepts, slopes, and errors in vaccine-induced antibody per antibiotic course for each vaccine are shown in Supplemental Table 2 for panel A (prebooster) and Supplemental Table 3 for panel B (postbooster). The effect of antibiotic days can be inferred because each of 1678 antibiotic courses was 10 days, except the subset of 187 (11%) children who received only 5 days of amoxicillin/clavulanate. Given the half-life of ceftriaxone, the 3 injections would constitute ∼10 days of antibiotics.