Although patients with eczema herpeticum often receive antibiotics for presumed bacterial coinfection, the effect of empiric antibiotic therapy is unknown. Our objective therefore was to determine the association between empiric antibiotics and outcomes in children hospitalized with eczema herpeticum. We conducted a multicenter retrospective cohort study of 1,150 children ages 2 months to 17 years admitted with eczema herpeticum between January 1, 2001, and March 31, 2010, to 42 tertiary care children's hospitals in the Pediatric Health Information System. All patients received antibiotics during the hospitalization. Multivariable linear regression models determined the association between empiric antibiotic therapy and the main outcome measure: hospital length of stay (LOS). There were no deaths during the study period. Receipt of empiric antibiotics was not associated with a change in the LOS on unadjusted or multivariable analysis. The class of empiric antibiotic was not associated with the LOS except for receipt of vancomycin, which was associated with a longer LOS (21% adjusted longer LOS, 95% confidence interval (CI) = 8-35%; p = 0.001). When restricted to patients with a bloodstream infection, receipt of empiric antibiotics was associated with a 51% adjusted shorter LOS (95% CI = −24 to −68%; p= 0.002). In children hospitalized with eczema herpeticum, empiric antibiotic therapy was not associated with a shorter LOS overall, but was associated with a shorter LOS in patients with a bloodstream infection. These findings highlight the importance of early recognition of systemic bacterial illness in children with eczema herpeticum. Empiric antibiotics did not affect mortality, which is low.