BACKGROUND: Group A Streptococcus (GAS) causes invasive infections in healthcare facility and community settings. We compared demographic, clinical and microbial characteristics of healthcare-associated (HA) and community-associated (CA) cases in Minnesota.
METHODS: Population- and laboratory-based active surveillance of invasive (i) GAS is conducted within CDC’s Active Bacterial Core surveillance. Cases are infections with GAS isolated from a normally sterile site, or wound if accompanied by necrotizing fasciitis. Cases are defined as HA if they occur in a person who had surgery (postsurgical) or delivered an infant (postpartum) ≤7 days before culture, or GAS was cultured ≥3 days after hospital admission (hospital-onset-[HO]) (categories not mutually exclusive). Clinical information is obtained by record review. Isolates are submitted to CDC for emm typing.
RESULTS: From January 1, 2000 to December 31, 2013, 2,428 iGAS cases were reported. Of these, 206 (8.5%) were classified as HA; 99 (48.1%) were HO, 76 (36.9%) postsurgical, and 47 (22.8%) postpartum. Among persons with HA iGAS, median age was 40 years, and 124 (60.2%) were female. Among persons with CA iGAS, median age was 53 years, and 1003 (45.1%) were female. HA iGAS cases were significantly younger (p<0.001), more often female (p<0.001), American Indian (p=0.019), or non-white (p=0.001) than CA iGAS cases. Excluding postpartum cases, the age and sex of both HO and postsurgical iGAS cases were similar to CA cases, although race differences persisted (p=0.008). Presence of comorbidities did not differ between HA (53.4%) and CA (57.2%) cases, but comorbidities were more common in HO vs. CA cases (p=0.021). GAS was more frequently isolated from peritoneal fluid (OR 3.52, p=0.009), bone (OR 2.55, p=0.014) and internal body sites (OR 2.94, p=0.008) of HA compared to CA cases. Case fatality ratios were lower for HA than CA cases (5.8% vs. 10.1%; p=0.049). Associations remained after adjusting for presence of comorbidities. Emm results were available for 2,178 (89.7%) GAS isolates; 48 emm types were identified. The distribution of emm types among HA and CA cases were similar.
CONCLUSIONS: After exclusion of postpartum cases, differences in patient race and specimen source between HA (HO and postsurgical cases) and CA iGAS cases persisted. Racial differences remained after adjusting for presence of comorbidities. These differences could be attributed to differences in socioeconomic status, access to healthcare or other risk factors not measured by surveillance. Further investigation may help guide future prevention efforts.