Minor neurological dysfunction in five year old very preterm children is associated with lower processing speed

Early Human Development 103 (2016) 55–60

Contents lists available at ScienceDirect

Early Human Development journal homepage: www.elsevier.com/locate/earlhumdev

Minor neurological dysfunction in five year old very preterm children is associated with lower processing speed Tinka Kurpershoek a, Eva S. Potharst-Sirag b, Cornelieke S.H. Aarnoudse-Moens b, Aleid G. van Wassenaer-Leemhuis a a b

Department of Neonatology, Emma Children's Hospital, Academic Medical Centre, Postbox 22660, 1100 DD, Amsterdam, The Netherlands Psychosocial Department, Emma Children's Hospital, Academic Medical Centre, Postbox 22660, 1100 DD, Amsterdam, The Netherlands

a r t i c l e

i n f o

Article history: Received 5 April 2016 Received in revised form 1 July 2016 Accepted 1 July 2016 Available online xxxx Keywords: Minor neurological dysfunction Prematurity Intelligence quotient Cognition Processing speed Reaction time

a b s t r a c t Background: Minor neurological dysfunction (MND) is present in one quarter to one third of children born very preterm (VP). The more severe form, complex (c)-MND has been associated with learning disabilities, behavioural and motor problems. Objective: To study the association between c-MND and neurocognitive and motor disabilities at age five in VP children without CP. Methods: Ninety-four children born with gestational age b 30 weeks and/or a birth weight b 1000 g were assessed at five years corrected age. MND was classified according to Touwen. The Wechsler Preschool and Primary School Scale of Intelligence (WPPSI-III-NL) was used to measure intelligence. Simple reaction time, focused attention and visuomotor coordination were measured using the Amsterdam Neuropsychological Tasks, and working memory using a Digit Span Task. For motor skills the Movement Assessment Battery for children (M-ABC2) was used. Results: Eighty-one percent was classified as ‘normal’ (no or simple (s-)-MND) and 19% as ‘abnormal’(c-MND or mild CP). The abnormal group had a significantly lower processing speed quotient (PSQ), M-ABC percentile score and slower simple Reaction Time than the normal group. Verbal IQ, Performance IQ, working memory, focused attention and visuomotor coordination did not differ between groups. Exclusion of the mild CP cases (n = 4) led to similar results. Conclusions: Five year old VP children with c-MND have lower PSQ, slower reaction time, and poorer motor skills, than those without c-MND. Neurological examination should include identification of MND to help identify children at risk for neurocognitive disabilities. © 2016 Published by Elsevier Ireland Ltd.

1. Introduction Minor neurological dysfunction (MND) is defined as the occurrence of neurological symptoms in the absence of evident neurological pathology i.e. cerebral palsy (CP). While the incidence of CP is decreasing in preterm infants [1], MND is still often occurring [2]. Seven to 20% of healthy term born children has a form of MND and prevalence before puberty increases with age [3]. The prevalence of simple (s-) and complex (c-) MND at the age of five in a population of preterm small and Abbreviations: MND, minor neurological dysfunction; c-MND, complex-minor neurological dysfunction; s-MND, simple minor neurological dysfunction; WPSSI-III-NL, Wechsler Preschool and Primary Scale of Intelligence third edition, Dutch version; IQ, intelligence quotient; PSQ, processing speed quotient; M-ABC-2, Movement Assessment Battery for children-2; VP, very preterm; SES, socio-economic status; ANT, Amsterdam Neuropsychological Tasks; RT, Reaction Time; NICU, neonatal intensive care unit. E-mail address: [email protected] (T. Kurpershoek).

http://dx.doi.org/10.1016/j.earlhumdev.2016.07.002 0378-3782/© 2016 Published by Elsevier Ireland Ltd.

