lable at ScienceDirect

Pediatric Neurology 51 (2014) 681e687

Contents lists avai

Pediatric Neurology

journal homepage: www.elsevier .com/locate/pnu

Original Article

Relationship Between Neurological Assessments of PretermInfants in the First 2 Years and Cognitive Outcome at School Age

Jana Kodric PhD a, Breda Sustersic MDb, Darja Paro-Panjan MD, PhD c,*

aDivision of Paediatrics, Department of Neurology, University Medical Centre Ljubljana, Ljubljana, SloveniabDevelopmental Department, Health Centre, Domzale, SloveniacDivision of Paediatrics, Department of Neonatology, University Medical Centre Ljubljana, Ljubljana, Slovenia

Article HistReceived Ju* Commu

of PaediatrLjubljana;

E-mail a

0887-8994/$http://dx.doi

abstract

BACKGROUND: The risk of cognitive disability in preterm

infants is higher than in general population. The Amiel-Tison neurological assessment could be a useful tool for early identification of preterm children at risk of cogni-tive disability in school age. This study investigated the value of categorization of neurological signs assessed bythe Amiel-Tison neurological assessment in the first 2 years of life in relation to cognitive performance at schoolage in a group of preterm children. METHODS: Preterm children with gestational age from 23 to 36 weeks wereincluded in the prospective study. From the initial group of 45 children, in whom the Amiel-Tison neurologicalassessment was performed at term age, at 3 months corrected age, and at 2 years, the Wechsler Intelligence Scalefor Childrenethird edition was performed in 39 children after school entry. RESULTS: Full scale IQ, Verbal IQ, andPerformance IQ of the whole group of preterm children were not significantly different from the normative data;most of the children had IQ scores in the normal range (�85). The mean cognitive results of children decreased asthe number of neurological signs increased. There was a significant correlation between the categories ofneurological signs at 2 years and later cognitive results. CONCLUSIONS: The grade of severity of neurological signs at2 years was associated with the cognitive results at school age. The categorization of neurological signs accordingto the Amiel-Tison neurological assessment in preterm children might have prognostic value for cognitive outcomeat school age.

Keywords: Amiel-Tison neurological assessment, preterm, cognitive, outcome

Pediatr Neurol 2014; 51: 681-687

� 2014 Elsevier Inc. All rights reserved.

Introduction

Medical and technological advances have improved thesurvival of preterm infants, but neurodevelopmentaldisabilities remain the most important consequence ofpreterm birth. Many preterm infants require additionalmedical and educational resources throughout childhoodinto adolescence, so it is important to identify thesechildren early. In recent years, the research focus in devel-opmental outcome of preterm infants has shifted from thestudy of major handicaps, such as intellectual disability,

ory:ne 9, 2014; Accepted in final form July 17, 2014nications should be addressed to: Dr. Paro-Panjan; Divisionics; Department of Neonatology; University Medical CentreBohoriceva 20; Ljubljana 1000, Slovenia.ddress: darja.paro@gmail.com

- see front matter � 2014 Elsevier Inc. All rights reserved..org/10.1016/j.pediatrneurol.2014.07.024

sensorineural hearing loss or blindness, cerebral palsy (CP),and epilepsy, to less severe and more frequent dysfunctionssuch as learning disabilities, borderline intelligence,neuropsychologic deficits, attention deficit hyperactivitydisorder, and other behavioral problems.1,2 Although newmeasures of brain structure and function have improved theability to identify preterm infants at risk of developmentaldisabilities, neurological assessment remains an importanttool in daily clinical practice.

