SEARCH

SEARCH BY CITATION

Keywords:

  • Numerical skills;
  • Preterm infants;
  • Risk factors

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

Aim

To unravel risk predictors for impaired numerical skills at 5 years of age in a population-based cohort of very preterm infants.

Methods

Between January 2003 and August 2006, we prospectively enrolled all infants born in Tyrol with <32 weeks of gestation. A total of 161 of 223 preterm infants (participation rate 72.2%) had a detailed examination at 5 years of age including cognitive assessment (Hannover-Wechsler Intelligence Test for preschool children, third edition (HAWIVA-III) or Snijders-Oomen Nonverbal Intelligence Test (SON-R)). In 135 children, numerical abilities were assessed with the dyscalculia test TEDI-MATH. The association between pre- and postnatal factors and impaired numerical skills was analyzed by means of logistic regression analysis.

Results

Dyscalculia test showed delayed numerical skills (TEDI-MATH Sum T-score <40) in 27 of 135 children tested (20.0%). In half of the children tested, delayed numerical abilities were related to lower IQ scores. Smoking in pregnancy, intracerebral haemorrhage and chronic lung disease were predictive of delayed numerical skills at 5 years of age in the multivariate analysis.

Conclusion

This study identified risk predictors for impaired numerical skills in preterm infants. Our data support the role of both pre- and perinatal factors in the evolution of mathematical deficits.

Abbreviations
CLD

Chronic lung disease

HAWIVA-III

Hannover-Wechsler Intelligence Test for preschool children, third edition

ICH

Intracerebral haemorrhage

IPE

Intraparenchymal echodense lesions

IQ

Intelligence quotient

NEC

Necrotizing enterocolitis

PPROM

Preterm premature rupture of membranes

PVL

Periventricular leucomalacia

RDS

Respiratory distress syndrome

ROP

Retinopathy of prematurity

SGA

Small for gestational age

SON-R

Snijders-Oomen Nonverbal Intelligence Test

Key notes

  • Information on preschool skills and risk predictors for the development numerical skills is lacking.
  • 20.0% of the children tested (dyscalculia test TEDI-MATH at 5 years of age) showed delayed numerical abilities. Smoking in pregnancy and risk factors indicating a vulnerable neonatal period such as intracerebral haemorrhage and chronic lung disease were predictive of delay in numerical skills.
  • Identification of children at risk and tailored educational programmes may help to prevent later school problems.

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

In European countries, 1.1–1.6% of live births are very preterm [1]. The number of neonates surviving very preterm birth has gradually increased due to advances in perinatal and neonatal care. However, increasing survival is linked to an increased awareness about morbidity regarding cognitive and behavioural outcomes of these children [2]. Even in children without obvious neurological deficits, subtle abnormalities occur, which include lower cognitive test scores in reading and mathematical abilities. Research has shown that school difficulties may be related to delay in preschool skills [3, 4]. However, information on preschool skills and risk predictors for the development of impaired abilities in these children is lacking. This is particularly true for deficits in numerical skills.

This study was designed to detect potential risk predictors for impaired numerical development in formerly preterm infants at the age of 5 years. To analyze this association, all children were individually assessed with the TEDI-MATH test, which is a multi-componential dyscalculia test especially designed for the assessment of preschoolers. To the best of our knowledge, this is the first study investigating this issue in a geographically defined cohort of preterm infants.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

Participants

The study survey area was Tyrol, a state in western Austria with 680000 inhabitants and about 7000 live births per year. All infants born before 32 completed weeks of pregnancy at Innsbruck Medical University, the only neonatal intensive care unit in the geographical region, were prospectively enrolled (survey period January 2003 till August 2006; 303 live births). Of these infants, 33 died and 37 were non-residents or moved out of region. The figure visualizes the inclusion procedure and follow-up. Disabilities were classified according to Wood et al. [5]. ‘Severe disability’ was defined as one that was likely to put the child in need of physical assistance to perform daily activities [5]. Children with severe disabilities who are not able to perform tests as used in the current study (n = 10) were excluded, because we were interested in the variables that contribute to numerical skills in those children for whom arithmetical problems could not be attributed to sensory or neurological handicap. Thus, 223 children were invited for a detailed follow-up visit at 5 years of age. A total of 62 parents (27.8%) did not accept invitation; thus, 161 children formed the current study population (Fig. 1).

