Summary - research.ed.ac.uk file · Web viewCognitive impairment in early onset epilepsy is...

Cognitive impairment in early onset

epilepsy is associated with reduced left

thalamic volume

Authors: Michael Yoong MBBCh MRCPCH PhD1,2, Matthew Hunter PhD1, Jacqueline

Stephen PhD3, Alan Quigley MBChB FRCR4, Jeremy Jones MBChB FRCR4, Jay Shetty

MBBS FRCPCH2, Ailsa McLellan MBChB FRCPCH2, Mark E. Bastin PhD1,5, Richard FM

Chin MRCPCH PhD1,2

1Muir Maxwell Epilepsy Centre, University of Edinburgh

2Department of Paediatric Neurology, Royal Hospital for Sick Children, Edinburgh, UK

3Edinburgh Clinical Trials Unit, Usher Institute of Population Health Sciences and

Informatics, University of Edinburgh, UK

4Department of Radiology, Royal Hospital for Sick Children, Edinburgh, UK

5Brain Research Imaging Centre, University of Edinburgh, UK

Corresponding author: Dr Yoong, Muir Maxwell Epilepsy Centre, Child Life and Health, 20

Sylvan Place, Edinburgh EH9 1UW, UK. E-mail: [email protected] Tel: +44 131 536

0801

Word count:

Summary: 239 words

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

mailto:[email protected]

Left thalamic volume reduction and cognitive impairment in preschool epilepsy Yoong

Page 2 of 24

Article: 22 pages comprising 2816 words, 3 tables, 2 figures

Supplementary material: 3 Tables

Summary

Objective: To investigate whether reduction of thalamic volumes in children with early onset

epilepsy (CWEOE) is associated with cognitive impairment.

Methods: Nested case-control study including a prospectively recruited cohort of 76 children

with newly-diagnosed early onset epilepsy (onset <5 years age) and 14 healthy controls

presenting to hospitals within NHS Lothian and Fife. Quantitative volumetric analysis of

subcortical structures was performed using volumetric T1-weighted MRI and correlated with

the results of formal neurocognitive and clinical assessment. False discovery rate was used to

correct for multiple comparisons as appropriate with q<0.05 used to define statistical

significance.

Results: Age, gender and ICV-adjusted left thalamic volumes were significantly reduced in

CWEOE with cognitive impairment compared to CWEOE without impairment (5295 mm3 vs

6418mm3, q=0.008) or healthy controls (5295mm3 vs 6410mm3, q<0.001). The differences in

left thalamic volume remained if grey matter or cortical/cerebellar volumes were used as co-

variates rather than ICV (q<0.05). The degree of volume reduction correlated with the

severity of cognitive impairment (q=0.048).

Significance: Reduced left thalamic volume may be a biomarker for cognitive impairment in

CWEOE and could help inform the need for further formal cognitive evaluations and

interventions.

Key Words: Epilepsy, Volumetric MRI, cognitive impairment, preschool, thalamus,

2

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 3 of 24

Abbreviations

Bayley III – Bayley Scales of Infant and Toddler Development III

CI – cognitive impairment

CWEOE – children with early onset epilepsy

EP+ - Children with epilepsy and cognitive impairment

EP- - Children with epilepsy with no cognitive impairment

FDR – False Discovery Rate

HC – Healthy Controls

ICV – Intracranial volume

WPPSI III – Wechsler Preschool and Primary Scales of Intelligence III

1 Introduction

Cognitive problems are common in children with epilepsy compared to both the

general population and children with other chronic health problems[1–3] with age of epilepsy

onset a recognised risk factor[4,5]. Forty percent of children with early onset epilepsy

(CWEOE; onset < 5 years) have cognitive impairment (CI) with cognitive scores at least two

standard deviations below population means[5]. CI is strongly associated with other

neurobehavioural impairments [6] and is a stronger predictor of long term social and quality

3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18


Page 4 of 24

of life outcomes than seizure control or severity[7,8] It is acknowledged that improving our

understanding and ability to predict CI remains a priority[9].

Although the first years of life are the age with the highest incidence of epilepsy[10]

as well as the highest risk of associated comorbidities, there are few systematic

neuroimaging studies on CI in this group. The majority of neurological models of cognition

focus predominantly on cortical-based measurements of grey matter volume, thickness and

connectivity[11], however there is increasing interest in the role of subcortical structures in

cognition, particularly in patients with epilepsy. Decreases in thalamic volume compared to

healthy controls have been demonstrated in several childhood epilepsy syndromes[12,13] and

are associated with poorer cognitive performance[14] and behavioural problems[15]. This has

not been explored in CWEOE, but changes in thalamic volume and/or other subcortical

structures may be potential biomarkers to help predict the risk of CI in CWEOE and inform

the need for further formal cognitive evaluations or interventions.

