A CaseControl 3'0 Tesla MRI Morphometric Analysis of Autism with and without AttentionDeficitHyperac

1
A Case-Control 3.0 Tesla MRI Morphometric
Analysis of Autism with and without
Attention-Deficit/Hyperactivity DisorderJed T.
Elison, BA, Erin D. Bigler, PhD, Judith Miller,
PhD, Jeffrey Lu, MD, William M. McMahon, MD,
Janet E. Lainhart, MD
Abstract Background Signs of ADHD frequently
occur in children with autism. Neuroimaging
studies have not yet identified brain correlates
of ADHD in autism or considered the effects of
ADHD-related morphometry on autism-control
studies. Objectives This pilot study compares
regional white and gray matter densities in the
brains of children with autism who have
co-occurring ADHD (autismADHD), autism without
ADHD (autism-ADHD), and typically developing
controls (TD). Methods Three groups of males,
6.5 to 13.5 years of age were matched on age,
performance IQ, and handedness 1) autism - ADHD
(n 10), 2) autism ADHD (n 11), and 3) TD (n
12). Autism was diagnosed using the ADI-R and
ADOS-G. Signs of ADHD were measured using the
Conners ADHD/DSM-IV Scale (parent version).
Imaging data were collected on a 3Tesla Siemens
Trio scanner and analyzed using statistical
parametric mapping and voxel-based morphometry
(statistical threshold plt.001). Results The
autismADHD group had decreased gray matter
density in the right inferior cerebellum compared
to autism-ADHD and TD. Conclusions The authors
reject the null hypothesis that there are no
white or gray matter density differences between
individuals with autism, with versus without
ADHD, and identify the right-inferior cerebellum
as a region of interest.

• Results (Continued)
• The autismADHD group had less gray matter
  density in the right-inferior cerebellum when
  compared with the autism-ADHD group and the
  typically developing control group.
• The autism-ADHD group had more gray matter
  density in the right-inferior cerebellum when
  compared with the autismADHD group and the
  typically developing control group.
• No significant differences were found in global
  or regional white matter concentration.
• The image shown below highlights the region of
  gray matter differences in the cerebellum (6,
  -62, -38), PFDR-corr 0.068, F(2,29) 21.67.

• Specific Study Aim
• The aim of this pilot study was to determine if
  attention-deficit/hyperactivity disorder (ADHD)
  symptoms are related to differences in total and
  regional white and gray matter density in autism.
  
• Background
• Neuroimaging findings in case-control studies of
  ADHD and separate studies of idiopathic autism
  are listed as follows
• ADHD
• An estimated 3.2-8.1 reduction in total brain
  volume in children and adults with
  ADHD3,6,8,9,13.
• Reduced prefrontal cortex volume8,9,13.
• Reduced cerebellum volumes3,6,8,9,12.
• Voxel-Based Morphometry studies in ADHD.
• Carmona et al. (2005)4 recently found a 5.2
  decrease in global gray matter concentration in
  25 children with ADHD between the ages of 6 and
  16they also reported reduced gray matter density
  in the prefrontal cortex and the cerebellum
  (bilateral posterior).
• AUTISM
• Increased total brain volume in children between
  the ages of 2 and 127,10,11,16.
• Increased total gray matter in sample of children
  between the ages of 7 and 15, no cerebral white
  matter enlargement14.
• Increased prefrontal cortex volume5.
• Increased white matter volume in the prefrontal
  cortices11.
• Increased gray matter volume in the
  cerebellum14.
• Voxel-Based Morphometry studies of autism.
• Increased gray matter density in the
  cerebellum1,15.
• One study15 found significant increases in gray
  matter density in a sample of children with
  autism between 8 and 18 years old in the inferior
  cerebellum (15, -62, -40).