image

Figure 1. Flow-chart of inclusion of very preterm infants. GA, gestational age; IQ, intelligence quotient; HAWIVA-III, Hannover-Wechsler Intelligence Test for preschool children, third edition; SON-R, Snijders-Oomen Nonverbal Intelligence Test.

Download figure to PowerPoint

Participants compared with non-participants were less likely to have a mother with an education of <12 years (p < 0.001), were more likely to speak German at home (p = 0.008), had a lower gestational age and a lower birth weight (p < 0.001, each), and more often required surfactant treatment after birth (p = 0.002). There was no difference regarding smoking in pregnancy, sex, SGA, early- and late-onset sepsis, CLD, severe ROP, ICH and NEC.

Maternal and neonatal data

Maternal and neonatal data included maternal age, maternal years of education, smoking in pregnancy, antenatal steroid use, timing of rupture of membranes, gestational age (full weeks of gestation), birth weight (grams), multiple birth, sex, postnatal surfactant use, small for gestational age (SGA), and diagnosis of early- and late-onset sepsis, chronic lung disease (CLD), necrotizing enterocolitis (NEC), intracerebral haemorrhage (ICH) and retinopathy of prematurity (ROP). Gestational age was calculated from the first day of the last menstrual period. This was compared with assessment of gestational age by ultrasound scans performed before 24 weeks. If there was a difference of more than 1 week between menstrual and ultrasound assessment, the scan assessment was preferred. Preterm premature rupture of membranes (PPROM) was defined when the rupture of membranes occurs before 37 weeks of gestation and more than 24 h before the onset of labour. CLD was defined as oxygen dependence at 36 weeks postconceptional age. NEC was defined according to Bell's criteria [6] and was classified as medical (clinical symptoms and signs plus evidence of pneumatosis on abdominal X-ray) or surgical (histological evidence of NEC on surgical specimens of intestine). ICH was classified according to the method of Papile et al. [7]. Infants with cystic periventricular leukomalacia (PVL) included those with ultrasonographic findings suggesting cystic degeneration of periventricular white matter. Intraparenchymal echodense lesions (IPE) detected by ultrasound were classified as yes or no. Ultrasound examinations were routinely performed on the second and fifth day of life, thereafter every week and finally every second or third week. Growth charts developed by Alexander et al. [8] were used to classify infants as SGA at birth, defined as a birth weight lower than the 10th percentile for sex and gestational age. A diagnosis of early-onset (≤72 h of birth) or late-onset (>72 h) sepsis required signs of generalized infection, a positive blood culture and antibiotic therapy for 5 or more days. Information on smoking habits and years of education was obtained during the stay at the neonatal department. Smoking habits in pregnancy (yes/no) were based on self-reported data. Those mothers who refused to provide information on smoking status were considered smokers.

Follow-up/neurodevelopmental outcome

All children with a gestational age of <32 weeks are regularly invited for follow-up visits till they enter school. The follow-up visit at 5 years of age includes an interview with the children's mothers or other primary caregivers, a physical and neurological examination, assessment of cognitive development using the Hannover-Wechsler Intelligence Test for preschool children, third edition (HAWIVA-III, German version of the WPPSI-III, Wechsler Preschool and Primary Scales of Intelligence) [9]. In children with language difficulties, the Snijders-Oomen Nonverbal Intelligence Test (SON-R) was used [10]. On the HAWIVA-III, six of the core-subtests were administered (Block Design, Information, Matrix Reasoning, Vocabulary, Symbol Search, Coding), and then verbal performance, processing speed and full-scale intelligence quotient (IQ) were calculated from the subscales. Each scale or index is age-standardized with a mean of 100 (SD, 15) and a scale of <85 (>1 SD below the mean) indicating delay and a scale of <70 (>2 SDs below the mean) indicating significant delay.

