# Shifting from an expected to an opportunistic pathogen: Comparison of cases of infant late and very late onset group B streptococcal (GBS) infection in a Canadian city over a 27-year period

## TL;DR

This study compares late and very late onset GBS infections in infants, finding that VLOD cases are more complex and may involve GBS acting as an opportunistic pathogen.

## Contribution

The study provides new insights into the clinical and epidemiological differences between late and very late onset GBS infections in infants.

## Key findings

- VLOD infants were more likely to be preterm and have complex medical histories compared to LOD infants.
- GBS VLOD cases showed higher rates of mechanical ventilation and co-infections with other pathogens.
- Serotype III was the most common in both LOD and VLOD cases, but VLOD had a lower proportion.

## Abstract

Despite decades of study, the characteristics of infants with invasive group B streptococcus (GBS) late onset disease (LOD) (onset day 7–89 of life) and in particular very late onset disease (VLOD) (after day 89 of life) are not well described.

This was a retrospective cohort study of infants hospitalized in four Edmonton hospitals April 1, 1994 through June 30, 2022 with LOD or VLOD GBS invasive disease. Data were collected on demographics, day of onset of infection, clinical manifestations and outcomes.

There were 115 episodes of LOD in 111 infants of which 50 infants (45%) were preterm. Onset of initial LOD infection was on median day 27 (IQR 19–40.5) of life. All but one infant was bacteremic while 38/111 (34%) had proven and 17/111 (15%) had possible GBS meningitis. Five of 111 (5%) died before hospital discharge with all deaths probably due to GBS. There were 11 episodes of VLOD in 11 infants (8 [73%] preterm) presenting on median day 116 (IQR 103–138) (range 93–207) of life. Three (27%) had GBS meningitis. All 11 survived to discharge. As compared with infants with LOD, those with VLOD were less likely to be born vaginally (n = 3/11 [27%] versus n = 72/111 [65%] p = 0.036), more likely to be mechanically ventilated during their birth hospitalization (n = 5/11 [45%] versus n = 17/111 [15%]; p = 0.039), and more likely to have serious infection with other pathogens during their GBS admission (n = 3/11 [27%] versus n = 4/111 [4%]); p = 0.032). Serotype III accounted for 78% of LOD and 64% of VLOD cases.

GBS remains a significant cause of infant morbidity and mortality. Infants with VLOD appear to have more complex medical histories prior to GBS infection than those with LOD. There is increasing evidence that GBS may sometimes be an opportunistic pathogen in infants over 89 days of age. This speaks of the need for increased surveillance for VLOD with further research into its epidemiology and underlying immunologic determinants.