Patterns of regional gray matter loss at different stages of schizophrenia: A multisite, cross-sectional VBM study in first-episode and chronic illness.
Patterns of regional gray matter loss at different stages of schizophrenia: A multisite, cross-sectional VBM study in first-episode and chronic illness.
复制标题
DOI：
10.1016/j.nicl.2016.06.002
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Busatto GF
中科院分区：
文献类型：
--
作者：
Torres US;Duran FL;Schaufelberger MS;Crippa JA;Louzã MR;Sallet PC;Kanegusuku CY;Elkis H;Gattaz WF;Bassitt DP;Zuardi AW;Hallak JE;Leite CC;Castro CC;Santos AC;Murray RM;Busatto GF
背景：磁共振成像(MRI)研究已多次证实精神分裂症患者的脑结构异常，但目前尚不清楚这些异常是静态的还是进行性的。虽然纵向MRI研究传统上用于评估精神分裂症患者大脑异常进展的问题，但直接将首发和慢性精神分裂症患者与健康对照组进行比较的横断面神经成像研究的信息也可能有助于进一步澄清这一问题。随着最近对结合多个中心的结构性MRI数据以提高统计能力的多部位巨型分析的兴趣，本研究进行了基于多部位体素的形态计量学(VBM)研究，以检查根据疾病的不同阶段大脑结构变化的模式，并确定这些结构异常中的哪些(如果有的话)将与潜在的临床调节因素特别相关，包括抗精神病药物的累积暴露、发病年龄、病程和总体疾病严重程度。方法：我们收集了大样本精神分裂症患者(161例，其中慢性精神分裂症患者99例，首发精神分裂症患者62例)和对照组(151例)，这些样本来自之前在巴西同一地理区域进行的四项形态计量学MRI研究(1.5T)。图像处理和分析使用统计参数映射(SPM8)软件，通过指数李代数(Dartel)算法进行差分解剖配准。使用协方差的一般线性模型分析(ANCOVA)通过全脑体素比较来研究群体对区域灰质(GM)体积的影响，总是将GM总体积、扫描方案、年龄和性别作为干扰变量。最后，对上述临床调节指标与局部和整体脑体积进行相关性分析。结果：首发精神分裂症受试者在仅在小体积校正(SVC)分析中识别的受限脑区域网络(包括脑岛、颞缘结构和纹状体)中显示出与对照组相比的微妙的体积缺陷(P<0.05)。另一方面，慢性精神分裂症患者的双侧额叶上、下、眶皮质、右侧额中皮质、双侧扣带前皮质、双侧胰岛和右侧上、中颞叶皮质均表现出广泛的GM体积减小模式(P<0.05，全脑未校正)。在连体(首发和慢性)精神分裂症组的SVC分析中，其中几个脑区的GM体积与发病年龄直接相关。病程和相对GM体积之间也存在显著负相关的广泛病灶，但在考虑了发病年龄的影响后，这种发现仅在右侧背外侧前额叶皮质有显著意义。最后，在精神分裂症患者中，抗精神病药物的终生累积暴露与总GM和白质体积之间存在显著的负相关，但在任何特定脑区的抗精神病药物使用指数与相对GM体积之间没有发现显著的相关性。结论：与首发精神分裂症患者相比，慢性精神分裂症患者与诊断精神分裂症相关的脑部改变更为普遍。我们的发现还表明，在发病年龄较早的病例中，相对GM体积缺陷可能更大(假设更严重)，并随特定额叶大脑区域的病程变化而变化。最后，我们的结果突显了持续使用抗精神病药物对精神分裂症患者结构性脑异常的潜在复杂影响，因为我们发现抗精神病药物的累积剂量影响全球脑体积，而不是选择性地在额叶-颞区。与首发精神分裂症相比，慢性精神分裂症患者的脑部结构改变更为普遍。在发病年龄较早的病例中，地区性GM缺陷可能更大。病程似乎对某些特定的额叶结构变化有影响。抗精神病药物似乎影响全球的脑容量，而不是区域性的。
Background: Structural brain abnormalities in schizophrenia have been repeatedly demonstrated in magnetic resonance imaging (MRI) studies, but it remains unclear whether these are static or progressive in nature. While longitudinal MRI studies have been traditionally used to assess the issue of progression of brain abnormalities in schizophrenia, information from cross-sectional neuroimaging studies directly comparing first-episode and chronic schizophrenia patients to healthy controls may also be useful to further clarify this issue. With the recent interest in multisite mega-analyses combining structural MRI data from multiple centers aiming at increased statistical power, the present multisite voxel-based morphometry (VBM) study was carried out to examine patterns of brain structural changes according to the different stages of illness and to ascertain which (if any) of such structural abnormalities would be specifically correlated to potential clinical moderators, including cumulative exposure to antipsychotics, age of onset, illness duration and overall illness severity. Methods: We gathered a large sample of schizophrenia patients (161, being 99 chronic and 62 first-episode) and controls (151) from four previous morphometric MRI studies (1.5 T) carried out in the same geographical region of Brazil. Image processing and analyses were conducted using Statistical Parametric Mapping (SPM8) software with the diffeomorphic anatomical registration through exponentiated Lie algebra (DARTEL) algorithm. Group effects on regional gray matter (GM) volumes were investigated through whole-brain voxel-wise comparisons using General Linear Model Analysis of Co-variance (ANCOVA), always including total GM volume, scan protocol, age and gender as nuisance variables. Finally, correlation analyses were performed between the aforementioned clinical moderators and regional and global brain volumes. Results: First-episode schizophrenia subjects displayed subtle volumetric deficits relative to controls in a circumscribed brain regional network identified only in small volume-corrected (SVC) analyses (p < 0.05, FWE-corrected), including the insula, temporolimbic structures and striatum. Chronic schizophrenia patients, on the other hand, demonstrated an extensive pattern of regional GM volume decreases relative to controls, involving bilateral superior, inferior and orbital frontal cortices, right middle frontal cortex, bilateral anterior cingulate cortices, bilateral insulae and right superior and middle temporal cortices (p < 0.05, FWE-corrected over the whole brain). GM volumes in several of those brain regions were directly correlated with age of disease onset on SVC analyses for conjoined (first-episode and chronic) schizophrenia groups. There were also widespread foci of significant negative correlation between duration of illness and relative GM volumes, but such findings remained significant only for the right dorsolateral prefrontal cortex after accounting for the influence of age of disease onset. Finally, significant negative correlations were detected between life-time cumulative exposure to antipsychotics and total GM and white matter volumes in schizophrenia patients, but no significant relationship was found between indices of antipsychotic usage and relative GM volume in any specific brain region. Conclusion: The above data indicate that brain changes associated with the diagnosis of schizophrenia are more widespread in chronic schizophrenia compared to first-episode patients. Our findings also suggest that relative GM volume deficits may be greater in (presumably more severe) cases with earlier age of onset, as well as varying as a function of illness duration in specific frontal brain regions. Finally, our results highlight the potentially complex effects of the continued use of antipsychotic drugs on structural brain abnormalities in schizophrenia, as we found that cumulative doses of antipsychotics affected brain volumes globally rather than selectively on frontal-temporal regions. Structural brain changes are more widespread in chronic than first-episode schizophrenia. Regional GM deficits may be greater in cases with earlier age of onset. Illness duration seems to impact in some specific frontal structural brain changes. Antipsychotics seem to affect brain volumes globally rather than regionally.