AN OUTBREAK OF M-SEROTYPE-1 GROUP-A STREPTOCOCCUS IN A NEONATAL INTENSIVE-CARE UNIT

Objective: To describe the investigation and control of an outbreak of M serotype 1, Streptococcus pyogenes (group A Streptococcus, GAS) inf ections in a neonatal intensive care unit (NICU). Study design: The st udy was conducted in an NICU in a large urban university-affiliated ho spital. Retrospective review was performed of all infants and health c are workers in the NICU, especially those either colonized or infected with GAS during the outbreak and the prospective surveillance period (July through September 1994). Prospective epidemiologic investigation , including cultures of throat, umbilicus, and anorectum (infants), or throat and anus (NICU personnel), identified a possible common source of the disease in case infants. Antimicrobial susceptibility testing and serotyping of all GAS strains were performed; M serotype 1 isolate s were examined by DNA analysis with restriction fragment length polym orphism. The M-1 GAS isolates were tested for streptococcal pyrogenic exotoxin (SPE) A and SPE B production. A retrospective chart review an d analysis of infants with GAS infection or colonization was conducted . Results: During a 1-week period, two very low birth weight infants m ore than 3 weeks of age had GAS septicemia and focal infection. Two ad ditional very low birth weight infants with asymptomatic throat coloni zation were identified during the first week of surveillance. Benzathi ne penicillin G was administered to all NICU infants, but failed to er adicate throat colonization in the four case subjects. Seven days afte r completing parenteral antibiotic therapy, the index patient had a re currence of GAS septicemia that was fatal. Eradication of throat colon ization in the remaining three infants was achieved with a 10-day cour se of intravenous clindamycin therapy. Among 103 NICU personnel, five (4.9%) had asymptomatic GAS colonization with strains that were unifor mly susceptible to penicillin. Each colonized adult was successfully t reated with oral clindamycin therapy. Serotyping revealed that five is olates of GAS from four infants and one NICU respiratory therapist wer e M-l isolates; DNA analysis confirmed that these were the same strain . The five M-1 isolates produced both SPE A and SPE B. Conclusions: Th e previously documented increase in prevalence of M-1 strains of GAS i n the United States is likely to be associated with their introduction Into closed populations including NICUs. Control of such outbreaks ma y be achieved by isolation, cohorting of case subjects and possible ca rriers, and successful eradication of colonization in case subjects an d carriers. Although GAS organisms are uniformly susceptible to penici llin G, eradication may require agents other than penicillin.