Cognitive delay was defined as a full-scale IQ <85 in either HAWIVA-III or SON-R test. Children in whom a score could not be calculated because of severe mental delay were also classified as having delayed cognitive abilities.

In addition, the children were assessed individually with the TEDI-MATH which is a multi-componential dyscalculia test based on cognitive neuropsychological models of number processing and calculation [11]. The TEDI-MATH consists of several subtests designed for the assessment of preschoolers: In the counting principles subtest, children's mastery of the verbal counting sequence and its flexibility is tested (e.g. counting in steps of two, and counting backwards). The object counting subtest evaluates how well children understand and use basic counting principles. The Arabic digits and number words subtests assess children's knowledge of digits and number words by testing whether they can discriminate them from non-numerical symbols and words). Finally, in the calculation subtest, simple addition and subtraction problems are given in a pictorially supported format. If the child was able to perform all these subtests, the Sum T-score was calculated. The score is age-standardized with a mean of 50 (SD, 10), and a score of <40 (>1 SD below the mean) indicates delay [11].

Consequently, delay in numerical skills was defined as a Sum T-score <40. Children in whom a score could not be calculated because of severe delay in numerical abilities were also classified as having delayed numerical skills. In addition, a separate definition of delay in numerical skills was based on a C-score lower than 3 (mean C-score = 5, SD, 2) in at least one of the subtests counting principles, object counting or calculation, because these three key TEDI-MATH subtests were considered most relevant for numerical abilities [12].

All cognitive tests were performed by one of the three experienced psychologists.

Statistical analysis

Data analysis was performed using SPSS software version 20.0 for Windows (SPSS Inc., Chicago, IL, USA). Comparison of categorical data was made using the chi-squared or Fischer's exact test. Multivariate risk profiles for impaired calculation abilities in the fifth year of life were computed by means of logistic regression analysis using a stepwise forward selection procedure with inclusion and exclusion criteria as follows (PI < 0.05 and PE > 0.10). This analysis allows for the following variables: maternal age, low educational level of mother, smoking in pregnancy, PPROM, multiple birth, antenatal steroids, gestational age, birth weight, SGA, male sex, surfactant treatment, CLD, ICH all grades and ICH grades 3 and 4, PVL, IPE, NEC, ROP grades 3 and 4, and early- and late-onset sepsis.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

Full assessment of cognitive abilities at 5 years of age was available in 153 of the 161 children. In 140 children, HAWIVA-III was used, and in 13 children, SON-R was used (Fig. 1). A total of 28 of these 153 children (18.3%) showed delayed cognitive performance.

In 135 of 161 children, numerical abilities were assessed (Fig. 1) and 126 children completed all TEDI-Math subtests. In 9 children, a score could not be calculated because of severe delay in numerical abilities. A total of 27 of the 135 (20.0%) showed delayed numerical skills (TEDI-MATH Sum T-score < 40). In 13 of these children, numerical delay was related to lower IQ scores. Finally, 39 children (28.5%) showed impaired performance in one of the three key TEDI-MATH subtests (see 'Patients and Methods').

Table 1 summarizes maternal, peri- and neonatal data for the population of children born <32 gestational weeks with and without impaired arithmetical abilities (TEDI-Math Sum T-score <40). Smoking in pregnancy (p = 0.014), multiple birth (p = 0.026), CLD (p = 0.005), PVL (p = 0.039), ICH all grades (p < 0.001) and ICH grades 3 and 4 (p = 0.025) were related to an increased risk of impaired numerical skills at an age of 5 years in the univariate analysis. The number of smokers was too low to calculate robust risks in categories of smoking severity. Only eight women of 135 (5.9%) reported to smoke more than 10 cigarettes/day during pregnancy.