The structural, standardized, and age-dependent Amiel-Tison neurological assessment (ATNA) covers growthparameters and cranial morphology, neurosensory aspects,passive and active muscle tone, spontaneous motor activity,and primary reflexes. It considers signs that depend on theintegrity of the upper structures (cranial signs, alertness,and axial tone) and enables a clinician to recognize minorneurological signs from the neonatal period to the age of 6years.3,4 Minor neurological signs, which reflect minor brain

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687682

damage occurring before, during, or after birth, present awindow into central nervous system disruption, which maypresent as intellectual, academic, and behavioral problems.2

Amiel-Tison et al. studied three minor neurological andcranial signs, two of which refer to passive tone (imbalancein axial tone with excessive dorsal extension and a phasicstretch reflex in one or both gastrocnemius muscles), and apalpable ridge on the squamous suturesdthe Amiel-Tisontriad (ATT). They demonstrated that these signs wereimportant in documenting the relationship between mildbrain damage and possible future learning disabilities.4

Significant correlations between the categories of neuro-logical signs and the Bayley-II Mental Developmental Indexhave been found at age 2 years.5,6 At age 2 to 5 years, sig-nificant differences in the coordination, language, andpractical reasoning according to categories of ATNA werepresent.4 Parents of children with abnormal neurologicalsigns on ATNA reported significantly more problems inmotor and/or praxis skills, language development, andattention at age 3 to 5 years in comparison with parents ofchildren with optimal neurological status.7 In the longitu-dinal study, in which preterm children were classified intothree groups (normal, impaired without disability, and withassociated disability) at age 1 year, the results of ATNAwerepredictive of the cognitive performance on GriffithsDevelopmental Test at age 4, 8, and 14-15 years.8-10

The aim of the study was to evaluate the usefulness ofATNA for predicting developmental disabilities and toanalyze the correlation between neurological signs assessedat three periods (newborn, infant, and toddler age) andcognitive performance at school age in a heterogeneousgroup of preterm children. The study is a part of a longi-tudinal follow-up of preterm children in which ATNAproved to be a useful clinical tool for predicting cognitiveoutcome at 2 years.5

Materials and Methods

Participants

In a single-centre longitudinal cohort study, we monitored a group of45 preterm infants (23 boys and 22 girls) born in the period 2001-2004.All except three children were born at the Ljubljana Maternity Hospital.The inclusion criteriawere (1) gestational age of 36weeks or less; (2) oneor more neonatal complications; and (3) first neurological examinationat the corrected age of 40 weeks (�5 days). Children with dysmorphicsyndromes and chromosomal abnormalities were excluded from thestudy.

Measures

Neurological status was assessed using ATNA. The assessment ofneurological status is age dependent, with different items being assessedat different ages. The interpretation of findings relies on the pattern ofmaturation of the subcorticospinal and corticospinal motor controlsystems.3 The scoring is based on a three-point scale, where 0 indicates atypical result for that age, 1 indicates a moderately abnormal result forthat age, and 2 indicates a definitely abnormal result. The final synthesisof the neurological status relies on clusters of neurological signs thatproduce a categorization of the severity of neurological signs. The cate-gorization varies at different ages: at term age, there are three categoriesof neurological signs for preterm infants (normal, minor to moderate,and severe degrees of neurological signs); at 3 months, there are fourcategories (normal, minor, moderate, and severe degrees of neurologicalsigns); and at 2 years, there are six categories (normal, intermediate with

one ATT sign, intermediate with two ATT signs, ATT, minimal CP, and CP).The detailed procedures, scoring, and categorization of neurologicalsigns are described elsewhere.3,11,12

Cognitive abilities at school age were assessed using the WechslerIntelligence Scale for Childrenethird edition.13 Three composite scoreswere calculated: Full Scale IQ (FIQ), Verbal IQ (VIQ), and Performance IQ(PIQ). The results of IQ scores were considered as severely abnormalwhen the S.D. values were below �2, as mildly abnormal whenbetween �1 and �2, and normal when �1 below the mean or higher.

After the cognitive assessment at school age, parents were asked iftheir child was recognized as having special education needs accordingto the Slovenian Special Education Act.