appropriate for gestational age neonates is much higher, between 26 and 49% [2,4–6]. The presence of s-MND is considered to reflect the lower end of the normal distribution of physiological brain function [3]. C-MND is strongly related to pre- and perinatal risk factors and is considered as a borderline form of CP [3]. Known risk factors in prematurity (i.e. perinatal infection, intra-ventricular haemorrhage (IVH), chronic lung disease etc.) increase the risk of MND [3,6]. These risk factors also lead to diffuse white matter damage or localised damage in cerebellum and/or basal ganglia [3]. MRI studies have demonstrated that mild to moderate basal ganglia lesions and/or marked white matter damage are associated with MND [7]. Basal ganglia and cerebellum have thalamo-cortical connections to the pre-motor cortex as well as to prefrontal areas. Damage to these circuitries therefore may not only become apparent in slow processing speed but also in motor and learning difficulties that are characteristic in children with MND.

56

T. Kurpershoek et al. / Early Human Development 103 (2016) 55–60

In previous studies in a preterm population and also in a population of female healthy term born children, the presence of c-MND was indeed found to increase the risk of concurrent learning disabilities and behavioural problems [6,8,9]. Within the spectrum of MND, especially fine manipulative and coordination problems were associated with lower IQ [10]. Also, the development of autism spectrum and other psychiatric disorders in children born before 32 weeks gestational age was found to be related to MND [11,12]. Moreover, in previous papers we have demonstrated that abnormal motor scores were associated with c-MND [13]. Thus, in literature c-MND is a risk factor for a broad array of developmental problems. To get further insight in the association between cMND and its associated developmental problems, we set out to study, in a cohort of very preterm born children, the relationship between MND, neurocognitive measures and motor function with special attention to speed measures. We expected to find worse neurocognitive and motor skills in children with c-MND but not with s-MND.

invited at our outpatient clinic for long term follow-up assessments between December 2007 and June 2009. 2.2. Assessment schedule For every participant two separate visits were planned shortly after the child reached the corrected age of five years. During the first appointment one of three trained child psychologists assessed intelligence, inhibition, sustained attention and visual-motor coordination. In the second session within three months of the first visit, the psychologist assessed working memory, focused attention and processing speed, while one of three trained paediatricians assessed motor and neurological status. Perinatal and socio-economic characteristics were taken from the ongoing NICU database. A combined measure of parental education level was used as a proxy of socio-economic status (SES) [2]. Standard scores of the outcome measures were calculated from raw scores based on the corrected age at testing.

2. Methods 2.3. Outcome measures 2.1. Patients The present study is a single centre prospective cohort study as part of the follow-up program of the Neonatal Intensive Care Unit (NICU) from Emma's Children's Hospital in the Academic medical Centre, Amsterdam, the Netherlands. The study population consisted of children born at gestational age of b30 weeks or with a birth weight of b1000 g. Exclusion criteria were severe handicaps, as a result of which an age appropriate IQ test was not feasible, a genetic syndrome, participation in other studies that required a different study protocol or no previous follow-up visits to our clinic. Moreover, for the current study, the presence of CP N GMFSC N 1 was an exclusion criterion (see Fig. 1 for details of patient flow). For detailed description of the protocol see Potharst et al. [2]. At the corrected age of five years old, children were

2.3.1. Neurological development Neurological development was assessed using the standardised and age specific neurological examination according to Touwen [14]. Eight different domains were tested: posture, reflexes, fine manipulation, involuntary and associated movements, coordination, sensory deficits and cranial nerve dysfunction. According to the classification of HaddersAlgra [3], children were classified as neurologically normal if having no abnormal domains. If 1 or 2 domains were scored abnormally this was classified as s-MND and 3 or more abnormal domains as c-MND. CP was diagnosed when definite abnormalities in posture, tone and reflexes were found. Classification of CP was done according to Gross Motor Function Classification Scale (GMFSC) [15]. We divided the patients into two groups. The first (‘neuro-normal’) group consisted of

Fig. 1. Patient flow of inclusion.

