- Email: [email protected]
Entorhinal cortical thinning affects perceptual and cognitive functions in adolescents born preterm with very low birth weight (VLBW)
Early Human Development 88 (2012) 103–109
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Entorhinal cortical thinning affects perceptual and cognitive functions in adolescents born preterm with very low birth weight (VLBW) Jon Skranes a, e,⁎, Gro C.C. Løhaugen a, e, Kari Anne I. Evensen a, d, Marit S. Indredavik b, d, Olav Haraldseth c, Anders M. Dale c, f, Ann-Mari Brubakk a, d, Marit Martinussen a, d a
Dept. of Lab. Medicine, Children's and Women's Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Dept. of Neuroscience, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Dept. of Circulation and Medical Imaging, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway d St Olav University Hospital, Trondheim, Norway e Sørlandet Hospital, Arendal, Norway f MIL, Dept. of Neurosciences and Radiology, University of San Diego, La Jolla, CA, USA b c
a r t i c l e
i n f o
Article history: Received 9 February 2011 Received in revised form 22 June 2011 Accepted 3 July 2011 Keywords: VLBW Preterm birth Brain Cortical thickness Entorhinal cortex VMI WISC-III IQ Executive function
a b s t r a c t Background: The entorhinal cortex serves as an important gateway between the cerebral cortex and the hippocampus by receiving afferent information from limbic, modality sensory-specific, and multimodal association fibers from all the brain lobes. Aim: To investigate whether thinning of entorhinal cortex is associated with reduced perceptual, cognitive and executive skills in very low birth weight (VLBW) adolescents. Study design: Prospective, geographically based follow-up study of three year cohorts of preterm born VLBW children. Subjects: Forty-nine VLBW (birth weight ≤ 1500 g) and 58 term-born control adolescents were examined at the age of 14–15 years. Outcome measures: Perceptual and cognitive functions were assessed with Visual motor integration test, Grooved Pegboard test, Wechsler Intelligence Scale for Children-III and different executive function tests (Wisconsin card sorting test, Trail Making test, Knox cube test). An automated MRI technique at 1.5 T for morphometric analyses of cortical thickness was performed. Areas with cortical thinning in left and right entorhinal cortex in the VLBW group were chosen as regions of interest to look for associations between cortical thickness and clinical findings. Results: Thinning of the entorhinal cortex was correlated with low performance on perceptual and cognitive scores in the VLBW adolescents, but not in controls. In addition, thinning of the entorhinal cortices correlated with reduced performance on several executive tests, including perceptual speed and aspects of working memory. Conclusions: Entorhinal cortical thinning is related with low IQ and reduced perceptual and executive functions in VLBW adolescents. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The medial temporal lobe includes a system of anatomically related structures where the hippocampus lies at the end of a cortical processing hierarchy; the entorhinal cortex (ERC) is the major source of its cortical projections. This system is principally concerned with long-term memory, but also critically involved in complex functions such as sensory representation and spatial orientation [1]. Recently it was stated that the ERC may act as the nodal point between ⁎ Corresponding author at: Department of Laboratory Medicine, Children's and Women's Health, Medical Faculty, Norwegian University of Science and Technology, Medisinsk teknisk forskningssenter, N-7489 Trondheim, Norway. Tel.: + 47 99390285. E-mail address: [email protected] (J. Skranes). 0378-3782/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2011.07.017
hippocampus and a variety of multimodal association areas of the cortex such as parietal, temporal, and prefrontal cortex [2]. Computational automated morphometric MRI methods with high reliability have detected correlates between cortical thinning and cognitive task performance in healthy individuals and in different patient groups (schizophrenia, Alzheimer's disease (AD), epilepsy) [3–5], and entorhinal shrinkage has been associated with cognitive decline in healthy elderly persons and in AD patients [6]. Children born very preterm or with very low birth weight (VLBW: birth weight ≤ 1500 g) are prone to perinatal brain injury and have an increased risk of developing motor and cognitive impairments during childhood and adolescence [7]. These impairments include reduced skills in perceptual and higher cognitive tasks that measure executive functions, attention and memory functions [8,9]. Quantitative MR
104
J. Skranes et al. / Early Human Development 88 (2012) 103–109
studies have revealed both white and gray matter abnormalities in VLBW children, and the MR pathology seems to persist into adolescence and early adulthood, as do clinical problems [10–12]. We have previously published areas of cortical thinning and thickening in a cohort of VLBW children at 14–15 years of age and found that regional thinning in areas corresponding to the ERC was more pronounced in those with low IQ compared with those with normal IQ in the VLBW group, and in the most immature of the preterm born children [13]. The ERC area therefore seemed interesting to investigate further with quantitative measurements of mean thickness versus different aspects of cognitive functions. The aim of the present study was to investigate whether ERC thinning is associated with reduced perceptual, cognitive and executive/attentional skills in VLBW adolescents. We hypothesized that aberrant development of this important cortical area would interfere with higher order cognitive and perceptual functions in these adolescents. 2. Methods This study is part of a hospital-based follow-up study from birth where three year cohorts of prematurely born VLBW children from a defined geographic region in Norway were compared with randomly selected term born controls from the same geographic area examined at regular intervals with a battery of clinical tests and different MRI techniques. 2.1. Participants The study population consisted of 49 VLBW and 58 control adolescents examined at 15 years of age. The Regional Committee for Medical Research Ethics approved the study protocol and written informed consent was obtained. (See Appendix for more information about the study population). 2.2. Neuropsychological assessment battery An estimate of the adolescents' intelligence quotient (IQest) was calculated using four subtests of Wechsler Intelligence Scale for Children (WISC)-third edition: Arithmetic, Vocabulary, Block design and Picture arrangement [15]. We defined low IQ as an estimated IQ score b 2 standard deviations (SD) of the mean in the control group. Visual–motor function was assessed by the Developmental Test of Visual–Motor Integration (VMI–IV), including supplementary tests to evaluate visual perception and motor coordination [14]. Poor performance/impairment was defined as score b − 2 SD from the mean scores of the control group. Attention and executive function were assessed with the, Wisconsin Card Sorting Test (WCST), version III [17], Knox-Cube test [18], and Trail Making Test [19], while the Grooved Pegboard (GP) test [16] was used to assess both motor coordination and processing speed. (See Appendix for more information about each of the neuropsychological tests). 2.3. MR imaging The scanning was performed on a 1.5 T Siemens Symphony Sonata, Siemens AG, Erlangen, Germany. The imaging for the morphometric analyses was a 3D inversion recovery MPRAGE sequence with 128 sagittal partitions, slice thickness of 1.33 mm, square FOV of 256 mm, and acquisition duration of 8.5 min. An automated cortical surface reconstruction method developed by Dale and Fischl was performed to construct cortical thickness for each subject's entire brain [20,21]. The method has been described before, and statistical significant cortical thickness difference maps between the two groups of adolescents were generated [13]. Several significantly thinner areas were found in the VLBW adolescents (Fig. 1). Thinner areas in the right and left medial
Fig. 1. Statistical difference maps of cortical thickness between 49 VLBW and 58 control adolescents are overlaid on the surface reconstruction with uninflated (upper panel) and inflated (middle and lower panel) brains. Dark gray areas correspond to sulci and light gray areas are gyri. Areas with statistically significant thickness difference between groups are shown in color, and the color scale shows the dynamic range of the changes, red to yellow represents an increasing thinning of the cortex in the VLBW group, and full yellow corresponds to a statistical difference in cortical thickness with a p-value ≤ 0.0001. Blue represents a statistically significant thickening of the cortex in the VLBW group; light blue corresponds to a statistical difference in cortical thickness with a p-value ≤ 0.0001. Areas of regional thinning of the entorhinal cortex (regions of interest) are marked in white in the lower panel.
