ساختار مغزی در اسکیزوفرنی بالاتر هوشی

The current study aims to fill a gap in the knowledge base by investigating the structural brain characteristics of individuals with schizophrenia and superior intellectual abilities. Subcortical volumes, cortical thickness and cortical surface area were examined in intellectually normal and intellectually superior participants with schizophrenia and their IQ-matched healthy controls, as well as in intellectually low schizophrenia participants. We replicated significant diagnostic group effects on hippocampal and ventricular size after correction for multiple comparisons. There were no statistically significant effects of intellectual level or of the interaction between diagnostic group and intellectual level. Effect sizes indicated that differences between schizophrenia and healthy control participants were of similar magnitude at both intellectual levels for all three types of morphological data. A secondary analysis within the schizophrenia group, including participants with low intellectual abilities, yielded numerical, but no statistically significant differences on any structural brain measure. The present findings indicate that the brain structure abnormalities in schizophrenia are present at all intellectual levels, and individuals with schizophrenia and superior intellectual abilities have brain structure abnormalities of the same magnitude as individuals with schizophrenia and normal intellectual abilities.

Schizophrenia (SZ) is associated with impaired cognition as well as structural brain abnormalities. Gray matter abnormalities include cortical volume reductions and cortical thinning in frontal and temporal regions (Kuperberg et al., 2003, Nesvåg et al., 2008, Arnone et al., 2009 and Ellison-Wright and Bullmore, 2010). Subcortical structural abnormalities such as increased ventricular and basal ganglia volumes, and reduced hippocampal, amygdalar and thalamic volumes, have been consistently reported across studies (Honea et al., 2005, Arnone et al., 2009 and Ellison-Wright and Bullmore, 2010). A meta-analysis of volumetric brain alterations in over 18,000 participants (Haijma et al., 2013) found that intracranial and total brain volumes were significantly reduced in SZ compared with healthy control participants (HC). The largest effect sizes were seen for gray matter structures. This meta-analysis noted that there was a trend towards larger volume reductions in studies that did not match SZ and HC groups on IQ, indicating that studies of brain structure characteristics in SZ should take IQ into account. Although impaired cognition is considered an important feature of SZ, about one quarter of individuals with SZ present with near-normal scores on neuropsychological tests (Kremen et al., 2000, Weickert et al., 2000 and Rund et al., 2006). This has led to a longstanding debate on whether it is possible to have schizophrenia and be neuropsychologically normal (Palmer et al., 1997 and Wilk et al., 2005). Research evidence indicates that it is not possible, at least not on a group level. Individuals with SZ who score within the normal range on neuropsychological tests have reductions of one kind or another, such as deficits in attention and executive functioning in spite of preserved IQ (Weickert et al., 2000), or lower processing speed and memory scores compared with IQ-matched HC participants (Wilk et al., 2005). Neurocognitive decrements are present in practically all SZ cases (Keefe et al., 2005), even in individuals with SZ and superior intellectual abilities (MacCabe et al., 2012 and Gray et al., 2013). In a recent study (Vaskinn et al., 2014), we found that individuals diagnosed with SZ with IQ scores>120 had the same magnitude of neurocognitive decrements as those with normal or low intelligence when compared with IQ-matched HC individuals. Also, symptom profiles and functional deficits were similar across the three IQ strata. To our knowledge, the brain characteristics of people with SZ and superior intellectual abilities (IQ>120) have not previously been investigated, although a handful of studies have investigated SZ samples defined as cognitively preserved or neuropsychologically near-normal. There are several ways to investigate brain structure characteristics in (intellectually superior) SZ. One approach focuses on anomalies in brain structure compared with HC individuals (who can be matched on intelligence or not)(Wexler et al., 2009, Cobia et al., 2011 and Ortiz-Gil et al., 2011), whereas another looks at differences in brain structure within the SZ population (across cognitive or intellectual level) (Wexler et al., 2009, Cobia et al., 2011, Ortiz-Gil et al., 2011 and Ayesa-Arriola et al., 2013). Wexler et al. (2009) found that neuropsychologically near-normal SZ participants had markedly less gray matter volume and larger third ventricles than HC participants in spite of almost intact cognition. A neuropsychologically impaired SZ group had similar gray matter reductions, but in addition had smaller white matter volumes and larger lateral ventricles compared with the HC group. Cobia et al. (2011) found no significant cortical thinning patterns in neuropsychologically near-normal SZ compared with HC participants, but effect sizes indicated mild cortical thinning with moderate effects for several brain areas, in particular bilateral frontal and left temporal regions. Ortiz-Gil et al. (2011) found similar brain volume and gray matter volume reductions and ventricular enlargement in SZ participants who were cognitively preserved and cognitively impaired. Similarly, a recent longitudinal study of first episode psychosis found no morphometric differences between cognitively preserved and cognitively impaired participants at baseline, although greater volume decrease for parietal tissue volume appeared for the cognitively impaired subgroup over a 3-year period (Ayesa-Arriola et al., 2013). In summary, the literature on brain structure in neuropsychologically near-normal/cognitively preserved SZ suggests structural abnormalities, and that some of these abnormalities may be similar to the ones seen in cognitively impaired SZ participants. The main goal of the present study is to explore whether SZ with superior intellectual abilities is characterized by structural brain abnormalities. This is a follow-up of our recent neuropsychological study (Vaskinn et al., 2014), using an overlapping, but smaller, sample for which structural magnetic resonance imaging (MRI) data were available. Here we examine the brain characteristics of intellectually superior SZ using the two above-mentioned approaches, i.e., by comparing them to HC participants as well as to SZ participants with low or normal intellectual abilities. Earlier studies were limited to either subcortical volumes or cortical thickness. We performed a comprehensive investigation of both cortical thickness and surface area, which together constitute cortical volume, and subcortical volumes, and compare three sets of structural brain measures in SZ and HC participants: (a) subcortical volumetric measures, (b) surface-based measures of cortical area, and (c) surface-based measures of cortical thickness. First, we investigated the presence of MRI abnormalities in SZ versus HC participants matched for level of intelligence in participants with normal and superior intellectual abilities. Previous MRI studies on neuropsychologically near-normal SZ samples (Wexler et al., 2009 and Cobia et al., 2011) did not perform such matching. We ask whether intellectually superior SZ participants have abnormalities compared with IQ-matched HC participants, and whether abnormalities, if present, are of the same magnitude as in intellectually normal SZ. Second, we compare MRI characteristics in SZ participants with low, normal and superior intellectual abilities. Based on the findings reviewed above of brain abnormalities in SZ participants with near-normal neuropsychological scores (Wexler et al., 2009) and of similar neurocognitive decrements and symptom and function profiles in SZ participants across the IQ spectrum (Vaskinn et al., 2014), we expected to find relative brain abnormalities of the same degree in SZ participants, regardless of their intellectual level, for all three types of structural brain measures. Because results regarding brain structure in SZ participants with different cognitive abilities have been mixed, we had no specific hypothesis for the intra-diagnostic comparison.