T. Kurpershoek et al. / Early Human Development 103 (2016) 55–60

children with normal neurological examination or s-MND. The second (‘neuro-abnormal’) group included children with c-MND or nondisabling CP (GMFCS 1). 2.3.2. Intelligence Intelligence was measured using the Dutch version of third edition of the Wechsler Preschool and Primary scale of Intelligence (WPSSIIII-NL). Verbal intelligence quotient (VIQ) and performance IQ (PIQ) were calculated and also Processing Speed Quotient (PSQ) [16]. These scores were classified as abnormal or severely abnormal if they were more than one or two SD below the mean (m = 100, sd = 15) respectively. 2.3.3. Reaction time Reaction time (RT) was measured using the Baseline Speed task of the Amsterdam Neuropsychological Tasks (ANT) [17]. This test requires children to respond as quickly and accurately as possible to a stimulus change shown on a computer screen. RT was calculated as mean of the response time of both dominant and non-dominant hand responses. Scores were considered abnormal if more than one SD below the mean of the control group as previously described [2]. 2.3.4. Working memory Working memory was assessed using the Digit Span-task of the Wechsler Intelligence Scale for Children, third version in which children have to verbally repeat a line of numbers in increasing length first forwards and then backwards [18]. Dependent measure was the total number of correct trials when repeating backwards. 2.3.5. Focused attention Focused attention was measured using the Focused Attention Objects of the ANT battery [17]. This task requires the child to respond to the presence and location of an object by pushing one of two buttons. Dependent measures were the reaction time on correct responses and the proportion of incorrect responses. 2.3.6. Visual-motor coordination Visual-motor coordination was assessed using the Tracking and Pursuit tasks of the ANT battery [17]. In the Tracking task the child was asked to trace a circle with the mouse cursor. In the Pursuit task, the child had to follow a moving target with the mouse cursor. Dependent measures were the distance between the mouse and circle or target and the variability in this distance, measuring the precision and stability of precision. 2.3.7. Motor development Motor development was assessed using the Movement Assessment Battery for children (M-ABC) [19]. Nineteen percent (n = 18) of the children was assessed with the first and 81% (n = 76) with the second edition of the test. The M-ABC is a standardised age related test to identify motor problems in children using different tasks for different age bands. The eight tasks are grouped into three components: manual dexterity, aiming and catching and balance. Raw scores are converted into standard and percentile scores. M-ABC scores equal or below the 15th percentile for age are considered abnormal and equal or below the 5th percentile as severely abnormal. Because two different versions of the M-ABC were administered, we used the percentile scores as the dependent variable to enable comparison of scores. The original UK-norms were used. 2.4. Parental education Parents who had less than or equal to the lowest type of college were rated “low level of education” (total years postelementary schooling: b6) Parents who graduated from the middle level of college were rated with “middle level of education” (total years after elementary

57

schooling: 6–8). Parents who had highest level college or university were rated with “high level of education” (total years post-elementary schooling: N8). The combined parental education score was low if one or both parents had low level of education, middle if both parents had middle, and high if one or both parents had high education.

3. Statistical analyses Pearson χ2 test or Fisher's Exact Test and t-test were used to determine differences in the perinatal and sociodemographic characteristics, and all developmental outcomes, if appropriate. Effect sizes in terms of Cohen's d or Odds Ratio were calculated for all developmental outcome measures. One way analysis of variance (ANOVA) was used to establish differences in parental educational level between groups. All tests were repeated comparing children with normal neurological examination versus all forms of MND and CP, and also separate analyses were done excluding the mild CP cases (n = 4).

Table 1 Patient characteristics and sociodemographics. Perinatal characteristics

‘Neuro-normal’(n ‘Neuro-abnormal p-Value (n = 18)’ = 76)

Gestational age (wk), mean (SD) Birth weight(g), mean (SD) Sex: male, n (%) Multiple births, n (%) Small for gestational agea (bp2.3), n (%) Apgar score 5′,mean (SD) PROM, n (%) Surfactant, n (%) Sepsis, n (%) Postnatal steroids, n (%) Indomethacin for patent ductus arteriosus, n (%) Requiring ventilation, n(%) Days on ventilation, median(max) Days on O2, mean(SD) Bronchopulmonary dysplasia (oxygen at 36 weeks post menstrual age) Necrotising enterocolitis stage 2