In this study of a cohort of newly diagnosed CWEOE, we hypothesised that:

1) CWEOE and CI would show reductions in the volume of the thalamus but

not other striatal structures compared to CWEOE without CI and healthy

controls;

2) That in CWEOE, the degree of reduction in volumes is positively correlated

with cognitive scores;

2. Materials and Methods

2.1 Patient recruitment

In this nested case-control study, 76 newly diagnosed CWEOE were recruited as part of a

prospective population-based study of early onset epilepsy across NHS Fife and Lothian

4

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23


Page 5 of 24

(NEUROPROFILES) between 2012-2015. Inclusion criteria were: (1) having a physician

confirmed new diagnosis of epilepsy using the 2014 ILAE definition [16], (2) less than five

years old at epilepsy diagnosis, (3) attending hospital in NHS Lothian or Fife for epilepsy

management. Exclusion criteria were: (1) having febrile seizures and or acute symptomatic

seizures only, (2), non-English speaking. Fourteen healthy controls (HC) with normal clinical

neurological examination were recruited from children attending Royal Hospital for Sick

Children, Edinburgh (RHSC) for MRI under general anaesthesia for other clinical reasons

(Table e-1) with no history of epilepsy, febrile seizures or developmental problems.

2.2 MRI and cognitive assessments

All study participants underwent MRI (1.5T, Siemens Espree) under general anaesthesia at

RHSC as part of their routine clinical care including volumetric T1-weighted sequences

(MPRAGE – TR 1870ms, TE 3.76ms, slice thickness 0.9mm, FOV 256mm). They also

underwent cognitive assessment with age-appropriate standardised tools (< 30 months,

Bayley Scales of Infant and Toddler Development III[17] (Bayley III); > 30 months,

Wechsler Preschool and Primary Scales of Intelligence III[18] (WPPSI III)) . Clinical details

were collected using a standardised proforma by direct interview of care-givers and where

possible patients themselves when they attended for MRI and/or cognitive assessment.

All MRI scans were reviewed by two consultant paediatric radiologists (AQ and JJ) blinded

to the clinical history of each participant. Scans were categorised by consensus agreement

into major/minor abnormalities or normal as previously defined [19]. Examples of

abnormalities found are given in Table e-2.

z-scores were generated from the cognitive scale (Bayley III) or full-scale IQ (WPPSI III).

Children who scored 2 standard deviations below population means (z-score < -2) were

5

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23


Page 6 of 24

considered to have CI. CWEOE were divided into those with CI (EP+) and those without CI

(EP-).

2.3 Image analysis

Image processing and analysis of volumetric T1-weighted MRI sequences was performed

using FSL 5.0 (Oxford Centre for Functional MRI of the Brain,

http://www.fmrib.ox.ac.uk/fsl). The FIRST segmentation tool was used to segment left and

right thalami, caudates, putamen and pallidi. Each segmentation was visually assessed for

errors by one of the authors (MY) and individual structure volumes recorded from these

segmentations (Figure 1). Subjects that failed segmentation on one or more structures were

noted and excluded from the analysis (Figure 2). Intracranial volume (ICV) was measured

using the FAST segmentation tool to combine grey matter, white matter and cerebrospinal

fluid volumes. Due to the changes in myelination and water content that occur during early

infancy, it is not possible to apply a single algorithm to reliably and consistently segment

grey/white matter across the entire age range of children in this study. Inspection of

segmentation results showed that children aged less than 10 months did not reliably segment

grey/white matter with the FAST algorithm, therefore analysis involving grey and white

matter volumes were restricted to children aged over 10 months.

2.4 Statistical analysis

Data was analysed using SPSS 22.0 (IBM) for Windows. ANCOVA and Fishers exact test

were used to compare demographic and clinical variables between CWEOE and HC groups

and EP+ and EP- groups as appropriate. Intergroup differences of ICV, grey/white matter

volume and subcortical volumes amongst EP+, EP- and HC were assessed using univariate

6

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

http://www.fmrib.ox.ac.uk/fsl


Page 7 of 24

ANCOVA with age, gender and ICV entered as co-variates and post-hoc pairwise

comparison of all group-pairs. Linear regression adjusting for age, gender and ICV was used

to test for overall correlations of structure volumes with cognitive z-scores. False discovery

rate (FDR) was used to reduce the risk of Type 1 error when considering multiple

comparisons[20] with q<0.05 used to define statistical significance.