temporal lobe corresponded to an area that covered most of the ERC. These areas, labeled white in Fig. 1C and D, were used as regions of interest (ROIs) in the present study. The regions of interest were mapped back to each participant by using a high-dimensional spherical morphing procedure to find the homologous regions across subjects. A mean thickness for each of the two entorhinal cortical areas was then recorded for every VLBW participant. 2.4. Statistics The associations between the individual mean thickness calculations in the ROI and the neuropsychological test scores in the VLBW group were explored by linear regression analysis. For continuous variables Pearson's r was used for correlation analyses between ERC thickness and IQ measurements, while Spearman's rho were performed for the neuropsychological test results since these are not normally distributed. For the correlation analyses the two-tailed significance threshold was set to p b 0.02 to adjust for multiple comparisons. In addition, we explored the relationship between thickness and neuropsychological test scores when the scores were dichotomized into those indicating normal and abnormal/low function (more than 2 SD from mean value of controls) using the Mann–Whitney U test. Adjustments for socioeconomic status (SES) and IQ were performed for the neuropsychological tests, since low test results may be influenced by both these factors. Here p-values≤ 0.05 were considered significant. 3. Results 3.1. Clinical and MRI findings The clinical characteristics, test results and MRI findings for the study groups are shown in Table 1. Age at MRI examination was about 12 months higher in both groups of children when comparing with
**p ≤ 0.01, *p ≤ 0.05, #p = 0.09 (normal versus low test results). Mann–Whitney U-test, General Linear Model for adjusted means. Abbreviations: SD: standard deviation; VLBW: very low birth weight; VMI: Visual Motor Integration test; WISC: Wechsler Intelligence Scale for Children; WCST persev. errors: Wisconsin Card Sorting Test–perseverative error score; SES: socioeconomic status; SE: standard error. a Scores higher/equal to or lower than −2 SD from the mean score in the control group.
2.43 (0.04)* 2.40 (0.05) 2.24 (0.08) 2.25 (0.10) 2.44 (0.04)** 2.44 (0.04)** 2.13 (0.11) 2.01 (0.12) 2.41 (0.04) 2.38 (0.05) 2.26 (0.13) 2.25 (0.14) 2.43 (0.05) 2.39 (0.05) 2.28 (0.08) 2.30 (0.05) 2.44 (0.04)** 2.42 (0.04)** 2.16 (0.09) 2.12 (0.09) 2.43 (0.04)* 2.41 (0.05)* 2.14 (0.11) 2.13 (0.12) 2.44 (0.04)* 2.42 (0.05)* for SES only): 2.42 2.23 (0.04) (0.09) 2.39 2.17 (0.05) (0.10) Mean values after adjustment for SES and IQ (WISC-III results adjusted ERC left: 2.27 2.44 2.33 2.42 2.29 (SE 0.08) (0.05) (0.08) (0.05) (0.10) ERC right: 2.16 2.45 2.21 2.42 2.29 (0.08) (0.05)** (0.09) (0.05)* (0.11)
ERC right:
(39)
2.44 (0.26)** 2.41 (0.27) 2.19 (0.27) 2.22 (0.37) 2.45 (0.26)** 2.43 (0.25)** 2.09 (0.13) 2.03 (0.29) 2.43 (0.27)* 2.40 (0.28) 2.15 (0.22) 2.16 (0.33) 2.45 (0.26)** 2.41 (0.28) 2.23 (0.25) 2.25 (0.33) 2.44 (0.27)** 2.42 (0.27)** 2.17 (0.19) 2.12 (0.29) 2.44 (0.26)** 2.41 (0.27)* 2.08 (0.18) 2.10 (0.30) 2.45 (0.26)** 2.43 (0.27)* 2.18 (0.24) 2.16 (0.29) 2.43 (0.25)* 2.40 (0.29) 2.23 (0.33) 2.23 (0.32) 2.44 (0.27) 2.43 (0.28)**
b−2 SD
(10) (41)
≥−2 SD b−2 SD
(8) (43)
≥−2 SD b−2 SD
(6) (36)
≥−2 SD b−2 SD
(13) (40) (9) (42)
≥−2 SDa b−2 SD ≥−2 SD b−2 SD
(7) (38) (11) (40) (9) (36) (12)
2.27 (0.25) 2.18 (0.29) 2.46 (0.28)** 2.46 (0.25)** 2.22 (SD 0.19) 2.15 (0.29)
In this study we report that entorhinal cortical (ERC) thinning is associated with general cognitive ability, visual–motor and executive function deficits in VLBW adolescents. To our knowledge, this is the first study that reports a possible relationship between higher order cognitive deficits and ERC thinning in preterm born subjects.
ERC left:
4. Discussion
(35)
Table 2 shows the associations between cortical thickness and dichotomized test results (means unadjusted and adjusted for SES and IQ) in the VLBW group. Thinning of the entorhinal cortex (ERC) was associated with low scores on VMI test, motor coordination test (right ERC only), estimated IQ, Grooved pegboard (left ERC only) and WCST perseverative errors. Correlation analyses (Fig. 2) showed that there was a linear correlation between right ERC thickness and motor coordination score and Trail Making Test score, and between left ERC thickness and performance and full scale IQ, Knox Cube Test and WCST perseverative error score. A full table of all the correlations investigated is presented as Table A in the Appendix.
(14)
3.2. Relationship between neuropsychological function and entorhinal cortical thickness
≥−2 SD
age at clinical examination. The reason for this was that the initial MRI taken at the same time as the clinical assessment had to be repeated because of technical reasons due to incorrect DTI sequencing.
b−2 SD
*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (VLBW versus controls). t-test for parametric data, Mann–Whitney U-test for non-parametric data. Abbreviations: ERC: Entorhinal cortex; VLBW: very low birth weight; SD: standard deviation; SES: socio-economic status; MRI: magnetic resonance imaging; VMI: Visual Motor Integration test; WCST: Wisconsin card sorting test; persev: perseverative; TMT: Trail making test.
≥−2 SD
2.63 (0.28) 2.56 (0.34)
b−2 SD
WISC-III
2.39 (0.27)*** 2.36 (0.29)**
(6.5) (6.2) (10.8) (2.0) (5.0) (13.6)
≥−2 SD
13.6 11.5 14.7 13.8 14.1 29.7
(37.6) (12.5)*** (16.9)** (2.7)*** (8.5)*** (25.0)***
b−2 SD
27.6 18.8 23.2 12.0 19.4 50.3
≥−2 SD
65.1 (8) 73.5 (11)
105
b−2 SDa
79.8 (48)* 89.7 (43)**
WCST persev. errors
22.7 (2.7) 24.2 (2.2) 24.6 (2.5) 96 (17) 93 (16) 101 (19)
WCST trials to complete
19.5 (3.7)*** 21.4 (3.1)*** 22.5 (3.7)*** 79 (23)*** 82 (19)*** 81 (28)***
Knox cube
15.3 (0.5) 14.2 (0.3) 167.8 (8.3) 58.0 (11.4)
Grooved pegboard non-dom
15.1 (0.6) 14.2 (0.3) 162.2 (9.2)** 50.7 (13.2)***
Grooved pegboard dom
3707 (486) 39.6 (1.1) 23/35 3.6 (1.1)
Visual perception
1195 (239)*** 29.1 (2.7)*** 25/24 3.2 (1.3)
Motor coordination test
n = 58
Anatomical VMI region
Controls
n = 49
Table 2 Entorhinal cortical (ERC) thickness, mean (SD/SE) in VLBW adolescents with low and normal neuropsychological test results.
Perinatal variables Birth weight (grams) Gestational age (weeks) Boys/girls SES At age 14–15 Age at MRI examination (years) Age at clinical testing (years) Length (cm) Weight (kg) Clinical test results VMI Motor coordination Visual perception Full scale IQ –Verbal IQ –Performance IQ Grooved pegboard (seconds) – Writing hand – Non-dominant hand WCST – Trial 1st category – Persev. errors – Non-persev. errors Knox cube (seconds) TMT A (seconds) TMT B (seconds) MRI findings ERC thickness left (mm) ERC thickness right (mm)
VLBW
Trail making test A
Table 1 Child characteristics, clinical test results and ERC thickness (mean, SD) in the study groups.
≥−2 SD
J. Skranes et al. / Early Human Development 88 (2012) 103–109
106 J. Skranes et al. / Early Human Development 88 (2012) 103–109 Fig. 2. Correlation plots (Spearman's rho) of right entorhinal cortical thickness and motor coordination score (upper left) and Trail making test score (upper middle), respectively. Correlation plots of left entorhinal cortical thickness and performance IQ (upper right), full scale IQ (lower left) (Pearson's r), Knox cube test (lower middle) and WCST perseverative errors (lower right) (Spearman's rho), respectively in 49 VLBW adolescents. Abbreviations: WCST: Wisconsin card sorting test; VLBW: very low birth weight.