Procedure

After discharge from the maternity hospital, children were routinelyreferred to their regional outpatient Developmental Centre. Infantswhose parents agreed to participate and signed an informed consentform were consequently enrolled in the study. ATNA was performed byan experienced pediatrician at term age and repeated every 3 months upto the age of 2 years when the final categorization was made. Whenchildren started school, they were invited for cognitive assessment; ifthere was no response, another postal and then telephone contact wereattempted; and if there was still no response, a final invitation letter wassent a few months later. An experienced psychologist performed thecognitive assessment individually.

The National Medical Ethics Committee in Slovenia approved thestudy.

Statistical analysis

The mean and the S.D. of FIQ, VIQ, and PIQ were used in analysis. Themeasures of skewness and kurtosis were calculated for the distributionof the IQ scores. The z scores for skewness and kurtosis were 0.30and �0.51, respectively, for the FIQ; 0.55 and 0.32, respectively, for theVIQ; and 0.21 and �0.53, respectively, for the PIQ. Both measures werebelow the limit of z score of �2.58, the level at which the distribution isconsidered to be significantly skewed or kurtic.14 A one-group t test wasused to compare the results of the studied group with the referencepopulation. P values <0.05 were considered significant. The effect sizewas computed using Cohen d, where a d of 0.2 represents a small, d of 0.5a medium, and d of 0.8 a large effect size.14 Spearman’s rho (r) correla-tion coefficient was used to assess the relationships between categoriesof children according to neurological signs with cognitive outcome (FIQ,NIQ, and PIQ). Statistical analysis was performed using the IBM SPSSStatistics, version 20. Categorization of neurological status as normal andabnormal was used for the computation of sensitivity, specificity, andpositive predictive value and negative predictive value. This was per-formed using MedCalc for Windows, version 12.5 (MedCalc Software,Ostend, Belgium).

Results

Participants

Initially, 45 infants were included in the study. Theywereborn with a gestational age from 23 to 36 weeks (meangestational age, 31.6 weeks; S.D., 3.3 weeks) and birthweight from 525 to 3240 g (mean,1788 g; S.D., 718 g). Seveninfants had extremely low birth weights, 17 infants hadbirth weights below 1500 g. Six infants were small forgestational age, two were large for gestational age, and theothers were appropriate for gestational age. There wereeight pairs of twins. Twenty-six children experienced oneadditional neonatal complication besides prematurity,whereas 19 experienced two or more. The perinatal andneonatal characteristics of the participants are presented inTable 1. Neurological assessments using ATNA were

TABLE 1.Characteristics of Participants

Characteristics Number of Children (%)

Boys/girls 23/22 (51/49)Gestational age (wk)23-27 8 (18)28-32 15 (33)33-36 22 (49)

Birth weight (g)�999 7 (16)1000-1499 10 (22)�1500 28 (62)

Apgar score, �7 at 5 min 12 (27)Mechanical ventilation, >1 wk 28 (62)Sepsis 18 (40)CNS infection 3 (7)Seizures 6 (13)Hyperbilirubinemia exchange transfusion 4 (9)Surgical intervention 2 (4)CNS hemorrhage 6 (13)

Abbreviation:CNS ¼ Central nervous system

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687 683

successively preformed from term age to 2 years. Thenumbers of children in different categories of the classifi-cation according to neurological signs are presented inTable 2.

From the initial group, 39 children participated in thecognitive assessment between 6 years 10 months and9 years 8 months (mean age, 7.9 years; S.D., 0.8 years). Theparents of four children refused to participate, one child wasuncontactable, and one child was unable to cooperate withthe assessment because of moderate intellectual disability.

