Mortality, neurodevelopmental impairments, and economic outcomes after invasive group B streptococcal disease in early infancy in Denmark and the Netherlands: a national matched cohort study

Abstract

Background: Group B Streptococcus (GBS) disease is a leading cause of neonatal death, but its long-term effects have not been studied after early childhood. The aim of this study was to assess long-term mortality, neurodevelopmental impairments (NDIs), and economic outcomes after infant invasive GBS (iGBS) disease up to adolescence in Denmark and the Netherlands.

Methods: For this cohort study, children with iGBS disease were identified in Denmark and the Netherlands using national medical and administrative databases and culture results that confirmed their diagnoses. Exposed children were defined as having a history of iGBS disease (sepsis, meningitis, or pneumonia) by the age of 89 days. For each exposed child, ten unexposed children were randomly selected and matched by sex, year and month of birth, and gestational age. Mortality data were analysed with the use of Cox proportional hazards models. NDI data up to adolescence were captured from discharge diagnoses in the National Patient Registry (Denmark) and special educational support records (the Netherlands). Health care use and household income were also compared between the exposed and unexposed cohorts.

Findings: 2258 children-1561 in Denmark (born from Jan 1, 1997 to Dec 31, 2017) and 697 in the Netherlands (born from Jan 1, 2000 to Dec 31, 2017)-were identified to have iGBS disease and followed up for a median of 14 years (IQR 7-18) in Denmark and 9 years (6-11) in the Netherlands. 366 children had meningitis, 1763 had sepsis, and 129 had pneumonia (in Denmark only). These children were matched with 22 462 children with no history of iGBS disease. iGBS meningitis was associated with an increased mortality at age 5 years (adjusted hazard ratio 4·08 [95% CI 1·78-9·35] for Denmark and 6·73 [3·76-12·06] for the Netherlands). Any iGBS disease was associated with an increased risk of NDI at 10 years of age, both in Denmark (risk ratio 1·77 [95% CI 1·44-2·18]) and the Netherlands (2·28 [1·64-3·17]). A history of iGBS disease was associated with more frequent outpatient clinic visits (incidence rate ratio 1·93 [95% CI 1·79-2·09], p<0·0001) and hospital admissions (1·33 [1·27-1·38], p<0·0001) in children 5 years or younger. No differences in household income were observed between the exposed and unexposed cohorts.

Interpretation: iGBS disease, especially meningitis, was associated with increased mortality and a higher risk of NDIs in later childhood. This previously unquantified burden underlines the case for a maternal GBS vaccine, and the need to track and provide care for affected survivors of iGBS disease.

Funding: The Bill & Melinda Gates Foundation.

Translations: For the Dutch and Danish translations of the abstract see Supplementary Materials section.

Figure 1

Study flowcharts showing children with iGBS disease and gestational age-matched unexposed cohorts in Denmark (A) and the Netherlands (B) Of 1525 iGBS cases, 1398 were included in the NDI and economic analyses (after including only those who survived iGBS and excluding pneumonia). This cohort was further restricted in the NDI analyses to those who were alive at age of 5 years (n=1293), 7 years (n=1179), and 10 years (n=969). iGBS=invasive group B Streptococcus. NDI=neurodevelopmental impairments.

Figure 2

Proportion of children with NDIs among those with invasive GBS disease compared with unexposed children matched on gestational age in Denmark (A) and the Netherlands (B) (A) Proportions of children with NDIs of different severities at different ages in the Danish study population are shown for the exposed and unexposed cohorts. For these four panels, the NDI outcomes include: cognitive domain, motor domain, overall NDI, and multi-domain NDIs. Of the 1525 children who survived GBS disease in Denmark, 1293 (by age 5 years) and 969 (by age 10 years) were included in analyses of long-term impairment. (B) The proportion of children with NDIs in the study population from the Netherlands. Children enrolled in special needs schools are presented as having moderate or severe NDI. 489 of the 646 patients who survived GBS in the Netherlands reached the age of mandatory education; 16 (3·2%) were not found in the national database and were considered lost to follow-up. The diagnosis of NDIs was defined cumulatively—namely, that each child was considered to have impairments based on all follow-up information available up to the relevant age limit. GBS=group B Streptococcus. GBS-M=GBS meningitis. GBS-S=GBS sepsis. NC=not categorised. NDIs=neurodevelopmental impairments.

Figure 3

Health-care use among children with invasive GBS disease compared with unexposed children that were matched by gestational age in Denmark (A) Outpatients clinic visits. (B) Number of hospital admissions (bars) and mean number of days in hospital (red line corresponds to exposed children; blue line corresponds to unexposed children). 95% CIs of age-specific means are presented for each bar. Hospital admissions for acute GBS were excluded in the analysis of number of days of hospitalisation in the first year of life. GBS=group B Streptococcus disease.