J. Skranes et al. / Early Human Development 88 (2012) 103–109
107
4.1. Methodological considerations
4.3. Clinical implications
The automated MR segmentation tool in this study has been widely used for measurement of cortical thickness (http://freesurfersoftware.org). However, using an automated cortical reconstruction method for measurement of cortical thickness always raises the possibility of inaccuracies in the normalization process and in defining the gray-white matter border. The segmentation method has been validated by performing analysis on post-mortem brains [22], and by manual measurements [4]. All surface models in our study were visually inspected for accuracy. However, the entorhinal cortex is a difficult area to map out in the Freesurfer program, so some manual correction of the segmentation was performed, but this was done blinded to group adherence and by one person. The region of interest (ROI) in our study constitutes a rather large portion of the medial temporal lobe, covering parts of the entorhinal cortex, parts of the temporal pole as well as the parahippocampal gyrus. The ROI therefore most likely includes cortical areas with different functionality. This could obviously interfere with the possibility of detecting structural–functional relationship. The results could also be due to chance, since the number of participants was rather low. However, then we would have expected both positive and negative correlations between structure and function, and not only what we found: the thinner the cortex, the lower the test results. Nevertheless, the structure–function association, identified for the entorhinal regions in the present study should be interpreted cautiously and subjected to replication in larger samples before generalization of the results.
Our findings of an association between entorhinal thinning and a rather broad spectrum of cognitive, visual–motor, and executive deficits support the view of the ERC as a gateway and major source of cortical projections to and from the hippocampus. We have already reported reduced hippocampal volume and an association between hippocampal volume and performance IQ in the same group of VLBW children [27]. We speculate that ERC thinning may negatively affect higher order cognitive functioning in VLBW subjects by influencing the hippocampal–entorhinal–neocortical dialog that probably is necessary for normal cognitive functioning. We are also concerned whether this renders preterm born children at increased risk of early or worsen aging, or even dementia like Alzheimer's disease (AD), since our findings resemble the entorhinal shrinkage seen in AD patients. However, this is a speculation and long term follow up studies of VLBW adults into aging has to be done to give an answer to this concern. Aging has also been associated with decreased FA in the CC and in association tracts that connect frontal cortex to more posterior brain regions [28]. Our previous report of reduced FA in these tracts in the same group of VLBW subjects, may give them an additional increased risk [12]. Long term follow up of the VLBW adults is warranted to rule out health complications like early or pathological aging like dementia later in life. However, the observed relationship between entorhinal cortical thickness and neuropsychological functioning may not be causal and other unknown factors that may affect ERC growth negatively in VLBW children may also impact negatively on perceptual and cognitive function through a more general effect on brain development and maturation.
4.2. Entorhinal cortical thinning and cognition Rodrigue et al. found that even mild longitudinal age-related shrinkage of the ERC was associated with reduced memory performance [6]. ERC shrinkage may be a sensitive predictor of memory decline, and volume reduction in the ERC is viewed as the earliest indicator of incipient conversion from preclinical cognitive impairment to dementia [3,23]. Thinning of the entorhinal cortex resulting in impaired cognitive function may be due to neuronal loss, loss of dendritic branching, or reduced intracortical myelination, or a combination of these events. The entorhinal cortical thinning seen in the VLBW adolescents in our study may be related to aberrant cortical maturation that has been described due to prematurity [24]. Whether this is caused by a primary influence on cortical development or secondary to perinatal white matter (WM) injury in preterm children, is unknown. Inder et al. reported that premature infants with concomitant perinatal WM injury had significantly reduced cortical gray matter, and that these infants exhibited moderate to severe neurodevelopmental disability at one year of age [25]. Entorhinal cortical thinning seen in our VLBW adolescents, secondary to WM injury and influencing connectivity to and from this important area, is supported by the fact that we have reported reduced fractional anisotropy (FA) values on diffusion tensor imaging in widespread WM areas in the same group of VLBW adolescents, indicating deviations in white matter microstructure [12]. Children with low IQ had reduced FA values in association tracts in the external capsule and inferior and middle superior fascicles [12]. These tracts may take part in networks of efferent and afferent information to and from the hippocampus through the entorhinal cortex. Andiman et al. have described a “dying-back” mechanism for neuronal loss and secondary cortical reduction in preterms with white matter injury [26]. Axonal disruption caused by perinatal white matter injury (periventricular leukomalacia) may lead to death of the neuronal somata when it loses its connections — a so-called dyingback mechanism. One could speculate that ERC areas that receive afferent input may be influenced by such a mechanism, if the tracts are affected.
5. Conclusion Entorhinal cortical thinning is related to low IQ and reduced scores on perceptual and executive function tests in VLBW adolescents, which may be caused by compromised hippocampal–entorhinal– prefrontal cortical dialog necessary for normal cognitive functioning. The structural–functional relationship found in this study is worrying, and we speculate whether there is an increased risk of early aging and dementia in adults born with VLBW. Conflict of interest statement The authors report no conflicts of interest. Ethical statement The Regional Committee for Medical Research Ethics approved the study protocol (reference number 78-00; May 29th 2000). The Data Inspectorate assigned the license for keeping a data register with personal information. The study population consisted of 49 VLBW and 58 control adolescents examined at 15 years of age. The Regional Committee for Medical Research Ethics approved the study protocol and written informed consent was obtained from both the adolescents and their parents. Acknowledgements We want to thank the participants and neuropsychologist Siri Kulseng, St Olav's University Hospital, Trondheim for the clinical assessments of the children. The study was funded by grants from The Research Council of Norway (NevroNor), the Norwegian University of Science and Technology, and Research Funds at St. Olav University Hospital, Trondheim, Norway.
108
J. Skranes et al. / Early Human Development 88 (2012) 103–109
Appendix A Material and methods Participants VLBW group. In 1986–88, 121 children were admitted to the NICU at the University Hospital in Trondheim. Thirty-three died in the neonatal period, one child with trisomy 21 was excluded, and six had moved before follow-up. Of the remaining 81, 55 (68%) agreed to participate in follow up and were assessed with cerebral MRI. Due to unsuccessful sequence acquisition and image artifacts six MRI examinations had to be excluded resulting in 49 studies suitable for morphometric analysis with the Freesurfer program. Control group. The control group consisted of 120 term born children with a birth weight ≥10th percentile for gestational age, born to mothers in a 10% random sample from another follow-up study. At follow-up 10 had moved and 27 did not consent to participate. Of the remaining 83, 65 (78%) underwent MRI scanning. Seven MRI scans had to be excluded because of inferior quality, leaving 58 MRI investigations suitable for segmentation and cortical thickness calculations. Non-participants. There were no significant differences in mothers' age at childbirth, duration of pregnancy, or the infant's birth weight, body length and head circumference between those who participated in the follow up study and those who did not consent to participation in any of the study groups. Neuropsychological tests Tests of Visual–motor integration. The Grooved Pegboard (GP) test requires manual dexterity. The adolescent is instructed to insert, successively and as quickly as possible, 25 pegs in the keyhole-shaped holes, ordered in a 5 × 5 matrix and pointing in different directions. The task was performed with the writing hand and thereafter with the non-writing hand. The scores were the times (seconds) used to complete each task. Adolescents who did not manage to finish the task were assigned a score three standard deviations (SD) above the mean of the controls. Poor performance was defined as any value 2 SD or more above (i.e. longer than) the mean time of the controls. The Developmental Test of Visual–Motor Integration (VMI–IV) comprises 27 geometric designs in increasing order of difficulty that has to be copied. No time limit was imposed. Visual perception and motor coordination are supplementary tasks requiring the subject to identify depictions of the designs that he/she has copied earlier among several similar shapes (VMI visual) and to trace the same designs with a pencil without leaving double-lined paths in which the designs were presented (VMI motor coordination). Time limits to complete the tasks are three and five minutes. Scores were given according to the manual (Beery, 1997) and raw scores were used in the analysis. Poor performance was defined as 2 SD below the mean of the control group. Tests of attention/executive functions/processing speed The Wisconsin Card Sorting Test (WCST), Research Edition (Computer Version 3 for Windows) was used to assess the problem-solving and cognitive flexibility aspects of executive functions. The adolescent was told to sort 128 cards with designs of geometrical figures, having to find the rule for sorting based on feedback from the examiner (color, shape or number). The participant is not told that the sorting rule changes through the task following specific rules based on the participants responses. Trials to complete the 1st category, number of errors of perseveration (a sorting response according to the previous sorting rule) and non-perseveration (all other errors) were registered
and raw scores were used in the analysis. Low performance was defined as scores more than 2 SD from the mean score in the control group. The Trail Making Test (TMT A and B) (intermediate form, 9–14 years) was chosen to assess mental flexibility, divided attention, visual scanning and psychomotor speed. In the first task (Part A) 15 encircled numbers randomly arranged on sheet of paper have to be connected in ascending order by drawing lines between them. In the second task (Part B) the subject is supposed to connect eight encircled numbers and eight encircled letters in ascending alternating order. Time to complete each test was computed for both groups. Low performance was defined as scores more than 2 SD from the mean score in the control group. The Knox-Cube test was used to assess immediate memory span for nonverbal material. The subject is instructed to tap a row of four identical cubes in the same sequence as the examiner did immediately before him. The sequences start at four taps and can increase up to 16 taps. Two successive identical trials are given, and the number of correctly imitated sequences was recorded in each of two trials; in addition the sum score was calculated. Low performance was defined as scores lower than − 2 SD from the mean score in the control group.
