- Email: [email protected]
Meta-analysis of magnetic resonance imaging studies in chromosome 22q11.2 deletion syndrome (velocardiofacial syndrome)
Schizophrenia Research 115 (2009) 173–181
Contents lists available at ScienceDirect
Schizophrenia Research 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 / s c h r e s
Meta-analysis of magnetic resonance imaging studies in chromosome 22q11.2 deletion syndrome (velocardiofacial syndrome) Giles M. Tan a,1, Danilo Arnone b,d,⁎,1, Andrew M. McIntosh c, Klaus P. Ebmeier c,d a
Section of Brain Maturation, PO Box 50, Division of Psychological Medicine & Psychiatry, Institute of Psychiatry at the Maudsley, King's College London, De Crespigny Park, London, SE5 8AF, United Kingdom Neuroscience and Psychiatry Unit, Stopford Building, Manchester, United Kingdom c University of Edinburgh, Royal Edinburgh Hospital, Edinburgh, United Kingdom d University Department of Psychiatry, Oxford, United Kingdom b
a r t i c l e
i n f o
Article history: Received 24 May 2009 Received in revised form 12 August 2009 Accepted 7 September 2009 Available online 9 October 2009 Keywords: Chromosome 22q11.2 deletion Velocardiofacial syndrome Magnetic resonance imaging Neuroimaging
a b s t r a c t Objectives: 22q11.2 deletion syndrome (22q11.2DS), also known as velocardiofacial syndrome (VCFS) or DiGeorge Syndrome, is a genetic disorder due to a micro deletion on chromosome 22q11.2. VCFS is associated with abnormalities in brain structure and with an increased risk of psychiatric disorders, particularly schizophrenia. The aim of this review was to statistically summarize the structural imaging literature on VCFS which due to the relatively rarity of the disorder tends to consider small sample sizes. Method: A systematic review and meta-analysis of region of interest (ROI) studies comparing VCFS with healthy controls was carried out. Significant heterogeneity was explored using meta-regression. Results: Subjects with VCFS were characterised by global brain volumetric reduction including several cortical regions, cerebellum and hippocampus. The area of the corpus callosum was increased. Conclusions: Many regions extensively studied in schizophrenia were not covered in the existing VCFS literature. However, the studies considered support volumetric abnormalities which may help explain why VCFS is associated with a greatly increased risk of psychosis and other psychiatric disorders. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Chromosome 22q11.2 deletion syndrome (22q11.2DS), affects approximately 1 in 5000 live births (Tézenas Du Montcel et al., 1996) and refers to a group of previously distinct genetically determined disorders (velocardiofacial syndrome, Shprintzen syndrome, DiGeorge syndrome, conotruncal anomaly face syndrome, Cayler cardiofacial syndrome, and CATCH 22 syndrome) caused by a microdeletion
⁎ Corresponding author. Neuroscience and Psychiatry Unit, University of Manchester, G810 Stopford Building, Oxford Road, Manchester, M13 9PT, United Kingdom. E-mail addresses: [email protected] (G.M. Tan), [email protected] (D. Arnone). 1 Authors contributed equally to the work. 0920-9964/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2009.09.010
on the long arm of chromosome 22 (Scambler et al., 1992). VCFS is characterised by a distinct phenotype including a typical facial appearance, cleft palate, velopharyngeal insufficiency, cardiac anomalies (McDonald-McGinn et al., 1999; Swillen et al., 1997), learning disabilities and by an increased risk for psychotic disorders (Henry et al., 2002; Gothelf et al., 2004a,b; Baker and Skuse, 2005; Papolos et al., 1996; Niklasson et al., 2002) particularly schizophrenia (Murphy et al., 1999). More specifically, psychotic symptoms have been diagnosed in up to 30% of adolescents and adults with 22q11.2DS (Murphy et al., 1999). Moreover, the deletion has been found in 1–6% of people with schizophrenia (Murphy et al., 1998; Yan et al., 1998). This association has led to considerable research to morphologically characterise 22q11.2DS brain abnormalities and potential similarities with schizophrenia. Magnetic
174
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
resonance imaging (MRI) has allowed detailed in vivo examination of brain structures in people with 22q11.2DS but the evidence is inconclusive (Eliez et al., 2000; Kates et al., 2001; Simon et al., 2005; Eliez et al., 2002; Kates et al., 2006; Campbell et al., 2006; Gothelf et al., 2007; DeBoer et al., 2007; Shashi et al., 2004). This is in part due to the relative rarity of the disorder and the tendency for studies to be small and potentially underpowered. Nevertheless, even taking this into account, the literature appears heterogeneous and published significant effect sizes differ in both direction and magnitude. The aim of this meta-analysis was to apply quantitative methods to characterise morphometric differences in 22q11.2DS in comparison with healthy controls by 1) identifying region of interest (ROI) studies which offered measurements of brain regions in 22q11.2DS and 2) synthesizing findings using random effects meta-analysis. We also assessed heterogeneity to clarify the role of a number of variables including year of publication, age, sex, IQ, scanner strength and slice thickness. 2. Materials and methods
Medicine, Oxford, UK. Standardised mean differences were calculated using Cohen's d statistic: 0
Cohen s d =
X 1 −X 2 SDp
where X 1 and X 2 are the mean volumes from the first and second groups respectively and SDp is the pooled standard deviation estimated from both groups: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðn1 −1ÞSD21 + ðn2 −1ÞSD22 SDp = ðn1 + n2 −2Þ where ni and SDi are the mean and standard deviation of the ‘ith’ group. Standardised effect sizes were then combined using the inverse variance method. The standard deviation of Cohen's d is estimated as: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N d2 SDðdÞ + n1 n2 2ðN−2Þ
2.1. Search strategy and inclusion/exclusion criteria A comprehensive search from a range of electronic databases, including Medline, EMBASE, PsycINFO, and PubMED was conducted up to March 2008 that reported structural MRI data in people with 22q11.2DS and unaffected controls. Search terms used to identify the studies included ‘Velocardiofacial syndrome’, ‘DiGeorge syndrome’, ‘22q11.2 deletion syndrome’ and related terms combined using the AND operator with ‘Magnetic Resonance Imaging’. The search was also supplemented by a manual and bibliographic crossreferencing. We placed no limit on year of publication. Studies were included if they were published as a peer reviewed article and presented original data which compared a sample of genetically diagnosed 22q11.2DS subjects with a group of healthy controls. Only studies reporting structural MRI data where means and standard deviations were available (or could be extracted) for each group were included. Researchers were contacted if this information was not immediately available in an otherwise acceptable report. Studies were excluded if data were included in more recent larger studies. Data were extracted from all included studies by one author (GMT) and checked by another author (DA). Characteristics including age, sex and IQ of the 22q11.2DS and control groups which may have confounded any observed difference were also recorded as was the year of publication, scanner strength and slice thickness. Data were utilised when volume or area results were available for a region of interest (ROI) from three or more studies. Whereas volumes were acquired from more than one slice, callosal area was measured from one mid-sagittal slice. Volume and area data were never combined in the same analysis. 2.2. Statistical analysis Statistical analysis was conducted using STATA 9.0 (Stata Corp, College Station, Texas) supplemented by ‘Metan’ software downloadable from the Centre for Statistics in
where N is the total sample size for the study, d is Cohen's d and n1 and n2 are as defined above. Random effects analyses (DerSimonian and Laird, 1986) were used throughout to weight each study. The presence of heterogeneity was tested using the Q-test and its magnitude estimated using I2 and can be interpreted as the proportion of variance in effect size due to heterogeneity (Higgins et al., 2003). When the Q-test was significant, a Galbraith plot was used to identify those studies contributing the greatest amount to that heterogeneity to investigate potential causes. Publication bias which describes the excess of low-precision studies providing effect sizes of magnitude greater than the average was tested with the Egger's test (Egger et al., 1997). Significance level was set at P b 0.05. To further investigate causes for heterogeneity, metaregression analyses were performed for the following variables: mean age, sex, IQ, scanner strength, slice thickness and year of publication. The STATA program “metareg.ado” was used throughout and the REML (restricted maximum likelihood) method used to estimate the model parameters. 3. Results 3.1. Systematic search The literature search identified 156 publications of which 48 were retrieved in full text format. Fig. 1 summarises the study flow and reasons for exclusion. Twenty-two studies met inclusion criteria and were included in the final analysis which provided information on 35 regions of interest (Table 1). Images were manually traced reflecting similar anatomical borders and used different software programs to aid image processing. In some cases a semiautomated stereotactic based parcellation method was used (e.g. Eliez et al., 2000). All the studies used the same in-situ hybridization technique but did not report on the extent of the microdeletion. Three of these studies were included after
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
175
Fig. 1. Study flow and reasons for exclusion.
authors were contacted to supplement information published (DeBoer et al., 2007; Bearden et al., 2007; Kates et al., 2004). Year of publication ranged from 2000 to 2007. Table 1 lists the key demographic and descriptive data for each of the included studies. Three of the studies included both unrelated age matched individuals and siblings as controls while one study included only siblings as the control subjects. Across the 22 studies included in the meta-analysis, there were a total of 626 22q11.2DS subjects and 576 controls. The mean age and age range were 13.3 (9.9–31) for 22q11.2DS and 13.5 (10–36) for control subjects. The mean and median number of subjects across the studies were 28.5 and 21 respectively for the 22q11.2DS subjects and 26.2 and 20.5 for the control subjects. Except for one study (Eliez et al., 2002), all studies reported sex of subjects with the mean percentage of male 22q11.2DS and control subjects as 51.5% and 55.7% respectively. All the studies used a 1.5 Tesla magnet and diagnosis of 22q11.2DS was confirmed by fluorescent in-situ hybridisation. All the reports included in the analysis measured brain regions as volumes except the corpus callosum which was available as area. Sixteen studies reported IQs in cases and controls but they were matched only in a minority of studies.
3.2. Meta-analysis Random effect analyses were conducted for the 36 regions of interest (ROI) and results are presented in Tables 2 and 3 and Fig. 2: 1) Total brain The findings confirm global brain volumetric reduction including total brain (N=7, estimate: −1.02, 95% CI: −1.26, −0.78), combined left and right hemisphere (N=4, estimate: −0.73; 95% CI: −1.08, −0.39), left hemisphere (N=4, estimate: −0.68; 95% CI:−1.02, −0.34), right hemisphere (N=4, estimate: −0.68; 95% CI: −1.02, −0.34), left and right hemisphere white matter (N=4, estimate: −1.09; 95% CI: −1.56, −0.62), left and right hemisphere gray matter (N=5, estimate: −0.52; 95% CI: −0.83, −0.22), whole brain gray matter (N = 5, estimate: −0.92; 95% CI: −1.77, −0.68), and whole brain white matter (N = 5, estimate: −0.85, 95% CI: −1.20, −0.50). Possible duplicate publications (Eliez et al., 2001b; Kates et al., 2005) did not affect the finding of global brain volumetric reduction when these studies were excluded (N= 5, estimate: −1.01, 95% CI: −1.28, −0.75).