We found significant SZ–HC differences for 2 of the total 25 variables: hippocampus (F(1,139)=12.8, p<0.001, η2=0.09) and total ventricular size (F(1,139)=8.7, p=0.004, η2=0.06). There were no significant effects of intellectual level or significant interaction effects. Effect sizes for SZ–HC differences within each intellectual level are reported in Table 2 and Table 3. For the intellectually superior level, these were medium-sized for pallidum (d=0.48), left frontal cortical area (d=−0.50) and left frontal (d=−0.46) and right parietal cortical thickness (d=−0.56). None of the effect sizes differed significantly from the corresponding effect sizes in the intellectually normal level (pallidum: z=0.30, p=0.382; left frontal cortical area: z=0.79, p=0.224; left frontal cortical thickness: z=0.36, p=0.359; right parietal cortical thickness: z=0.54, p=0.295). Analyses of group differences within the SZ sample yielded no significant effects after Bonferroni corrections for multiple testing. Strongest effects were seen for total ventricular size (F(2,69)=2.8, p=0.071, η2=0.08) and frontal (F(2,69)=3.2, p=0.044, η2=0.09) and temporal (F(2,69)=4.7, p=0.013, η2=0.13) lobe cortical thickness in the left hemisphere (see Fig. 1, Fig. 2 and Fig. 3). The intellectually low SZ group diverged from the two other groups with larger ventricles and thinner cortex, albeit not significantly so after correction for multiple testing.