Table A Correlations between different test scores and left and right entorhinal cortical thickness in the VLBW adolescents.
VMI Motor coordination Visual perception Grooved pegboard dominant hand Grooved pegboard nondominant hand Full scale IQ Verbal IQ Performance IQ Knox cube WCST trials to complete 1st category WCST perseverative errors WCST nonperseverative errors Trail making test A Trail making test B
ERC thickness (left)
p-value
ERC thickness (right)
p-value
0.243 0.245 0.205 −0.404
0.092 0.093 0.163 0.004
0.321 0.342 0.255 −0.187
0.025 0.017 0.080 0.199
−0.373
0.008
−0.367
0.010
0.355 0.215 0.421 0.409 −0.441
0.012 0.139 0.003 0.004 0.002
0.201 0.149 0.280 0.149 −0.167
0.166 0.307 0.052 0.308 0.252
−0.486 −0.488
0.000 0.000
−0.307 −0.221
0.032 0.128
−0.405 −0.206
0.004 0.155
−0.464 0.004
0.001 0.978
Abbreviations: VLBW: very low birth weight; ERC: entorhinal cortex; VMI: Visual Motor Integration test; WCST: Wisconsin Card Sorting Test. p ≤ 0.01 (bold italic style), 0.01 b p ≤ 0.05 (bold style).
References [1] Squire LR, Stark CEL, Clark RE. The medial temporal lobe. Annu Rev Neurosci 2004;27:279–306. [2] Canto CB, Wouterlood FG, Witter MP. What does the anatomical organization of the entorhinal cortex tell us? Neural Plast 2008:381243 [Review.]. [3] De-Toledo-Morrell L, Goncharova I, Dickerson B, Wilson RS, Bennett DA. From healthy aging to early Alzheimer's disease: in vivo detection of entorhinal cortex atrophy. Ann NY Acad Sci 2000;911:240–53. [4] Kuperberg GR, Broome MR, McGuire PK, David AS, Eddy M, Ozawa F, et al. Regionally localized thinning of the cerebral cortex in schizophrenia. Arch Gen Psychiatry 2003;60:878–88. [5] Dickerson BC, Fenstermacher E, Salat DH, Wolk DA, Maguire RP, Desikan R, et al. Detection of cortical thickness correlates of cognitive performance: reliability across MRI scan sessions, scanners, and field strengths. Neuroimage 2008;39:10–8. [6] Rodrigue KM, Raz N. Shrinkage of the entorhinal cortex over five years predicts memory performance in healthy adults. J Neurosci 2004;24:956–63. [7] Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet 2009;8:110–24. [8] Kulseng S, Jennekens-Schinkel Aag, Naess P, Romundstad P, Indredavik M, Vik T, et al. Very-low-birthweight and term small-for-gestational-age adolescents: attention revisited. Acta Paediatr 2006;95:224–30.
J. Skranes et al. / Early Human Development 88 (2012) 103–109 [9] Aarnoudse-Moens CSH, Smidts DP, Oosterlaan J, Duivenvoorden HJ, WeisglasKuperus N. Executive function in very preterm children at early school age. J Abnorm Child Psychol 2009;37:981–93. [10] Hack M, Flannery DJ, Schluchter M, Cartar L, Borawski E, Klein N. Outcomes in young adulthood for very-low-birth-weight infants. N Engl J Med 2002;346: 149–57. [11] Kapellou O, Counsell SJ, Kennea N, Dyet L, Saeed N, Stark J, et al. Abnormal cortical development after premature birth shown by altered allometric scaling of brain growth. PLoS Med 2006;3:e265. [12] Skranes J, Vangberg TR, Kulseng S, Indredavik MS, Evensen KA, Martinussen M, et al. Clinical findings and white matter abnormalities seen on diffusion tensor imaging in adolescents with very low birth weight. Brain 2007;130:654–66. [13] Martinussen M, Fischl B, Larsson HB, Skranes J, Kulseng S, Vangberg TR, et al. Cerebral cortex thickness in 15-year-old adolescents with low birth weight measured by an automated MRI-based method. Brain 2005;128:2588–96. [14] Beery KE. The Beery-Buktencia. Developmental test of visual–motor integration. Administration, scoring and teaching manual. 4th ed. Modern Curriculum Press; 1997. [15] Wechsler D. Wechsler intelligence scale for children-third edition, Swedish version 1999. Stockholm: Psykologiförlaget AB; 1999. [16] Kløve H. Grooved Pegboard. Lafayette Instruments, model 32025; 1963. [17] Heaton RK, Thompson JA, Gomez E. Wisconsin Card Sorting Test. Computer Version 3 for Windows. Odessa, Florida: PAR Psychological assessment resources; 1999. [18] Stone MH, Wright BD. Knox's Cube test: junior and senior version. Wood Dale: Instruction manual. Stoelting Co.; 1980.
109
[19] Spreen O, Strauss E. A compendium of neuropsychological tests: administration, norms, and commentary. 2nd edition. New York: Oxford University Press; 1998. [20] Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis I. Segmentation and surface reconstruction. Neuroimage 1999;9:179–94. [21] Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis II. Inflation, flattening, a surface based coordinate system. Neuroimage 1999;9:195–207. [22] Rosas HD, Liu AK, Hersch S, Glessner M, Ferrante RJ, Salat DH, et al. Regional and progressive thinning of the cortical ribbon in Huntington's disease. Neurology 2002;58:695–701. [23] Dickerson BC, Goncharova I, Sullivan MP, Forchetti C, Wilson R, Bennett DA, et al. MRI-derived entorhinal and hippocampal atrophy in incipient and very mild Alzheimer's disease. Neurobiol Aging 2001;22:747–54. [24] Giménez M, Junqué C, Vendrell P, Narberhaus A, Bargalló N, Botet F, et al. Abnormal orbitofrontal development due to prematurity. Neurology 2006;67:1818–22. [25] Inder TE, Warfield SK, Wang H, Hüppi PS, Volpe JJ. Abnormal cerebral structure is present at term in premature infants. Pediatrics 2005;115:286–94. [26] Andiman SE, Haynes RL, Trachtenberg FL, Billiards SS, Folkerth RD, Volpe JJ, et al. The cerebral cortex overlying periventricular leukomalacia: analysis of pyramidal neurons. Brain Pathol 2010;20:803–14. [27] Martinussen M, Flanders DW, Fischl B, Busa E, Løhaugen GC, Skranes J, et al. Segmental brain volumes and cognitive and perceptual correlates in 15-year-old adolescents with low birth weight. J Pediatr 2009;155:848–53. [28] Perry ME, McDonald CR, Hagler Jr DJ, Gharapetian L, Kuperman JM, Koyama AK, et al. White matter tracts associated with set-shifting in healthy aging. Neuropsychologia 2009;47:2835–42.