28.7 (1.3) 1073.9 (250.3) 32 (42.1) 24 (31.6) 6 (7.9)

+ 3, n (%) Subependymal haemorrhage, n (%) Intraventricular haemorrhage 2 + 3, n (%) Periventricular leucomalacia, n (%) Social background characteristics Maternal age at date of birth (SD) Paternal age at date of birth (SD) Maternal Education, n (%) Low Middle High Paternal Education, n (%) Low Middle High Parental education combined, n (%) Low Middle High

29.0 (1.97) 953.3 (261) 10 (55.6) 2 (11.1) 6 (33.3)

0.71 0.072 0.302 0.140 0.01*

8.3 (1.8) 8 (10.5) 33 (43.3) 20 (26.3) 2 (2.6%) 20 (26.3)

8.3 (1.6) 0 (0.0) 7 (38.9) 3 (16.7) 2 (11.1) 3 (16.7)

0.935 0.346 0.727 0.546 0.164 0.546

40 (41.6) 1.0 (53)

10 (55.6) 3.0 (57)

0.823 0.083

27.8 (21.9) 10 (13.2)

39.3 (33.7) 4 (22.2)

0.077 0.459

1 (1.3)

1( 5.6)

0.348

18 (23.7)

5 (27.8)

0.763

4 (5.2)

1 (5.6)

0.983

2 (2.6)

2 (11.1)

0.164

31.3 (5.9) 34.2 (6.9)

32.3 (4.7) 34.5 (6.3)

20 (26.7) 30 (39.5) 25 (33.3)

6 (35.3) 5 (29.4) 6 (35.3)

18 (25.4) 21 (29.6) 32 (45.1)

5 (31.2) 4 (25.0) 7 (43.8)

0.524 0.861 0.674

0.873

0.977 24 (31.6) 19 (25.0) 33 (43.4)

7 (38.9) 2 (11.1) 9 (50.0)

* Indicates significant p-value. a SGA was defined as bp2.3 according to Dutch reference data.

58

T. Kurpershoek et al. / Early Human Development 103 (2016) 55–60

To avoid type I error due to multiple comparisons a two-tailed pvalue of b 0.01 was considered statistically significant. Outcomes with p-values between 0.01 and 0.05 were described as a trend. Statistical analyses were carried out using IBM SPSS, version 21.0 (SPSS Inc., Chicago, Il).

4. Results For detailed description of the research protocol see Potharst et al. [2]. For this paper we included all children with complete data on neurological development and intelligence. In total, 115 children were eligible for the study. Twelve children were lost to follow-up or refused to participate and in nine children no neurological examination was performed due to patient refusal or impossibility to plan an appointment, resulting in a cohort of 94 children. With respect to perinatal baseline characteristics listed in Table 1, participants (n = 94) did not differ from non-participants (n = 21). Forty-nine (52%) VP children had a normal neurological examination at the corrected age of five and 27 (28.7%) had s-MND. These children formed the ‘neuro-normal’ group. Fourteen VP children (14.9%) had c-MND and 4 (4.2%) VP children had CP and formed part of the ‘neuro-abnormal’ group. Table 1 shows perinatal characteristics. These were comparable for the neuro-normal and the neuro-abnormal group except for the percentage of children that were very small for gestational age (SGA). There was a trend towards more respiratory complications in the neuro-abnormal group. Importantly, socio-demographic variables were comparable between the groups. Table 2 shows the results for all neurodevelopmental assessments in the two study groups. Statistically significant differences between groups were found for PSQ and motor performance. There was a trend towards slower simple RT in the neuro-abnormal group. The children in the neuro-abnormal group were slower and also showed worse performance on M-ABC. Other measures were either slightly and nonsignificantly worse or comparable. Table 3 shows results on impairment level of the most important dependent measures. Significantly more VP children in the neuroabnormal group had abnormal simple RT and motor performance and there was a trend towards more children with an abnormal PSQ in the neuro-abnormal group, while differences in low PIQ did not reach statistical significance and VIQ was comparable. Exclusion of the children with CP (GMFCS1) led to comparable results. When we repeated analyses comparing all children with s- and