Table 1. Sociodemographic and neonatal characteristics of preterm infants with a gestational age of <32 weeks according to delayed numerical skills at an age of 5 years
Variable Not delayed (n = 108) n (%) or mean ± SD Delayed (n = 27) n (%) or mean ± SD p-value
  1. SGA, small for gestational age; CLD, chronic lung disease; ICH: intracerebral haemorrhage; IPE, intraparenchymal echodense lesions; NEC, necrotizing enterocolitis; PPROM, preterm premature rupture of membranes; PVL, periventricular leucomalacia; p-values are from Fischer's exact test or t-test, as appropriate. In all variables, the proportion of missing data was ≤5%.

Maternal age <23 years (vs. ≥23 years)6 (5.6)3 (11.1)0.38
Low educational level of mother (<12 years vs. ≥12 years)43 (39.8)17 (63.0)0.087
Smoking in pregnancy (yes vs. no)22 (20.4)12 (44.4)0.014
PPROM24 (22.2)8 (29.6)0.78
Multiple birth (yes vs. no)46 (42.6)5 (18.5)0.026
Antenatal steroids (yes vs. no)94 (87.0)21 (77.8)0.37
Gestational age (weeks)29.7 ± 1.828.9 ± 2.20.073
Birth weight (g)1322 ± 3641165 ± 3220.80
SGA (yes vs. no)8 (7.4)3 (11.1)0.46
Male sex (yes vs. no)59 (54.6)18 (66.7)0.29
Surfactant treatment (yes vs. no)61 (56.5)17 (63.0)0.66
CLD (yes vs. no)6 (5.6)7 (25.9)0.005
ICH (all grades) (yes vs. no)13 (12.0)11 (40.7)<0.001
ICH (grades 3 and 4)1 (0.9)3 (11.1)0.025
PVL (yes vs. no)0 (0)2 (7.4)0.039
IPE (yes vs. no)2 (1.9)0 (0)1.000
NEC (yes vs. no)6 (5.6)4 (14.8)0.11
ROP (grades 3 and 4) (yes vs. no)3 (2.8)3 (11.1)0.094
Early-onset sepsis (yes vs. no)1 (0.9)1 (3.7)0.36
Late-onset sepsis (yes vs. no)15 (13.9)2 (7.4)0.52

Multivariate analysis fitted with a stepwise selection procedure identified smoking in pregnancy, CLD and ICH all grades as significant risk predictors for delayed numerical skills (Table 2). When excluding non-responders regarding smoking in pregnancy (n = 8), the association between smoking and delayed numerical skills was further strengthened: OR 4.59 [1.57–13.44], p = 0.005).

Table 2. Multivariable association between risk variables and delayed numerical skills at 5 years of age
VariableOR (95% CI)p-value
  1. CLD, chronic lung disease; CI, confidence interval; ICH, intracerebral haemorrhage; OR, odds ratio derived from logistic regression analysis of risk variables for delayed numerical skills. The multivariate model was fitted with a forward stepwise selection procedure.

Smoking in pregnancy (yes vs. no)4.26 (1.56–11.65)0.004
ICH all grades (yes vs. no)4.66 (1.56–13.93)0.007
CLD (yes vs. no)4.35 (1.11–17.01)0.038

When using an alternative definition of delay in numerical abilities (C-score < 3 in at least one of the three key TEDI-MATH subtests), results were very similar. Again, smoking in pregnancy, CLD and ICH all grades were independent risk predictors (multivariable odds ratios 2.75 [95% confidence intervals 1.13–6.71], p = 0.026, 4.37 [1.22–15.56], p = 0.023, and 3.96 [1.48–10.58], p = 0.006).