Contents lists available at ScienceDirect
Early Human Development j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e a r l h u m d ev
Entorhinal cortical thinning affects perceptual and cognitive functions in adolescents born preterm with very low birth weight (VLBW) Jon Skranes a, e,⁎, Gro C.C. Løhaugen a, e, Kari Anne I. Evensen a, d, Marit S. Indredavik b, d, Olav Haraldseth c, Anders M. Dale c, f, Ann-Mari Brubakk a, d, Marit Martinussen a, d a
Dept. of Lab. Medicine, Children's and Women's Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Dept. of Neuroscience, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway Dept. of Circulation and Medical Imaging, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway d St Olav University Hospital, Trondheim, Norway e Sørlandet Hospital, Arendal, Norway f MIL, Dept. of Neurosciences and Radiology, University of San Diego, La Jolla, CA, USA b c
a r t i c l e
i n f o
Article history: Received 9 February 2011 Received in revised form 22 June 2011 Accepted 3 July 2011 Keywords: VLBW Preterm birth Brain Cortical thickness Entorhinal cortex VMI WISC-III IQ Executive function
a b s t r a c t Background: The entorhinal cortex serves as an important gateway between the cerebral cortex and the hippocampus by receiving afferent information from limbic, modality sensory-specific, and multimodal association fibers from all the brain lobes. Aim: To investigate whether thinning of entorhinal cortex is associated with reduced perceptual, cognitive and executive skills in very low birth weight (VLBW) adolescents. Study design: Prospective, geographically based follow-up study of three year cohorts of preterm born VLBW children. Subjects: Forty-nine VLBW (birth weight ≤ 1500 g) and 58 term-born control adolescents were examined at the age of 14–15 years. Outcome measures: Perceptual and cognitive functions were assessed with Visual motor integration test, Grooved Pegboard test, Wechsler Intelligence Scale for Children-III and different executive function tests (Wisconsin card sorting test, Trail Making test, Knox cube test). An automated MRI technique at 1.5 T for morphometric analyses of cortical thickness was performed. Areas with cortical thinning in left and right entorhinal cortex in the VLBW group were chosen as regions of interest to look for associations between cortical thickness and clinical findings. Results: Thinning of the entorhinal cortex was correlated with low performance on perceptual and cognitive scores in the VLBW adolescents, but not in controls. In addition, thinning of the entorhinal cortices correlated with reduced performance on several executive tests, including perceptual speed and aspects of working memory. Conclusions: Entorhinal cortical thinning is related with low IQ and reduced perceptual and executive functions in VLBW adolescents. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The medial temporal lobe includes a system of anatomically related structures where the hippocampus lies at the end of a cortical processing hierarchy; the entorhinal cortex (ERC) is the major source of its cortical projections. This system is principally concerned with long-term memory, but also critically involved in complex functions such as sensory representation and spatial orientation [1]. Recently it was stated that the ERC may act as the nodal point between ⁎ Corresponding author at: Department of Laboratory Medicine, Children's and Women's Health, Medical Faculty, Norwegian University of Science and Technology, Medisinsk teknisk forskningssenter, N-7489 Trondheim, Norway. Tel.: + 47 99390285. E-mail address: [email protected] (J. Skranes). 0378-3782/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.earlhumdev.2011.07.017
hippocampus and a variety of multimodal association areas of the cortex such as parietal, temporal, and prefrontal cortex [2]. Computational automated morphometric MRI methods with high reliability have detected correlates between cortical thinning and cognitive task performance in healthy individuals and in different patient groups (schizophrenia, Alzheimer's disease (AD), epilepsy) [3–5], and entorhinal shrinkage has been associated with cognitive decline in healthy elderly persons and in AD patients [6]. Children born very preterm or with very low birth weight (VLBW: birth weight ≤ 1500 g) are prone to perinatal brain injury and have an increased risk of developing motor and cognitive impairments during childhood and adolescence [7]. These impairments include reduced skills in perceptual and higher cognitive tasks that measure executive functions, attention and memory functions [8,9]. Quantitative MR
104
J. Skranes et al. / Early Human Development 88 (2012) 103–109
studies have revealed both white and gray matter abnormalities in VLBW children, and the MR pathology seems to persist into adolescence and early adulthood, as do clinical problems [10–12]. We have previously published areas of cortical thinning and thickening in a cohort of VLBW children at 14–15 years of age and found that regional thinning in areas corresponding to the ERC was more pronounced in those with low IQ compared with those with normal IQ in the VLBW group, and in the most immature of the preterm born children [13]. The ERC area therefore seemed interesting to investigate further with quantitative measurements of mean thickness versus different aspects of cognitive functions. The aim of the present study was to investigate whether ERC thinning is associated with reduced perceptual, cognitive and executive/attentional skills in VLBW adolescents. We hypothesized that aberrant development of this important cortical area would interfere with higher order cognitive and perceptual functions in these adolescents. 2. Methods This study is part of a hospital-based follow-up study from birth where three year cohorts of prematurely born VLBW children from a defined geographic region in Norway were compared with randomly selected term born controls from the same geographic area examined at regular intervals with a battery of clinical tests and different MRI techniques. 2.1. Participants The study population consisted of 49 VLBW and 58 control adolescents examined at 15 years of age. The Regional Committee for Medical Research Ethics approved the study protocol and written informed consent was obtained. (See Appendix for more information about the study population). 2.2. Neuropsychological assessment battery An estimate of the adolescents' intelligence quotient (IQest) was calculated using four subtests of Wechsler Intelligence Scale for Children (WISC)-third edition: Arithmetic, Vocabulary, Block design and Picture arrangement [15]. We defined low IQ as an estimated IQ score b 2 standard deviations (SD) of the mean in the control group. Visual–motor function was assessed by the Developmental Test of Visual–Motor Integration (VMI–IV), including supplementary tests to evaluate visual perception and motor coordination [14]. Poor performance/impairment was defined as score b − 2 SD from the mean scores of the control group. Attention and executive function were assessed with the, Wisconsin Card Sorting Test (WCST), version III [17], Knox-Cube test [18], and Trail Making Test [19], while the Grooved Pegboard (GP) test [16] was used to assess both motor coordination and processing speed. (See Appendix for more information about each of the neuropsychological tests). 2.3. MR imaging The scanning was performed on a 1.5 T Siemens Symphony Sonata, Siemens AG, Erlangen, Germany. The imaging for the morphometric analyses was a 3D inversion recovery MPRAGE sequence with 128 sagittal partitions, slice thickness of 1.33 mm, square FOV of 256 mm, and acquisition duration of 8.5 min. An automated cortical surface reconstruction method developed by Dale and Fischl was performed to construct cortical thickness for each subject's entire brain [20,21]. The method has been described before, and statistical significant cortical thickness difference maps between the two groups of adolescents were generated [13]. Several significantly thinner areas were found in the VLBW adolescents (Fig. 1). Thinner areas in the right and left medial
Fig. 1. Statistical difference maps of cortical thickness between 49 VLBW and 58 control adolescents are overlaid on the surface reconstruction with uninflated (upper panel) and inflated (middle and lower panel) brains. Dark gray areas correspond to sulci and light gray areas are gyri. Areas with statistically significant thickness difference between groups are shown in color, and the color scale shows the dynamic range of the changes, red to yellow represents an increasing thinning of the cortex in the VLBW group, and full yellow corresponds to a statistical difference in cortical thickness with a p-value ≤ 0.0001. Blue represents a statistically significant thickening of the cortex in the VLBW group; light blue corresponds to a statistical difference in cortical thickness with a p-value ≤ 0.0001. Areas of regional thinning of the entorhinal cortex (regions of interest) are marked in white in the lower panel.