176
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Table 1 Summary of the studies included in the meta-analysis. VCFS cases
Normal controls
Diagnostic
Slice
Study (year)
n
Age % (years) male
IQ
n
Age % (years) male
IQ
Type
Diagnosis criteria
Scanner Thickness Measure
DeBoer et al. (2007) Glaser et al. (2007) Gothelf et al. (2007) Bearden et al. (2007) Machado et al. (2007) Campbell et al. (2006) Debbane et al. (2006) Kates et al. (2006)
36 42 29 21 18 39 43 47
10.75 13.98 12.3 11.7 9.9 11 16.7 11.7
77
36 54 29 13 18 26 40 33
10.5 13.44 12.7 10.9 10.4 11 15.1 11.5
110
Unrelated Unrelated Unrelated Unrelated Unrelated Siblings Unrelated Unrelated siblings Unrelated siblings Unrelated siblings Unrelated Unrelated
and
VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS
FISH FISH FISH FISH FISH FISH FISH FISH
1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T
1.0 mm 1.5 mm 1.5 mm 1.0 mm 1.0 mm 1.5 mm 1.5 mm 1.5 mm
Volume Volume Volume Volume Area Volume Volume Volume
and
VCFS
FISH
1.5 T
1.5 mm
Area
and
VCFS
FISH
1.5 T
1.5 mm
Volume
LD
VCFS VCFS
FISH FISH
1.5 T 1.5 T
1.5 mm 1.5 mm
Volume Volume
VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS
FISH FISH FISH FISH FISH FISH FISH FISH FISH FISH
1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T
5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm
Volume Volume Area Volume Volume Volume Volume Volume Volume Volume
0.47 0.43 0.69 0.48 0.39 0.51 0.37 0.468
0.64 0.43 0.69 0.54 0.67 0.62 0.425 0.52
72.8 74.5 73.9 67 69.4 71.2
115.3 111.3 109.3 102 111.1 95.8
Antshel et al. (2005b) 60 11.1
0.52
72.3 52 10.89
0.48
99.3
Antshel et al. (2005a) 90 10.97
0.556
77 10.93
0.53
Kates et al. (2005) van Amelsvoort et al. (2004) Bearden et al. (2004) Kates et al. (2004) Shashi et al. (2004) Eliez et al. (2002) Chow et al. (2002) Eliez et al. (2001b) Eliez et al. (2001a) Kates et al. (2001) Eliez et al. (2001c) Eliez et al. (2000)
19 11.81 12 31
0.421 0.333
72.4 18 11.97 74 14 36
0.444 0.429
13 10 13 30 14 23 18 10 24 15
0.462 0.7 0.62
76.6 9 73.2 10 13 69.5 30 71.3 14 23 18 73 10 69.5 24 15
0.556 116.2 Unrelated 0.7 116.5 Unrelated 0.62 Unrelated 115.9 Unrelated 0.5 116.2 Unrelated 0.652 Unrelated 0.61 Unrelated 0.3 96.6 Unrelated 0.667 117.4 Unrelated 0.667 Unrelated
12.31 10.1 10.04 12.1 27.5 12.7 11.95 10.1 12.5 10.5
0.5 0.652 0.61 0.3 0.667 0.667
12.22 10 10.8 12.2 28.2 12.9 12.5 10.1 12.7 10.8
2) Cortical regions There were widespread cortical volumetric reductions in 22q11.2DS in comparison with controls affecting most regions including frontal lobes (N = 7, estimate: − 0.35, 95% CI:−0.56,− 0.14), right frontal lobe (N = 4, estimate: −0.27, 95% CI: − 0.52,−0.02), frontal gray matter (N = 3, estimate: −0.78, 95% CI:−1.41,− 0.15), temporal lobes (N = 5, estimate: − 0.83, 95% CI:− 1.17,− 0.48), left temporal lobe (N = 3, estimate: −0.51, 95% CI:− 0.87,−0.14), right temporal lobe (N = 3, estimate: −0.66, 95% CI:−1.03,−0.29), left temporal gray matter (N = 3, − 0.94, 95% CI: − 1.37, − 0.51), right temporal gray matter (N = 3, estimate: − 0.83, 95% CI: −1.26,−0.41), temporal lobes gray matter (N = 4, estimate: −0.92, 95% CI:−1.25,−0.59), temporal lobes white matter (N = 3, estimate: −0.76, 95% CI:−1.46,−0.05), parietal lobes gray matter (N = 3, estimate: −1.22, 95%CI:−1.83,−0.61), parietal lobes white matter (N = 3, −0.72, 95% CI:−1.43, −0.02), occipital lobes gray matter (N = 3, estimate: −1.11, 95% CI: −1.66, −0.56) and occipital lobes white matter (N = 3, estimate: −1.14, 95%CI:−1.78,−0.49). 3) Sub-cortical regions There were volumetric reductions in the hippocampi (N = 5, estimate: −0.86, 95% CI:−1.13, −0.59), right hippocampus (N = 6, estimate: −0.62, 95% CI:−0.92,−0.31), left hippocampus (N = 6, estimate: −0.5, 95% CI:−0.79,−0.21), cerebellum (N = 4, estimate: −1.11, 95%CI: −1.56, −0.66) in 22q11.2DS subjects compared with controls. The area of the corpus callosum was larger in 22q11.2DS but not in controls (N = 4, estimate: 0.8, 95% CI: 0.49, 1.11).
97.6 75
3.3. Meta-regression analyses and publication bias The only significant level of heterogeneity was detected in the parietal lobes (I2: 0.86, P = 0.001) although volumetric differences between cases and controls did not differ in the analysis. Effect sizes for this structure were significantly related to the proportion of male 22q11.2DS subjects (C: −7.44, Z: −3.73, P = b0.001, 95% CI: −11.35–3.53) with larger decrements in parietal lobes volume in male 22q11.2DS subjects. Age and year of publication did not show a significant association (P = 0.82 and P = 0.65 respectively). The contribution of IQ was not possible to calculate because it was not sufficiently reported in the primary studies. Scanner strength did not differ and the three studies included in this analysis reported the same slice thickness (1.5 mm). Egger's test indicated that there was significant publication bias for the right frontal lobe (Egger P = 0.027) and left frontal lobe (Egger P = 0.016). This effect was attributable to one single study by Antshel et al. (2005a,b) which being the least precise study acted as an outlier. When this study was excluded from the analysis, the results for right and left frontal lobes did not change but there was no longer evidence of publication bias (Egger P = 0.12 and P = 0.13 respectively). 4. Discussion Findings from this meta-analysis show that 22q11.2DS is associated with global brain volumetric reduction affecting both gray and white matter, and with reductions of the prefrontal cortex and hippocampi, while the corpus callosum area was increased. These relatively large effect sizes were
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
177
Table 2 Regions of interest that yielded significant standardized mean differences (SMDs) between VCFS and control groups. Brain region
Volumes Total brain volume Hemispheres (both) Hemisphere (L) Hemisphere (R) Hemispheres GM (both) Hemispheres WM (both) Gray matter whole brain White matter whole brain Frontal lobes Frontal lobe (R) Frontal gray matter (total) Parietal gray matter (total) Parietal white matter (total) Temporal lobes Temporal lobe (L) Temporal lobe (R) Temporal gray matter (total) Temporal white matter (total) Temporal gray (L) Temporal gray (R) Occipital gray matter (total) Occipital white matter (total) Hippocampus (total) Hippocampus (L) Hippocampus (R) Cerebellum Areas Corpus callosum
No. of studies
No. of subjects
Volume or area difference
VCFS
Controls
SMD
95%CI
7 4 4 4 5 4 5 5 7 4 3 3 3 5 3 3 4 3 3 3 3 3 5 6 6 4
227 79 79 79 88 58 147 147 217 156 60 60 60 116 66 66 83 60 47 47 60 60 146 200 200 95
221 64 64 64 86 56 136 136 158 102 52 52 52 101 55 55 75 52 47 47 52 52 132 172 172 84
− 1.02 − 0.73 − 0.68 − 0.68 − 0.52 − 1.09 − 0.92 − 0.85 − 0.35 − 0.27 − 0.78 − 1.22 − 0.72 − 0.83 − 0.51 − 0.66 − 0.92 − 0.76 − 0.94 − 0.83 − 1.11 − 1.14 − 0.86 − 0.5 − 0.62 − 1.11
− 1.26, − 1.08, − 1.02, − 1.02, − 0.83, − 1.56, − 1.17, − 1.20, − 0.56, − 0.52, − 1.41, − 1.83, − 1.43, − 1.17, − 0.87, − 1.03, − 1.25, − 1.46, − 1.37, − 1.26, − 1.66, − 1.78, − 1.13, − 0.79, − 0.92, − 1.56,
3
91
83
0.8
Heterogeneity 2
− 0.78 − 0.39 − 0.34 − 0.34 − 0.22 − 0.62 − 0.68 − 0.50 − 0.14 − 0.02 − 0.15 − 0.61 − 0.02 − 0.48 − 0.14 − 0.29 − 0.59 − 0.05 − 0.51 − 0.41 − 0.56 − 0.49 − 0.59 − 0.21 − 0.31 − 0.66
0.49, 1.11
Publication bias
Studies
I , P-value
Egger, P-value
0.28, P = 0.2 0, P = 0.6 0, P = 0.75 0, P = 0.75 0, P = 0.5 0.26, P = 0.26 0, P = 0.71 0.47, P = 0.11 0, P = 0.85 0, P = 0.66 0.57, P = 0.1 0.49, P = 0.14 0.66, P = 0.053 0.29, P = 0.23 0, P = 0. 84 0, P = 0.9 0, P = 0.91 0.66, P = 0.053 0,P = 0.4 0, P = 0.9 0.40, P = 0.2 0.56, P = 0.1 0. 14, P = 0.32 0.45, P = 0.11 0.51, P = 0.07 0.45, P = 0.14
0.6 0.8 0.75 0.75 0.82 0.06 0.3 0.6 0.07 0.027 0.92 0.9 0.9 0.8 0.4 0.7 0.75 0.9 0.52 0.82 0.5 0.55 0.2 0.7 0.6 0.32
2,3,7,8,11,18,21 6,12,13,22 6,12,13,22 6,12,13,22 12,13,16,18,20 12,13,18,20 1,3,4,7,19 1,3,4,7,19 3,6,10,11,12,13,22 6,10,12,22 3,4,20 3,4,20 3,4,20 3,6,12,13,18 6,12,22 6,12,22 3,4,18,20 3,4,20 17,18,20 17,18,20 3,4,20 3,4,20 3,6,7,12,18 1,6,7,8,12,18 1,6,7,8,12,18 3,6,12,22
0, P = 0.4
0.6
5,9,15
Studies: [1] DeBoer et al., 2007; [2]Glaser et al., 2007; [3]Gothelf et al., 2007; [4]Bearden et al., 2007; [5]Machado et al., 2007; [6]Campbell et al., 2006; [7]Debbane et al., 2006; [8]Kates et al., 2006; [9]Antshel et al., 2005b; [10]Antshel et al., 2005a; [11]Kates et al., 2005; [12]van Amelsvoort et al., 2004; [13]Bearden et al., 2004; [14]Kates et al., 2004; [15]Shashi et al., 2004; [16]Eliez et al., 2002; [17]Chow et al., 2002; [18]Eliez et al., 2001b; [19]Eliez et al., 2001a; [20]Kates et al., 2001; [21] Eliez et al., 2001c; [22]Eliez et al., 2000.
found in the absence of either heterogeneity or publication bias. Diffuse global brain volumetric reduction is consistent with the common occurrence of a learning disability in the 22q11.2DS population and the known association with a high incidence of psychiatric morbidity. A rostro-caudal gradient
has been described in the brain volumetric reductions of 22q11.2DS children, progressing into more diffuse gray and white matter loss later in adulthood (see Gothelf et al., 2008 for review). The studies we included were almost exclusively conducted in children with 22q11.2DS. We generally found an ascending order of magnitude of effect sizes in the total
Table 3 Regions of interest that did not yield significant standardized mean differences (SMDs) between VCFS and control groups. Brain region
Volumes Frontal lobe (L) Frontal white matter (total) Parietal lobes Caudate Caudate (L) Caudate (R) Amygdala Amygdala (L) Amygdala (R)
No. of studies
4 3 3 5 3 3 4 5 5
No. of subjects
Volume difference
Heterogeneity
Publication bias
VCFS
Controls
SMD
95%CI
I2, P-value
Egger, P-value
156 60 83 120 81 81 107 161 161
102 52 70 109 70 70 106 146 146
− 0.18 − 0.21 − 0.82 0.14 0.3 0.28 − 0.1 0.1 0.01
− 0.43, − 0.58, − 1.75, − 0.13, − 0.07, − 0.05, − 0.37, − 0.25, − 0.23,
0, P = 0.65 0, P = 0.44 0.86, P = 0.001 0. 04, P = 0.4 0.23, P = 0.27 0, P = 0.7 0, P = 0.4 0.55, P = 0.063 0.11, P = 0.34
0.016 0.66 0.5 0.73 0.7 0.3 0.6 0.82 0.93
0.07 0.17 0.12 0.41 0.68 0.60 0.17 0.45 0.25
Studies
6,10,12,22 3,4,20 3,6,22 3,6,12,14,16 6,12,16 6,12,16 3,7,12,18 1,7,8,12,18 1,7,8,12,18
Studies: [1] DeBoer et al., 2007; [2]Glaser et al., 2007; [3]Gothelf et al., 2007; [4]Bearden et al., 2007; [5]Machado et al., 2007; [6]Campbell et al., 2006; [7]Debbane et al., 2006; [8]Kates et al., 2006; [9]Antshel et al., 2005b; [10]Antshel et al., 2005a; [11]Kates et al., 2005; [12]van Amelsvoort et al., 2004; [13]Bearden et al., 2004; [14]Kates et al., 2004; [15]Shashi et al., 2004; [16]Eliez et al., 2002; [17]Chow et al., 2002; [18]Eliez et al., 2001b; [19]Eliez et al., 2001a; [20]Kates et al., 2001; [21] Eliez et al., 2001c; [22]Eliez et al., 2000.