2.5 Ethics

All participants provided written consent for study entry. The study was approved by the

South East Scotland Research Ethics Committee (Ref: 14/SS/1010), NHS Lothian Research

and Development, and NHS Fife Research and Development.

3. Results

3.1 Demographics

53/76 CWEOE and 14/14 HC had MRI data potentially suitable for volumetric analysis. The

remaining 23 CWEOE either did not have volumetric T1 weighted MRI sequences performed

or such sequences had significant motion artefact. There were no significant differences

between those subjects with and without suitable MRI data in age, cognitive score or epilepsy

aetiology. All MRI and clinical assessments were performed within 4 months of initial

epilepsy diagnosis.

On visual inspection 3/53 CWEOE and 2/14 HC failed segmentation and were excluded from

further analysis (Figure 2). Demographic and clinical characteristics of the participants are

given in Table 1. Overall CWEOE had significantly lower cognitive scores than HC (p =

7

1

2

3

4

5

6

7

8

9

10

11

1213

14

15

16

17

18

19

20

21

22


Page 8 of 24

0.001); 13/53 CWEOE met criteria for CI compared to 0/14 HC. Significant differences were

found in age between groups (p=0.031), post-hoc analysis showed that EP+ children were

significantly younger than EP- children (q=0.027) but differences in age between EP+ and

HC (q=0.098) or EP- and HC children (q=0.698) did not reach statistical significance after

correction for multiple comparisons. 50 children (12 HC, 38 CWEOE of whom 5 had CI)

were > 10 months age and had reliable measurements of grey and white matter volume

available.

3.2 Clinical features

Compared to EP- children, EP+ children had higher seizure frequency (p = 0.005), were more

likely to be on multiple anti-epileptic medications (p = 0.020), have an abnormal neurological

examination (p < 0.001) or have existing developmental concerns (p < 0.001). There were no

significant differences in seizure focality (p = 0.251) or age at first seizure (p=0.102).

Although a higher proportion of EP+ children had an abnormal scan or a structural/metabolic

cause for their epilepsy, differences in clinical MRI findings (p = 0.061) or aetiology (p =

0.070) did not reach statistical significance.

8

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17


Page 9 of 24

Our cohort was recruited close to the time of their initial diagnosis of epilepsy, often

with some aetiological investigations still ongoing (e.g. in-depth genetic analyses).

Information on their epilepsy syndromes were those available at their last follow-up

and are listed in Table 2.

3.3 Subcortical structure volumes

Comparison of ICV between HC/EP-/EP+ groups showed that there was no significant

difference between any of the groups after adjustment for age and gender (p=0.21). The ICV

in the EP+ group was smaller compared to EP- (Mean difference (95%CI): -1950 (-3870, -

20) cm3, q=0.1065) and HC groups (Mean difference (95%CI): -1900 (-3960, +170),

q=0.1065) cm3 but the difference did not meet statistical significance after FDR correction In

children aged over 10 months, EP+ children (n=5) showed significant reductions in overall

grey matter volume compared to both HC (n=12) (Mean difference (95% CI): -126 (-233, 19)

cm3, q=0.033) and EP- (n=33) (Mean difference (95%CI): -122 (-222, -22) cm3, q=0.033)

groups. There were no significant intergroup differences in white matter (p=0.597) or CSF

volumes (p=0.904). To investigate potential drivers of the grey matter volume difference, the

total volume of subcortical structures was subtracted from total grey matter volume to give an

estimate of cortical and cerebellar grey matter volume alone; this was also significantly

reduced in EP+ children vs HC (Mean difference (95%CI): -123 (-226, -19) cm3, q=0.0315)

and EP- (Mean difference (95%CI): -120 (-217, -23) cm3, q=0.0315) groups.

Total subcortical volume was not significantly different between groups in children over 10

months (p=0.167) or in the cohort as a whole (p=0.114).

Table 3 details adjusted volumes of individual subcortical structures according to each study

group, and intergroup comparison of adjusted volumes. EP+ children had significantly

9

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23


Page 10 of 24

lower left thalamic volumes than HC (Mean difference (95%CI): -1.123 (-1.765, -0.481) cm3,

q = 0.008) and EP- (Mean difference (95%CI): -1.115 (-1.725, -0.505) cm3, q = 0.008) after

correction for age, gender and ICV. The differences in left thalamic volume between

EP+/HC and EP+/EP- remained if grey matter or cortical/cerebellar volumes were used as co-

variates rather than ICV (q<0.05). EP+ had significantly smaller right thalamic (Mean

difference (95%CI): -0.891 (-1.581, -0.202) cm3, q = 0.032) and right pallidum volumes

(Mean difference (95%CI): -0.232 (-0.400, -0.064) cm3, q = 0.032) than HC. In EP+

children, point estimates of the adjusted mean volumes of the right thalamus and right

pallidum were smaller than EP- children but the difference was not statistically significant.

