C-reactive protein levels and behavioural outcomes in mid-childhood in very preterm and term born children

Abstract

Background

Preterm birth has been associated with long-term physical, cognitive, behavioural and emotional challenges. Preterm infants are often exposed to significant inflammation in early life, which has been studied in relation to adverse neurodevelopmental outcomes. In later childhood, inflammation has been associated with psychiatric symptoms. However, the developmental processes underlying the association between inflammation and mental health outcomes, particularly in preterm children, are not fully understood. We aimed to investigate the association between inflammation and behavioural outcomes in mid-childhood in preterm- and term-born children.

Methods

This study used data collected as part of the Brain, Immunity and Psychopathology following very Preterm Birth (BIPP) study, a prospective cohort study assessing very preterm-born (<33 weeks’ gestation) and term-born children at a median age of 9.2 years (range 7.0 – 13.7 years). Children received behavioural and cognitive assessments and provided saliva samples for C-reactive protein (CRP) analysis as a marker of chronic inflammation. Clinical neonatal discharge summaries were available for the very preterm participants, from which a perinatal inflammatory risk score was devised. Missing behavioural and cognitive outcome scores were imputed. Generalised linear mixed models were used to test the association between CRP or perinatal inflammatory risk score and each behavioural/cognitive outcome variable in turn, controlling for age, sex, neighbourhood deprivation, ethnicity and prematurity. In models that showed an effect of CRP, the interaction between prematurity and CRP was also examined.

Results

Participants were 264 children (n = 195 very preterm and n = 69 control). After excluding participants with high salivary CRP levels (predicted serum CRP > 10 mg/L), indicative of acute infection/inflammation (n = 3), 202 children had complete salivary CRP data: 47.9 % were male. There was no significant difference in salivary CRP value in mid-childhood between very preterm and term participants. When controlling for covariates, neither very preterm birth nor perinatal inflammatory risk score were associated with salivary CRP level in mid-childhood. Salivary CRP in mid-childhood was associated with lower externalising symptoms (B = -0.26, p = 0.037) and higher anxiety symptoms (B = 0.06, p = 0.022). Perinatal inflammatory risk score was not associated with behavioural or cognitive outcome measures.

Conclusion

We show no evidence that prematurity or perinatal inflammation are associated with salivary CRP in mid-childhood, nor that perinatal inflammatory insults are related to later behavioural outcomes in our cohort of preterm children. Our findings support an association between salivary CRP and behavioural outcomes in mid-childhood. Further research is needed to understand the interplay between inflammation and behavioural and emotional regulation throughout childhood.