When separately focusing on children with a gestational age of <30 weeks (n = 63) and 30 weeks and above (n = 74), CLD and ICH all grades were significantly associated with a delay in numerical abilities in the younger age group (5.05 [1.16–21.98], p = 0.031, and 4.59 [1.20–17.69], p = 0.026), whereas smoking in pregnancy remained significant in the older ones (5.26 [1.33–20.82], p = 0.018). When delayed IQ was included in the multivariate model, smoking remained significant (4.06 [1.30–12.68], p = 0.016). In line, after exclusion of subjects with delayed IQ, 14 children had isolated delay of numerical skills and smoking remained a significant risk predictor (6.65 [1.88–23.52], p = 0.003). In patients with delay in numerical abilities, impaired IQ was the most prominent risk predictor (7.32 [1.91–28.09], p = 0.004).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

Very preterm birth places children at high risk for cognitive and learning difficulties in later life. In our cohort, 18.3% of the children enrolled showed delayed cognitive performance (full-scale IQ score < 85), 20.0% had delayed numerical skills (TEDI-MATH Sum T-score < 40) and 28.5% showed impaired performance in one of the three key TEDI-MATH subtests at the age of 5 years. In the EPICure cohort, up to 44% had a serious impairment in the core subjects of reading and mathematics, and 50% had a performance below the average range for the respective age. Extremely preterm children had a 13-fold risk for special educational needs involving additional learning support [13]. High prevalence of cognitive deficits in extremely preterm survivors was also found in other cohorts [14, 15]. In line with a Finnish cohort [16], our prevalence data of cognitive deficits are much lower due to improvements in neonatal care during the last 10 years and because not only extremely preterm children but also those with a higher gestational age (up to 32 weeks) were included. Moreover, most of our children were regularly seen for follow-up visits and were offered special care, if necessary. In some studies, deficits in reading scores disappeared after adjustment for general cognitive ability [17, 18]. However, global cognitive deficits cannot account for all learning difficulties [17, 19]. In our cohort, 13 of 27 children with impaired mathematical skills had an IQ score < 85 in HAWIVA-III or SON-R, the other 14 children showed delayed numerical skills and a normal IQ score. Deficits in numerical skills may result from impairment of specific brain areas [20, 21]. In particular, brain images of children with calculation difficulties point to functional impairment and structural and microstructural alterations in the parietal brain regions [22, 23]. The perinatal clinical course of preterm infants may explain these structural abnormalities. Animal models have shown that factors related to prematurity can promote neuronal cell death in the immature brain [24]. In our study, postnatal complications such as ICH and CLD – reflecting extreme prematurity – were risk predictors for delayed numerical skills in the multivariate analysis. When focusing on children with a gestational age of <30 weeks, ICH and CLD remained significant risk predictors for delayed numerical skills. In addition, very preterm infants are exposed to an extrauterine environment and subjected to multiple painful procedures. Increased rates of neuronal cell death lead to volumetric losses in specific brain regions and may partially explain cognitive abnormalities such as delayed numerical skills in these children [20].

Nicotine in pregnancy was reported to induce abnormalities in cell proliferation and differentiation leading to abnormalities in cell number and macromolecular content [25]. This study is among the first to report an association between maternal smoking and delayed numerical preschool skills. An association between smoking and poor cognitive outcome was evident in our study population already at a corrected age of 24 months [26] and also in a number of prospective studies [27, 28]. In the current study, the effect of smoking on impaired numerical abilities remained significant when subjects with delayed IQ were excluded from multivariate analysis. Thus, smoking in pregnancy may not only have an impact on cognitive function but may also alter specific brain regions involved in the development of numerical abilities. However, in former studies on term infants it remained unclear, whether associations between in utero exposure to tobacco smoke and cognitive function are causal or confounded by socioeconomic status [29, 30].

As strength of the current study, information on maternal sociodemographic status, as well as birth characteristics, that could explain the relation between smoking and impaired numerical skills, is available. However, residual confounding by unmeasured characteristics that differ between smokers and non-smokers cannot be fully excluded.

