Serotypes, sequence types, and erythromycin resistance of maternal colonizing group B Streptococcus: implications for neonatal early-onset disease prevention

Intrapartum antibiotic prophylaxis for maternal group B Streptococcus (GBS) colonization depends on the assumption that a woman identified as colonized receives an antibiotic active against her own isolate, an assumption that routine susceptibility reporting may overestimate when inducible clindamycin resistance is present. This study characterized the capsular serotypes, sequence types, macrolide-lincosamide resistance phenotypes and the intrapartum antibiotic actually administered for maternal colonizing GBS isolates at a single tertiary maternity centre. Banked isolates from antenatal screening were revived and analysed by capsular multiplex PCR, multilocus sequence typing, broth microdilution, D-test and resistance gene PCR, and each isolate was linked to the intrapartum regimen recorded in the corresponding delivery record. Among 120 isolates, serotype III predominated at 35.8%, and the six serotypes included in a hexavalent capsular vaccine accounted for 94.9% of typeable isolates, representing a serotype-based ceiling rather than expected vaccine effectiveness. Thirteen sequence types were resolved, and CC17/ST17 accounted for 11.7% of the collection. All isolates were susceptible to penicillin, ampicillin and vancomycin, whereas erythromycin non-susceptibility reached 63.3%, comprising constitutive (40.0%), inducible (8.3%) and M (15.0%) phenotypes, with a further 3.3% showing the L phenotype. Routine testing classified 56.7% of isolates as clindamycin-susceptible, but effective in-vitro clindamycin availability after reclassification of D-test-positive isolates was 48.3%, an absolute reduction of 8.3 percentage points. The inducible phenotype was more frequent in CC17/ST17 than in other lineages (28.6% versus 5.7%), although this subgroup contained only 14 isolates and should be interpreted as exploratory. Among the women who had actually received clindamycin during labour, half were colonized by isolates against which clindamycin lacked in-vitro activity. These findings support routine D-testing of erythromycin non-susceptible maternal GBS isolates and reporting that states whether clindamycin is suitable for intrapartum prophylaxis.