High incidence of invasive group B streptococcal infections in uninfected infants born to HIV-1-infected mothers

Background

Practice guidelines recommend the follow up of all children born to HIV-infected women to ascertain the absence of vertical transmission and to detect possible adverse effects of exposure to ART. The occurrence of an unusual number of invasive group B streptococcal (GBS) infections in uninfected HIV-exposed infants in our center prompted this study.

Objective

To describe the incidence and clinical presentation of GBS infections in a cohort of uninfected HIV-exposed infants born between 2001 and 2008 in comparison to the population of infants not exposed to HIV born in the same hospital during this period.

Methods

The medical charts of all uninfected HIV-exposed infants prospectively followed since birth in our center and the microbiology laboratory records were reviewed to identify invasive GBS infections.

Results

8 episodes of invasive GBS infection occurred in 7 out of 397 uninfected HIV-exposed infants. Ninety seven % of the mothers were treated with antiretroviral agents during pregnancy, including 6 of the 7 mothers of children with GBS infection. The median gestational age of these infants was 36 weeks. Five episodes occurred later than 7 days of life (days 9, 26, 33, 64 and 72). One of the infants had a recurrent episode 28 days after completion of the antibiotic treatment of a first GBS infection. When the analysis was restricted to the infants born in our centre, GBS invasive infection occurred in 5/322 (15.5/1000 live births) HIV-exposed infants compared to 16/20158 (0.79/1000 live births) infants in the control population (OR = 19.6 p < 0.0001). In the latter, median age of onset of GBS infection was 1 day and the median gestational age 40 weeks.

Conclusion

Between 2001 and 2008 the incidence of invasive GBS infection was significantly higher in the uninfected infants born to HIV-infected mothers than in the control population born in our centre. The majority of GBS infections in HIV-exposed infants were late or very late onset and 1 child had a recurrence; 2 features that were strikingly different than in the general population.