temporal lobe corresponded to an area that covered most of the ERC. These areas, labeled white in Fig. 1C and D, were used as regions of interest (ROIs) in the present study. The regions of interest were mapped back to each participant by using a high-dimensional spherical morphing procedure to find the homologous regions across subjects. A mean thickness for each of the two entorhinal cortical areas was then recorded for every VLBW participant. 2.4. Statistics The associations between the individual mean thickness calculations in the ROI and the neuropsychological test scores in the VLBW group were explored by linear regression analysis. For continuous variables Pearson's r was used for correlation analyses between ERC thickness and IQ measurements, while Spearman's rho were performed for the neuropsychological test results since these are not normally distributed. For the correlation analyses the two-tailed significance threshold was set to p b 0.02 to adjust for multiple comparisons. In addition, we explored the relationship between thickness and neuropsychological test scores when the scores were dichotomized into those indicating normal and abnormal/low function (more than 2 SD from mean value of controls) using the Mann–Whitney U test. Adjustments for socioeconomic status (SES) and IQ were performed for the neuropsychological tests, since low test results may be influenced by both these factors. Here p-values≤ 0.05 were considered significant. 3. Results 3.1. Clinical and MRI findings The clinical characteristics, test results and MRI findings for the study groups are shown in Table 1. Age at MRI examination was about 12 months higher in both groups of children when comparing with
**p ≤ 0.01, *p ≤ 0.05, #p = 0.09 (normal versus low test results). Mann–Whitney U-test, General Linear Model for adjusted means. Abbreviations: SD: standard deviation; VLBW: very low birth weight; VMI: Visual Motor Integration test; WISC: Wechsler Intelligence Scale for Children; WCST persev. errors: Wisconsin Card Sorting Test–perseverative error score; SES: socioeconomic status; SE: standard error. a Scores higher/equal to or lower than −2 SD from the mean score in the control group.
2.43 (0.04)* 2.40 (0.05) 2.24 (0.08) 2.25 (0.10) 2.44 (0.04)** 2.44 (0.04)** 2.13 (0.11) 2.01 (0.12) 2.41 (0.04) 2.38 (0.05) 2.26 (0.13) 2.25 (0.14) 2.43 (0.05) 2.39 (0.05) 2.28 (0.08) 2.30 (0.05) 2.44 (0.04)** 2.42 (0.04)** 2.16 (0.09) 2.12 (0.09) 2.43 (0.04)* 2.41 (0.05)* 2.14 (0.11) 2.13 (0.12) 2.44 (0.04)* 2.42 (0.05)* for SES only): 2.42 2.23 (0.04) (0.09) 2.39 2.17 (0.05) (0.10) Mean values after adjustment for SES and IQ (WISC-III results adjusted ERC left: 2.27 2.44 2.33 2.42 2.29 (SE 0.08) (0.05) (0.08) (0.05) (0.10) ERC right: 2.16 2.45 2.21 2.42 2.29 (0.08) (0.05)** (0.09) (0.05)* (0.11)
ERC right:
(39)
2.44 (0.26)** 2.41 (0.27) 2.19 (0.27) 2.22 (0.37) 2.45 (0.26)** 2.43 (0.25)** 2.09 (0.13) 2.03 (0.29) 2.43 (0.27)* 2.40 (0.28) 2.15 (0.22) 2.16 (0.33) 2.45 (0.26)** 2.41 (0.28) 2.23 (0.25) 2.25 (0.33) 2.44 (0.27)** 2.42 (0.27)** 2.17 (0.19) 2.12 (0.29) 2.44 (0.26)** 2.41 (0.27)* 2.08 (0.18) 2.10 (0.30) 2.45 (0.26)** 2.43 (0.27)* 2.18 (0.24) 2.16 (0.29) 2.43 (0.25)* 2.40 (0.29) 2.23 (0.33) 2.23 (0.32) 2.44 (0.27) 2.43 (0.28)**
b−2 SD
(10) (41)
≥−2 SD b−2 SD
(8) (43)
≥−2 SD b−2 SD
(6) (36)
≥−2 SD b−2 SD
(13) (40) (9) (42)
≥−2 SDa b−2 SD ≥−2 SD b−2 SD
(7) (38) (11) (40) (9) (36) (12)
2.27 (0.25) 2.18 (0.29) 2.46 (0.28)** 2.46 (0.25)** 2.22 (SD 0.19) 2.15 (0.29)
In this study we report that entorhinal cortical (ERC) thinning is associated with general cognitive ability, visual–motor and executive function deficits in VLBW adolescents. To our knowledge, this is the first study that reports a possible relationship between higher order cognitive deficits and ERC thinning in preterm born subjects.
ERC left:
4. Discussion
(35)
Table 2 shows the associations between cortical thickness and dichotomized test results (means unadjusted and adjusted for SES and IQ) in the VLBW group. Thinning of the entorhinal cortex (ERC) was associated with low scores on VMI test, motor coordination test (right ERC only), estimated IQ, Grooved pegboard (left ERC only) and WCST perseverative errors. Correlation analyses (Fig. 2) showed that there was a linear correlation between right ERC thickness and motor coordination score and Trail Making Test score, and between left ERC thickness and performance and full scale IQ, Knox Cube Test and WCST perseverative error score. A full table of all the correlations investigated is presented as Table A in the Appendix.
(14)
3.2. Relationship between neuropsychological function and entorhinal cortical thickness
≥−2 SD
age at clinical examination. The reason for this was that the initial MRI taken at the same time as the clinical assessment had to be repeated because of technical reasons due to incorrect DTI sequencing.
b−2 SD
*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (VLBW versus controls). t-test for parametric data, Mann–Whitney U-test for non-parametric data. Abbreviations: ERC: Entorhinal cortex; VLBW: very low birth weight; SD: standard deviation; SES: socio-economic status; MRI: magnetic resonance imaging; VMI: Visual Motor Integration test; WCST: Wisconsin card sorting test; persev: perseverative; TMT: Trail making test.
≥−2 SD
2.63 (0.28) 2.56 (0.34)
b−2 SD
WISC-III
2.39 (0.27)*** 2.36 (0.29)**
(6.5) (6.2) (10.8) (2.0) (5.0) (13.6)
≥−2 SD
13.6 11.5 14.7 13.8 14.1 29.7
(37.6) (12.5)*** (16.9)** (2.7)*** (8.5)*** (25.0)***
b−2 SD
27.6 18.8 23.2 12.0 19.4 50.3
≥−2 SD
65.1 (8) 73.5 (11)
105
b−2 SDa
79.8 (48)* 89.7 (43)**
WCST persev. errors
22.7 (2.7) 24.2 (2.2) 24.6 (2.5) 96 (17) 93 (16) 101 (19)
WCST trials to complete
19.5 (3.7)*** 21.4 (3.1)*** 22.5 (3.7)*** 79 (23)*** 82 (19)*** 81 (28)***
Knox cube
15.3 (0.5) 14.2 (0.3) 167.8 (8.3) 58.0 (11.4)
Grooved pegboard non-dom
15.1 (0.6) 14.2 (0.3) 162.2 (9.2)** 50.7 (13.2)***
Grooved pegboard dom
3707 (486) 39.6 (1.1) 23/35 3.6 (1.1)
Visual perception
1195 (239)*** 29.1 (2.7)*** 25/24 3.2 (1.3)
Motor coordination test
n = 58
Anatomical VMI region
Controls
n = 49
Table 2 Entorhinal cortical (ERC) thickness, mean (SD/SE) in VLBW adolescents with low and normal neuropsychological test results.
Perinatal variables Birth weight (grams) Gestational age (weeks) Boys/girls SES At age 14–15 Age at MRI examination (years) Age at clinical testing (years) Length (cm) Weight (kg) Clinical test results VMI Motor coordination Visual perception Full scale IQ –Verbal IQ –Performance IQ Grooved pegboard (seconds) – Writing hand – Non-dominant hand WCST – Trial 1st category – Persev. errors – Non-persev. errors Knox cube (seconds) TMT A (seconds) TMT B (seconds) MRI findings ERC thickness left (mm) ERC thickness right (mm)
VLBW
Trail making test A
Table 1 Child characteristics, clinical test results and ERC thickness (mean, SD) in the study groups.
≥−2 SD
J. Skranes et al. / Early Human Development 88 (2012) 103–109
106 J. Skranes et al. / Early Human Development 88 (2012) 103–109 Fig. 2. Correlation plots (Spearman's rho) of right entorhinal cortical thickness and motor coordination score (upper left) and Trail making test score (upper middle), respectively. Correlation plots of left entorhinal cortical thickness and performance IQ (upper right), full scale IQ (lower left) (Pearson's r), Knox cube test (lower middle) and WCST perseverative errors (lower right) (Spearman's rho), respectively in 49 VLBW adolescents. Abbreviations: WCST: Wisconsin card sorting test; VLBW: very low birth weight.