178
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Fig. 2. Forest plot (random effects) showing the mean difference (95% CI) in total brain volume in 22q11.2DS subjects and normal controls in seven published studies.
volumes of the regions analysed with a rostro-caudal gradient: frontalb temporalb cerebellarb occipital. Based on this observation, it is possible to speculate that 22q11.2DS brain volumetric reduction may be the expression of an aberrant developmental trajectory possibly altered early during the developmental years. There is uncertainty about the brain regions and morphometric changes associated with vulnerability to psychosis and schizophrenia. Recent studies investigating samples at high risk of developing psychosis indicate that in comparison with controls, vulnerability could be associated with volumetric reductions in brain regions including right inferior frontal and superior temporal gyrus (Pantelis et al., 2003; Witthaus et al., 2009), hippocampus/para-hippocampal gyrus, and cingulate cortex (Pantelis et al., 2003). Moreover, high risk individuals show a progressive volumetric reduction when progression into first episode schizophrenia occurs, affecting several brain regions including cingulate cortex (Pantelis et al., 2003; Witthaus et al., 2009), hippocampus/para-hippocampal gyrus (Pantelis et al., 2003; Witthaus et al., 2009), right insula, the adjacent part of the right anterior superior temporal gyrus (Borgwardt et al., 2007), fusiform gyrus, orbitofrontal cortex (Pantelis et al., 2003; Witthaus et al., 2009), amygdala (Witthaus et al., 2009) and cerebellum (e.g. Pantelis et al. 2003). Further evidence emerged from a recent meta-analysis which demonstrated that anatomical changes in first episode schizophrenia affect the basal ganglia-thalamo cortical circuitry whereas progression into chronic schizophrenia is mostly characterised by further cortical grey matter loss not affecting temporo-limbic regions (Ellison-Wright et al., 2008). Although volumetric reductions demonstrated in this meta-analysis could be related to different psychiatric disorders and impairments, the fronto-temporal volumetric decrease may be associated with the increased relative risk of developing schizophrenia. This risk is approximately 20 to 25
times higher than the 1% described as lifetime risk for the general population (e.g. Bassett and Chow, 2008). Linkage studies have suggested an association between 22q11.2 and schizophrenia (Williams et al., 2003). The region deleted in 22q11.2DS contains several risk associated genetic variants including COMT, PRODH, ZDHHC8, CLDN5, DGCR14 and DGCR2 which have been proposed as candidate genes for association with schizophrenia (Arinami, 2006). It has been suggested that haplo-insufficiency of a gene mapping to 22q11 could lead to disturbed neuronal migration and pruning, predisposing to psychosis (Murphy and Owen, 2001). Among all the genes, COMT, which codes for catechol-O-methyl-transferase, has been the most extensively investigated. The COMT gene is located on 22q11 and is associated with optimal dopaminergic neurotransmission, and risk of developing schizophrenia-like symptoms (Boot et al., 2008). van Amelsvoort et al. (2008) demonstrated that COMT Val/Met polymorphism affects the volume of the frontal lobes in subjects with 22q11.2DS. Gothelf et al. (2005) investigated a sample of individuals with 22q11.2DS and not only confirmed the association between COMT low activity allele and progressive prefrontal volume loss but also the link with the risk of developing psychotic symptoms with progressive volumetric loss. Similarly Takahashi et al. (2009) showed that progressive volumetric loss in the superior temporal gyrus precedes the first expression of florid psychosis. These observations suggest that fronto-temporal volumetric reductions may be responsible for the predisposition to psychotic symptoms in 22q11.2DS. The hippocampus has specific functional links with the mesial temporal lobe network and it has been hypothesised to be related to the increased risk of developing psychopathology and memory impairments (Kates et al., 2006). Hippocampal volume reduction has been described in schizophrenia (e.g.
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Nelson et al., 1998) and the hippocampal volumetric reduction in 22q11.2DS could be expression of similar morphometric changes in the two conditions. However, several brain regions that have been extensively studied in schizophrenia (e.g. thalamus) were not covered in the existing VCFS studies and we are therefore not able to comment on their potential contribution. We found an increased area of the corpus callosum in this meta-analysis. There are limitations associated with studying the area of the corpus callosum, including what constitutes the best mid-sagittal slice and technical differences in measurements including the choice of anatomical boundaries. Meta-analyses of corpus callosum area in schizophrenia and bipolar disorder suggest that psychosis is associated with reduced callosal area (Arnone et al., 2008a,b). It may be that an enlarged corpus callosal area is an expression of an aberrant developmental trajectory, indicating abnormalities in the cortical areas connected by the white matter bundles crossing the corpus callosum. Morphometric changes may occur in the corpus callosum with the emergence of psychotic symptoms. This could be secondary to grey matter loss associated with schizophrenia (Ellison-Wright et al., 2008). It could also be a consequence of primary disruption of white matter bundles as indicated by diffusion tensor imaging and spectroscopy studies (Tang et al., 2007). A recent metaanalysis of diffusion tensor imaging studies in schizophrenia demonstrated a significant reduction in fractional anisotropy in the deep white matter of the left frontal and left temporal lobes supporting the possibility of ‘disconnectivity’ in schizophrenia (Ellison-Wright and Bullmore, 2009). It is difficult to be certain about the progression of such changes with time and effects on psychopathology. Longitudinal studies could help clarify the natural evolution of brain abnormalities in 22q11.2DS. Gothelf et al. (2007) for instance studied a sample of individuals with 22q11.2DS and showed a significant association between cortical/subcortical volumetric changes, verbal IQ and the emergence of psychotic symptoms. Due to the paucity of available reports it was not possible to include studies in this meta-analysis comparing 22q11.2DS subjects with and without schizophrenia. Therefore the possible association between 22q11.2DS and schizophrenia remains speculative. Future longitudinal studies could include a third comparator of subjects with schizophrenia with and/or without 22q11.2DS. Such a design could significantly improve our understanding of the disorder and its association with schizophrenia. Differences in defining regions of interest could significantly influence findings on a particular brain area. Voxel based morphometry studies could overcome the limitation of ‘a priory’ hypothesised regions, and provide the so much needed unbiased reporting of volumetric differences in 22q11.2DS. We found evidence of significant heterogeneity for the parietal lobes, and for no other brain region. This heterogeneity could be attributable to sex differences, although there was no evidence that this structure differed in individuals with 22q11.2DS compared to controls. However, we adopted an overall random effect analyses model which takes heterogeneity into account when calculating all summary effect sizes. Selective reporting and publication of positive results is a potential limitation to all meta-analyses and cannot be definitively excluded. We found evidence of
179
publication bias only in the analysis of the frontal lobes. Publication bias for these structures was related to a specific study which acted as an outlier and the effect size of the right frontal lobe was still significant once it was removed. In summary, we found that 22q11.2DS is associated with brain volumetric reduction but also with specific abnormalities in multiple cortical and sub-cortical brain regions. We found evidence of volumetric reductions in regions implicated in schizophrenia. These abnormalities may help explain why 22q11.2DS is associated with a greatly increased risk of psychosis. Role of funding source GMT is currently supported by the Baily Thomas Charitable Fund. DA is currently supported by the UK Medical Research Council. AM is currently supported by the Health Foundation. Contributors GMT and DA were involved in literature searches, data extraction and analyses, and in writing the first draft of the report. AM provided supervision, statistical expertise, offered guidance in the interpretation of the results and participated at all stages of development of the final report. KPE had a role in overall supervision and final drafting of the report. All the authors contributed to and have approved the final manuscript. Conflict of interest Nothing to declare. Acknowledgements GMT and DA would like to thank Drs. Amelsvoort, Bearden, DeBoer, Kates and Simon for supplementing their work with unpublished data to complete this meta-analysis. GMT would like to thank the Baily Thomas Charitable Fund for their support and Professor Declan Murphy for his guidance and encouragement.
References Antshel, K.M., AbdulSabur, N., Roizen, N., Fremont, W., Kates, W.R., 2005a. Sex differences in cognitive functioning in velocardiofacial syndrome (VCFS). Dev. Neuropsychol. 28 (3), 849–869. Antshel, K.M., Conchelos, J., Lanzetta, G., Fremont, W., Kates, W.R., 2005b. Behavior and corpus callosum morphology relationships in velocardiofacial syndrome (22q11.2 deletion syndrome). Psychiatry Res. 138 (3), 235–245. Arinami, T., 2006. Analyses of the associations between the genes of 22q11 deletion syndrome and schizophrenia. J. Hum. Genet. 51 (12), 1037–1045. Arnone, D., McIntosh, A.M., Chandra, P., Ebmeier, K.P., 2008a. Meta-analysis of magnetic resonance imaging studies of the corpus callosum in bipolar disorder. Acta Psychiatr. Scand. 118 (5), 357–362. Arnone, D., McIntosh, A.M., Tan, G.M., Ebmeier, K.P., 2008b. Meta-analysis of magnetic resonance imaging studies of the corpus callosum in schizophrenia. Schizophr. Res. 101 (1–3), 124–132. Baker, K.D., Skuse, D.H., 2005. Adolescents and young adults with 22q11 deletion syndrome: psychopathology in an at-risk group. Br. J. Psychiatry 186, 115–120. Bassett, A.S., Chow, W.C., 2008. Schizophrenia and 22q11.2 deletion syndrome. Curr. Psychiatry Rep. 10 (2), 148–157. Bearden, C.E., van Erp, T.G., Monterosso, J.R., Simon, T.J., Glahn, D.C., Saleh, P.A., Hill, N.M., McDonald-McGinn, D.M., Zackai, E., Emanuel, B.S., Cannon, T.D., 2004. Regional brain abnormalities in 22q11.2 deletion syndrome: association with cognitive abilities and behavioral symptoms. Neurocase 10 (3), 198–206. Bearden, C.E., van Erp, T.G., Dutton, R.A., Tran, H., Zimmermann, L., Sun, D., Geaga, J.A., Simon, T.J., Glahn, D.C., Cannon, T.D., Emanuel, B.S., Toga, A.W., Thompson, P.M., 2007. Mapping cortical thickness in children with 22q11.2 deletions. Cereb. Cortex 17 (8), 1889–1898. Boot, E., Booij, J., Zinkstok, J., Abeling, N., de Haan, L., Baas, F., Linszen, D., van Amelsvoort, T., 2008. Disrupted dopaminergic neurotransmission in 22q11 deletion syndrome. Neuropsychopharmacology 33 (6), 1252–1258. Borgwardt, S.J., Riecher-Rossler, A., Dazzan, P., Chitnis, X., Aston, J., Drewe, M., Gschwandtner, U., Haller, S., Pfluger, M., Rechsteiner, E., D'Souza, M., Stieglitz, R.D., Radu, E.W., McGuire, P.K., 2007. Regional gray matter
