151 Cluster of Group B Streptococcus in a Neonatal Intensive Care Unit

Monday, June 10, 2013
Exhibit Hall A (Pasadena Convention Center)
Ashley Moore , Georgia Department of Public Health, Atlanta, GA
Eileen Usman , Georgia Department of Public Health-West Central Health District, Columbus, GA
Susan Harp , Columbus Regional Medical Center, Columbus, GA
Emily Weston , Centers for Disease Control and Prevention, Atlanta, GA
Lesley McGee , Centers for Disease Control and Prevention, Atlanta, GA
Robert Gertz , Centers for Disease Control and Prevention, Atlanta, GA
Cherie Drenzek , Georgia Department of Public Health, Atlanta, GA
Matthew Crist , Georgia Department of Public Health, Atlanta, GA

BACKGROUND: Neonatal sepsis caused by Group B Streptococcus (GBS) infections have been a public health concern since the 1970s. The risk factors and transmission routes for late-onset (LO) GBS are less well understood than that of early onset (EO) GBS disease (0-6 days of life), which is generally caused by vertical transmission from the mother. In April 2012, the District Epidemiologist and the Georgia Department of Health were notified of a cluster of four LO GBS infections in a neonatal intensive care unit (NICU). Our objectives were to identify any additional cases through medical record review and to identify the outbreak strain.

METHODS: The GBS isolates from the 4 case-patients were sent to CDC for serotyping and multi-locus sequence typing. We defined a case as any infant in the NICU in the 3 months prior to the outbreak with a positive invasive GBS culture.  We conducted retro­spective case finding via patient medical record review. To determine potential sources of transmission and risk factors, we interviewed infection preventionists (IP) and medical providers in the NICU and conducted an evaluation of infection control practices.

RESULTS:  Three GBS isolates were serotype 1B; one was serotype III. Molecular testing indicated that the three serotype 1B isolates had identical genetic sequences. Temporary overcrowding of infants in the NICU was noted to have occurred in the weeks prior to the onset of the cluster. Analysis of staffing patterns demonstrated multiple shared caregivers between those infants with the same serotype. The infant with the unique serotype did not have a shared caregiver shift with the other infants. 

CONCLUSIONS: While no single source was identified, a lapse in hand hygiene was most likely the cause of the spread among the infants. We recommend that IP staff monitor NICUs for clusters of LOS GBS by increased communication with the microbiology lab and periodic retrospective review of medical records.  Other measures to potentially prevent transmission of GBS in NICUS include vigilant hand hygiene, avoidance of staff overlap when feasible, and ensuring no overcrowding. Following enhanced infection control reviews by the hospital IP, there have been no further cases of invasive LOS GBS in this NICU.