J. Skranes et al. / Early Human Development 88 (2012) 103–109
107
4.1. Methodological considerations
4.3. Clinical implications
The automated MR segmentation tool in this study has been widely used for measurement of cortical thickness (http://freesurfersoftware.org). However, using an automated cortical reconstruction method for measurement of cortical thickness always raises the possibility of inaccuracies in the normalization process and in defining the gray-white matter border. The segmentation method has been validated by performing analysis on post-mortem brains [22], and by manual measurements [4]. All surface models in our study were visually inspected for accuracy. However, the entorhinal cortex is a difficult area to map out in the Freesurfer program, so some manual correction of the segmentation was performed, but this was done blinded to group adherence and by one person. The region of interest (ROI) in our study constitutes a rather large portion of the medial temporal lobe, covering parts of the entorhinal cortex, parts of the temporal pole as well as the parahippocampal gyrus. The ROI therefore most likely includes cortical areas with different functionality. This could obviously interfere with the possibility of detecting structural–functional relationship. The results could also be due to chance, since the number of participants was rather low. However, then we would have expected both positive and negative correlations between structure and function, and not only what we found: the thinner the cortex, the lower the test results. Nevertheless, the structure–function association, identified for the entorhinal regions in the present study should be interpreted cautiously and subjected to replication in larger samples before generalization of the results.
Our findings of an association between entorhinal thinning and a rather broad spectrum of cognitive, visual–motor, and executive deficits support the view of the ERC as a gateway and major source of cortical projections to and from the hippocampus. We have already reported reduced hippocampal volume and an association between hippocampal volume and performance IQ in the same group of VLBW children [27]. We speculate that ERC thinning may negatively affect higher order cognitive functioning in VLBW subjects by influencing the hippocampal–entorhinal–neocortical dialog that probably is necessary for normal cognitive functioning. We are also concerned whether this renders preterm born children at increased risk of early or worsen aging, or even dementia like Alzheimer's disease (AD), since our findings resemble the entorhinal shrinkage seen in AD patients. However, this is a speculation and long term follow up studies of VLBW adults into aging has to be done to give an answer to this concern. Aging has also been associated with decreased FA in the CC and in association tracts that connect frontal cortex to more posterior brain regions [28]. Our previous report of reduced FA in these tracts in the same group of VLBW subjects, may give them an additional increased risk [12]. Long term follow up of the VLBW adults is warranted to rule out health complications like early or pathological aging like dementia later in life. However, the observed relationship between entorhinal cortical thickness and neuropsychological functioning may not be causal and other unknown factors that may affect ERC growth negatively in VLBW children may also impact negatively on perceptual and cognitive function through a more general effect on brain development and maturation.
4.2. Entorhinal cortical thinning and cognition Rodrigue et al. found that even mild longitudinal age-related shrinkage of the ERC was associated with reduced memory performance [6]. ERC shrinkage may be a sensitive predictor of memory decline, and volume reduction in the ERC is viewed as the earliest indicator of incipient conversion from preclinical cognitive impairment to dementia [3,23]. Thinning of the entorhinal cortex resulting in impaired cognitive function may be due to neuronal loss, loss of dendritic branching, or reduced intracortical myelination, or a combination of these events. The entorhinal cortical thinning seen in the VLBW adolescents in our study may be related to aberrant cortical maturation that has been described due to prematurity [24]. Whether this is caused by a primary influence on cortical development or secondary to perinatal white matter (WM) injury in preterm children, is unknown. Inder et al. reported that premature infants with concomitant perinatal WM injury had significantly reduced cortical gray matter, and that these infants exhibited moderate to severe neurodevelopmental disability at one year of age [25]. Entorhinal cortical thinning seen in our VLBW adolescents, secondary to WM injury and influencing connectivity to and from this important area, is supported by the fact that we have reported reduced fractional anisotropy (FA) values on diffusion tensor imaging in widespread WM areas in the same group of VLBW adolescents, indicating deviations in white matter microstructure [12]. Children with low IQ had reduced FA values in association tracts in the external capsule and inferior and middle superior fascicles [12]. These tracts may take part in networks of efferent and afferent information to and from the hippocampus through the entorhinal cortex. Andiman et al. have described a “dying-back” mechanism for neuronal loss and secondary cortical reduction in preterms with white matter injury [26]. Axonal disruption caused by perinatal white matter injury (periventricular leukomalacia) may lead to death of the neuronal somata when it loses its connections — a so-called dyingback mechanism. One could speculate that ERC areas that receive afferent input may be influenced by such a mechanism, if the tracts are affected.
5. Conclusion Entorhinal cortical thinning is related to low IQ and reduced scores on perceptual and executive function tests in VLBW adolescents, which may be caused by compromised hippocampal–entorhinal– prefrontal cortical dialog necessary for normal cognitive functioning. The structural–functional relationship found in this study is worrying, and we speculate whether there is an increased risk of early aging and dementia in adults born with VLBW. Conflict of interest statement The authors report no conflicts of interest. Ethical statement The Regional Committee for Medical Research Ethics approved the study protocol (reference number 78-00; May 29th 2000). The Data Inspectorate assigned the license for keeping a data register with personal information. The study population consisted of 49 VLBW and 58 control adolescents examined at 15 years of age. The Regional Committee for Medical Research Ethics approved the study protocol and written informed consent was obtained from both the adolescents and their parents. Acknowledgements We want to thank the participants and neuropsychologist Siri Kulseng, St Olav's University Hospital, Trondheim for the clinical assessments of the children. The study was funded by grants from The Research Council of Norway (NevroNor), the Norwegian University of Science and Technology, and Research Funds at St. Olav University Hospital, Trondheim, Norway.
108
J. Skranes et al. / Early Human Development 88 (2012) 103–109
Appendix A Material and methods Participants VLBW group. In 1986–88, 121 children were admitted to the NICU at the University Hospital in Trondheim. Thirty-three died in the neonatal period, one child with trisomy 21 was excluded, and six had moved before follow-up. Of the remaining 81, 55 (68%) agreed to participate in follow up and were assessed with cerebral MRI. Due to unsuccessful sequence acquisition and image artifacts six MRI examinations had to be excluded resulting in 49 studies suitable for morphometric analysis with the Freesurfer program. Control group. The control group consisted of 120 term born children with a birth weight ≥10th percentile for gestational age, born to mothers in a 10% random sample from another follow-up study. At follow-up 10 had moved and 27 did not consent to participate. Of the remaining 83, 65 (78%) underwent MRI scanning. Seven MRI scans had to be excluded because of inferior quality, leaving 58 MRI investigations suitable for segmentation and cortical thickness calculations. Non-participants. There were no significant differences in mothers' age at childbirth, duration of pregnancy, or the infant's birth weight, body length and head circumference between those who participated in the follow up study and those who did not consent to participation in any of the study groups. Neuropsychological tests Tests of Visual–motor integration. The Grooved Pegboard (GP) test requires manual dexterity. The adolescent is instructed to insert, successively and as quickly as possible, 25 pegs in the keyhole-shaped holes, ordered in a 5 × 5 matrix and pointing in different directions. The task was performed with the writing hand and thereafter with the non-writing hand. The scores were the times (seconds) used to complete each task. Adolescents who did not manage to finish the task were assigned a score three standard deviations (SD) above the mean of the controls. Poor performance was defined as any value 2 SD or more above (i.e. longer than) the mean time of the controls. The Developmental Test of Visual–Motor Integration (VMI–IV) comprises 27 geometric designs in increasing order of difficulty that has to be copied. No time limit was imposed. Visual perception and motor coordination are supplementary tasks requiring the subject to identify depictions of the designs that he/she has copied earlier among several similar shapes (VMI visual) and to trace the same designs with a pencil without leaving double-lined paths in which the designs were presented (VMI motor coordination). Time limits to complete the tasks are three and five minutes. Scores were given according to the manual (Beery, 1997) and raw scores were used in the analysis. Poor performance was defined as 2 SD below the mean of the control group. Tests of attention/executive functions/processing speed The Wisconsin Card Sorting Test (WCST), Research Edition (Computer Version 3 for Windows) was used to assess the problem-solving and cognitive flexibility aspects of executive functions. The adolescent was told to sort 128 cards with designs of geometrical figures, having to find the rule for sorting based on feedback from the examiner (color, shape or number). The participant is not told that the sorting rule changes through the task following specific rules based on the participants responses. Trials to complete the 1st category, number of errors of perseveration (a sorting response according to the previous sorting rule) and non-perseveration (all other errors) were registered
and raw scores were used in the analysis. Low performance was defined as scores more than 2 SD from the mean score in the control group. The Trail Making Test (TMT A and B) (intermediate form, 9–14 years) was chosen to assess mental flexibility, divided attention, visual scanning and psychomotor speed. In the first task (Part A) 15 encircled numbers randomly arranged on sheet of paper have to be connected in ascending order by drawing lines between them. In the second task (Part B) the subject is supposed to connect eight encircled numbers and eight encircled letters in ascending alternating order. Time to complete each test was computed for both groups. Low performance was defined as scores more than 2 SD from the mean score in the control group. The Knox-Cube test was used to assess immediate memory span for nonverbal material. The subject is instructed to tap a row of four identical cubes in the same sequence as the examiner did immediately before him. The sequences start at four taps and can increase up to 16 taps. Two successive identical trials are given, and the number of correctly imitated sequences was recorded in each of two trials; in addition the sum score was calculated. Low performance was defined as scores lower than − 2 SD from the mean score in the control group.