180
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
volume abnormalities in the at risk mental state. Biol. Psychiatry 61, 1148–1156. Campbell, L.E., Daly, E., Toal, F., Stevens, A., Azuma, R., Catani, M., Ng, V., van Amelsvoort, T., Chitnis, X., Cutter, W., Murphy, D.G., Murphy, K.C., 2006. Brain and behaviour in children with 22q11.2 deletion syndrome: a volumetric and voxel-based morphometry MRI study. Brain 129 (Pt 5), 1218–1228. Chow, E.W., Zipursky, R.B., Mikulis, D.J., Bassett, A.S., 2002. Structural brain abnormalities in patients with schizophrenia and 22q11 deletion syndrome. Biol. Psychiatry 51 (3), 208–215. Debbane, M., Schaer, M., Farhoumand, R., Glaser, B., Eliez, S., 2006. Hippocampal volume reduction in 22q11.2 deletion syndrome. Neuropsychologia 44 (12), 2360–2365. Deboer, T., Wu, Z., Lee, A., Simon, T.J., 2007. Hippocampal volume reduction in children with chromosome 22q11.2 deletion syndrome is associated with cognitive impairment. Behav. Brain Funct. 3, 54. DerSimonian, R., Laird, N., 1986. Meta-analysis in clinical trials. Control. Clin. Trials 7 (3), 177–188. Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in metaanalysis detected by a simple, graphical test. BMJ 315, 629–634. Eliez, S., Schmitt, J.E., White, C.D., Reiss, A.L., 2000. Children and adolescents with velocardiofacial syndrome: a volumetric MRI study. Am. J. Psychiatry 157 (3), 409–415. Eliez, S., Antonarakis, S.E., Morris, M.A., Dahoun, S.P., Reiss, A.L., 2001a. Parental origin of the deletion 22q11.2 and brain development in velocardiofacial syndrome: a preliminary study. Arch. Gen. Psychiatry 58 (1), 64–68. Eliez, S., Blasey, C.M., Schmitt, E.J., White, C.D., Hu, D., Reiss, A.L., 2001b. Velocardiofacial syndrome: are structural changes in the temporal and mesial temporal regions related to schizophrenia? Am. J. Psychiatry 158 (3), 447–453. Eliez, S., Schmitt, J.E., White, C.D., Wellis, V.G., Reiss, A.L., 2001c. A quantitative MRI study of posterior fossa development in velocardiofacial syndrome. Biol. Psychiatry 49 (6), 540–546. Eliez, S., Barnea-Goraly, N., Schmitt, J.E., Liu, Y., Reiss, A.L., 2002. Increased basal ganglia volumes in velo-cardio-facial syndrome (deletion 22q11.2). Biol. Psychiatry 52 (1), 68–70. Ellison-Wright, I., Bullmore, E., 2009. Meta-analysis of diffusion tensor imaging studies in schizophrenia. Schizophr. Res. 108 (1–3), 3–10. Ellison-Wright, I., Glahn, D.C., Laird, A.R., Thelen, S.M., Bullmore, E., 2008. The anatomy of first-episode and chronic schizophrenia: an anatomical likelihood estimation meta-analysis. Am. J. Psychiatry 165 (8), 1015–1023. Glaser, B., Schaer, M., Berney, S., Debbane, M., Vuilleumier, P., Eliez, S., 2007. Structural changes to the fusiform gyrus: a cerebral marker for social impairments in 22q11.2 deletion syndrome? Schizophr. Res. 96 (1–3), 82–86. Gothelf, D, Presburger, G, Levy, D, Nahmani, A, Burg, M, Berant, M, Blieden, LC, Finkelstein, Y, Frisch, A, Apter, A, Weizman, A., 2004a. Genetic, developmental, and physical factors associated with attention deficit hyperactivity disorder in patients with velocardiofacial syndrome. Am. J. Med. Genet. B Neuropsychiatr. Genet. 126B (1), 116–121. Gothelf, D., Presburger, G., Zohar, A.H., Burg, M., Nahmani, A., Frydman, M., Shohat, M., Inbar, D., Aviram-Goldring, A., Yeshaya, J., Steinberg, T., Finkelstein, Y., Frisch, A., Weizman, A., Apter, A., 2004b. Obsessive– compulsive disorder in patients with velocardiofacial (22q11 deletion) syndrome. Am. J. Med. Genet. B Neuropsychiatr. Genet. 126B (1), 99–105. Gothelf, D., Eliez, S., Thompson, T., Hinard, C., Penniman, L., Feinstein, C., Kwon, H., Jin, S., Jo, B., Antonarakis, S.E., Morris, MA., Reiss, AL., 2005. COMT genotype predicts longitudinal cognitive decline and psychosis in 22q11.2 deletion syndrome. Nat. Neurosci. 8 (11), 1500–1502. Gothelf, D., Penniman, L., Gu, E., Eliez, S., Reiss, A.L., 2007. Developmental trajectories of brain structure in adolescents with 22q11.2 deletion syndrome: a longitudinal study. Schizophr. Res. 96 (1–3), 72–81. Gothelf, D., Schaer, M., Eliez, S., 2008. Genes, brain development and psychiatric phenotypes in velo-cardio-facial syndrome. Dev. Disabil. Res. Rev. 14 (1), 59–68. Henry, J.C., van Amelsvoort, T., Morris, R.G., Owen, M.J., Murphy, D.G., Murphy, K.C., 2002. An investigation of the neuropsychological profile in adults with velo-cardio-facial syndrome (VCFS). Neuropsychologia 40 (5), 471–478. Higgins, J.P., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring inconsistency in meta-analyses. BMJ 327, 557–560. Kates, W.R., Burnette, C.P., Jabs, E.W., Rutberg, J., Murphy, A.M., Grados, M., Geraghty, M., Kaufmann, W.E., Pearlson, G.D., 2001. Regional cortical white matter reductions in velocardiofacial syndrome: a volumetric MRI analysis. Biol. Psychiatry 49 (8), 677–684. Kates, W.R., Burnette, C.P., Bessette, B.A., Folley, B.S., Strunge, L., Jabs, E.W., Pearlson, G.D., 2004. Frontal and caudate alterations in velocardiofacial syndrome (deletion at chromosome 22q11.2). J. Child Neurol. 19 (5), 337–342.
Kates, W.R., Antshel, K., Willhite, R., Bessette, B.A., AbdulSabur, N., Higgins, A.M., 2005. Gender-moderated dorsolateral prefrontal reductions in 22q11.2 Deletion Syndrome: implications for risk for schizophrenia. Child Neuropsychol 11 (1), 73–85. Kates, W.R., Miller, A.M., Abdulsabur, N., Antshel, K.M., Conchelos, J., Fremont, W., Roizen, N., 2006. Temporal lobe anatomy and psychiatric symptoms in velocardiofacial syndrome. J. Am. Acad. Child Adolesc. Psych. 45 (5), 587–595. Machado, A.M., Simon, T.J., Nguyen, V., McDonald-McGinn, D.M., Zackai, E.H., Gee, J.C., 2007. Corpus callosum morphology and ventricular size in chromosome 22q11.2 deletion syndrome. Brain Res. 1131 (1), 197–210. McDonald-McGinn, D.M., Kirschner, R., Goldmuntz, E., Sullivan, K., Eicher, P., Gerdes, M., Moss, E., Solot, C., Wang, P., Jacobs, I., Handler, S., Knightly, C., Heher, K., Wilson, M., Ming, J.E., Grace, K., Driscoll, D., Pasquariello, P., Randall, P., Larossa, D., Emanuel, B.S., Zackai, E.H., 1999. The Philadelphia story: the 22q11.2 deletion: report on 250 patients. Genet. Couns. 10 (1), 11–24. Murphy, K.C., Owen, M.J., 2001. Velo-cardio-facial syndrome: a model for understanding the genetics and pathogenesis of schizophrenia. Br. J. Psychiatry 179, 397–402. Murphy, K.C., Jones, R.G., Griffiths, E., Thompson, P.W., Owen, M.J., 1998. Chromosome 22q11 deletions: an under recognized cause of idiopathic learning disability. Br. J. Psychiatry 172, 180–183. Murphy, K.C., Jones, L.A., Owen, M.J., 1999. High rates of schizophrenia in adults with velo-cardio-facial syndrome. Arch. Gen. Psychiatry 56, 940–945. Nelson, M.D., Saykin, A.J., Riordan, H.J., 1998. Hippocampal volume reduction in schizophrenia as assessed by magnetic resonance imaging: a metaanalytic study. Arch. Gen. Psychiatry 55 (5), 433–440. Niklasson, L., Rasmussen, P., Oskarsdóttir, S., Gillberg, C., 2002. Chromosome 22q11 deletion syndrome (CATCH 22): neuropsychiatric and neuropsychological aspects. Dev. Med. Child Neurol. 44 (1), 44–50. Pantelis, C., Velakoulis, D., McGorry, P.D., Wood, S.J., Suckling, J., Phillips, L.J., Yung, A.R., Bullmore, E.T., Brewer, W., Soulsby, B., Desmond, P., McGuire, P.K., 2003. Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet 361 (9354), 281–288. Papolos, D.F., Faedda, G.L., Veit, S., Goldberg, R., Morrow, B., Kucherlapati, R., Shprintzen, R.J., 1996. Bipolar spectrum disorders in patients diagnosed with velo-cardio-facial syndrome: does a hemizygous deletion of chromosome 22q11 result in bipolar affective disorder? Am. J. Psychiatry 153 (12), 1541–1547. Scambler, P.J., Kelly, D., Lindsay, E., Williamson, R., Goldberg, R., Shprintzen, R., Wilson, D.I., Goodship, J.A., Cross, I.E., Burn, J., 1992. Velo-cardio-facial syndrome associated with chromosome 22 deletions encompassing the DiGeorge locus. Lancet 339 (8802), 1138–1139. Shashi, V., Muddasani, S., Santos, C.C., Berry, M.N., Kwapil, T.R., Lewandowski, E., Keshavan, M.S., 2004. Abnormalities of the corpus callosum in nonpsychotic children with chromosome 22q11 deletion syndrome. NeuroImage 21 (4), 1399–1406. Simon, T.J., Ding, L., Bish, J.P., McDonald-McGinn, D.M., Zackai, E.H., Gee, J., 2005. Volumetric, connective, and morphologic changes in the brains of children with chromosome 22q11.2 deletion syndrome: an integrative study. NeuroImage 25 (1), 169–180. Swillen, A., Devriendt, K., Legius, E., Eyskens, B., Dumoulin, M., Gewillig, M., Fryns, J.P., 1997. J. Med. Genet. 34 (6), 453–458. Takahashi, T., Wood, S.J., Yung, A.R., Soulsby, B., McGorry, P.D., Suzuki, M., Kawasaki, Y., Phillips, L.J., Velakoulis, D., Pantelis, C., 2009. Progressive gray matter reduction of the superior temporal gyrus during transition to psychosis. Arch. Gen. Psychiatry 66 (4), 366. Tang, C.Y., Friedman, J., Shungu, D., Chang, L., Ernst, T., Stewart, D., Hajianpour, A., Carpenter, D., Ng, J., Mao, X., Hof, P.R., Buchsbaum, M.S., Davis, K., Gorman, J.mM., 2007. Correlations between Diffusion Tensor Imaging (DTI) and Magnetic Resonance Spectroscopy (1H MRS) in schizophrenic patientsmand normal controls. BMC Psychiatry 7, 25. Tézenas Du Montcel, S., Mendizabai, H., Aymé, S., Lévy, A., Philip, N., 1996. Prevalence of 22q11 microdeletion. J. Med. Genet. 33 (8), 719. van Amelsvoort, T., Daly, E., Henry, J., Robertson, D., Ng, V., Owen, M., Murphy, K.C., Murphy, D.G., 2004. Brain anatomy in adults with velocardiofacial syndrome with and without schizophrenia: preliminary results of a structural magnetic resonance imaging study. Arch. Gen. Psychiatry 61 (11), 1085–1096. van Amelsvoort, T., Zinkstok, J., Figee, M., Daly, E., Morris, R., Owen, M.J., Murphy, K.C., De Haan, L., Linszen, D.H., Glaser, B., Murphy, D.G., 2008. Effects of a functional COMT polymorphism on brain anatomy and cognitive function in adults with velo-cardio-facial syndrome. Psychol. Med. 38 (1), 89–100. Williams, N.M., Norton, N., Williams, H., Ekholm, B., Hamshere, M.L., Lindblom, Y., Chowdari, K.V., Cardno, A.G., Zammit, S., Jones, L.A., Murphy, K.C., Sanders, R.D., McCarthy, G., Gray, M.Y., Jones, G., Holmans,
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181 P., Nimgaonkar, V., Adolfson, R., Osby, U., Terenius, L., Sedvall, G., O'Donovan, M.C., Owen, M.J., 2003. A systematic genomewide linkage study in 353 sib pairs with schizophrenia. Am. J. Hum. Genet. 73 (6), 1355–1367. Witthaus H., Kaufmann C., Bohner G., Ozgürdal S., Gudlowski Y., Gallinat J., Ruhrmann S., Brüne M., Heinz A., Klingebiel R., Juckel G., 2009. Gray matter abnormalities in subjects at ultra-high risk for schizophrenia and
181
first-episode schizophrenic patients compared to healthy controls. Psychiatry Res. 173, 163–169. Yan, W., Jacobsen, L.K., Krasnewich, D.M., Guan, X.Y., Lenane, M.C., Paul, S.P., Dalwadi, H.N., Long, R.T., Kumra, S., Martin, B.M., Scrambler, P.J., Trent, J.M., Sidrandky, E., Ginns, E.I., Rapoport, J.L., 1998. Chromosome 22q11.2 interstizial deletions among childhood onset schizophrenics and ‘multidimensionally impaired’. Am. J. Med. Genet. 81, 41–43.