Comparison of individual subcortical structure volumes showed no significant differences

between EP- and HC children for any subcortical structures.

Linear regression showed a significant positive correlation between left thalamic volume

(cm3) (β= 0.78 (95%CI: 0.32 – 1.24), q=0.048, overall model fit R2=0.323) and cognitive z-

score after correction for age, gender and ICV in children with CWEOE. Other structures

were not significantly correlated after FDR correction (Table e-3) and no correlations were

found in HC for any subcortical structures.

4. Discussion

In this novel quantitative MRI study of CWEOE, we found: (1) EP+ children showed

significantly smaller left thalamic volumes compared to EP- or HC children; (2) that the

degree of volume reduction correlated with the cognitive score even after adjusting for ICV

or grey matter volume. These data suggest that the left thalamus could play a role in

10

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 11 of 24

cognition in CWEOE and that quantitative MRI may help to identify CWEOE with cognitive

impairment. Since the lower thalamic volume appears to be independent of the cortical grey

matter volume or ICV, this implies that there may be a specific injury/developmental

abnormality rather than a result of global brain pathology.

The thalamus plays a critical role in integrating and modulating the flow of

information in different cortical areas[21], and has long been recognised to be involved in

many different types of epilepsy. It plays an important role in primary and secondary

generalised seizures[22] with recruitment of cortico-thalamic circuits one of the main drivers

of seizure generalisation[23]. Thalamic volumes have been shown to be reduced compared to

healthy controls in studies of adult temporal lobe epilepsy[24], genetic generalised

epilepsy[25], childhood absence epilepsy[26] and juvenile myoclonic epilepsy[12]. Our

finding that thalamic volumes are only reduced in EP+ but not in EP- children is in some

contrast with these findings, however the majority of these studies do not differentiate

between patients with and without CI or other neurobehavioural comorbidity and are

conducted in older subjects with chronic epilepsy. This may suggest distinct pathological

processes at work affecting the thalamus: one occurring during brain

development/maturation, prior to seizure onset and strongly linked to CI; and one linked to

ongoing seizures/epilepsy that has a weaker link to CI.

There is increasing evidence that subcortical structures, such as the thalamus also

have an important role to play in cognition[27,28] particularly in disease states. Thalamic

atrophy in sickle cell disease[29], Parkinson’s disease[30], multiple sclerosis[31] and

ischaemic stroke[32] have all been associated with reductions in IQ and poorer performance

on cognitive tasks. The rich connectivity of the thalamus makes it an important relay node in

many cortical networks and hence disruption of the cognitive loops through the thalamus has

11

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24


Page 12 of 24

the potential to significantly affect network function and impact cognitive performance.

Modest correlations between thalamic volumes and cognitive ability have been demonstrated

in healthy adult twins[27], healthy children and adolescents[28,33] and aging populations,

nevertheless these are mostly overshadowed by cortical factors. Other studies in ex-preterm

infants have found associations between thalamo-cortical connectivity and cognitive

outcomes[34]. We speculate that the relative contribution of the thalamus to cognition in

healthy populations may be modest because an intact thalamus is not a limiting step for

normal cognitive processing. However, in subjects that have sustained damage to, or have

abnormal development of the thalamus this may have a more significant impact on cognitive

processes.

It remains unclear why the left thalamus appears to be more significantly involved

than the right, but this parallels similar findings in Childhood Absence Epilepsy and children

with focal seizures with impairment of consciousness by Lin et al[15]. They hypothesised

that the left thalamus may have a greater vulnerability to injury during early development

than the right. Stimulation and lesion studies suggest that the left thalamus plays a greater

role in language and verbal memory[35,36] which may explain the greater cognitive impact

of reductions in left thalamic volume as opposed to right in our study.

While the lack of involvement of other basal ganglia structures is consistent with

other studies of newly diagnosed childhood [15] and adult[25] epilepsy, this is in contrast to

studies of adults with established focal[24] or generalised epilepsy[37], which show changes

within basal ganglia structures such as putamen and pallidum. It may be hypothesised that,

similar to the late thalamic atrophy seen in established epilepsy, other basal ganglia structures

are only affected after seizures have been present for some time.

12

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23


Page 13 of 24

4.1 Limitations

Although our study is nested within a population-based study, and one of the largest

cohorts of CWEOE reported to date, the sample size remains comparatively modest,

particularly the number of children with CI. Despite this, our finding of smaller thalamus and

an association with lowered cognition is consistent with other studies of childhood epilepsy,

albeit of different types of epilepsy[15]. The modest number of CWEOE and HC with CI and

overall sample size precluded any logistic regression analysis for factors associated with CI.