c-MND or mild CP (n = 45) to children with normal neurological examination (n = 49), no differences between groups were found. 5. Discussion This study showed that VP children with c-MND at the corrected age of five have significantly worse PSQ, RT and motor performance than VP children with a normal neurological examination or with s-MND. Focused attention and visuomotor coordination were somewhat, but not significantly, worse in children with c-MND, while PIQ- and VIQ as well as working memory were comparable between groups. To our knowledge, this study is the first to report on the relationship between slower processing speed and RT, and c-MND. C-MND can be considered as a distinct form of perinatally acquired brain dysfunction, which is likely associated with the cortico-striato-thalamo-cortical and cerebello-thalamo-cortical pathways [3]. In children with encephalopathy at birth, MND was associated with mild or moderate basal ganglia lesions or more marked white matter lesions on MRI at age 5 [7]. White matter injury occurs after premature birth due to harmful events i.e. inflammation/infection or hypoxia-ischaemia [20]. Also processing speed was found to be related to diffuse white matter injury and lower white matter volume [21]. It is thus conceivable that damage to cortico-striato-thalamo-cortical and cerebello-thalamo-cortical pathways as a result of wider spread white matter loss are causal to cMND and slower speed measures. In our study, in particular motor measures and speed measures with strong motor demands were found to be impaired in children with cMND, in contrast to verbal, attentional and executive functions that were not impaired. These results suggest that in particular the neurocognitive functions that require motor control under time pressure are vulnerable for damage in c-MND children. Slower processing speed may be an important factor underlying academic disabilities in VP children, and is found to be reduced in one third of preterm children born b 32 weeks at the age of five [22]. Lower PSQ and slower reaction time may lead to poorer executive functioning at a later age and are important factors in academic underachievement and behavioural problems in VPT [23–25]. Thus carrying out a neurological examination that includes diagnosis of both CP and simple and complex MND has several benefits. It contributes to diagnosing the neurological basis for motor impairments but also neurocognitive impairments. The prevalence of c-MND in other studies with VP children is variable and dependent on the risk profile of the population [6,26]. Differences in differentiation between either s-or c-MND in these studies

Table 2 Neurocognitive function tests.

Intelligence (WPPSI-III-NL) Verbal IQ, mean (SD) Performance IQ, mean (SD) Processing speed quotient, mean (SD) Reaction time (ANT) Baseline speed, mean (SD) Focused attention (ANT) Reaction time, mean(SD) Errors,% of trials (SD) Visuomotor coordination(ANT) Tracking precision, mean (SD) Tracking stability in mm (SD) Pursuit precision, mean (SD) Pursuit stability in mm (SD) Working memory (WISC) Number of correct trials (SD) Motor development (M-ABC) Percentile score, mean(SD)

‘Neuro-normal’ (n = 76)

‘Neuro-abnormal’ (n = 18)

p-Value

Effect size (Cohen's d)

96.5 (15.5) 95.4 (13.6) 97.9(16.4) (n = 73) 620.4 (157.0) (n = 67) 2092 (503) 12.4 (11.2) (n = 73) 7.08 (3.8) 5.3 (3.8) 12.1 (4.5) 10.7 (7.0) (n = 74) 6.35 (2.28)

97.9 (17.7) 90.9 (15.3) 85.5 (15.9) (n = 18) 709.2 (160.8) (n = 17) 2275 (584) 10.3(6.6) (n = 18) 9.88 (7.3) 7.7 (6.1) 13.8 (5.7) 11.4 (10.2) (n = 18) 6.28 (1.97)

0.731 0.221 0.005*

−0.08 0.31 0.76

44.7 (27.5)

16.1 (19.4)

Note: Positive effect sizes indicate better performance for the Neuro-normal group than for the Neuro-abnormal group. ⁎ Indicates significant p-value.