Contents lists available at ScienceDirect
Schizophrenia Research 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 / s c h r e s
Meta-analysis of magnetic resonance imaging studies in chromosome 22q11.2 deletion syndrome (velocardiofacial syndrome) Giles M. Tan a,1, Danilo Arnone b,d,⁎,1, Andrew M. McIntosh c, Klaus P. Ebmeier c,d a
Section of Brain Maturation, PO Box 50, Division of Psychological Medicine & Psychiatry, Institute of Psychiatry at the Maudsley, King's College London, De Crespigny Park, London, SE5 8AF, United Kingdom Neuroscience and Psychiatry Unit, Stopford Building, Manchester, United Kingdom c University of Edinburgh, Royal Edinburgh Hospital, Edinburgh, United Kingdom d University Department of Psychiatry, Oxford, United Kingdom b
a r t i c l e
i n f o
Article history: Received 24 May 2009 Received in revised form 12 August 2009 Accepted 7 September 2009 Available online 9 October 2009 Keywords: Chromosome 22q11.2 deletion Velocardiofacial syndrome Magnetic resonance imaging Neuroimaging
a b s t r a c t Objectives: 22q11.2 deletion syndrome (22q11.2DS), also known as velocardiofacial syndrome (VCFS) or DiGeorge Syndrome, is a genetic disorder due to a micro deletion on chromosome 22q11.2. VCFS is associated with abnormalities in brain structure and with an increased risk of psychiatric disorders, particularly schizophrenia. The aim of this review was to statistically summarize the structural imaging literature on VCFS which due to the relatively rarity of the disorder tends to consider small sample sizes. Method: A systematic review and meta-analysis of region of interest (ROI) studies comparing VCFS with healthy controls was carried out. Significant heterogeneity was explored using meta-regression. Results: Subjects with VCFS were characterised by global brain volumetric reduction including several cortical regions, cerebellum and hippocampus. The area of the corpus callosum was increased. Conclusions: Many regions extensively studied in schizophrenia were not covered in the existing VCFS literature. However, the studies considered support volumetric abnormalities which may help explain why VCFS is associated with a greatly increased risk of psychosis and other psychiatric disorders. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Chromosome 22q11.2 deletion syndrome (22q11.2DS), affects approximately 1 in 5000 live births (Tézenas Du Montcel et al., 1996) and refers to a group of previously distinct genetically determined disorders (velocardiofacial syndrome, Shprintzen syndrome, DiGeorge syndrome, conotruncal anomaly face syndrome, Cayler cardiofacial syndrome, and CATCH 22 syndrome) caused by a microdeletion
⁎ Corresponding author. Neuroscience and Psychiatry Unit, University of Manchester, G810 Stopford Building, Oxford Road, Manchester, M13 9PT, United Kingdom. E-mail addresses: [email protected] (G.M. Tan), [email protected] (D. Arnone). 1 Authors contributed equally to the work. 0920-9964/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2009.09.010
on the long arm of chromosome 22 (Scambler et al., 1992). VCFS is characterised by a distinct phenotype including a typical facial appearance, cleft palate, velopharyngeal insufficiency, cardiac anomalies (McDonald-McGinn et al., 1999; Swillen et al., 1997), learning disabilities and by an increased risk for psychotic disorders (Henry et al., 2002; Gothelf et al., 2004a,b; Baker and Skuse, 2005; Papolos et al., 1996; Niklasson et al., 2002) particularly schizophrenia (Murphy et al., 1999). More specifically, psychotic symptoms have been diagnosed in up to 30% of adolescents and adults with 22q11.2DS (Murphy et al., 1999). Moreover, the deletion has been found in 1–6% of people with schizophrenia (Murphy et al., 1998; Yan et al., 1998). This association has led to considerable research to morphologically characterise 22q11.2DS brain abnormalities and potential similarities with schizophrenia. Magnetic
174
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
resonance imaging (MRI) has allowed detailed in vivo examination of brain structures in people with 22q11.2DS but the evidence is inconclusive (Eliez et al., 2000; Kates et al., 2001; Simon et al., 2005; Eliez et al., 2002; Kates et al., 2006; Campbell et al., 2006; Gothelf et al., 2007; DeBoer et al., 2007; Shashi et al., 2004). This is in part due to the relative rarity of the disorder and the tendency for studies to be small and potentially underpowered. Nevertheless, even taking this into account, the literature appears heterogeneous and published significant effect sizes differ in both direction and magnitude. The aim of this meta-analysis was to apply quantitative methods to characterise morphometric differences in 22q11.2DS in comparison with healthy controls by 1) identifying region of interest (ROI) studies which offered measurements of brain regions in 22q11.2DS and 2) synthesizing findings using random effects meta-analysis. We also assessed heterogeneity to clarify the role of a number of variables including year of publication, age, sex, IQ, scanner strength and slice thickness. 2. Materials and methods
Medicine, Oxford, UK. Standardised mean differences were calculated using Cohen's d statistic: 0
Cohen s d =
X 1 −X 2 SDp
where X 1 and X 2 are the mean volumes from the first and second groups respectively and SDp is the pooled standard deviation estimated from both groups: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ðn1 −1ÞSD21 + ðn2 −1ÞSD22 SDp = ðn1 + n2 −2Þ where ni and SDi are the mean and standard deviation of the ‘ith’ group. Standardised effect sizes were then combined using the inverse variance method. The standard deviation of Cohen's d is estimated as: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N d2 SDðdÞ + n1 n2 2ðN−2Þ
2.1. Search strategy and inclusion/exclusion criteria A comprehensive search from a range of electronic databases, including Medline, EMBASE, PsycINFO, and PubMED was conducted up to March 2008 that reported structural MRI data in people with 22q11.2DS and unaffected controls. Search terms used to identify the studies included ‘Velocardiofacial syndrome’, ‘DiGeorge syndrome’, ‘22q11.2 deletion syndrome’ and related terms combined using the AND operator with ‘Magnetic Resonance Imaging’. The search was also supplemented by a manual and bibliographic crossreferencing. We placed no limit on year of publication. Studies were included if they were published as a peer reviewed article and presented original data which compared a sample of genetically diagnosed 22q11.2DS subjects with a group of healthy controls. Only studies reporting structural MRI data where means and standard deviations were available (or could be extracted) for each group were included. Researchers were contacted if this information was not immediately available in an otherwise acceptable report. Studies were excluded if data were included in more recent larger studies. Data were extracted from all included studies by one author (GMT) and checked by another author (DA). Characteristics including age, sex and IQ of the 22q11.2DS and control groups which may have confounded any observed difference were also recorded as was the year of publication, scanner strength and slice thickness. Data were utilised when volume or area results were available for a region of interest (ROI) from three or more studies. Whereas volumes were acquired from more than one slice, callosal area was measured from one mid-sagittal slice. Volume and area data were never combined in the same analysis. 2.2. Statistical analysis Statistical analysis was conducted using STATA 9.0 (Stata Corp, College Station, Texas) supplemented by ‘Metan’ software downloadable from the Centre for Statistics in
where N is the total sample size for the study, d is Cohen's d and n1 and n2 are as defined above. Random effects analyses (DerSimonian and Laird, 1986) were used throughout to weight each study. The presence of heterogeneity was tested using the Q-test and its magnitude estimated using I2 and can be interpreted as the proportion of variance in effect size due to heterogeneity (Higgins et al., 2003). When the Q-test was significant, a Galbraith plot was used to identify those studies contributing the greatest amount to that heterogeneity to investigate potential causes. Publication bias which describes the excess of low-precision studies providing effect sizes of magnitude greater than the average was tested with the Egger's test (Egger et al., 1997). Significance level was set at P b 0.05. To further investigate causes for heterogeneity, metaregression analyses were performed for the following variables: mean age, sex, IQ, scanner strength, slice thickness and year of publication. The STATA program “metareg.ado” was used throughout and the REML (restricted maximum likelihood) method used to estimate the model parameters. 3. Results 3.1. Systematic search The literature search identified 156 publications of which 48 were retrieved in full text format. Fig. 1 summarises the study flow and reasons for exclusion. Twenty-two studies met inclusion criteria and were included in the final analysis which provided information on 35 regions of interest (Table 1). Images were manually traced reflecting similar anatomical borders and used different software programs to aid image processing. In some cases a semiautomated stereotactic based parcellation method was used (e.g. Eliez et al., 2000). All the studies used the same in-situ hybridization technique but did not report on the extent of the microdeletion. Three of these studies were included after
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
175
Fig. 1. Study flow and reasons for exclusion.
authors were contacted to supplement information published (DeBoer et al., 2007; Bearden et al., 2007; Kates et al., 2004). Year of publication ranged from 2000 to 2007. Table 1 lists the key demographic and descriptive data for each of the included studies. Three of the studies included both unrelated age matched individuals and siblings as controls while one study included only siblings as the control subjects. Across the 22 studies included in the meta-analysis, there were a total of 626 22q11.2DS subjects and 576 controls. The mean age and age range were 13.3 (9.9–31) for 22q11.2DS and 13.5 (10–36) for control subjects. The mean and median number of subjects across the studies were 28.5 and 21 respectively for the 22q11.2DS subjects and 26.2 and 20.5 for the control subjects. Except for one study (Eliez et al., 2002), all studies reported sex of subjects with the mean percentage of male 22q11.2DS and control subjects as 51.5% and 55.7% respectively. All the studies used a 1.5 Tesla magnet and diagnosis of 22q11.2DS was confirmed by fluorescent in-situ hybridisation. All the reports included in the analysis measured brain regions as volumes except the corpus callosum which was available as area. Sixteen studies reported IQs in cases and controls but they were matched only in a minority of studies.
3.2. Meta-analysis Random effect analyses were conducted for the 36 regions of interest (ROI) and results are presented in Tables 2 and 3 and Fig. 2: 1) Total brain The findings confirm global brain volumetric reduction including total brain (N=7, estimate: −1.02, 95% CI: −1.26, −0.78), combined left and right hemisphere (N=4, estimate: −0.73; 95% CI: −1.08, −0.39), left hemisphere (N=4, estimate: −0.68; 95% CI:−1.02, −0.34), right hemisphere (N=4, estimate: −0.68; 95% CI: −1.02, −0.34), left and right hemisphere white matter (N=4, estimate: −1.09; 95% CI: −1.56, −0.62), left and right hemisphere gray matter (N=5, estimate: −0.52; 95% CI: −0.83, −0.22), whole brain gray matter (N = 5, estimate: −0.92; 95% CI: −1.77, −0.68), and whole brain white matter (N = 5, estimate: −0.85, 95% CI: −1.20, −0.50). Possible duplicate publications (Eliez et al., 2001b; Kates et al., 2005) did not affect the finding of global brain volumetric reduction when these studies were excluded (N= 5, estimate: −1.01, 95% CI: −1.28, −0.75).