Epilepsy aetiology in our cohort was heterogeneous, with no single syndrome making

up more than 10% of the cohort. While this means that no syndrome-specific conclusions can

be made, it suggests that perhaps common underlying mechanisms may exist for CI in

CWEOE, regardless of aetiology. It should also be noted that a higher proportion of EP+

children were on more than one anti-epileptic medication and it is possible that this

polytherapy may account for a proportion of the cognitive deficit. As this study was

conducted relatively early in the course of their epilepsy (within 4 months of diagnosis) it

was not possible to judge whether any children had responded to medication yet or what

proportion were going to be medically refractory, though other markers associated with

refractoriness, such as multiple seizure types, seizure frequency and symptomatic aetiology

were addressed in our analysis.

The cross-sectional nature of this study means that it is not possible to determine if

the smaller thalamic volumes are a result of injury, pre-existing hypoplasia/malformation or

growth failure. All study participants were imaged within four months of their first reported

seizure, so it is possible that these abnormalities preceded the emergence of clinical epilepsy

and are not due to direct injury from recurrent or prolonged seizures. However, growth arrest

13

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23


Page 14 of 24

from sub-clinical or clinical epileptiform activity would remain a possibility. Longitudinal

studies are needed to clarify the pathophysiological process.

There is evidence that the cognitive effect from thalamic lesions results primarily

from damage to anterior and mesio-dorsal thalamic nuclei; however the imaging used in this

study was not of sufficient resolution to distinguish different thalamic sub-regions. We have

been able to demonstrate an effect of the left thalamus as a whole, but it may be that

measurement of specific thalamic nuclei with higher resolution MRI at higher field strengths

and/or connectivity-based parcellation of thalamic sub-regions [38] would help to refine our

understanding of the underlying neuropathology.

5. Conclusion

Reduced volume of the left thalamus after adjusting for age, gender and intracranial volume

or grey matter volume is associated with cognitive impairment in CWEOE. If this is

replicated in other cohorts of CWEOE, then quantitative volumetric measurement of the left

thalamus may be useful on its own or in combination with other factors such as MRI/EEG

connectomics as a biomarker for cognitive impairment.

6. Acknowledgements

We are grateful to the children and their families who participated in this project,

without whom this study would not have been possible. We would like to thank the medical

and nursing staff who were instrumental in helping to recruit children for this study and the

MRI radiographers at RHSC for their patients and dedication in assisting with data collection.

14

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21


Page 15 of 24

Funding for NEUROPROFILES was generously given by the Muir Maxwell Epilepsy

Trust, with additional funding for control recruitment from the British Paediatric Neurology

Association and the RS MacDonald Trust. MY was funded by an NHS Research Scotland

Clinical Lectureship. JS was funded by NHS Research Scotland.

7. Disclosures

No authors have any financial conflicts of interest with the conduct of or results of this

study.

References

[1] Davies S, Heyman I, Goodman R. A population survey of mental health problems in

children with epilepsy. Dev Med Child Neurol 2003;45:292–5.

[2] Selassie A, Wilson D, Martz G. Epilepsy beyond seizure: A population-based study of

comorbidities. Epilepsy Res 2014;108:305–15. doi:10.1016/j.eplepsyres.2013.12.002.

[3] Reilly C, Atkinson P, Das K, Chin R. Neurobehavioral Comorbidities in Children With

Active Epilepsy: A Population-Based Study. Pediatrics 2014;133:1586–03.

doi:10.1542/peds.2013-3787.

[4] Berg AT, Zelko F a., Levy SR, Testa FM. Age at onset of epilepsy,

pharmacoresistance, and cognitive outcomes: A prospective cohort study. Neurology

2012;79:1384–91. doi:10.1212/WNL.0b013e31826c1b55.

[5] Rantanen K, Eriksson K, Nieminen P. Cognitive impairment in preschool children

with epilepsy. Epilepsia 2011;52:1499–505. doi:10.1111/j.1528-1167.2011.03092.x.

15

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21


Page 16 of 24

[6] Berg AT, Plioplys S. Epilepsy and autism: Is there a special relationship? Epilepsy

Behav 2012;23:193–8. doi:10.1016/j.yebeh.2012.01.015.

[7] Baca CB, Vickrey BG, Caplan R, Vassar SD, Berg AT. Psychiatric and medical

comorbidity and quality of life outcomes in childhood-onset epilepsy. Pediatrics

2011;128:e1532-43. doi:10.1542/peds.2011-0245.

[8] Camfield CS, Camfield PR. Long-term social outcomes for children with epilepsy.