A short-coming of our single-centre study is the limited sample (n = 161) eligible for follow-up. However, all preterm infants of a geographically well-determined region were included, and thus, results may be regarded representative for the general community. Information on smoking was based on the mother's self-report probably leading to an underestimation of smoking. An additional weakness was the lack of information on breastfeeding and exposure of the child to passive smoke. Although there was no significant association between socioeconomic variables such as maternal age or maternal education and the risk of delayed numerical abilities, undetermined social variables might have an impact on numerical outcome partly due to the fact that families with lower socioeconomic status might have more difficulties to access the special care offered.

In conclusion, the current study unravels smoking in pregnancy and postnatal factors such as CLD and ICH as risk predictors for delayed preschool numerical skills. Identification of children at risk and tailored educational programmes may help to prevent later school problems. Moreover, our results and these about other adverse effects of smoking in pregnancy provide a strong rationale for smoking cessation in pregnancy.

Conflicts of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References

The authors have no potential conflict of interest relevant to this article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Conflicts of Interests
  8. References
  • 1
    Buitendijk S, Zeitlin J, Cuttini M, Langhoff-Roos J, Bottu J. Indicators of fetal and infant health outcomes. Eur J Obstet Gynecol Reprod Biol 2003; 111(Suppl 1): S6677.
  • 2
    Bhutta AT, Cleves MA, Casey PH, Cradock MM, Anand KJ. Cognitive and behavioral outcomes of school-aged children who were born preterm: a meta-analysis. JAMA 2002; 288: 72837.
  • 3
    Krajewski K, Schneider W. Exploring the impact of phonological awarness, visual-spatial working memory, and preschool quantity-number competencies on mathematics achievement in elementary school: findings from a 3-year longitudinal study. J Exp Child Psychol 2009; 103: 51631.
  • 4
    Jordan NC, Kaplan D, Ramineni C, Locuniak MN. Early math matters: kindergarten number competence and later mathematics outcomes. Dev Psychol 2009; 45: 85067.
  • 5
    Wood NS, Costeloe K, Gibson AT, Hennessy EM, Marlow N, Wilkinson AR; EPICure study Group. The EPICure study: associations and antecedents of neurological and developmental disability at 30 months of age following extremely preterm birth. Arch Dis Child Fetal Neonatal Ed 2005; 90: F13440.
  • 6
    Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Ann Surg 1978; 187: 17.
  • 7
    Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1500 gm. J Pediatr 1978; 92: 52934.
  • 8
    Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996; 87: 1638.
  • 9
    Ricken G, Fritz A, Schuck KD, Preuß U. (Hrsg.). (2007). HAWIVA®-III, Hannover-Wechsler-Intelligenztest für das Vorschulalter-III. Manual zur Durchführung und Auswertung. Verlag Hans Huber 2007, Göttingen, Deutschland.
  • 10
    Snijders JT, Snijders-Oomen N. Snijders-Oomen Non-verbale Intelligence Scale, S.O.N. 2 ½-7. HD Tjeenk Willink, 1976, Groningen, Holland.
  • 11
    Kaufmann L, Nuerk H-C, Graf M, Krinzinger H, Delazer M, Willmes K. TEDI-MATH-Test zur Erfassung numerisch-rechnerischer Fertigkeiten vom Kindergarten bis zur 3. Klasse. Bern, Verlag Hans Huber 2009, Bern Schweiz.
  • 12
    Geary DC, Hamson CO, Hoard MK. Numerical and arithmetical cognition: a longitudinal study of process and concept deficits in children with learning disability. J Exp Child Psychol 2000; 77: 23663.