TABLE 2.Results of WISC-III in Different Categories of ATNA at Three Different Assessment Period

ATNA n WISC-III

FIQ

Mean S.D. d

At term ageNormal 10 103.30 13.42 0.22Mild/moderate 26 98.27 17.08 �0.11Severe 3 87.33 0.58 �0.84

At 3 mo corrected ageNormal 13 104.54 13.43 0.30Mild 11 100.18 18.16 0.01Moderate 12 93.92 16.30 �0.40Severe 3 87.33 0.57 �0.84

At 2 yrNormal 16 106.81 12.51 0.45Intermediate one sign 6 102.83 15.20 0.19Intermediate two signs 9 96.78 16.00 �0.21Amiel-Tison triad 3 79.67 11.59 �1.35MCP 2 81.00 8.48 �1.27CP 3 84.00 6.08 �1.07

TOTAL 39 98.72 15.82 �0.08

Abbreviations:ATNA ¼ Amiel-Tison neurological assessmentCP ¼ Cerebral palsyFIQ ¼ Full Scale IQMCP ¼ Minimal cerebral palsyPIQ ¼ Performance IQVIQ ¼ Verbal IQWISC-III ¼ Wechsler Intelligence Scale for Childrenethird edition

She had a severe degree of neurological signs at term ageand at 3 months corrected age, and CP was diagnosed at2 years. Her Mental Developmental Index on the BayleyScales of Infant Developmentesecond edition was below50, and she was attending a special education program atthe time of the follow-up.

Cognitive outcome at school age

Results of intelligence test (FIQ, VIQ, and PIQ) of allchildren were in the normal or mildly abnormal range,whereas none had results in the severely abnormal range(Figure). The FIQ scores ranged from 72 to 132, VIQ rangedfrom 72 to 140, and PIQ ranged from 71 to 133. The mean IQscores of the whole group of preterm children at 7 yearswere in the normal range (Table 2). They were not signifi-cantly different from the normative data (FIQ: t(38) ¼ �0.51,P ¼ 0.62; VIQ: t(38) ¼ �0.49, P ¼ 0.63; PIQ: t(38) ¼ �0.36,P ¼ 0.72). The percentage of children with IQ 85 or higherwas 79.49, 87.18, and 76.92 for FIQ, VIQ, and PIQ,respectively.

The FIQ, VIQ, and PIQ scores were the highest in thegroups of children with normal neurological status at termage, at 3 months corrected age, and at 2 years. The IQ scoresdecreased according to the number of neurological signs(Table 2). Children with a severely abnormal neurologicalstatus achieved the lowest IQ scores. The average IQ scoresfor three childrenwith severe neurological signs at term ageand at 3 months corrected age (two of them had CP and onehad MCP at 2 years) were significantly lower than thenormative mean. The children in whom the ATT was diag-nosed at age 2 years achieved the lowest scores on FIQ and

s and the Effect Size Compared With Normative Data

VIQ PIQ

Mean S.D. d Mean S.D. d

102.30 12.11 0.15 104.10 15.79 0.2798.46 16.27 �0.10 98.65 16.08 �0.0990.67 2.52 �0.62 86.00 4.00 �0.93

103.77 14.87 0.25 104.46 13.19 0.30100.00 16.28 0.00 100.73 17.47 0.0594.50 14.14 �0.37 95.00 17.11 �0.3490.67 2.52 �0.62 86.00 4.00 �0.93

105.88 14.06 0.39 106.88 12.31 0.46102.83 10.76 0.19 101.67 17.59 0.1197.67 13.42 �0.15 96.67 15.29 �0.2276.67 4.51 �1.56 88.33 19.30 �0.7884.50 12.02 �1.03 80.00 2.83 �1.3388.67 2.08 �0.75 83.00 8.89 �1.1398.85 14.75 �0.08 99.08 15.82 �0.06

FIGURE.Distribution of FIQ scores at different ages according to categories of neurological signs assessed by ATNA. ATNA, Amiel-Tison neurological assessment;ATT1, intermediate with one sign; ATT2, intermediate with two signs; ATT3, Amiel-Tison triad; CP, cerebral palsy; FIQ, Full Scale IQ; MCP, minimal cerebralpalsy; PIQ, Performance IQ; VIQ, Verbal IQ.

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687684

VIQ. Children with the ATT, MCP, and CP had significantlylower results compared with the normative mean. Themean score for verbal abilities (VIQ) was higher than themean score for nonverbal abilities (PIQ) in the group ofchildrenwith MCP and CP, whereas the profile was reversedin the group of children with the ATT.

