Executive function deficits in preterm subjects are a combination of social risk factors and brain maldevelopment

Executive functions are important higher order cognitive functions that plan, organise, control and regulate goal directed behaviour and provide cognitive flexibility, including working memory, goal setting, attentional control and information processing [1]. Deficits in executive functions influence problem-solving abilities and can have a substantial impact on the individual's cognitive, social and academic functioning and quality of life.

In addition to the increased risk of inferior general cognitive ability, reflected by lower total intelligent quotient (IQ) and IQ indices scores, subjects born very preterm, with a gestational age of less than 31 weeks or with very-low-birthweight (VLBW) of <1500 g, have a higher prevalence of executive and attentional deficits, often combined with a slow processing speed [2]. One study showed that adolescents and young adults who were born preterm with VLBW had significantly lower scores than term born controls with regard to the attention and executive function domains. VLBW young adults also had an increased risk of impairment in their attention domain, with an odds ratio (OR) of 2.16 and 95% confidence interval (95% CI) of 1.46–3.15, and in their executive function domain (OR: 2.76, 95% CI: 2.19–3.48) [3].

In this issue of Acta Paediatrica, O'Meagher et al. assessed 141 preterm children and 77 control children aged four to five years using standardised measures of IQ and executive functions. Perinatal and social risk factors were investigated as possible predictors using multivariate regression analyses. The authors found that the preterm born children had significantly inferior scores than their term born peers for all the intelligence and executive function tests [4]. This was consistent with reports on older children, adolescents and young adults born preterm. However, it is really interesting that these deficits in executive function were reported at such an early age, when many executive functions are still very immature and in a developmental phase.

In the same study, the parents of the preterm children did not report more executive function problems using the Behavioral Rating Inventory of Executive Function (BRIEF) tool than the control parents, while the teachers reported more problems in the preterm group. These results are different from a study by Grunewaldt et al., where the parents of 10-year-old children born with extremely low-birthweight (ELBW) of less than 1000 g reported significantly higher scores on several of the BRIEF measures compared with parents of term born control children, when socio-economic status was taken into account [5]. The discrepancy may have been due to the older age of the subjects in this study, as they had acquired more advanced executive functions, making it more likely that the parents would notice any deficits in the preterm group. However, the neuropsychological test results did not differ significantly between the preterm subjects and term born controls, suggesting that questionnaires and neuropsychological tests assess different aspects of executive functions. The opposite results were reported in a study by Sølsnes et al., where self-reported BRIEF scores in VLBW young adults at the age of 20 did not differ from the controls. Despite this, the VLBW participants obtained lower scores on eight of the 18 neuropsychological tests that assessed attention and executive functions. The effect sizes for the tests with significant group differences ranged from 0.54 to 0.88. When the processing speed index from the Wechsler Adult Intelligence Scale was included as a covariate, the VLBW subjects still achieved poorer scores on several of the executive function tests when compared with the controls [6]. The authors speculated that the lack of consistency could be explained by the assumption that measurements obtained from behavioural observations, self-reports and neuropsychological test results were based on different aspects of executive functions. Another explanation suggested by the authors was an underreporting of executive problems among the preterm born young adults, either caused by lack of insight or as a result of environmental adjustment, which could make them more able to handle the executive function requirements of daily life activities.

When they looked at risk factors for executive function deficits, O'Meagher et al. found that higher social risk, in particular the lower educational level of the main caregiver, was the strongest predictor for the IQ score and executive function results in the preterm group. Other significant risk factors that had a more limited impact on outcome were male sex, gestational age, as a reflection of immaturity and the length of stay in the neonatal intensive care unit (NICU), as a reflection of both immaturity and perinatal morbidity [4]. As preterm groups age the influence of perinatal risk factors on executive functioning seem to be even lower than the levels that were reported in these preschool children. For example, in a study by Østgård et al. [3] of young adults born VLBW, none of perinatal risk factors – birth weight, gestational age, days in the NICU and days on mechanical ventilator – correlated with the executive function or the attention domain score, while both scores correlated with special educational needs in school.

Even when group differences in socioeconomic status have been controlled for, preterm born participants exhibit inferior executive function and attentional deficits compared with term born peers. This indicates that perinatal brain injuries or brain maldevelopment contribute to these neuroimpairments. Several studies have tried to investigate the possible brain correlates of reduced executive functions in preterm born subjects. It seems that both deviations in white matter integrity and connectivity and cortical abnormalities are associated with these deficits in higher order cognitive functioning. Our research group at the Norwegian University of Science and Technology in Trondheim, Norway has investigated such possible structure-function relationships in several cohorts of preterm born very and extremely low birthweight children and adolescents. This showed that perinatal white matter injuries from focal noncystic or diffuse periventricular leukomalacia could result in white matter loss and reduced white matter volumes seen on later structural magnetic resonance imaging. In the study by Grunewaldt et al., full IQ correlated with volumes of putamen and globus pallidus and better working memory index scores correlated with larger volumes of putamen and cerebellar white matter. The authors also found correlations between higher total scores for attention deficit hyperactivity disorder (ADHD), indicating more hyperactivity, and smaller volumes of putamen, cerebellar white matter and corpus callosum. They also reported correlations between higher BRIEF main indices, indicating more executive function problems, and smaller volumes of putamen, cerebellar white matter and corpus callosum [5]. It is possible that the relationship between executive function and cerebellar white matter emphasises the role of cerebellar-frontal connectivity for optimal executive functioning.

However, perinatal white matter injuries may also influence the myelination process that causes the deviations in white matter microstructure that are seen on diffusion tensor imaging. Allin et al. found that executive function and attention skills relied on white matter tract integrity within subcortical and cortical regions [7]. In a voxel-based diffusor tensor imaging study of VLBW adolescents, we found correlations between the results of an executive function test, namely the Wisconsin Card Sorting Test, and fractional anisotropy values in the left cingulum and both of the inferior fronto-occipital fascicles, which are known to contain long association tracts. We speculated that disturbed connectivity between the posterior and prefrontal cortical areas could have had an impact on executive functioning [8].

Cortical abnormalities have also been linked to deficits in executive functions in preterm born subjects. Thinning of the entorhinal cortices correlated with reduced performances in several executive tests in VLBW teenagers at 14–15 years of age. There was a linear correlation between the thickness of the entorhinal cortices and scores on several executive tests like the Trail Making Test, the Knox Cube Test and the Wisconsin Card Sorting Test [9]. Based on this study, we could not tell whether the structure-function relationship that was observed was causal or whether some unknown factors that negatively affected the growth of the entorhinal cortices may also have caused executive function deficits due to a general effect on brain development. However, the negative influence on executive functioning might have been explained by the compromised hippocampal–entorhinal–prefrontal cortical dialogue, which is necessary for higher order cognitive functioning. In VLBW young adults from the same study population, positive correlations between the executive function domain score and cortical surface area were reported, especially in the antero-medial frontal and the temporal lobes of the brain [3].

In summary, executive function deficits are common in preterms born subjects and, as the paper by O'Meagher et al. shows, may become apparent before the child goes to school [4]. The early detection of such problems may enable children at risk to be included in intervention programmes that aim to improve higher order cognitive functions and academic performance. One example is working memory training, which has shown promising results in preterm children [10]. The aetiology of these deficits may be the sum of genetic and early environmental and epigenetic factors, together with aberrant brain development due to preterm birth.