  • 13
    Johnson S, Hennessy E, Smith R, Trikic R, Wolke D, Marlow N. Academic attainment and special educational needs in extremely preterm children at 11 years of age: the EPICure study. Arch Dis Child Fetal Neonatal Ed 2009; 94: F2839.
  • 14
    Marlow N, Wolke D, Bracewell MA, Samara M; EPICure Study Group. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med 2005; 352: 919.
  • 15
    Hack M, Taylor HG, Drotar D, Schluchter M, Cartar L, Andreias L, et al. Chronic conditions, functional limitations, and special health care needs of school-aged children born with extremely low-birth-weight in the 1990s. JAMA 2005; 294: 31825.
  • 16
    Lind A, Korkman M, Lehtonen L, Lapinleimu H, Parkkola R, Matomäki J, et al. Cognitive and neuropsychological outcomes at 5 years of age in preterm children born in the 2000s. Dev Med Child Neurol 2011; 53: 25662.
  • 17
    Anderson P, Doyle LW. Neurovbehavioral outcomes of school-age children born extremely low birth weight or very preterm in the 1990s. JAMA 2003; 289: 326472.
  • 18
    Wolke D, Samara M, Bracewell M, Marlow N; EPICure Study Group. Specific language difficulties and school achievement in children born at 25 weeks of gestation or less. J Pediatr 2008; 152: 25662.
  • 19
    Aarnoudse-Moens CS, Oosterlaan J, Duivenvoorden HJ, van Goudoever JB, Weisglas-Kuperus N. Development of preschool and academic skills in children born very preterm. J Pediatr 2011; 158: 516.
  • 20
    Peterson BS, Vohr B, Staib LH, Cannistraci CJ, Dolberg A, Schneider KC, et al. Regional brain volume abnormalities and longterm cognitive outcome in preterm infants. JAMA 2000; 284: 193947.
  • 21
    Kaufmann L, Wood G, Rubinsten O, Henrik A. Meta-analysis of developmental fMRI studies investigatin typical and atypical trajectories of number processing and calculation. Dev Neuropsychol 2011; 36: 76387.
  • 22
    Kaufmann L, Vogel SE, Starke M, Wood G. Developmental dyscalculia: compensatory mechanisms in left intraparietal regions in response to nonsymbolic magnitudes. Behav Brain Funct 2009; 5: 35.
  • 23
    Kucian K, Loenneker T, Dietrich T, Dosch M, Martin E, von Aster M. Impaired neuronal networks for approximate calculation in dyscalculic children: a functional MRI study. Behav Brain Funct 2006; 2: 31.
  • 24
    Bhutta AT, Anand KJS. Abnormal cognition and behavior in preterm neonates linked to smaller brain volumes. Trends Neurosci 2001; 24: 12932.
  • 25
    Ernst M, Moolchan ET, Robinson ML. Behavioral and neural consequences of prenatal exposure to nicotine. J Am Acad Child Adolesc Psychiatry 2001; 40: 63041.
  • 26
    Kiechl-Kohlendorfer U, Ralser E, Pupp Peglow U, Reiter G, Griesmaier E, Trawöger R. Smoking in pregnancy: a risk factor for adverse neurodevelopmental outcome in preterm infants? Acta Paediatr 2010; 99: 10169.
  • 27
    Frydman M. The smoking addiction of pregnant women and the consequences on their offspring's intellectual development. J Environ Pathol Toxicol Oncol 1996; 15: 16972.
  • 28
    Mortensen EL, Michaelsen KF, Sanders SA, Reinisch JM. A dose-response relationship between maternal smoking during late pregnancy and adult intelligence in male offspring. Paediatr Perinat Epidemiol 2005; 19: 411.
  • 29
    Lambe M, Hultman C, Torrang A, MacCabe J, Cnatttingius S. Maternal smoking during pregnancy and school performance at age 15. Epidemiology 2006; 17: 52430.
  • 30
    Larroque B. Prenatal exposure to tobacco smoke and cognitive development: epidemiology studies. J Gynecol Obstet Biol Reprod 2003; 32(Suppl 1): S239.