TABLE 1
Characteristic sd Autism - ADHD
Autism ADHD Control-No
ADHD range
n 10 n
11 n 12 Mean Age
(year.month) 10.1026.9-13.4
10.81.58.1-12.5
10.92.46.7-13.7 Mean PIQ
97.320.271-130
100.514.879-123 110.913.790-136
Mean VIQ
85.222.551-124
99.819.870-136
109.213.283-127 ADI B Total
18.34.911-29
20.55.911-27 NA ADI
Communication 16.24.211-23
16.63.911-22
NA ADI D Total
6.42.13-11
7.72.24-12 NA ADOS
Communication 5.61.93-9
5.01.53-7
.75.750-2 ADOS Social
10.92.17-14
9.92.07-13
1.21.50-4 ADOS Social
16.43.610-23
14.93.112-20
1.92.00-6 Communication Connors ADHD Index
56.25.747-64
74.94.770-82
46.56.740-60 Handedness Index
64.450.9-67 100 54.673.3-100
100 69.143.6-60 100
Left-handers 1
Left-handers 2
Left-handers 1 no significant differences in
age, PIQ, or handedness between the three groups.
Results Differences in gray matter concentration
as shown by the statistical parametric map are
listed below.

• Conclusions
• Our voxel-based morphometric analysis suggests
  that the cerebellum may be involved with the ADHD
  symptoms observed in autism. The finding of less
  gray matter density in the right cerebellum may
  be part of the neurobiological explanation of why
  some children with autism have symptoms of ADHD
  and some do not.
• Heterogeneity of autism based on ADHD
  and other psychiatric comorbidity may be a reason
  for the inconsistent results of past neuroimaging
  studies of the cerebellum in autism.
• More research is needed to elucidate the nature
  of brain pathology associated with psychiatric
  comorbidity in autism. Finally, our findings
  need to be confirmed by replication.

• Methods
• Participants
  Measures
• Autism ADHD, n 10 1. Autism
  Diagnostic Interview-Revised
• Autism ADHD, n 11 2. Autism
  Diagnostic Observation Schedule-Generic
• Normal Controls, n 12 3. Connors
  ADHD/DSM-IV Scale (parent Version)
• 
  4. Differential Ability Scales-School
  Age
• 4 children had the WISC-III, 1 had the
  DAS-Preschool
• 5. Edinburgh Handedness Inventory
• The children were scanned on a 3-T Siemens Trio
  scanner using a sagittal T1-weighted 3D MPRAGE
  sequence with the following parameters TR, 1800
  ms TE 2.93 ms flip angle 12 and voxel size,
  1.0 x 1.0 x 1.0 mm.
• Voxel-based morphometric analyses were conducted
  by spatially normalizing the structural images to
  the same stereotactic space, parsing out the gray
  and white matter, and smoothing the images. A
  voxel-wise statistical analysis then yielded a
  statistical parametric map that indicates
  significant differences in gray or white matter
  concentration between groups (Ashburner
  Friston, 2000).
• See TABLE 1 for descriptive statistics of the
  sample.

References 1. Abell et al., NeuroReport 10,
1647-1651 (1999). 2. Ashburner
and Friston, NeuroImage 11, (2000). 3. Berquin
et al., Neurology 50(4), 1087-1093 (1998).
4. Carmona et al., Neuroscience Letters 389,
88-93 (2005). 5. Carper et al., NeuroImage
16, 1038-1051 (2002). 6.
Castellanos et al., JAMA 288(14), 1740-1748
(2002).
7.
Courchesne et al., Neurology 57, 245-254 (2001).

8.
Durston et al., J. Am. Acad Child Adolesc. Psy.
43(3), 332-340.
9. Filipek et al., Neurology 48(3), 589-601
(1998).

10. Hazlett et
al., Biol. Psychiatry 59, 1-6 (2006).

11.
Herbert et al., Brain 126, 1182-1192 (2003).


12. Mostofsky et al., J. Child Neurology 3,
434-439 (1998). 13. Mostofsky et al., Biol
Psychiatry 52, 785-794 (2002). 14. Palmen et
al., Psychological Med. 35, 561-570, (2005).
15. Salmond et al., Eur. J. of Neuroscience 22,
764-772 (2005). 16. Sparks et al.,
Neurology 59, 184-192 (2002).
Acknowledgements This research was funded by
the NICHD/NICDC Collaborative Programs of
Excellence in Autism grant number U19 HD035476.
This research was made possible by those
children and families who willingly participated
and supported the study. We also appreciate
Tracy Abildskovs time and effort and the
constant support that the staff at the Utah
Autism Research Program contributed to the
current report.