176
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Table 1 Summary of the studies included in the meta-analysis. VCFS cases
Normal controls
Diagnostic
Slice
Study (year)
n
Age % (years) male
IQ
n
Age % (years) male
IQ
Type
Diagnosis criteria
Scanner Thickness Measure
DeBoer et al. (2007) Glaser et al. (2007) Gothelf et al. (2007) Bearden et al. (2007) Machado et al. (2007) Campbell et al. (2006) Debbane et al. (2006) Kates et al. (2006)
36 42 29 21 18 39 43 47
10.75 13.98 12.3 11.7 9.9 11 16.7 11.7
77
36 54 29 13 18 26 40 33
10.5 13.44 12.7 10.9 10.4 11 15.1 11.5
110
Unrelated Unrelated Unrelated Unrelated Unrelated Siblings Unrelated Unrelated siblings Unrelated siblings Unrelated siblings Unrelated Unrelated
and
VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS
FISH FISH FISH FISH FISH FISH FISH FISH
1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T
1.0 mm 1.5 mm 1.5 mm 1.0 mm 1.0 mm 1.5 mm 1.5 mm 1.5 mm
Volume Volume Volume Volume Area Volume Volume Volume
and
VCFS
FISH
1.5 T
1.5 mm
Area
and
VCFS
FISH
1.5 T
1.5 mm
Volume
LD
VCFS VCFS
FISH FISH
1.5 T 1.5 T
1.5 mm 1.5 mm
Volume Volume
VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS VCFS
FISH FISH FISH FISH FISH FISH FISH FISH FISH FISH
1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T 1.5 T
5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm 1.5 mm
Volume Volume Area Volume Volume Volume Volume Volume Volume Volume
0.47 0.43 0.69 0.48 0.39 0.51 0.37 0.468
0.64 0.43 0.69 0.54 0.67 0.62 0.425 0.52
72.8 74.5 73.9 67 69.4 71.2
115.3 111.3 109.3 102 111.1 95.8
Antshel et al. (2005b) 60 11.1
0.52
72.3 52 10.89
0.48
99.3
Antshel et al. (2005a) 90 10.97
0.556
77 10.93
0.53
Kates et al. (2005) van Amelsvoort et al. (2004) Bearden et al. (2004) Kates et al. (2004) Shashi et al. (2004) Eliez et al. (2002) Chow et al. (2002) Eliez et al. (2001b) Eliez et al. (2001a) Kates et al. (2001) Eliez et al. (2001c) Eliez et al. (2000)
19 11.81 12 31
0.421 0.333
72.4 18 11.97 74 14 36
0.444 0.429
13 10 13 30 14 23 18 10 24 15
0.462 0.7 0.62
76.6 9 73.2 10 13 69.5 30 71.3 14 23 18 73 10 69.5 24 15
0.556 116.2 Unrelated 0.7 116.5 Unrelated 0.62 Unrelated 115.9 Unrelated 0.5 116.2 Unrelated 0.652 Unrelated 0.61 Unrelated 0.3 96.6 Unrelated 0.667 117.4 Unrelated 0.667 Unrelated
12.31 10.1 10.04 12.1 27.5 12.7 11.95 10.1 12.5 10.5
0.5 0.652 0.61 0.3 0.667 0.667
12.22 10 10.8 12.2 28.2 12.9 12.5 10.1 12.7 10.8
2) Cortical regions There were widespread cortical volumetric reductions in 22q11.2DS in comparison with controls affecting most regions including frontal lobes (N = 7, estimate: − 0.35, 95% CI:−0.56,− 0.14), right frontal lobe (N = 4, estimate: −0.27, 95% CI: − 0.52,−0.02), frontal gray matter (N = 3, estimate: −0.78, 95% CI:−1.41,− 0.15), temporal lobes (N = 5, estimate: − 0.83, 95% CI:− 1.17,− 0.48), left temporal lobe (N = 3, estimate: −0.51, 95% CI:− 0.87,−0.14), right temporal lobe (N = 3, estimate: −0.66, 95% CI:−1.03,−0.29), left temporal gray matter (N = 3, − 0.94, 95% CI: − 1.37, − 0.51), right temporal gray matter (N = 3, estimate: − 0.83, 95% CI: −1.26,−0.41), temporal lobes gray matter (N = 4, estimate: −0.92, 95% CI:−1.25,−0.59), temporal lobes white matter (N = 3, estimate: −0.76, 95% CI:−1.46,−0.05), parietal lobes gray matter (N = 3, estimate: −1.22, 95%CI:−1.83,−0.61), parietal lobes white matter (N = 3, −0.72, 95% CI:−1.43, −0.02), occipital lobes gray matter (N = 3, estimate: −1.11, 95% CI: −1.66, −0.56) and occipital lobes white matter (N = 3, estimate: −1.14, 95%CI:−1.78,−0.49). 3) Sub-cortical regions There were volumetric reductions in the hippocampi (N = 5, estimate: −0.86, 95% CI:−1.13, −0.59), right hippocampus (N = 6, estimate: −0.62, 95% CI:−0.92,−0.31), left hippocampus (N = 6, estimate: −0.5, 95% CI:−0.79,−0.21), cerebellum (N = 4, estimate: −1.11, 95%CI: −1.56, −0.66) in 22q11.2DS subjects compared with controls. The area of the corpus callosum was larger in 22q11.2DS but not in controls (N = 4, estimate: 0.8, 95% CI: 0.49, 1.11).
97.6 75
3.3. Meta-regression analyses and publication bias The only significant level of heterogeneity was detected in the parietal lobes (I2: 0.86, P = 0.001) although volumetric differences between cases and controls did not differ in the analysis. Effect sizes for this structure were significantly related to the proportion of male 22q11.2DS subjects (C: −7.44, Z: −3.73, P = b0.001, 95% CI: −11.35–3.53) with larger decrements in parietal lobes volume in male 22q11.2DS subjects. Age and year of publication did not show a significant association (P = 0.82 and P = 0.65 respectively). The contribution of IQ was not possible to calculate because it was not sufficiently reported in the primary studies. Scanner strength did not differ and the three studies included in this analysis reported the same slice thickness (1.5 mm). Egger's test indicated that there was significant publication bias for the right frontal lobe (Egger P = 0.027) and left frontal lobe (Egger P = 0.016). This effect was attributable to one single study by Antshel et al. (2005a,b) which being the least precise study acted as an outlier. When this study was excluded from the analysis, the results for right and left frontal lobes did not change but there was no longer evidence of publication bias (Egger P = 0.12 and P = 0.13 respectively). 4. Discussion Findings from this meta-analysis show that 22q11.2DS is associated with global brain volumetric reduction affecting both gray and white matter, and with reductions of the prefrontal cortex and hippocampi, while the corpus callosum area was increased. These relatively large effect sizes were
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
177
Table 2 Regions of interest that yielded significant standardized mean differences (SMDs) between VCFS and control groups. Brain region
Volumes Total brain volume Hemispheres (both) Hemisphere (L) Hemisphere (R) Hemispheres GM (both) Hemispheres WM (both) Gray matter whole brain White matter whole brain Frontal lobes Frontal lobe (R) Frontal gray matter (total) Parietal gray matter (total) Parietal white matter (total) Temporal lobes Temporal lobe (L) Temporal lobe (R) Temporal gray matter (total) Temporal white matter (total) Temporal gray (L) Temporal gray (R) Occipital gray matter (total) Occipital white matter (total) Hippocampus (total) Hippocampus (L) Hippocampus (R) Cerebellum Areas Corpus callosum
No. of studies
No. of subjects
Volume or area difference
VCFS
Controls
SMD
95%CI
7 4 4 4 5 4 5 5 7 4 3 3 3 5 3 3 4 3 3 3 3 3 5 6 6 4
227 79 79 79 88 58 147 147 217 156 60 60 60 116 66 66 83 60 47 47 60 60 146 200 200 95
221 64 64 64 86 56 136 136 158 102 52 52 52 101 55 55 75 52 47 47 52 52 132 172 172 84
− 1.02 − 0.73 − 0.68 − 0.68 − 0.52 − 1.09 − 0.92 − 0.85 − 0.35 − 0.27 − 0.78 − 1.22 − 0.72 − 0.83 − 0.51 − 0.66 − 0.92 − 0.76 − 0.94 − 0.83 − 1.11 − 1.14 − 0.86 − 0.5 − 0.62 − 1.11
− 1.26, − 1.08, − 1.02, − 1.02, − 0.83, − 1.56, − 1.17, − 1.20, − 0.56, − 0.52, − 1.41, − 1.83, − 1.43, − 1.17, − 0.87, − 1.03, − 1.25, − 1.46, − 1.37, − 1.26, − 1.66, − 1.78, − 1.13, − 0.79, − 0.92, − 1.56,
3
91
83
0.8
Heterogeneity 2
− 0.78 − 0.39 − 0.34 − 0.34 − 0.22 − 0.62 − 0.68 − 0.50 − 0.14 − 0.02 − 0.15 − 0.61 − 0.02 − 0.48 − 0.14 − 0.29 − 0.59 − 0.05 − 0.51 − 0.41 − 0.56 − 0.49 − 0.59 − 0.21 − 0.31 − 0.66
0.49, 1.11
Publication bias
Studies
I , P-value
Egger, P-value
0.28, P = 0.2 0, P = 0.6 0, P = 0.75 0, P = 0.75 0, P = 0.5 0.26, P = 0.26 0, P = 0.71 0.47, P = 0.11 0, P = 0.85 0, P = 0.66 0.57, P = 0.1 0.49, P = 0.14 0.66, P = 0.053 0.29, P = 0.23 0, P = 0. 84 0, P = 0.9 0, P = 0.91 0.66, P = 0.053 0,P = 0.4 0, P = 0.9 0.40, P = 0.2 0.56, P = 0.1 0. 14, P = 0.32 0.45, P = 0.11 0.51, P = 0.07 0.45, P = 0.14
0.6 0.8 0.75 0.75 0.82 0.06 0.3 0.6 0.07 0.027 0.92 0.9 0.9 0.8 0.4 0.7 0.75 0.9 0.52 0.82 0.5 0.55 0.2 0.7 0.6 0.32
2,3,7,8,11,18,21 6,12,13,22 6,12,13,22 6,12,13,22 12,13,16,18,20 12,13,18,20 1,3,4,7,19 1,3,4,7,19 3,6,10,11,12,13,22 6,10,12,22 3,4,20 3,4,20 3,4,20 3,6,12,13,18 6,12,22 6,12,22 3,4,18,20 3,4,20 17,18,20 17,18,20 3,4,20 3,4,20 3,6,7,12,18 1,6,7,8,12,18 1,6,7,8,12,18 3,6,12,22
0, P = 0.4
0.6
5,9,15
Studies: [1] DeBoer et al., 2007; [2]Glaser et al., 2007; [3]Gothelf et al., 2007; [4]Bearden et al., 2007; [5]Machado et al., 2007; [6]Campbell et al., 2006; [7]Debbane et al., 2006; [8]Kates et al., 2006; [9]Antshel et al., 2005b; [10]Antshel et al., 2005a; [11]Kates et al., 2005; [12]van Amelsvoort et al., 2004; [13]Bearden et al., 2004; [14]Kates et al., 2004; [15]Shashi et al., 2004; [16]Eliez et al., 2002; [17]Chow et al., 2002; [18]Eliez et al., 2001b; [19]Eliez et al., 2001a; [20]Kates et al., 2001; [21] Eliez et al., 2001c; [22]Eliez et al., 2000.
found in the absence of either heterogeneity or publication bias. Diffuse global brain volumetric reduction is consistent with the common occurrence of a learning disability in the 22q11.2DS population and the known association with a high incidence of psychiatric morbidity. A rostro-caudal gradient
has been described in the brain volumetric reductions of 22q11.2DS children, progressing into more diffuse gray and white matter loss later in adulthood (see Gothelf et al., 2008 for review). The studies we included were almost exclusively conducted in children with 22q11.2DS. We generally found an ascending order of magnitude of effect sizes in the total
Table 3 Regions of interest that did not yield significant standardized mean differences (SMDs) between VCFS and control groups. Brain region
Volumes Frontal lobe (L) Frontal white matter (total) Parietal lobes Caudate Caudate (L) Caudate (R) Amygdala Amygdala (L) Amygdala (R)
No. of studies
4 3 3 5 3 3 4 5 5
No. of subjects
Volume difference
Heterogeneity
Publication bias
VCFS
Controls
SMD
95%CI
I2, P-value
Egger, P-value
156 60 83 120 81 81 107 161 161
102 52 70 109 70 70 106 146 146
− 0.18 − 0.21 − 0.82 0.14 0.3 0.28 − 0.1 0.1 0.01
− 0.43, − 0.58, − 1.75, − 0.13, − 0.07, − 0.05, − 0.37, − 0.25, − 0.23,
0, P = 0.65 0, P = 0.44 0.86, P = 0.001 0. 04, P = 0.4 0.23, P = 0.27 0, P = 0.7 0, P = 0.4 0.55, P = 0.063 0.11, P = 0.34
0.016 0.66 0.5 0.73 0.7 0.3 0.6 0.82 0.93
0.07 0.17 0.12 0.41 0.68 0.60 0.17 0.45 0.25
Studies
6,10,12,22 3,4,20 3,6,22 3,6,12,14,16 6,12,16 6,12,16 3,7,12,18 1,7,8,12,18 1,7,8,12,18
Studies: [1] DeBoer et al., 2007; [2]Glaser et al., 2007; [3]Gothelf et al., 2007; [4]Bearden et al., 2007; [5]Machado et al., 2007; [6]Campbell et al., 2006; [7]Debbane et al., 2006; [8]Kates et al., 2006; [9]Antshel et al., 2005b; [10]Antshel et al., 2005a; [11]Kates et al., 2005; [12]van Amelsvoort et al., 2004; [13]Bearden et al., 2004; [14]Kates et al., 2004; [15]Shashi et al., 2004; [16]Eliez et al., 2002; [17]Chow et al., 2002; [18]Eliez et al., 2001b; [19]Eliez et al., 2001a; [20]Kates et al., 2001; [21] Eliez et al., 2001c; [22]Eliez et al., 2000.
