The population structure of group B streptococcus (GBS) from a low-incidence region for invasive neonatal disease (Israel) was investigated using multilocus genotype data. The strain collection consisted of isolates from maternal carriage (n=104) and invasive neonatal disease (n=50), resolving into 46 sequence types. The most prevalent sequence types were ST-1 (17?5 %), , , and . Serotype III was the most common, accounting for 29?2 % of the isolates. None of the serotypes was significantly associated with invasive neonatal disease. BURST analysis resolved the 46 sequence types into seven lineages (clonal complexes), from which only lineage ST-17, expressing serotype III only, was significantly associated with invasive neonatal disease. Lineage ST-22 expressed mainly serotype II, and was significantly associated with carriage. The distribution of the various sequence types and lineages, and the association of lineage ST-17 with invasive disease, are consistent with the results of analyses from a global GBS isolate collection. These findings could imply that the global variation in disease incidence is independent of the circulating GBS populations, and may be more affected by other risk factors for invasive GBS disease, or by different prevention strategies.
INTRODUCTIONOver the last three decades, group B streptococcus (GBS) has become the leading cause of neonatal meningitis and septicaemia in North America, western Europe and Australia, causing considerable rates of morbidity and mortality (Baker, 2001; Isaacs et al., 1995;Schrag et al., 2000). The increasing incidence of invasive neonatal GBS disease has prompted measures to reduce its prevalence in the USA (American Academy of Pediatrics Committee on Infectious Diseases and Committee on Fetus and Newborn, 1997), Australia (Isaacs & Royle, 1999), and recently in the UK (www.hpa.org.uk/infections/). Interestingly, reports from many developing countries and some developed countries show rates of invasive neonatal GBS disease that are much lower than those reported from North America (Chong et al., 1993;Mayon-White, 1985;Schuchat, 1995;Stan et al., 2001;Stoll, 1997;Tsolia et al., 2003). An explanation for such variation is not apparent, and has not been systematically investigated. Low rates of GBS carriage among pregnant women have been suggested as a possible explanation, yet reports from developing countries have shown that, despite significant geographical variation in the proportion of GBS carriage, the range of colonization is similar to that identified in North America (Stoll & A possible explanation for such a low incidence of invasive neonatal GBS disease, despite the substantial carriage rate, is that invasive disease may be caused by a few virulent lineages (clones) within the total GBS population, such that measurement of the total GBS colonization rates may be misleading as to the true size of the population at risk. Secondly, virulent lineages may account for a different proportion of the vaginal GBS population in different geographical location...