Serologic Detection of Antibodies Targeting the Leukocidin LukAB Strongly Predicts Staphylococcus aureus in Children With Invasive Infection

BACKGROUND: Staphylococcus aureus is among the most commonly identified causes of invasive bacterial infection in children; however, reliable results from cultures of sterile-site samples often cannot be obtained, which necessitates prescription of a broad empiric antimicrobial agent(s). Children with invasive S aureus infection rapidly generate high antibody titers to the cytotoxin LukAB; therefore, the aim of this study was to assess the diagnostic utility of an anti-LukAB antibody assay for children with musculoskeletal infection (MSKI). METHODS: We conducted a 2-year prospective study of all eligible children admitted to Vanderbilt Children's Hospital with an MSKI. Acute and convalescent sera were obtained, and antibodies that target LukAB were measured by an enzyme-linked immunosorbent assay. RESULTS: Forty-two children were enrolled. The median concentrations of LukAB antibodies for children with S aureus infection were 130.3 U/mL in the acute phase and 455 U/mL in the convalescent phase (P < .001). The median concentrations of LukAB antibodies in children with a non-S aureus MSKI were 8.6 U/mL in the acute phase and 9.7 U/mL in the convalescent phase. The assay discriminated between S aureus and non-S aureus infection with areas under the receiver operating characteristic curve of 0.81 (95% confidence interval, 0.67-0.95; P < .001) and 0.95 (95% confidence interval, 0.86-1; P < .001) for samples tested in the acute and follow-up periods, respectively. With no false-negative results, the assay accurately ruled out S aureus in samples obtained during the convalescent phase. CONCLUSION: Culture-independent diagnostics have the potential to improve care by narrowing antimicrobial therapy on the basis of the likelihood of S aureus infection. The results of this proof-of-concept study suggest that a LukAB serologic assay might be useful in the diagnosis of invasive bacterial infections, and larger-scale validation studies are warranted.