Large-scale analysis of structural brain asymmetries in schizophrenia via the ENIGMA consortium
Dick Schijven, Merel C. Postema, Masaki Fukunaga, Junya Matsumoto, Kenichiro Miura, Sonja M. C. de Zwarte, Neeltje E.m. van Haren, Wiepke Cahn, Hilleke E. Hulshoff Pol, René S. Kahn, Rosa Ayesa‐Arriola, Víctor Ortiz‐García de la Foz, Diana Tordesillas‐Gutiérrez, Javier Vázquez-Bourgón, Benedicto Crespo‐Facorro, Dag Alnæs, Andreas Dahl, Lars T. Westlye, Ingrid Agartz, Ole A. Andreassen, Erik G. Jönsson, Peter Kochunov, Jason Bruggemann, Stanley V. Catts, Patricia T. Michie, Bryan Mowry, Yann Quidé, Paul E. Rasser, Ulrich Schall, Rodney J. Scott, Vaughan J. Carr, Melissa J. Green, Frans Henskens, Carmel M. Loughland, Christos Pantelis, Cynthia Shannon Weickert, Thomas W. Weickert, Lieuwe de Haan, Katharina Brosch, Julia‐Katharina Pfarr, Kai G. Ringwald, Frederike Stein, Andreas Jansen, Tilo Kircher, Igor Nenadić, Bernd Krämer, Oliver Gruber, Theodore D. Satterthwaite, Juan Bustillo, Daniel H. Mathalon, Adrian Preda, Vince D. Calhoun, Judith M. Ford, Steven G. Potkin, Jing Chen, Yunlong Tan, Zhiren Wang, Hong Xiang, Fengmei Fan, Fabio Bernardoni, Stefan Ehrlich, Paola Fuentes‐Claramonte, María Ángeles García‐León, Amalia Guerrero‐Pedraza, Raymond Salvador, Salvador Sarró, Edith Pomarol‐Clotet, Valentina Ciullo, Fabrizio Piras, Daniela Vecchio, Nerisa Banaj, Gianfranco Spalletta, Stijn Michielse, Thérèse van Amelsvoort, Erin W. Dickie, Aristotle N. Voineskos, Kang Sim, Simone Ciufolini, Paola Dazzan, Robin Murray, Woo‐Sung Kim, Young‐Chul Chung, Christina Andreou, André Schmidt, Stefan Borgwardt, Andrew M. McIntosh, Heather C. Whalley, Stephen M. Lawrie, Stefan S. du Plessis, Hilmar Luckhoff, Freda Scheffler, Robin Emsley, Dominik Grotegerd, Rebekka Lencer, Udo Dannlowski, Jesse T. Edmond, Kelly Rootes-Murdy, Julia M. Stephen, Andrew R. Mayer, Linda A. Antonucci
This large-scale study from the ENIGMA consortium provides the most definitive look to date at how schizophrenia relates to the physical "lopsidedness" of the human brain. By analyzing over 11,000 brain scans, the researchers found that while schizophrenia is associated with subtle shifts in brain asymmetry—particularly in regions tied to language and emotional control—these changes are far smaller than previously reported in smaller, less robust studies.
TL;DR
Using a massive global dataset (N = 11,095), researchers identified subtle but significant reductions in left-hemisphere cortical thickness in the language-processing and emotional-control regions of people with schizophrenia. While individual regional differences are tiny, a multivariate approach reveals that the overall pattern of brain asymmetry accounts for roughly 7% of the variance between patients and healthy controls.
What Problem Does This Paper Address?
The human brain is naturally asymmetrical; for example, the left hemisphere is typically dominant for language in most people. For decades, scientists have hypothesized that schizophrenia involves a "disruption" or even a reversal of this normal lateralization.
However, previous research has been plagued by "equivocal findings." Most studies used small sample sizes (often fewer than 100 participants) and varied widely in their imaging methods. This led to a "file drawer problem" where small, potentially random fluctuations were reported as significant findings that others couldn't replicate. The authors sought to provide a "ground truth" by using a massive, harmonized dataset to see which (if any) asymmetry differences actually hold up at scale.