The categories of neurological signs assessed by ATNA at2 years were significantly correlated with the FIQ, VIQ, andPIQ at school age (Table 3). The correlation between thecategory of neurological signs at 3 months and cognitiveoutcome at school age was significant for FIQ and PIQ andapproaching significance for VIQ, whereas the correlationbetween neurological signs at term age and cognitiveoutcome was not significant.

The sensitivity of neurological signs assessed by ATNA toidentify children with mildly abnormal cognitive outcomesand the negative predictive value were high at term age, 3months, and 2 years, whereas the specificity and positivepredictive value were low at all three ages (Table 4).

Eight children needed special education assistance inmainstream schools; none of whom had a normal

neurological status at 3months corrected age and at 2 years.Two of them had two neurological signs; three had the ATT,oneMCP, and two had CP. Parents of two other children (onein the intermediate group with two signs and one from thegroup with CP at 2 years) reported significant learningproblems, although the children had not yet been recog-nized as children with special education needs at school.

Discussion

The risk of cognitive disability in preterm infants ishigher than in term controls, although it is reported thatmost of the children born prematurely have IQs within thenormal range.15 The results of the cognitive outcome mea-sures in our group (FIQ, VIQ, and PIQ) are consistent withthis observation because they were not significantlydifferent from the normative data, and the variability was inthe expected range. Results of all children on the Full Scale,Verbal Scale, and Performance Scale were in the normal ormildly abnormal range (>70), with most of the childrenachieving scores of 85 or higher. Aylward described a

TABLE 3.The Correlations Between ATNA at Three Different Assessment Periods and FullScale, Verbal, and Nonverbal IQ of WISC-III

ATNA WISC-III

FIQ VIQ PIQ

At term ager 0.261 0.218 0.26195% CI �0.01 to 0.49 �0.07 to 0.45 �0.03 to 0.51P 0.11 0.18 0.11

At 3 mo corrected ager 0.355 0.303 0.32995% CI 0.07-0.56 0.04-0.52 0.02-0.55P 0.03 0.06 0.04

At 2 yrr 0.608 0.552 0.54395% CI 0.35-0.77 0.28-0.74 0.26-0.74P <0.001 <0.001 <0.001

Abbreviations:r ¼ Spearman’s rhoATNA ¼ Amiel-Tison neurological assessmentCI ¼ Confidence intervalFIQ ¼ Full Scale IQP ¼ Level of significancePIQ ¼ Performance IQVIQ ¼ Verbal IQWISC-III ¼ Wechsler Intelligence Scale for Childrenethird edition

TABLE 4.Sensitivity, Specificity, PPV, and NPV of ATNA at Three Different Assessment PeriodsWith Respect to Full Scale, Verbal, and Nonverbal IQ of WISC-III

Predictive Indexes Term Age

FIQ VIQ PIQ

Sensitivity 0.87 0.80 0.8995% CI 0.46-0.99 0.30-0.99 0.51-0.99Specificity 0.29 0.26 0.3095% CI 0.15-0.48 0.13-0.44 0.15-0.49PPV 0.24 0.14 0.2895% CI 0.11-0.44 0.04-0.32 0.13-0.47NPV 0.90 0.90 0.9095% CI 0.54-0.99 0.67-0.95 0.54-0.99

3 mo

FIQ VIQ PIQ

Sensitivity 1.00 1.00 0.8995% CI 0.60-1.00 0.46-1.00 0.51-0.99Specificity 0.42 0.38 0.4095% CI 0.25-0.61 0.23-0.56 0.23-0.59PPV 0.31 0.19 0.3195% CI 0.15-0.52 0.07-0.40 0.15-0.52NPV 1.00 1.00 0.9295% CI 0.72-1.00 0.72-1.00 0.62-0.99

