Abstract
Background
Cerebellum is not only responsible for motor functions but is also involving in cognitive and emotional processes (Strick et al. 2009). Recent advances in imaging technologies allow us to examine the cerebellum in a much more systematic way including functional parcellations of cerebellum (e.g. 7-Network parcellation based on the resting state functional connectivity, Buckner et al., 2011; and task-based functional parcellation, King et al., 2019). These functional parcellations further uncover the important roles of cerebellum in non-motor functions.
In the past two decades, altered cerebellar structure and function have been observed in schizophrenia patients, with the anterior (lobules III-V) and the posterior cerebellum (lobules VI, VIIa, Crus I, and Crus II, VIII) as regions commonly reported (Bernard & Mittal 2015). Using the 7-Network task-free parcellation, a recent multisite mega-analysis reported that patients with schizophrenia exhibited robust grey matter reduction in cerebellum (Moberget et al., 2018).
The present study aimed to examine the cerebellar grey matter volumes in schizophrenia patients using anatomical, task-free connectivity-based and task-based functional parcellations. We also explored how the cerebellar volumes changes along with age in schizophrenia patients and healthy controls.
Methods
Twenty-nine patients with schizophrenia (SCZ, mean age = 22.0±2.5 years) and 55 healthy controls (HC, mean age = 23.5±3.6 years) were recruited to undertake structural magnetic resonance imaging (MRI) brain scan on a GE 750 scanner. Cerebellar volumes were measured by a high-resolution Spatially Unbiased Infratentorial (SUIT) toolbox (Diedrichsen, 2006). Regional cerebellar grey matter volumes were calculated for each participant based on anatomical (SUIT atlas) and functional atlases (7 regions of task-free parcellation; 10 regions of task-based parcellation) of the cerebellum. The group comparisons on regional cerebellar volumes were conducted in SPSS v19.0, taking total ICV as a covariate. The correlations between age and cerebellar volumes of those regions showing significant group differences were conducted. The significance threshold was set at p < 0.05.
Results
Regarding the anatomical atlas, SCZ patients exhibited reduced grey matter volumes of the superior posterior cerebellum, mainly in lobules Crus II, VIIb and VIIIa. Regarding the task-free parcellation, SCZ patients exhibited smaller volumes in cerebellar regions showing functional connectivity with frontoparietal network and default mode network. Regarding the task-based parcellation, smaller volumes of cerebellar regions that activated during verbal fluency task (ROI9) was observed in SCZ patients.
Correlation analyses between age and cerebellar volumes of those regions with significant group differences further reported negative correlations between age and lobule Crus II in both SCZ patients and HC (SCZ: r = -0.44, p < 0.05; HC: r = -0.43, p = 0.001). Moreover, negative correlations between age and cerebellar regions showing functional connectivity with frontoparietal network was observed in HC only (r = -0.30, p < 0.05), whereas negative correlations between age and lobules VIIb (r = -0.48, p < 0.01), VIIIa (r = -0.39, p < 0.05) and ROI9 of task-based atlas (r = -0.43, p < 0.05) were observed in SCZ patients.
Discussion
Our findings showed reduced cerebellar grey matter volumes and the abnormal age-related changes of cerebellum in SCZ patients, especially in the regions related to frontoparietal network and cognitive function. These findings may highlight an important role of cerebellum underlying the mechanisms of cognitive dysfunction in mental disorders.
Effects of aging on emotion recognition from dynamic multimodal expressions and vocalizations