Epilepsia 2007;48 Suppl 9:3–5. doi:10.1111/j.1528-1167.2007.01390.x.

[9] Brooks-Kayal AR, Bath KG, Berg AT, Galanopoulou AS, Holmes GL, Jensen FE, et

al. Issues related to symptomatic and disease-modifying treatments affecting cognitive

and neuropsychiatric comorbidities of epilepsy. Epilepsia 2013;54 Suppl 4:44–60.

doi:10.1111/epi.12298.

[10] Christensen J, Vestergaard M, Pedersen MG, Pedersen CB, Olsen J, Sidenius P.

Incidence and prevalence of epilepsy in Denmark. Epilepsy Res 2007;76:60–5.

doi:10.1016/j.eplepsyres.2007.06.012.

[11] Deary IJ, Penke L, Johnson W. The neuroscience of human intelligence differences.

Nat Rev Neurosci 2010;11:201–11. doi:10.1038/nrn2793.

[12] Pulsipher DT, Seidenberg M, Guidotti L, Tuchscherer VN, Morton J, Sheth RD, et al.

Thalamofrontal circuitry and executive dysfunction in recent-onset juvenile myoclonic

epilepsy. Epilepsia 2009;50:1210–9. doi:10.1111/j.1528-1167.2008.01952.x.

[13] Cormack F, Gadian DG, Vargha-Khadem F, Cross JH, Connelly A, Baldeweg T.

Extra-hippocampal grey matter density abnormalities in paediatric mesial temporal

sclerosis. Neuroimage 2005;27:635–43. doi:10.1016/j.neuroimage.2005.05.023.

16

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 17 of 24

[14] Yoong M. Quantifying the deficit — imaging neurobehavioural impairment in

childhood epilepsy. Quant Imaging Med Surg 2015;5:225–37. doi:10.3978/j.issn.2223-

4292.2015.01.06.

[15] Lin J, Siddarth P, Riley J, Gurbani S. Neurobehavioral comorbidities of pediatric

epilepsies are linked to thalamic structural abnormalities. Epilepsia 2013;54:2116–24.

doi:10.1111/epi.12428.

[16] Fisher RS, Acevedo C, Arzimanoglou A, Bogacz A, Cross JH, Elger CE, et al. ILAE

official report: a practical clinical definition of epilepsy. Epilepsia 2014;55:475–82.

doi:10.1111/epi.12550.

[17] Bayley N. Bayley Scales of Intant and Toddler Development, Third Edition. 3rd ed.

Pearson; 2005.

[18] Wechsler D. Wechsler Preschool and Primary Scale of Intelligence - 3rd Edition. 3rd

ed. Pearson; 2002.

[19] Yoong M, Madari R, Martinos M, Clark C, Chong K, Neville B, et al. The role of

magnetic resonance imaging in the follow-up of children with convulsive status

epilepticus. Dev Med Child Neurol 2012;54:328–33. doi:10.1111/j.1469-

8749.2011.04215.x.

[20] Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and

Powerful Approach to Multiple Controlling the False Discovery Rate: a Practical and

Powerful Approach to Multiple Testing. J R Stat Soc 1995;57:289–300.

doi:10.2307/2346101.

[21] Sherman SM, Guillery RW. The role of the thalamus in the flow of information to the

17

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 18 of 24

cortex. Philos Trans R Soc Lond B Biol Sci 2002;357:1695–708.

doi:10.1098/rstb.2002.1161.

[22] Blumenfeld H. Cellular and network mechanisms of spike-wave seizures. Epilepsia

2005;46:21–33. doi:10.1111/j.1528-1167.2005.00311.x.

[23] Zhang CH, Sha Z, Mundahl J, Liu S, Lu Y, Henry TR, et al. Thalamocortical

relationship in epileptic patients with generalized spike and wave discharges - A

multimodal neuroimaging study. NeuroImage Clin 2015;9:117–27.

doi:10.1016/j.nicl.2015.07.014.

[24] Seidenberg M, Hermann B, Pulsipher D, Morton J, Parrish J, Geary E, et al. Thalamic

atrophy and cognition in unilateral temporal lobe epilepsy. J Int Neuropsychol Soc

2008;14:384–93. doi:10.1017/S1355617708080399.

[25] Perani S, Tierney TM, Centeno M, Shamshiri EA, Yaakub SN, O’Muircheartaigh J, et

al. Thalamic volume reduction in drug-naive patients with new-onset genetic

generalized epilepsy. Epilepsia 2017:1–9. doi:10.1111/epi.13955.