2 yr

FIQ VIQ PIQ

Sensitivity 1.00 1.00 0.8995% CI 0.60-1.00 0.46-1.00 0.51-0.99Specificity 0.52 0.47 0.5095% CI 0.33-0.69 0.30-0.64 0.31-0.68PPV 0.35 0.22 0.3595% CI 0.17-0.57 0.08-0.44 0.17-0.57NPV 1.00 1.00 0.9495% CI 0.75-1.00 0.76-1.00 0.68-1.00

Abbreviations:ATNA ¼ Amiel-Tison neurological assessmentCI ¼ Confidence intervalFIQ ¼ Full Scale IQNPV ¼ Negative predictive valuePIQ ¼ Performance IQPPV ¼ Positive predictive valueVIQ ¼ Verbal IQWISC-III ¼ Wechsler Intelligence Scale for Childrenethird edition

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687 685

gradient for the cognitive outcomes according to the bio-logical risk factors. He demonstrated that premature chil-dren with higher birth weights tend to achieve higher IQscores, whereas very and extremely lowebirth weightchildrenwithout any severe complications achieve a 0.3-0.6S.D. lower IQ compared with the control group (their resultsare in the borderline to average range, with the mode of lowaverage).2 Late preterm children without abnormal neuro-logical findings tend to have IQs within the normal range atpreschool age.16 Besides the biological factors, neuro-imaging evidence of brain injury, male sex, severity ofneonatal illness, chronic lung disease, and abnormalities inthe neurological assessment have been evident to be asso-ciated with cognitive outcomes.1,2 Because of the numberand interplay of risk factors, a systematic clinical assess-ment that would be practical for daily use in the follow-upof preterm infants would be a meaningful tool for earlyidentification of children with possible cognitive and aca-demic problems. Our study aimed to assess the predictivepower of ATNA at different ages to identify children withproblems in cognitive development.

We found that childrenwith severe neurological signs onATNA at term age, those with moderate and severe neuro-logical signs at age 3 months, and children with the ATT,mild CP, and CP at age 2 years achieved lower average IQs atschool age. Analysis of cognitive results in relation to ATNAat 2 years revealed that cognitive results declined accordingto the number of ATT signs, as the children with the ATTachieved the lowest results, and the results of children withthe ATT, mild CP, and CP were significantly lower than thenormative mean.5 We also demonstrated a correlation be-tween the progressive decrease in cognitive performanceand the increase in the number of neurological signs in thegroups of children with ATT, mild CP, and CP. This result isconsistent with the observation of Gosselin et al.4 whoclaim that the number of minor neurological signs reflects a

continuum in the degree of neurological impairments thatshare the same pathophysiologic background and antici-pates consequences. Lower cognitive abilities, even in thelower average range, can cause a child to be academicallydisadvantaged.2 All the children from our study whoneeded special education assistance during the first years ofschooling were from the groups with two or more neuro-logical signs at age 2 years.

Analysis of the cognitive results of preterm childrenaccording to their results of ATNA at different ages revealsthat childrenwith normal neurological status at term age, at3 months corrected age, and at 2 years have normalcognitive results. Cognitive results of the children withmild-to-moderate grades of neurological signs at term agewere not significantly different from the normative mean,whereas the group of children with the severe grade ofneurological signs had significantly lower results. It hasalready been demonstrated that ATNA at term age ispredictive of developmental outcome in preterm and terminfants. In our previous study of term infants with riskfactors, good agreement between ATNA at term age and the

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687686

developmental quotient at 12-15 months was established.17

Murray et al. also demonstrated a good predictive value ofATNA at term for neurodevelopmental outcome in thegroup of infants with hypoxic-ischemic encephalopathy.The authors also demonstrated changes in the neurologicalstatus at consecutive assessments and found the highestcorrelations between neurodevelopmental outcome andneurological assessment at discharge.18 Similar changes inthe neurological status at consecutive assessments of terminfants were also described in the study of Amiel-Tison.11

