Antibiotic Allergy Labels in Children Are Associated with Adverse Clinical Outcomes.

BACKGROUND: Self-reported antibiotic allergies are common among hospitalized adults and children. However, there is a paucity of studies investigating the impact of an antibiotic allergy label in childhood. OBJECTIVE: To investigate the impact of antibiotic allergy labeling on clinical outcomes in children. METHODS: A retrospective study was conducted in a major pediatric tertiary hospital to capture inpatient admissions (N = 1672) in April 2014 and April 2015. Data, collected by chart review, included documented antibiotic allergy labels, antibiotic prescriptions, admitting specialty, hospital length of stay, and hospital readmissions. RESULTS: Of the 1672 pediatric patients surveyed, 58.1% were male and 44.8% were prescribed antibiotics. Antibiotic allergy labels were recorded in 5.3% of patients; most were ß-lactam allergy labels (85%), mostly to unspecified penicillins. There was an increasing incidence of antibiotic allergy label with age, which was statistically significant (P < .001); no sex effect was seen. Patients with antibiotic allergy labels received more macrolide (P = .045), quinolones (P = .01), lincosamide (P < .001), and metronidazole (P = .009) antibiotics than did patients without an antibiotic allergy label. After adjusting for patient age, sex, principal diagnosis, and admitting specialty, children with any antibiotic or ß-lactam allergy label had longer hospital stays (odds ratio, 1.62; 95% CI, 1.05-2.50; P = .03) with a mean length of hospital stay of 3.8 days for those without a label and 5.2 days for those with a ß-lactam allergy label. CONCLUSIONS: This is the first study demonstrating the negative impact of antibiotic allergy labels on clinical outcomes in children, as evidenced by significant alternate antibiotic use and longer hospital stays.

BTN1A1, the mammary gland butyrophilin, and BTN2A2 are both inhibitors of T cell activation.

Butyrophilin (BTN) genes encode a set of related proteins. Studies in mice have shown that one of these, BTN1A1, is required for milk lipid secretion in lactation, whereas butyrophilin-like 2 is a coinhibitor of T cell activation. To understand these disparate roles of BTNs, we first compared the expression and functions of mouse Btn1a1 and Btn2a2. Btn1a1 transcripts were not restricted to lactating mammary tissue but were also found in virgin mammary tissue and, interestingly, spleen and thymus. In confirmation of this, BTN1A1 protein was detected in thymic epithelial cells. By contrast, Btn2a2 transcripts and protein were broadly expressed. Cell surface BTN2A2 protein, such as the B7 family molecule programmed death ligand 1, was upregulated upon activation of T cells. We next examined the potential of both BTN1A1 and BTN2A2 to interact with T cells. Recombinant Fc fusion proteins of murine BTN2A2 and, surprisingly BTN1A1, bound to activated T cells, suggesting the presence of one or more receptors on these cells. Immobilized BTN-Fc fusion proteins, but not MOG-Fc protein, inhibited the proliferation of CD4 and CD8 T cells activated by anti-CD3. BTN1A1 and BTN2A2 also inhibited T cell metabolism, IL-2, and IFN-gamma secretion. Inhibition of proliferation was not abrogated by exogenous IL-2 but could be overcome following costimulation with high levels of anti-CD28 Ab. These data are consistent with a coinhibitory role for mouse BTNs, including BTN1A1, the BTN expressed in the lactating mammary gland and on milk lipid droplets.