Methods
The researchers utilized a meta-analytical approach coordinated through the ENIGMA (Enhancing Neuro Imaging Genetics through Meta-Analysis) Consortium.
Standardized Pipeline: To avoid the "noise" of different software, all 46 datasets were processed using a single image analysis protocol (FreeSurfer) to measure cortical thickness, surface area, and subcortical volumes.
Asymmetry Index (AI): For every bilateral brain region, they calculated an AI using the formula: $AI = (L - R) / ((L + R) / 2)$. A positive value indicates left-dominance; a negative value indicates right-dominance.
Statistical Rigor: They performed random-effects meta-analyses across datasets to find common trends while controlling for age and sex.
Multivariate Analysis: In a subset of 2,029 individuals where raw data was available, they used MANCOVA to look at the entire pattern of brain asymmetries simultaneously, rather than looking at one region at a time.
Datasets
The study is a "mega-analysis" of 46 separate datasets from around the world, totaling 11,095 individuals (5,080 people with schizophrenia and 6,015 healthy controls). This includes well-known cohorts like COBRE, FBIRN, and various university-led studies across Europe, North America, Asia, and Africa. No new benchmarks were introduced, but the sheer scale represents the largest study of its kind in psychiatric neuroimaging.
Key Insights and Ideas
The "Subtlety" Realization: The effect sizes (Cohen’s $d$) for individual regions were very small (less than 0.1). This is a crucial insight: it suggests that schizophrenia isn't characterized by a massive "flip" in brain structure, but rather by very fine-grained deviations that require thousands of samples to detect reliably.
Left-Side Vulnerability: The significant findings in the rostral anterior cingulate and middle temporal gyrus were both driven by a thinner cortex in the left hemisphere specifically.
The Language Link: The middle temporal gyrus is a core hub for language. Finding reduced left-side thickness supports the long-standing theory that schizophrenia involves a failure of the left hemisphere to establish its typical dominance for language processing.
Age-Dependent Pallidum Shifts: Interestingly, the pallidum (a subcortical structure involved in reward and movement) showed a stronger leftward asymmetry in older patients compared to older controls, suggesting that some asymmetry changes might be related to the progression of the disease or aging with the disorder.
Key Results
Cortical Thickness: Two regions survived strict multiple-testing correction:
Multivariate Impact: When looking at all 76 asymmetry measures at once, the model was highly significant ($p = 1.25 \times 10^{-5}$), explaining 7% of the variance between cases and controls.
Clinical Variables: Surprisingly, these asymmetry shifts were not significantly associated with the use of antipsychotic medication, duration of illness, or the severity of symptoms (like hallucinations).
Limitations and Future Work
Cross-Sectional Design: Because the data is a "snapshot" in time, the authors cannot say for certain whether these asymmetries are present at birth or if they develop over the course of the illness.
Macro vs. Micro: The study used macro-anatomical atlases. The authors suggest that the "real" action might be happening at the molecular or cellular level (cytoarchitectonics), which MRI cannot see.
Future Directions: The authors call for longitudinal studies to track how these asymmetries change from the first psychotic episode through old age. They also suggest that vertex-wise mapping (looking at thousands of individual points on the brain surface rather than large predefined regions) might reveal more focal "hotspots" of asymmetry.