178
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Fig. 2. Forest plot (random effects) showing the mean difference (95% CI) in total brain volume in 22q11.2DS subjects and normal controls in seven published studies.
volumes of the regions analysed with a rostro-caudal gradient: frontalb temporalb cerebellarb occipital. Based on this observation, it is possible to speculate that 22q11.2DS brain volumetric reduction may be the expression of an aberrant developmental trajectory possibly altered early during the developmental years. There is uncertainty about the brain regions and morphometric changes associated with vulnerability to psychosis and schizophrenia. Recent studies investigating samples at high risk of developing psychosis indicate that in comparison with controls, vulnerability could be associated with volumetric reductions in brain regions including right inferior frontal and superior temporal gyrus (Pantelis et al., 2003; Witthaus et al., 2009), hippocampus/para-hippocampal gyrus, and cingulate cortex (Pantelis et al., 2003). Moreover, high risk individuals show a progressive volumetric reduction when progression into first episode schizophrenia occurs, affecting several brain regions including cingulate cortex (Pantelis et al., 2003; Witthaus et al., 2009), hippocampus/para-hippocampal gyrus (Pantelis et al., 2003; Witthaus et al., 2009), right insula, the adjacent part of the right anterior superior temporal gyrus (Borgwardt et al., 2007), fusiform gyrus, orbitofrontal cortex (Pantelis et al., 2003; Witthaus et al., 2009), amygdala (Witthaus et al., 2009) and cerebellum (e.g. Pantelis et al. 2003). Further evidence emerged from a recent meta-analysis which demonstrated that anatomical changes in first episode schizophrenia affect the basal ganglia-thalamo cortical circuitry whereas progression into chronic schizophrenia is mostly characterised by further cortical grey matter loss not affecting temporo-limbic regions (Ellison-Wright et al., 2008). Although volumetric reductions demonstrated in this meta-analysis could be related to different psychiatric disorders and impairments, the fronto-temporal volumetric decrease may be associated with the increased relative risk of developing schizophrenia. This risk is approximately 20 to 25
times higher than the 1% described as lifetime risk for the general population (e.g. Bassett and Chow, 2008). Linkage studies have suggested an association between 22q11.2 and schizophrenia (Williams et al., 2003). The region deleted in 22q11.2DS contains several risk associated genetic variants including COMT, PRODH, ZDHHC8, CLDN5, DGCR14 and DGCR2 which have been proposed as candidate genes for association with schizophrenia (Arinami, 2006). It has been suggested that haplo-insufficiency of a gene mapping to 22q11 could lead to disturbed neuronal migration and pruning, predisposing to psychosis (Murphy and Owen, 2001). Among all the genes, COMT, which codes for catechol-O-methyl-transferase, has been the most extensively investigated. The COMT gene is located on 22q11 and is associated with optimal dopaminergic neurotransmission, and risk of developing schizophrenia-like symptoms (Boot et al., 2008). van Amelsvoort et al. (2008) demonstrated that COMT Val/Met polymorphism affects the volume of the frontal lobes in subjects with 22q11.2DS. Gothelf et al. (2005) investigated a sample of individuals with 22q11.2DS and not only confirmed the association between COMT low activity allele and progressive prefrontal volume loss but also the link with the risk of developing psychotic symptoms with progressive volumetric loss. Similarly Takahashi et al. (2009) showed that progressive volumetric loss in the superior temporal gyrus precedes the first expression of florid psychosis. These observations suggest that fronto-temporal volumetric reductions may be responsible for the predisposition to psychotic symptoms in 22q11.2DS. The hippocampus has specific functional links with the mesial temporal lobe network and it has been hypothesised to be related to the increased risk of developing psychopathology and memory impairments (Kates et al., 2006). Hippocampal volume reduction has been described in schizophrenia (e.g.
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
Nelson et al., 1998) and the hippocampal volumetric reduction in 22q11.2DS could be expression of similar morphometric changes in the two conditions. However, several brain regions that have been extensively studied in schizophrenia (e.g. thalamus) were not covered in the existing VCFS studies and we are therefore not able to comment on their potential contribution. We found an increased area of the corpus callosum in this meta-analysis. There are limitations associated with studying the area of the corpus callosum, including what constitutes the best mid-sagittal slice and technical differences in measurements including the choice of anatomical boundaries. Meta-analyses of corpus callosum area in schizophrenia and bipolar disorder suggest that psychosis is associated with reduced callosal area (Arnone et al., 2008a,b). It may be that an enlarged corpus callosal area is an expression of an aberrant developmental trajectory, indicating abnormalities in the cortical areas connected by the white matter bundles crossing the corpus callosum. Morphometric changes may occur in the corpus callosum with the emergence of psychotic symptoms. This could be secondary to grey matter loss associated with schizophrenia (Ellison-Wright et al., 2008). It could also be a consequence of primary disruption of white matter bundles as indicated by diffusion tensor imaging and spectroscopy studies (Tang et al., 2007). A recent metaanalysis of diffusion tensor imaging studies in schizophrenia demonstrated a significant reduction in fractional anisotropy in the deep white matter of the left frontal and left temporal lobes supporting the possibility of ‘disconnectivity’ in schizophrenia (Ellison-Wright and Bullmore, 2009). It is difficult to be certain about the progression of such changes with time and effects on psychopathology. Longitudinal studies could help clarify the natural evolution of brain abnormalities in 22q11.2DS. Gothelf et al. (2007) for instance studied a sample of individuals with 22q11.2DS and showed a significant association between cortical/subcortical volumetric changes, verbal IQ and the emergence of psychotic symptoms. Due to the paucity of available reports it was not possible to include studies in this meta-analysis comparing 22q11.2DS subjects with and without schizophrenia. Therefore the possible association between 22q11.2DS and schizophrenia remains speculative. Future longitudinal studies could include a third comparator of subjects with schizophrenia with and/or without 22q11.2DS. Such a design could significantly improve our understanding of the disorder and its association with schizophrenia. Differences in defining regions of interest could significantly influence findings on a particular brain area. Voxel based morphometry studies could overcome the limitation of ‘a priory’ hypothesised regions, and provide the so much needed unbiased reporting of volumetric differences in 22q11.2DS. We found evidence of significant heterogeneity for the parietal lobes, and for no other brain region. This heterogeneity could be attributable to sex differences, although there was no evidence that this structure differed in individuals with 22q11.2DS compared to controls. However, we adopted an overall random effect analyses model which takes heterogeneity into account when calculating all summary effect sizes. Selective reporting and publication of positive results is a potential limitation to all meta-analyses and cannot be definitively excluded. We found evidence of
179
publication bias only in the analysis of the frontal lobes. Publication bias for these structures was related to a specific study which acted as an outlier and the effect size of the right frontal lobe was still significant once it was removed. In summary, we found that 22q11.2DS is associated with brain volumetric reduction but also with specific abnormalities in multiple cortical and sub-cortical brain regions. We found evidence of volumetric reductions in regions implicated in schizophrenia. These abnormalities may help explain why 22q11.2DS is associated with a greatly increased risk of psychosis. Role of funding source GMT is currently supported by the Baily Thomas Charitable Fund. DA is currently supported by the UK Medical Research Council. AM is currently supported by the Health Foundation. Contributors GMT and DA were involved in literature searches, data extraction and analyses, and in writing the first draft of the report. AM provided supervision, statistical expertise, offered guidance in the interpretation of the results and participated at all stages of development of the final report. KPE had a role in overall supervision and final drafting of the report. All the authors contributed to and have approved the final manuscript. Conflict of interest Nothing to declare. Acknowledgements GMT and DA would like to thank Drs. Amelsvoort, Bearden, DeBoer, Kates and Simon for supplementing their work with unpublished data to complete this meta-analysis. GMT would like to thank the Baily Thomas Charitable Fund for their support and Professor Declan Murphy for his guidance and encouragement.
References Antshel, K.M., AbdulSabur, N., Roizen, N., Fremont, W., Kates, W.R., 2005a. Sex differences in cognitive functioning in velocardiofacial syndrome (VCFS). Dev. Neuropsychol. 28 (3), 849–869. Antshel, K.M., Conchelos, J., Lanzetta, G., Fremont, W., Kates, W.R., 2005b. Behavior and corpus callosum morphology relationships in velocardiofacial syndrome (22q11.2 deletion syndrome). Psychiatry Res. 138 (3), 235–245. Arinami, T., 2006. Analyses of the associations between the genes of 22q11 deletion syndrome and schizophrenia. J. Hum. Genet. 51 (12), 1037–1045. Arnone, D., McIntosh, A.M., Chandra, P., Ebmeier, K.P., 2008a. Meta-analysis of magnetic resonance imaging studies of the corpus callosum in bipolar disorder. Acta Psychiatr. Scand. 118 (5), 357–362. Arnone, D., McIntosh, A.M., Tan, G.M., Ebmeier, K.P., 2008b. Meta-analysis of magnetic resonance imaging studies of the corpus callosum in schizophrenia. Schizophr. Res. 101 (1–3), 124–132. Baker, K.D., Skuse, D.H., 2005. Adolescents and young adults with 22q11 deletion syndrome: psychopathology in an at-risk group. Br. J. Psychiatry 186, 115–120. Bassett, A.S., Chow, W.C., 2008. Schizophrenia and 22q11.2 deletion syndrome. Curr. Psychiatry Rep. 10 (2), 148–157. Bearden, C.E., van Erp, T.G., Monterosso, J.R., Simon, T.J., Glahn, D.C., Saleh, P.A., Hill, N.M., McDonald-McGinn, D.M., Zackai, E., Emanuel, B.S., Cannon, T.D., 2004. Regional brain abnormalities in 22q11.2 deletion syndrome: association with cognitive abilities and behavioral symptoms. Neurocase 10 (3), 198–206. Bearden, C.E., van Erp, T.G., Dutton, R.A., Tran, H., Zimmermann, L., Sun, D., Geaga, J.A., Simon, T.J., Glahn, D.C., Cannon, T.D., Emanuel, B.S., Toga, A.W., Thompson, P.M., 2007. Mapping cortical thickness in children with 22q11.2 deletions. Cereb. Cortex 17 (8), 1889–1898. Boot, E., Booij, J., Zinkstok, J., Abeling, N., de Haan, L., Baas, F., Linszen, D., van Amelsvoort, T., 2008. Disrupted dopaminergic neurotransmission in 22q11 deletion syndrome. Neuropsychopharmacology 33 (6), 1252–1258. Borgwardt, S.J., Riecher-Rossler, A., Dazzan, P., Chitnis, X., Aston, J., Drewe, M., Gschwandtner, U., Haller, S., Pfluger, M., Rechsteiner, E., D'Souza, M., Stieglitz, R.D., Radu, E.W., McGuire, P.K., 2007. Regional gray matter