These observations, which stress the importance of long-term follow-up, were also proven in our study. ATNA atterm age and at 3 months corrected age had lower predic-tive values for cognitive outcome than ATNA at age 2 years.We also demonstrated a higher correlation betweencognitive outcome and neurological assessments at an olderage. As Aylward stated, it is difficult to determine the pre-dictive value of minor neurological signs identified in earlyinfancy because they may be transient and result fromcontinuing recovery or catch-up from the negative effects ofprematurity or reflect the emergence of a more permanenthandicap.2 Our observations could be explained by theplasticity of the brain, with the processes of recovery andreorganization that occur in preterm infants,2 and theeffects of neurodevelopmental interventions, as all infantswith abnormal neurological signs participated in an earlyneurodevelopmental program that stimulates not onlymotor but also cognitive development in the preschoolperiod.19 The results should also be interpreted in the lightof the observation of Amiel-Tison who reported poor sta-bility of the ATNA scores in the first weeks of life because ofrapid changes in the infant’s development rather than tounreliability of the clinical instrument. For this reason, shesuggested repeated assessments.11 As shown in otherstudies, the persistence of a severely abnormal neurologicalstatus is correlated with more adverse cognitive out-comes.18 Using normal results of ATNA as the criterion, highsensitivity and high negative predictive values and lowspecificity and low positive predictive values of theassessment method were found for all three assessmentperiods. This information is important for clinical workwithchildren at risk and their parents, reassuring them in casesof optimal ATNA, in spite of the existence of risk factorsassociated with prematurity, explaining the possibility ofchanges in the direction of a more favorable outcome incases with minor neurological signs, and providing thera-peutic and special education assistance from an early agefor the children with persistent and/or severely abnormalneurological signs.

We are aware of the limitations of our study created bythe small overall sample size and the small number ofchildren in individual groups; but in spite of this fact, webelieve that our results contribute to the validation of theassessment of neurological signs as a tool for the detectionof children with possible lower cognitive outcomes andpotential learning disabilities requiring special educationassistance at school. Although the use of IQ as an average ofcognitive function could mask specific cognitive deficits(e.g., sensorimotor functions, visuospatial functions, lan-guage, memory, executive functions) that could have anegative impact on later academic performance and qualityof life, it still presents a measure of a child’s global cognitive

functioning. Another limitation of the study is the hetero-geneity of the studied group with regard to the gestationalage of children and complications during the neonatalperiod. In our opinion, this bias should not reduce the valueof the results, because the purpose of the study was toanalyze neurological signs in relation to the developmentaloutcome, regardless of the etiology. On the other hand,many studies have proved that moderately preterm infants(macropremies or moderately low birth infants) are also atrisk of developmental problems, so follow-up should bebased on the neurological examination and not only ongestational age and other risk factors.20

Our results support the use of ATNA as an instrument foridentifying children at risk of adverse cognitive outcomes.Because there is evidence of decreasing IQ in the period afterchildhood,21 and the consequences of preterm birth areassociated with lower educational attainment and subse-quent occupational and economic status later in life,22 weintend to assess cognitive functions and quality of life inadolescence and further prove the valueof the current results.

References

1. Allen MC. Neurodevelopmental outcomes of preterm infants. CurrOpin Neurol. 2008;21:123-128.

2. Aylward GP. Neurodevelopmental outcomes of infants born pre-maturely. J Dev Behav Pediatr. 2005;26:427-440.

3. Amiel-Tison C, Gosselin J. Neurological Development From Birth to SixYears: Guide for Examination and Evaluation. Baltimore: The JohnsHopkins University Press; 2001.

4. Gosselin J, Amiel-TisonC, Infante-Rivard C, FouronC, Fouron JC.Minorneurological signs and developmental performance in high riskchildren at preschool age. Dev Med Child Neurol. 2002;44:323-328.

5. Paro-Panjan D, Kodric J, Sustersic B. Association between neuro-logical signs and developmental outcome: pilot results in pretermgroup. Croat Med J. 2009;50:345-350.

6. Simard MN, Lambert J, Lachance C, Audibert F, Gosselin J. Predictionof developmental performance in preterm infants at two years ofcorrected age: contribution of the neurological assessment at termage. Early Hum Dev. 2011;87:799-804.

