
Methods: We defined early-onset disease (EOD) cases as GBS infections in infants aged 0-6 days and late-onset disease (LOD) as infections among infants aged 7-89 days. To calculate incidence, we used cases identified by Active Bacterial Core surveillance (ABCs), a surveillance system covering nearly 10% of US births, for numerators, and live births from state vital records for denominators. We used the Cochrane-Armitage test to analyze the significance of trends in annual incidence rates. We analyzed serotype data from 7 of 10 ABCs sites that collect GBS isolates. We used national estimates of GBS disease among infants and data from ABCs on serotype distribution in a model with assumptions about efficacy to estimate potential impact of a vaccine.
Results: During 2006-2014, ABCs identified 1,180 EOD and 1,246 LOD cases. Overall EOD incidence initially declined, from 0.38 (2006) to 0.27 (2010) cases per 1000 live births (P<.0001), but stabilized at ~0.26/1000 live births since 2011. LOD incidence has largely remained unchanged over the nine year period at ~0.30/1000 live births. Serotype data were available for 1564 (91.4%) of 1711 EOD and LOD cases from sites that collected GBS isolates. The five most common serotypes (Ia, Ib, II, III and V) caused 93% of EOD and 94% of LOD. Assuming an efficacy of 80%, we estimated that a vaccine against these serotypes could prevent around 600 cases of EOD and 580 cases of LOD each year nationally.
Conclusion: Despite use of IAP, perinatal GBS remains a concern. EOD rates have plateaued in recent years while LOD rates remain unchanged. A vaccine containing the five most common serotypes, if effective, offers an opportunity to further reduce EOD, and reduce LOD for which there is no public health intervention.

S. Nanduri,
None
J. Baumbach, None
A. Reingold, None
L. Miller, None
L. Harrison, None
R. Lynfield, None
S. M. Zansky, None
W. Schaffner, None
A. Thomas, None
M. Farley, None
S. Schrag, None
G. E. Langley, None