0.035

0.56

0.199 0.334

0.34 0.22

0.140 0.118 0.187 0.711

−0.48 −0.47 −0.33 −0.08

0.891

0.05

0.000*

1.2

T. Kurpershoek et al. / Early Human Development 103 (2016) 55–60

59

Table 3 Neurocognitive function tests according to disability level.

Intelligence: WPPSI Verbal IQ, n(%) b85 b70 Performance IQ, n(%) b85 b70 Processing speed quotient, n(%) b85 b70 Reaction time (ANT), n(%) b-1 SD b-2 SD Motor development (M-ABC), n(%) ≤p15 ≤p5

‘Neuro-normal’ (n = 76)

‘Neuro-abnormal’ (n = 18)

p-Value

19 (25.0) 5 (6.6)

4 (22.2) 1(5.6)

1.0 1.0

0.86 (0.25–2.92) 0.80 ( 0.09–7.6)

15 (19.7) 4 (5.3)

6 (33.3) 1 (5.6)

0.222 1.0

2.03 (0.66–6.30) 1.1 (0.1–10.1)

15(19.7) 5 (6.6) n = 73 19 (26.0) 10 (13.7)

9 (50.0) 2 (11.1) n = 18 11 (61.1) 5 (27.8)

0.014 0.616

4.07 (1.38–12.01) 1.78 (0.3–9.9)

0.005* 0.166

4.47 (1.51–13.18)* 2.42 (0.71–8.28)

14(18.4) 6 (7.9)

13(72.2) 7 (38.9)

0.000* 0.003*

11.51 (3.5–37.6)* 7.42 (2.10–26.23)*

Odds ratio (95% CI)

⁎ Indicates significant p-value.

may play a role in this variable prevalence. Patients that are, for example, classified as s-MND in the cohort of Mikkola et al. [26] would possibly be classified as c-MND in our study. Birth year of the cohort plays a role, since the prevalence of CP is decreasing which may lead to a relatively higher prevalence of MND [1]. Mikkola et al. found an incidence of CP of 14%, while 7% had c-MND in a cohort born in 1996 [26]. In our cohort born in 2002–2003 the CP rate was 7% and 15% had c-MND [2]. Using a short version of the Touwen examination in the EPIPAGE study which included 1237 children born at GA ≤ 32 weeks, Arnaud and colleagues [6] found that any form of MND was present in 44% of the preterm infants and in 29% of a term control group. MND was significantly associated with a total IQ b 70 or learning- and behavioural problems as reported by parents at the age of 5. A total IQ b 70 was found in 12% of children with any form of MND compared to 6% of children without MND. In a population of 341 nine-year old term born children, MND was found to be associated with PIQ and VIQ [10]. These term born children with c-MND had a ten point lower PIQ and a seven point lower VIQ than children without c-MND. We did not find a difference in VIQ or PIQ between groups. Lower numbers of children in our cohort may have led to reduced power, but the lack of differences between IQ-scores between the neuro-normal and neuro-abnormal children could also be due to the younger age of the children in our cohort. Since we did find important differences in speed measures, a next step research would be to study whether slower processing speed precedes a decline in IQ at a later age. The strengths of our study were, firstly, the broad array of neuropsychological tests. We thus found that especially speed and motor measures were associated with c-MND. Secondly, we chose to divide our patients into two groups based on presence of c-MND or mild CP, strengthening the case made by Hadders-Algra that s-MND reflects the lower part of normal brain development [3]. Indeed s-MND was not associated with worse PSQ, reaction time or M-ABC. Thirdly, to reduce a type I error the significant p-value was set at b0.01. Limitations of our study were the small sample size that could have resulted in a type II error, and the small range in GA in our patient group, that was probably too small to detect GA related differences in prevalence of MND [4,6]. However, we were able to confirm the finding that intrauterine growth restriction is correlated with the development of minor neurological sequelae [27,28]. Prevention of MND lies probably in decreasing perinatal infections and hypoxic-ischaemic events and therefore preventing brain damage. Once c-MND is diagnosed, referral to occupational therapy, physiotherapy and educational interventions to improve speed and motor functioning could be subject to further study in terms of secondary prevention strategies.

