Use of Single-dose Azithromycin to Control a Community Outbreak of emm26.3 Group A Streptococcus Invasive Disease—Alaska, 2017

BACKGROUND: In July 2016, invasive infections caused by a rare subtype of group A Streptococcus (iGAS; subtype emm26.3) were detected among the estimated 700–1000 homeless persons in Anchorage, Alaska. An increase in case numbers of emm26.3 iGAS was detected in October, including one death. We implemented a mass antibiotic intervention at homeless service facilities in Anchorage to prevent further cases of emm26.3 iGAS. METHODS: We defined cases as the isolation of emm26.3 GAS from a normally sterile body site, or nonsterile sites in the case of necrotizing fasciitis or toxic shock syndrome. We identified cases through routine laboratory-based surveillance and conducted antimicrobial susceptibility testing on all invasive isolates. From February 13–18, 2017, we evaluated persons accessing homeless services at six facilities in Anchorage and offered a single oral dose of 1 gram of azithromycin for iGAS prophylaxis. We concurrently collected oropharyngeal (OP) and wound swab specimens on a subset of participants. The swab collection was repeated at the same locations 4 weeks after the intervention. Swabs were cultured for GAS and emm-typed. RESULTS: From October 1, 2016 through February 18, 2017, we detected 31 cases among homeless persons. All emm26.3 iGAS isolates were erythromycin susceptible. We evaluated 484 persons at homeless services facilities and provided azithromycin to 394 (81%). Of 289 swab participants, 9 (3.1%) had baseline emm26.3 OP colonization. Of participants with wounds, 3/71 (4.2%) had emm26.3 wound colonization. At follow-up, 3/298 (1.0%) participants had emm26.3 OP colonization and 1/63 (1.6%) had emm26.3 wound colonization (P-value for change in any colonization = 0.05). Colonization by other emm-types, primarily erythromycin non-susceptible emm11, was 5.1% at baseline and 5.0% at follow-up. In the 6 weeks post-intervention, we detected 1 case among homeless persons (0.2 cases/week post- vs. 1.6 cases/week pre-intervention, P = 0.01 for change). CONCLUSION: We reached a substantial proportion of the Anchorage homeless population with an antibiotic intervention to prevent iGAS. While possible that the outbreak was waning, the intervention was temporally associated with reduced case counts and colonization prevalence. DISCLOSURES: All authors: No reported disclosures.