Summary updated
Abstract
Left-right asymmetry is an important organizing feature of the healthy brain that may be altered in schizophrenia, but most studies have used relatively small samples and heterogeneous approaches, resulting in equivocal findings. We carried out the largest case-control study of structural brain asymmetries in schizophrenia, with MRI data from 5,080 affected individuals and 6,015 controls across 46 datasets, using a single image analysis protocol. Asymmetry indexes were calculated for global and regional cortical thickness, surface area, and subcortical volume measures. Differences of asymmetry were calculated between affected individuals and controls per dataset, and effect sizes were meta-analyzed across datasets. Small average case-control differences were observed for thickness asymmetries of the rostral anterior cingulate and the middle temporal gyrus, both driven by thinner left-hemispheric cortices in schizophrenia. Analyses of these asymmetries with respect to the use of antipsychotic medication and other clinical variables did not show any significant associations. Assessment of age- and sex-specific effects revealed a stronger average leftward asymmetry of pallidum volume between older cases and controls. Case-control differences in a multivariate context were assessed in a subset of the data (N = 2,029), which revealed that 7% of the variance across all structural asymmetries was explained by case-control status. Subtle case-control differences of brain macrostructural asymmetry may reflect differences at the molecular, cytoarchitectonic, or circuit levels that have functional relevance for the disorder. Reduced left middle temporal cortical thickness is consistent with altered left-hemisphere language network organization in schizophrenia.
BibTeX
@inproceedings{schijven2023large,
author = {Dick Schijven and Merel C. Postema and Masaki Fukunaga and Junya Matsumoto and Kenichiro Miura and Sonja M. C. de Zwarte and Neeltje E.M. van Haren and Wiepke Cahn and Hilleke E. Hulshoff Pol and René S. Kahn and Rosa Ayesa‐Arriola and Víctor Ortiz‐García de la Foz and Diana Tordesillas‐Gutiérrez and Javier Vázquez-Bourgón and Benedicto Crespo‐Facorro and Dag Alnæs and Andreas Dahl and Lars T. Westlye and Ingrid Agartz and Ole A. Andreassen and Erik G. Jönsson and Peter Kochunov and Jason Bruggemann and Stanley V. Catts and Patricia T. Michie and Bryan Mowry and Yann Quidé and Paul E. Rasser and Ulrich Schall and Rodney J. Scott and Vaughan J. Carr and Melissa J. Green and Frans Henskens and Carmel M. Loughland and Christos Pantelis and Cynthia Shannon Weickert and Thomas W. Weickert and Lieuwe de Haan and Katharina Brosch and Julia‐Katharina Pfarr and Kai G. Ringwald and Frederike Stein and Andreas Jansen and Tilo Kircher and Igor Nenadić and Bernd Krämer and Oliver Gruber and Theodore D. Satterthwaite and Juan Bustillo and Daniel H. Mathalon and Adrian Preda and Vince D. Calhoun and Judith M. Ford and Steven G. Potkin and Jing Chen and Yunlong Tan and Zhiren Wang and Hong Xiang and Fengmei Fan and Fabio Bernardoni and Stefan Ehrlich and Paola Fuentes‐Claramonte and María Ángeles García‐León and Amalia Guerrero‐Pedraza and Raymond Salvador and Salvador Sarró and Edith Pomarol‐Clotet and Valentina Ciullo and Fabrizio Piras and Daniela Vecchio and Nerisa Banaj and Gianfranco Spalletta and Stijn Michielse and Thérèse van Amelsvoort and Erin W. Dickie and Aristotle N. Voineskos and Kang Sim and Simone Ciufolini and Paola Dazzan and Robin Murray and Woo‐Sung Kim and Young‐Chul Chung and Christina Andreou and André Schmidt and Stefan Borgwardt and Andrew M. McIntosh and Heather C. Whalley and Stephen M. Lawrie and Stefan S. du Plessis and Hilmar Luckhoff and Freda Scheffler and Robin Emsley and Dominik Grotegerd and Rebekka Lencer and Udo Dannlowski and Jesse T. Edmond and Kelly Rootes-Murdy and Julia M. Stephen and Andrew R. Mayer and Linda A. Antonucci},
title = {{Large-scale analysis of structural brain asymmetries in schizophrenia via the ENIGMA consortium}},
booktitle = {Proceedings of the National Academy of Sciences},
year = {2023},
doi = {10.1073/pnas.2213880120},
}