In follow-up of VP infants a neurological examination is customary. We advise to use an assessment that can diagnose both CP and forms of MND. The Touwen test to establish neurological dysfunction takes only 20–30 min and is done with pleasure by most children. In conclusion, c-MND is often seen in VP children and was found to be associated with reduced speed measures and motor problems. Thus using a neurological assessment that can diagnose both CP and MND, is useful in neonatal follow up at age five.

References [1] B.R. Vohr, L.L. Wright, W.K. Poole, S.A. McDonald, Neurodevelopmental outcomes of extremely low birth weight infants b32 weeks' gestation between 1993 and 1998, Pediatrics 116 (3) (2005) 635–643. [2] E.S. Potharst, A.G. van Wassenaer, B.A. Houtzager, J.W. van Hus, B.F. Last, J.H. Kok, High incidence of multi-domain disabilities in very preterm children at five years of age, J. Pediatr. 159 (1) (2011) 79–85. [3] M. Hadders-Algra, Two distinct forms of minor neurological dysfunction: perspectives emerging from a review of data of the Groningen perinatal project, Dev. Med. Child Neurol. 44 (8) (2002) 561–571. [4] M. Hadders-Algra, H.J. Huisjes, B.C. Touwen, Preterm or small-for-gestational-age infants. Neurological and behavioural development at the age of 6 years, Eur. J. Pediatr. 147 (5) (1988) 460–467. [5] A. Fily, P. Truffert, A. Ego, M.H. Depoortere, C. Haquin, V. Pierrat, Neurological assessment at five years of age in infants born preterm, Acta Paediatr. 92 (12) (2003) 1433–1437. [6] C. Arnaud, L. Daubisse-Marliac, M. White-Koning, V. Pierrat, B. Larroque, H. Grandjean, et al., Prevalence and associated factors of minor neuromotor dysfunctions at age 5 years in prematurely born children: the EPIPAGE study, Arch. Pediatr. Adolesc. Med. 161 (11) (2007) 1053–1061. [7] A. Barnett, E. Mercuri, M. Rutherford, L. Haataja, M.F. Frisone, S. Henderson, et al., Neurological and perceptual-motor outcome at 5–6 years of age in children with neonatal encephalopathy: relationship with neonatal brain MRI, Neuropediatrics 33 (5) (2002) 242–248. [8] H.K. Kikkert, D.E. JC, v. den Heuvel ER, M. Hadders-Algra, Minor neurological dysfunction and behaviour in 9-year-old children born at term: evidence for sex dimorphism, Dev. Med. Child Neurol. 55 (11) (2013) 1023–1029. [9] H.K. Kikkert, C. de Jong, M. Hadders-Algra, Minor neurological dysfunction and cognition in 9-year-olds born at term, Early Hum. Dev. 89 (5) (2013) 263–270. [10] H.K. Kikkert, D.E. JC, M. Hadders-Algra, Minor neurological dysfunction and IQ in 9year-old children born at term, Dev. Med. Child Neurol. 53 (4) (2011) e16–e25. [11] M. De Jong, M. Punt, E. De Groot, R.B. Minderaa, M. Hadders-Algra, Minor neurological dysfunction in children with autism spectrum disorder, Dev. Med. Child Neurol. 53 (7) (2011) 641–646. [12] P.F. Schothorst, H. Swaab-Barneveld, E.H. van Engeland, Psychiatric disorders and MND in non-handicapped preterm children. Prevalence and stability from school age into adolescence, Eur. Child Adolesc. Psychiatry 16 (7) (2007) 439–448. [13] J.W. Van Hus, E.S. Potharst, M. Jeukens-Visser, J.H. Kok, A.G. Van WassenaerLeemhuis, Motor impairment in very preterm-born children: links with other developmental deficits at 5 years of age, Dev. Med. Child Neurol. 56 (6) (2014) 587–594. [14] M. Hadders-Algra, Neurological Examination of the Child With Minor Neurological Dysfunction, 3 ed. Mac Keith Press, London, 2010. [15] R. Palisano, P. Rosenbaum, S. Walter, D. Russell, E. Wood, B. Galuppi, Development and reliability of a system to classify gross motor function in children with cerebral palsy, Dev. Med. Child Neurol. 39 (4) (1997) 214–223.