180
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181
volume abnormalities in the at risk mental state. Biol. Psychiatry 61, 1148–1156. Campbell, L.E., Daly, E., Toal, F., Stevens, A., Azuma, R., Catani, M., Ng, V., van Amelsvoort, T., Chitnis, X., Cutter, W., Murphy, D.G., Murphy, K.C., 2006. Brain and behaviour in children with 22q11.2 deletion syndrome: a volumetric and voxel-based morphometry MRI study. Brain 129 (Pt 5), 1218–1228. Chow, E.W., Zipursky, R.B., Mikulis, D.J., Bassett, A.S., 2002. Structural brain abnormalities in patients with schizophrenia and 22q11 deletion syndrome. Biol. Psychiatry 51 (3), 208–215. Debbane, M., Schaer, M., Farhoumand, R., Glaser, B., Eliez, S., 2006. Hippocampal volume reduction in 22q11.2 deletion syndrome. Neuropsychologia 44 (12), 2360–2365. Deboer, T., Wu, Z., Lee, A., Simon, T.J., 2007. Hippocampal volume reduction in children with chromosome 22q11.2 deletion syndrome is associated with cognitive impairment. Behav. Brain Funct. 3, 54. DerSimonian, R., Laird, N., 1986. Meta-analysis in clinical trials. Control. Clin. Trials 7 (3), 177–188. Egger, M., Davey Smith, G., Schneider, M., Minder, C., 1997. Bias in metaanalysis detected by a simple, graphical test. BMJ 315, 629–634. Eliez, S., Schmitt, J.E., White, C.D., Reiss, A.L., 2000. Children and adolescents with velocardiofacial syndrome: a volumetric MRI study. Am. J. Psychiatry 157 (3), 409–415. Eliez, S., Antonarakis, S.E., Morris, M.A., Dahoun, S.P., Reiss, A.L., 2001a. Parental origin of the deletion 22q11.2 and brain development in velocardiofacial syndrome: a preliminary study. Arch. Gen. Psychiatry 58 (1), 64–68. Eliez, S., Blasey, C.M., Schmitt, E.J., White, C.D., Hu, D., Reiss, A.L., 2001b. Velocardiofacial syndrome: are structural changes in the temporal and mesial temporal regions related to schizophrenia? Am. J. Psychiatry 158 (3), 447–453. Eliez, S., Schmitt, J.E., White, C.D., Wellis, V.G., Reiss, A.L., 2001c. A quantitative MRI study of posterior fossa development in velocardiofacial syndrome. Biol. Psychiatry 49 (6), 540–546. Eliez, S., Barnea-Goraly, N., Schmitt, J.E., Liu, Y., Reiss, A.L., 2002. Increased basal ganglia volumes in velo-cardio-facial syndrome (deletion 22q11.2). Biol. Psychiatry 52 (1), 68–70. Ellison-Wright, I., Bullmore, E., 2009. Meta-analysis of diffusion tensor imaging studies in schizophrenia. Schizophr. Res. 108 (1–3), 3–10. Ellison-Wright, I., Glahn, D.C., Laird, A.R., Thelen, S.M., Bullmore, E., 2008. The anatomy of first-episode and chronic schizophrenia: an anatomical likelihood estimation meta-analysis. Am. J. Psychiatry 165 (8), 1015–1023. Glaser, B., Schaer, M., Berney, S., Debbane, M., Vuilleumier, P., Eliez, S., 2007. Structural changes to the fusiform gyrus: a cerebral marker for social impairments in 22q11.2 deletion syndrome? Schizophr. Res. 96 (1–3), 82–86. Gothelf, D, Presburger, G, Levy, D, Nahmani, A, Burg, M, Berant, M, Blieden, LC, Finkelstein, Y, Frisch, A, Apter, A, Weizman, A., 2004a. Genetic, developmental, and physical factors associated with attention deficit hyperactivity disorder in patients with velocardiofacial syndrome. Am. J. Med. Genet. B Neuropsychiatr. Genet. 126B (1), 116–121. Gothelf, D., Presburger, G., Zohar, A.H., Burg, M., Nahmani, A., Frydman, M., Shohat, M., Inbar, D., Aviram-Goldring, A., Yeshaya, J., Steinberg, T., Finkelstein, Y., Frisch, A., Weizman, A., Apter, A., 2004b. Obsessive– compulsive disorder in patients with velocardiofacial (22q11 deletion) syndrome. Am. J. Med. Genet. B Neuropsychiatr. Genet. 126B (1), 99–105. Gothelf, D., Eliez, S., Thompson, T., Hinard, C., Penniman, L., Feinstein, C., Kwon, H., Jin, S., Jo, B., Antonarakis, S.E., Morris, MA., Reiss, AL., 2005. COMT genotype predicts longitudinal cognitive decline and psychosis in 22q11.2 deletion syndrome. Nat. Neurosci. 8 (11), 1500–1502. Gothelf, D., Penniman, L., Gu, E., Eliez, S., Reiss, A.L., 2007. Developmental trajectories of brain structure in adolescents with 22q11.2 deletion syndrome: a longitudinal study. Schizophr. Res. 96 (1–3), 72–81. Gothelf, D., Schaer, M., Eliez, S., 2008. Genes, brain development and psychiatric phenotypes in velo-cardio-facial syndrome. Dev. Disabil. Res. Rev. 14 (1), 59–68. Henry, J.C., van Amelsvoort, T., Morris, R.G., Owen, M.J., Murphy, D.G., Murphy, K.C., 2002. An investigation of the neuropsychological profile in adults with velo-cardio-facial syndrome (VCFS). Neuropsychologia 40 (5), 471–478. Higgins, J.P., Thompson, S.G., Deeks, J.J., Altman, D.G., 2003. Measuring inconsistency in meta-analyses. BMJ 327, 557–560. Kates, W.R., Burnette, C.P., Jabs, E.W., Rutberg, J., Murphy, A.M., Grados, M., Geraghty, M., Kaufmann, W.E., Pearlson, G.D., 2001. Regional cortical white matter reductions in velocardiofacial syndrome: a volumetric MRI analysis. Biol. Psychiatry 49 (8), 677–684. Kates, W.R., Burnette, C.P., Bessette, B.A., Folley, B.S., Strunge, L., Jabs, E.W., Pearlson, G.D., 2004. Frontal and caudate alterations in velocardiofacial syndrome (deletion at chromosome 22q11.2). J. Child Neurol. 19 (5), 337–342.
Kates, W.R., Antshel, K., Willhite, R., Bessette, B.A., AbdulSabur, N., Higgins, A.M., 2005. Gender-moderated dorsolateral prefrontal reductions in 22q11.2 Deletion Syndrome: implications for risk for schizophrenia. Child Neuropsychol 11 (1), 73–85. Kates, W.R., Miller, A.M., Abdulsabur, N., Antshel, K.M., Conchelos, J., Fremont, W., Roizen, N., 2006. Temporal lobe anatomy and psychiatric symptoms in velocardiofacial syndrome. J. Am. Acad. Child Adolesc. Psych. 45 (5), 587–595. Machado, A.M., Simon, T.J., Nguyen, V., McDonald-McGinn, D.M., Zackai, E.H., Gee, J.C., 2007. Corpus callosum morphology and ventricular size in chromosome 22q11.2 deletion syndrome. Brain Res. 1131 (1), 197–210. McDonald-McGinn, D.M., Kirschner, R., Goldmuntz, E., Sullivan, K., Eicher, P., Gerdes, M., Moss, E., Solot, C., Wang, P., Jacobs, I., Handler, S., Knightly, C., Heher, K., Wilson, M., Ming, J.E., Grace, K., Driscoll, D., Pasquariello, P., Randall, P., Larossa, D., Emanuel, B.S., Zackai, E.H., 1999. The Philadelphia story: the 22q11.2 deletion: report on 250 patients. Genet. Couns. 10 (1), 11–24. Murphy, K.C., Owen, M.J., 2001. Velo-cardio-facial syndrome: a model for understanding the genetics and pathogenesis of schizophrenia. Br. J. Psychiatry 179, 397–402. Murphy, K.C., Jones, R.G., Griffiths, E., Thompson, P.W., Owen, M.J., 1998. Chromosome 22q11 deletions: an under recognized cause of idiopathic learning disability. Br. J. Psychiatry 172, 180–183. Murphy, K.C., Jones, L.A., Owen, M.J., 1999. High rates of schizophrenia in adults with velo-cardio-facial syndrome. Arch. Gen. Psychiatry 56, 940–945. Nelson, M.D., Saykin, A.J., Riordan, H.J., 1998. Hippocampal volume reduction in schizophrenia as assessed by magnetic resonance imaging: a metaanalytic study. Arch. Gen. Psychiatry 55 (5), 433–440. Niklasson, L., Rasmussen, P., Oskarsdóttir, S., Gillberg, C., 2002. Chromosome 22q11 deletion syndrome (CATCH 22): neuropsychiatric and neuropsychological aspects. Dev. Med. Child Neurol. 44 (1), 44–50. Pantelis, C., Velakoulis, D., McGorry, P.D., Wood, S.J., Suckling, J., Phillips, L.J., Yung, A.R., Bullmore, E.T., Brewer, W., Soulsby, B., Desmond, P., McGuire, P.K., 2003. Neuroanatomical abnormalities before and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet 361 (9354), 281–288. Papolos, D.F., Faedda, G.L., Veit, S., Goldberg, R., Morrow, B., Kucherlapati, R., Shprintzen, R.J., 1996. Bipolar spectrum disorders in patients diagnosed with velo-cardio-facial syndrome: does a hemizygous deletion of chromosome 22q11 result in bipolar affective disorder? Am. J. Psychiatry 153 (12), 1541–1547. Scambler, P.J., Kelly, D., Lindsay, E., Williamson, R., Goldberg, R., Shprintzen, R., Wilson, D.I., Goodship, J.A., Cross, I.E., Burn, J., 1992. Velo-cardio-facial syndrome associated with chromosome 22 deletions encompassing the DiGeorge locus. Lancet 339 (8802), 1138–1139. Shashi, V., Muddasani, S., Santos, C.C., Berry, M.N., Kwapil, T.R., Lewandowski, E., Keshavan, M.S., 2004. Abnormalities of the corpus callosum in nonpsychotic children with chromosome 22q11 deletion syndrome. NeuroImage 21 (4), 1399–1406. Simon, T.J., Ding, L., Bish, J.P., McDonald-McGinn, D.M., Zackai, E.H., Gee, J., 2005. Volumetric, connective, and morphologic changes in the brains of children with chromosome 22q11.2 deletion syndrome: an integrative study. NeuroImage 25 (1), 169–180. Swillen, A., Devriendt, K., Legius, E., Eyskens, B., Dumoulin, M., Gewillig, M., Fryns, J.P., 1997. J. Med. Genet. 34 (6), 453–458. Takahashi, T., Wood, S.J., Yung, A.R., Soulsby, B., McGorry, P.D., Suzuki, M., Kawasaki, Y., Phillips, L.J., Velakoulis, D., Pantelis, C., 2009. Progressive gray matter reduction of the superior temporal gyrus during transition to psychosis. Arch. Gen. Psychiatry 66 (4), 366. Tang, C.Y., Friedman, J., Shungu, D., Chang, L., Ernst, T., Stewart, D., Hajianpour, A., Carpenter, D., Ng, J., Mao, X., Hof, P.R., Buchsbaum, M.S., Davis, K., Gorman, J.mM., 2007. Correlations between Diffusion Tensor Imaging (DTI) and Magnetic Resonance Spectroscopy (1H MRS) in schizophrenic patientsmand normal controls. BMC Psychiatry 7, 25. Tézenas Du Montcel, S., Mendizabai, H., Aymé, S., Lévy, A., Philip, N., 1996. Prevalence of 22q11 microdeletion. J. Med. Genet. 33 (8), 719. van Amelsvoort, T., Daly, E., Henry, J., Robertson, D., Ng, V., Owen, M., Murphy, K.C., Murphy, D.G., 2004. Brain anatomy in adults with velocardiofacial syndrome with and without schizophrenia: preliminary results of a structural magnetic resonance imaging study. Arch. Gen. Psychiatry 61 (11), 1085–1096. van Amelsvoort, T., Zinkstok, J., Figee, M., Daly, E., Morris, R., Owen, M.J., Murphy, K.C., De Haan, L., Linszen, D.H., Glaser, B., Murphy, D.G., 2008. Effects of a functional COMT polymorphism on brain anatomy and cognitive function in adults with velo-cardio-facial syndrome. Psychol. Med. 38 (1), 89–100. Williams, N.M., Norton, N., Williams, H., Ekholm, B., Hamshere, M.L., Lindblom, Y., Chowdari, K.V., Cardno, A.G., Zammit, S., Jones, L.A., Murphy, K.C., Sanders, R.D., McCarthy, G., Gray, M.Y., Jones, G., Holmans,
G.M. Tan et al. / Schizophrenia Research 115 (2009) 173–181 P., Nimgaonkar, V., Adolfson, R., Osby, U., Terenius, L., Sedvall, G., O'Donovan, M.C., Owen, M.J., 2003. A systematic genomewide linkage study in 353 sib pairs with schizophrenia. Am. J. Hum. Genet. 73 (6), 1355–1367. Witthaus H., Kaufmann C., Bohner G., Ozgürdal S., Gudlowski Y., Gallinat J., Ruhrmann S., Brüne M., Heinz A., Klingebiel R., Juckel G., 2009. Gray matter abnormalities in subjects at ultra-high risk for schizophrenia and
181
first-episode schizophrenic patients compared to healthy controls. Psychiatry Res. 173, 163–169. Yan, W., Jacobsen, L.K., Krasnewich, D.M., Guan, X.Y., Lenane, M.C., Paul, S.P., Dalwadi, H.N., Long, R.T., Kumra, S., Martin, B.M., Scrambler, P.J., Trent, J.M., Sidrandky, E., Ginns, E.I., Rapoport, J.L., 1998. Chromosome 22q11.2 interstizial deletions among childhood onset schizophrenics and ‘multidimensionally impaired’. Am. J. Med. Genet. 81, 41–43.