Table A Correlations between different test scores and left and right entorhinal cortical thickness in the VLBW adolescents.
VMI Motor coordination Visual perception Grooved pegboard dominant hand Grooved pegboard nondominant hand Full scale IQ Verbal IQ Performance IQ Knox cube WCST trials to complete 1st category WCST perseverative errors WCST nonperseverative errors Trail making test A Trail making test B
ERC thickness (left)
p-value
ERC thickness (right)
p-value
0.243 0.245 0.205 −0.404
0.092 0.093 0.163 0.004
0.321 0.342 0.255 −0.187
0.025 0.017 0.080 0.199
−0.373
0.008
−0.367
0.010
0.355 0.215 0.421 0.409 −0.441
0.012 0.139 0.003 0.004 0.002
0.201 0.149 0.280 0.149 −0.167
0.166 0.307 0.052 0.308 0.252
−0.486 −0.488
0.000 0.000
−0.307 −0.221
0.032 0.128
−0.405 −0.206
0.004 0.155
−0.464 0.004
0.001 0.978
Abbreviations: VLBW: very low birth weight; ERC: entorhinal cortex; VMI: Visual Motor Integration test; WCST: Wisconsin Card Sorting Test. p ≤ 0.01 (bold italic style), 0.01 b p ≤ 0.05 (bold style).
References [1] Squire LR, Stark CEL, Clark RE. The medial temporal lobe. Annu Rev Neurosci 2004;27:279–306. [2] Canto CB, Wouterlood FG, Witter MP. What does the anatomical organization of the entorhinal cortex tell us? Neural Plast 2008:381243 [Review.]. [3] De-Toledo-Morrell L, Goncharova I, Dickerson B, Wilson RS, Bennett DA. From healthy aging to early Alzheimer's disease: in vivo detection of entorhinal cortex atrophy. Ann NY Acad Sci 2000;911:240–53. [4] Kuperberg GR, Broome MR, McGuire PK, David AS, Eddy M, Ozawa F, et al. Regionally localized thinning of the cerebral cortex in schizophrenia. Arch Gen Psychiatry 2003;60:878–88. [5] Dickerson BC, Fenstermacher E, Salat DH, Wolk DA, Maguire RP, Desikan R, et al. Detection of cortical thickness correlates of cognitive performance: reliability across MRI scan sessions, scanners, and field strengths. Neuroimage 2008;39:10–8. [6] Rodrigue KM, Raz N. Shrinkage of the entorhinal cortex over five years predicts memory performance in healthy adults. J Neurosci 2004;24:956–63. [7] Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet 2009;8:110–24. [8] Kulseng S, Jennekens-Schinkel Aag, Naess P, Romundstad P, Indredavik M, Vik T, et al. Very-low-birthweight and term small-for-gestational-age adolescents: attention revisited. Acta Paediatr 2006;95:224–30.
J. Skranes et al. / Early Human Development 88 (2012) 103–109 [9] Aarnoudse-Moens CSH, Smidts DP, Oosterlaan J, Duivenvoorden HJ, WeisglasKuperus N. Executive function in very preterm children at early school age. J Abnorm Child Psychol 2009;37:981–93. [10] Hack M, Flannery DJ, Schluchter M, Cartar L, Borawski E, Klein N. Outcomes in young adulthood for very-low-birth-weight infants. N Engl J Med 2002;346: 149–57. [11] Kapellou O, Counsell SJ, Kennea N, Dyet L, Saeed N, Stark J, et al. Abnormal cortical development after premature birth shown by altered allometric scaling of brain growth. PLoS Med 2006;3:e265. [12] Skranes J, Vangberg TR, Kulseng S, Indredavik MS, Evensen KA, Martinussen M, et al. Clinical findings and white matter abnormalities seen on diffusion tensor imaging in adolescents with very low birth weight. Brain 2007;130:654–66. [13] Martinussen M, Fischl B, Larsson HB, Skranes J, Kulseng S, Vangberg TR, et al. Cerebral cortex thickness in 15-year-old adolescents with low birth weight measured by an automated MRI-based method. Brain 2005;128:2588–96. [14] Beery KE. The Beery-Buktencia. Developmental test of visual–motor integration. Administration, scoring and teaching manual. 4th ed. Modern Curriculum Press; 1997. [15] Wechsler D. Wechsler intelligence scale for children-third edition, Swedish version 1999. Stockholm: Psykologiförlaget AB; 1999. [16] Kløve H. Grooved Pegboard. Lafayette Instruments, model 32025; 1963. [17] Heaton RK, Thompson JA, Gomez E. Wisconsin Card Sorting Test. Computer Version 3 for Windows. Odessa, Florida: PAR Psychological assessment resources; 1999. [18] Stone MH, Wright BD. Knox's Cube test: junior and senior version. Wood Dale: Instruction manual. Stoelting Co.; 1980.
109
[19] Spreen O, Strauss E. A compendium of neuropsychological tests: administration, norms, and commentary. 2nd edition. New York: Oxford University Press; 1998. [20] Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis I. Segmentation and surface reconstruction. Neuroimage 1999;9:179–94. [21] Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis II. Inflation, flattening, a surface based coordinate system. Neuroimage 1999;9:195–207. [22] Rosas HD, Liu AK, Hersch S, Glessner M, Ferrante RJ, Salat DH, et al. Regional and progressive thinning of the cortical ribbon in Huntington's disease. Neurology 2002;58:695–701. [23] Dickerson BC, Goncharova I, Sullivan MP, Forchetti C, Wilson R, Bennett DA, et al. MRI-derived entorhinal and hippocampal atrophy in incipient and very mild Alzheimer's disease. Neurobiol Aging 2001;22:747–54. [24] Giménez M, Junqué C, Vendrell P, Narberhaus A, Bargalló N, Botet F, et al. Abnormal orbitofrontal development due to prematurity. Neurology 2006;67:1818–22. [25] Inder TE, Warfield SK, Wang H, Hüppi PS, Volpe JJ. Abnormal cerebral structure is present at term in premature infants. Pediatrics 2005;115:286–94. [26] Andiman SE, Haynes RL, Trachtenberg FL, Billiards SS, Folkerth RD, Volpe JJ, et al. The cerebral cortex overlying periventricular leukomalacia: analysis of pyramidal neurons. Brain Pathol 2010;20:803–14. [27] Martinussen M, Flanders DW, Fischl B, Busa E, Løhaugen GC, Skranes J, et al. Segmental brain volumes and cognitive and perceptual correlates in 15-year-old adolescents with low birth weight. J Pediatr 2009;155:848–53. [28] Perry ME, McDonald CR, Hagler Jr DJ, Gharapetian L, Kuperman JM, Koyama AK, et al. White matter tracts associated with set-shifting in healthy aging. Neuropsychologia 2009;47:2835–42.