60

T. Kurpershoek et al. / Early Human Development 103 (2016) 55–60

[16] J. Hendriksen, P. Hurks, WPPSI-III-NL Wechsler Preschool and Primary Scale of intelligence-Third Edition. Nederlandse bewerking, Pearson Assesment and Information B.V, Amsterdam, 2009. [17] L. de Sonneville, Amsterdam neuropsychological tasks: a computer-aided assessment program, in: B. den Brinker, P. Bek, A. Brand, S. Maarse, L. Mulder (Eds.),Cognitive Ergonomics, Clinical Assessment and Computer-Assisted Learning: Computers in Psychology, 6 1999, pp. 187–203. [18] W. Kort, M. Schittekatte, P. Dekker, P. Verhaege, E. Compaan, M. Bosmans, Wechsler intelligence scale for children(Derde Editie NL) in: D. Wechsler (Ed.), Handleiding en Verantwoording, third ed.Harcourt Test Publishers/NIP Dienstencentrum, 2005. [19] S.E. Henderson, D.A. Sugden, A.L. Barnett, Movement Assessment Battery for Children, 2 ed. Harcourt Assessment, London, 2007. [20] A.R. Hart, E.W. Whitby, P.D. Griffiths, M.F. Smith, Magnetic resonance imaging and developmental outcome following preterm birth: review of current evidence, Dev. Med. Child Neurol. 50 (9) (2008) 655–663. [21] S. Soria-Pastor, M. Gimenez, A. Narberhaus, C. Falcon, F. Botet, N. Bargallo, et al., Patterns of cerebral white matter damage and cognitive impairment in adolescents born very preterm, Int. J. Dev. Neurosci.:Off. J. Int. Soc. Dev. Neurosci. 26 (7) (2008) 647–654. [22] M. Gnigler, V. Neubauer, E. Griesmaier, S. Zotter, K. Kager, U. Kiechl-Kohlendorfer, Very preterm children risk reduced processing speed at five-years-of-age, predicted

[23]

[24]

[25]

[26]

[27]

[28]

by typical complications of prematurity and prenatal smoking, Acta Paediatr. (2014). H. Mulder, N.J. Pitchford, N. Marlow, Processing speed and working memory underlie academic attainment in very preterm children, Arch. Dis. Child. Fetal Neonatal Ed. 95 (4) (2010) F267–F272. H. Mulder, N.J. Pitchford, N. Marlow, Processing speed mediates executive function difficulties in very preterm children in middle childhood, J. Int. Neuropsychol. Soc.: JINS. 17 (3) (2011) 445–454. S.A. Rose, J.F. Feldman, J.J. Jankowski, Modeling a cascade of effects: the role of speed and executive functioning in preterm/full-term differences in academic achievement, Dev. Sci. 14 (5) (2011) 1161–1175. K. Mikkola, N. Ritari, V. Tommiska, T. Salokorpi, L. Lehtonen, O. Tammela, et al., Neurodevelopmental outcome at 5 years of age of a national cohort of extremely low birth weight infants who were born in 1996–1997, Pediatrics 116 (6) (2005) 1391–1400. J.H. Kok, A.L. den Ouden, S.P. Verloove-Vanhorick, R. Brand, Outcome of very preterm small for gestational age infants: the first nine years of life, Br. J. Obstet. Gynaecol. 105 (2) (1998) 162–168. I. Guellec, A. Lapillonne, S. Renolleau, M.L. Charlaluk, J.C. Roze, S. Marret, et al., Neurologic outcomes at school age in very preterm infants born with severe or mild growth restriction, Pediatrics 127 (4) (2011) e883–e891.