An Outbreak of Group A Streptococcus Invasive Infections among Pediatric Patients in Columbus, Ohio

BACKGROUND: A cluster of invasive disease caused by group A Streptococcus (GAS) was detected among children admitted to Nationwide Children’s Hospital (NCH) in Columbus, Ohio. Clinical characteristics and molecular epidemiology of this cluster were studied to find out whether the observed cases were caused by closely related GAS strains. METHODS: Patient electronic medical records were reviewed for demographic, epidemiologic and clinical data. Cases were defined as patients with GAS isolated from a normally sterile site, or from a non-sterile site in combination with clinical signs of severe streptococcal illness. Susceptibility testing (E-test), analysis of the GAS emm types, multilocus sequence typing (MLST) and pulse field gel electrophoresis (PFGE) from isolates were performed. RESULTS: Thirteen children were admitted to NCH with invasive GAS disease between February 19 and March 23, 2017, from seven different Ohio counties (total population of approximately 1.8 million). The spectrum of illness was broad, including STSS (3 patients), GAS sepsis (3), orbital cellulitis with epidural abscess (3), bacteremia and subdural empyema (2), peritonsillar abscess (1), and septic arthritis (1). One patient with STSS died. Only two patients had chronic illnesses. One patient was diagnosed with acute myeloid leukemia on admission. Median age was eight years (range 0.1–16 years). Of six patients with STSS and sepsis only one received clindamycin and four were given IVIG. Hypocalcemia (67%) was common. Isolates from 12 patients were available for emm gene analysis and belonged to nine different emm types: 1, 2, 3, 6, 12, 22, 89, 118, 227. Two emm 1 and one emm 227 isolates had the same PFGE pattern. All isolates were susceptible to macrolides. CONCLUSION: In a population of 1.8 million, 13 pediatric patients with invasive GAS disease during a five-week period represented an apparent outbreak with polyclonal GAS isolates. There was no reported epidemiologic association among the patients and the outbreak was not preventable. A proposed 30-valent M-protein vaccine would have provided protection against 93% of outbreak isolates. DISCLOSURES: A. Leber, BioFIre Diagnostics: Research Contractor and Scientific Advisor, Research support, Speaker honorarium and Travel expenses