[26] Chan CH, Briellmann RS, Pell GS, Scheffer IE, Abbott DF, Jackson GD. Thalamic

atrophy in childhood absence epilepsy. Epilepsia 2006;47:399–405.

doi:10.1111/j.1528-1167.2006.00435.x.

[27] Bohlken MM, Brouwer RM, Mandl RCW, van Haren NEM, Brans RGH, van Baal

GCM, et al. Genes contributing to subcortical volumes and intellectual ability

implicate the thalamus. Hum Brain Mapp 2014;35:2632–42. doi:10.1002/hbm.22356.

[28] Xie Y, Chen YA, De Bellis MD. The relationship of age, gender and IQ with the

brainstem and thalamus in healthy children and adolescents: a magnetic resonance

18

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 19 of 24

imaging volumetric study. J Child Neurol 2012;27:325–31.

doi:10.1177/0883073811419260.

[29] Mackin RS, Insel P, Truran D, Vichinsky EP, Neumayr LD, Armstrong FD, et al.

Neuroimaging abnormalities in adults with sickle cell anemia: Associations with

cognition. Neurology 2014;82:835–41. doi:10.1212/WNL0000000000000188.

[30] Mak E, Bergsland N, Dwyer MG, Zivadinov R, Kandiah N. Subcortical atrophy is

associated with cognitive impairment in mild parkinson disease: A combined

investigation of volumetric changes, cortical thickness, and vertex-based shape

analysis. Am J Neuroradiol 2014;35:2257–64. doi:10.3174/ajnr.A4055.

[31] Batista S, Zivadinov R, Hoogs M, Bergsland N, Heininen-Brown M, Dwyer MG, et al.

Basal ganglia, thalamus and neocortical atrophy predicting slowed cognitive

processing in multiple sclerosis. J Neurol 2012;259:139–46. doi:10.1007/s00415-011-

6147-1.

[32] Mori E. Impact of subcortical ischemic lesions on behavior and cognition. Ann N Y

Acad Sci 2002;977:141–8. doi:10.1111/j.1749-6632.2002.tb04809.x.

[33] Frangou S, Chitins X, Williams SCR. Mapping IQ and gray matter density in healthy

young people. Neuroimage 2004;23:800–5. doi:10.1016/j.neuroimage.2004.05.027.

[34] Ball G, Pazderova L, Chew A, Tusor N, Merchant N, Arichi T, et al. Thalamocortical

connectivity predicts cognition in children born preterm. Cereb Cortex 2015;25:4310–

8. doi:10.1093/cercor/bhu331.

[35] Wester K, Irvine DRF, Hugdahl K. Auditory laterality and attentional deficits after

thalamic haemorrhage. J Neurol 2001;248:676–83. doi:10.1007/s004150170113.

19

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22


Page 20 of 24

[36] Ojemann GA. Asymmetric Function of the Thalamus in Man. Ann N Y Acad Sci

1977;299:380–96. doi:10.1111/j.1749-6632.1977.tb41923.x.

[37] Keller SS, Ahrens T, Mohammadi S, Möddel G, Kugel H, Bernd Ringelstein E, et al.

Microstructural and volumetric abnormalities of the putamen in juvenile myoclonic

epilepsy. Epilepsia 2011;52:1715–24. doi:10.1111/j.1528-1167.2011.03117.x.

[38] Serra L, Cercignani M, Carlesimo GA, Fadda L, Tini N, Giulietti G, et al.

Connectivity-Based Parcellation of the Thalamus Explains Specific Cognitive and

Behavioural Symptoms in Patients with Bilateral Thalamic Infarct. PLoS One

2013;8:4–12. doi:10.1371/journal.pone.0064578.

20

1

2

3

4

5

6

7

8

9

10

11

12


Page 21 of 24

HC (n=12) EP+ (n=12) EP- (n=38) All CWEOE (n=50)

p-value

Age in years (range, s.d.) 2.34 (0.66 – 5.03, 1.51)

1.19 (0.09 – 4.5, 1.27)

2.54 (0.02 – 5.05, 1.57)

2.22 (0.2 – 5.05, 1.60)

p=0.031

Male:Female ratio 7:5 5:7 26:12 31:19 p=0.306Age at first seizure (months) (range, s.d.)