7. Amiel-Tison C, Njiokiktjien C, Vaivre-Douret L, Verschoor CA,Chavanne E, Garel M. Relation of early neuromotor and cranial signswith neuropsychological outcome at 4 years. Brain Dev. 1996;18:280-286.

8. Roth S, Wyatt J, Baudin J, et al. Neurodevelopmental status at 1 yearpredicts neuropsychiatric outcome at 14-15 years of age in verypreterm infants. Early Hum Dev. 2001;65:81-89.

9. Roth SC, Baudin J, Pezzani-Goldsmith M, Townsend J, Reynolds EO,Stewart AL. Relation between neurodevelopmental status of verypreterm infants at one and eight years. Dev Med Child Neurol. 1994;36:1049-1062.

10. Stewart AL, Costello AM, Hamilton PA, et al. Relationship betweenneurodevelopmental status of very preterm infants at one and fouryears. Dev Med Child Neurol. 1989;31:756-765.

11. Amiel-Tison C. Update of the Amiel-Tison neurologic assessment forthe term neonate or at 40 weeks corrected age. Pediatr Neurol.2002;27:196-212.

12. Gosselin J, Gahagan S, Amiel-Tison C. The Amiel-Tison neurologicalassessment at term: conceptual and methodological continuity inthe course of follow-up. Ment Retard Dev Disabil Res Rev. 2005;11:34-51.

13. Wechsler D.Wechsler Intelligence Scale for Children. Slovenian version.3rd ed. Ljubljana: Center za psihodiagnosti�cna sredstva; 2001.

14. Clark-Carter D. Quantitative Psychological Research: A Student’sHandbook. 2nd ed. Hove: Psychology Press; 2004.

15. Allen MC. Outcomes after brain injury in the preterm infant. In:Michael S, Steven M, eds. Acquired Brain Injury in the Fetus andNewborn. London: Mac Keith Press; 2012:99-120.

J. Kodric et al. / Pediatric Neurology 51 (2014) 681e687 687

16. Romeo DM, Guzzardi S, Ricci D, et al. Longitudinal cognitiveassessment in healthy late preterm infants. Eur J Paediatr Neurol.2012;16:243-247.

17. Paro-Panjan D, Neubauer D, Kodric J, Bratanic B. Amiel-Tisonneurological assessment at term age: clinical application, correla-tion with other methods, and outcome at 12 to 15 months. Dev MedChild Neurol. 2005;47:19-26.

18. Murray DM, Bala P, O’Connor CM, Ryan CA, Connolly S,Boylan GB. The predictive value of early neurological examina-tion in neonatal hypoxic-ischaemic encephalopathy and neuro-developmental outcome at 24 months. Dev Med Child Neurol.2010;52:e55-59.

If mankind is to profit freely from the small and sporadic cropdelicate art of handling ideas. Psychology is now able to tell us wfreedom of thought and to new ideas is fear; and fear which careasoned skepticism, or on occasion even as courage.

19. Orton J, Spittle A, Doyle L, Anderson P, Boyd R. Do early interventionprogrammes improve cognitive and motor outcomes for preterminfants after discharge? A systematic review. Dev Med Child Neurol.2009;51:851-859.

20. Amiel-Tison C, Allen MC, Lebrun F, Rogowski J. Macropremies:underprivileged newborns. Ment Retard Dev Disabil Res Rev. 2002;8:281-292.

21. Isaacs EB, Edmonds CJ, Chong WK, Lucas A, Morley R, Gadian DG.Brain morphometry and IQ measurements in preterm children.Brain. 2004;127(Pt 12):2595-2607.

22. Breslau N, Paneth NS, Lucia VC. The lingering academic deficits oflow birth weight children. Pediatrics. 2004;114:1035-1040.

of heroically gifted it produces, it will have to cultivate theith reasonable assurance that the most influential obstacle ton with inimitable art disguise itself as caution, or sanity, or

Wilfred Trotter, 1941