- 14.3 (0-48, 15.3)

23.3 (0-59, 16.7)

21 (0-59, 16.7)

p=0.102

Seizure frequency

> daily 9 (75%) 8 (21%) 17 (34%)Weekly 3 (25%) 13 (34%) 16 (32%) p=0.005Monthly 0* 8 (21%) 8 (16%)< monthly 0* 9 (24%) 9 (18%)

Antiepileptic drugs

None 0 5 (13%) 5 (10%)

Monotherapy 7 (58%) 30 (79%) 37 (74%) p=0.020Polytherapy 5 (42%) 3 (8%) 8 (16%)

Focal seizures - 3 (25%) 19 (50%) 22 (44%) p=0.251Generalised seizures - 9 (75%) 16 (42%) 25 (50%)Generalised and Focal - 0 3 (8%) 3 (6%)Abnormal neurological examination

0 8 (67%) 4 (11%) 12 (24%) p<0.001

MRI findings

Major abnormality

0 5 (42%) 5 (13%) 10 (20%)

Minor abnormality

1 (8%) 0 9 (24%) 9 (18%) p=0.061

Normal 11 (92%) 7 (58%) 24 (63%) 31 (62%)Previous developmental concerns at initial

0 11 (92%) 10 (26%) 21 (42%) p<0.001

21

1


Page 22 of 24

assessmentCognitive z-score (s.d.) 0.26 (0.51) -3.31 (0.52) -0.43 (0.77) -1.12 (1.43) p<0.001Epilepsy aetiology classificationStructural/metabolic - 4 (33%) 5 (13%) 9 (18%)Identified genetic - 2 (17%) 1 (3%) 3 (6%) p=0.070Suspected genetic - 3 (25%) 11 (29%) 14 (28%)Unknown - 3 (25%) 21 (55%) 24 (48%)

p-value compares HC/EP+/EP- using ANOVA or Fishers Exact Test as appropriate for demographic variables, clinical variables are compared between EP+/EP- only.

Table 1: Demographics and clinical features of CWEOE and HC.

22

1

2


Page 23 of 24

Epilepsy Syndrome EP- (n=38) EP+ (n=12) All CWEOE (n=50)

Childhood absence epilepsy 2 0 2

GEFS+ 2 0 2

Genetic generalised epilepsy 0 1 1

West syndrome 1 6 7

Benign focal infantile epilepsy 4 0 4

Temporal lobe epilepsy with hippocampal sclerosis 1 0 1

Focal epilepsy with other focal structural brain lesion 4 1 5

Focal epilepsy of unknown cause 11 1 12

Generalised epilepsy of unknown cause 13 3 16

Table 2: Epilepsy syndrome diagnosis for EP- and EP+ children

23

1


Page 24 of 24

Mean structure volume (mm3) (Standard Error)

Mean volume

difference EP+ vs HC (95%CI)

(mm3)

Mean volume

difference EP- vs HC (95%CI)

(mm3)

Mean volume

difference EP+ vs EP-

(95%CI) (mm3)

HC EP+ EP-

Left thalamus

6418 (177)*

q = 0.008

5295 (265)

6410 (119)*

q = 0.008

-1123 (-1765 , -481)

-8 (-438, 422)

-1115 (-1725 - -505)

Right thalamus

6315 (204)*

q = 0.032

5424 (280)

6159 (131)

q = 0.096

-891 (-1580, -201)

-156 (-650 , 338)

-735 (-1390, -81)

Left caudate

3166 (112)

q = 0.335

2987 (136)

3015 (71)q = 0.86

-179 (-534, 177)

-150 (-420, 119)

-28 (-342, 287)

Right caudate

3487 (148)

q = 0.07

2967 (186)

3162 (87)q = 0.568

-520 (-995, -44)

-325 (-672, 22)

-194 (-613, 225)

Left putamen

3886 (161)

q = 0.335

3634 (198)

3521 (98)q = 0.757

-252 (-770, 267)

-365 (-744, 14)

113 (-338, 564)

Right putamen

3806 (165)

q = 0.066

3250 (202)

3510 (101)

q = 0.528

-555 (-1086, -25)

-295 (-684, 93)

-260 (-722, 202)

Left pallidum

1316 (55)q = 0.335

1213 (68)

1179 (33)q = 0.757

-103 (-280, 75)

-136 (-267, -7)

34 (-121, 189)

Right pallidum

1350 (52)*

q = 0.032

1118 (64)

1275 (31)q = 0.096

-232 (-400, -64)

-75 (-198, 48)

-157 (-304, -11)

Volumes adjusted for age, gender and intracranial volumeq-values quoted are statistical comparison to EP+ after FDR correction

* significantly different from EP+ after FDR correction (q < 0.05)

Table 3: Subcortical volumes comparing healthy controls (HC), children with epilepsy without cognitive impairment (EP-) and children with epilepsy with cognitive impairment (EP+)

24

1

2

Date post:	02-May-2019
Category:	Documents
Upload:	truongnhan
View:	215 times
Download:	0 times

Summary - research.ed.ac.uk file · Web viewCognitive impairment in early onset epilepsy is...

Documents