Changes in the Functional Connectivity of Cerebellar Subregions in Anti- N -Methyl-d-Aspartate Receptor Encephalitis

Abstract

Background:

In anti-N-methyl-d-aspartate receptor (anti-NMDAR) encephalitis, the cerebellum, a characteristic brain region, exhibits abnormal functioning and structure. However, the relationship between resting-state activities in the cerebellar subregions and core symptoms of cognitive dysfunction is unclear.

Methods:

In this study, a total of 23 patients with anti-NMDAR encephalitis and 23 healthy controls were included, and 19 patients (mean age 30.05 ± 13.03 years) and 21 healthy controls (mean age 28.42 ± 9.47 years) were enrolled based on image quality and head movement. Seed-based functional connectivity (FC) analysis was used to investigate changes in FC of cerebellar subregions, and the association between changes in cerebellar subregion FC and cognitive dysfunction was explored in conjunction with cognitive scales.

Results:

Patients with anti-NMDAR encephalitis exhibited poorer cognitive performance than the healthy controls did. In the patient group, the FC between the right cerebellar Crus I and the left thalamus was significantly reduced and showed a negative correlation with disease duration (p < 0.05); however, it showed positive correlations with attention and information processing speed according to Symbol Digit Modalities Test (p < 0.01), as well as with verbal learning and memory (according to California Verbal Learning Test (CVLT; p < 0.05). The FCs between the left cerebellar Crus Ia and the right medial superior frontal gyrus, as well as between the left cerebellar Crus Ib and the right middle temporal gyrus, were decreased. The connectivity between the cerebellar vermis Crus II and the left putamen, along with the opercular part of the inferior frontal gyrus, showed a significant reduction. Decreased FC between the left cerebellar X lobule and the left putamen showed positive correlation with CVLT (p < 0.01).

Conclusion:

Cerebellar subregion and brain abnormalities FC in patients with anti-NMDAR encephalitis are linked to cognitive deficits. These results provide insights into the neurophysiological mechanisms underlying cognitive impairment in these patients.

Impact Statement

This study reveals significant alterations in functional connectivity (FC) between cerebellar subregions and specific brain regions (e.g., thalamus, default mode network, putamen) in patients with anti-N-methyl-d-aspartate receptor (anti-NMDAR) encephalitis, which are closely associated with cognitive deficits in attention, memory, and language study. Using seed-based resting-state functional magnetic resonance imaging analysis focused on cerebellar subregions, it uncovers changes in FC in these patients. The findings complement previous research and highlight the critical role of the cerebellum in higher cognitive regulation, providing novel neuroimaging insights and evidence for a better understanding of the neurophysiological mechanisms underlying cognitive dysfunction in anti-NMDAR encephalitis.

Introduction

Anti-N-methyl-d-aspartate receptor (anti-NMDAR) encephalitis is a rare neuroimmune condition recognized as the most prevalent form of autoimmune encephalitis, primarily affecting young females (Dalmau and Graus, 2018; Tanguturi et al., 2019). It was first identified by Dalmau and colleagues in 2007 (Dalmau et al., 2007). Patients experience acute or subacute onset with various severe neuropsychiatric symptoms, and approximately 75% of patients experience irreversible neurological dysfunction, primarily in cognitive functions, including memory, attention, and executive control (McKeon et al., 2018; Zhang et al., 2018). Even during the onset of the disorder, approximately 50%–70% of patients have normal performance on conventional brain magnetic resonance imaging (MRI), and a few patients may present with nonspecific small white matter lesions (Finke et al., 2016; Xu et al., 2022). There are limitations to the interpretation of cognitive deficits using conventional imaging; thus, new imaging evidence is required to provide a deeper understanding.

Current research indicates that the pathogenic mechanism of anti-NMDAR encephalitis involves the production of autoantibodies called IgGs, which specifically bind to the GluN1 subunit of the NMDAR and bring about a notable reduction in the NMDAR density on the synaptic membrane (Dalmau and Graus, 2018; Hughes et al., 2010). The decreased NMDAR density directly weakens the efficiency of synaptic transmission and simultaneously inhibits the critical plasticity process of long-term potentiation (LTP), which is universally regarded as the primary crucial molecular mechanism for the formation of advanced cognitive functions such as learning and memory (Lau and Zukin, 2007).

Modern neuroimaging has yielded many significant findings that reveal the mechanisms of cognitive dysfunction in anti-NMDAR encephalitis. Functional neuroimaging studies have indicated a decrease in the amplitude of low-frequency fluctuation in the bilateral cerebellum, left precuneus, and bilateral posterior cingulate gyrus, accompanied by enhanced functional connectivity (FC) between the bilateral posterior cingulate gyrus and multiple other brain regions, such as the right posterior central gyrus, bilateral lingual gyrus, calcarine, fusiform gyrus, and cuneus, which may cause cognitive and emotional impairments in patients with anti-NMDAR encephalitis (Cai et al., 2020). Structural neuroimaging researches have indicated that in multivariate pattern analyses of anti-NMDAR encephalitis, brain regions with greater fractional anisotropy and mean diffusivity weight values overlap in the temporal lobe, cerebellum, and hippocampus, suggesting that the hippocampus, temporal lobe, and cerebellum are possibly the characteristic brain regions of anti-NMDAR encephalitis (Liang et al., 2020). In some patients, diffuse brain atrophy can be reversed, whereas cerebellar atrophy is irreversible in all patients. The reduction in cerebellar volume reflects the disease burden and degree of progression and correlates with poor outcomes in multiple functional brain regions (Lee et al., 2022).

The primary cells of cerebellar neurons (granulosa cells) are rich in NMDARs (Lee et al., 2022), and cerebellar LTP requires NMDAR activation, which is essential for synaptic LTP between parallel fibers and Purkinje cells (Armano et al., 2000; Mapelli et al., 2015). Consequently, it is reasonable to speculate that the cerebellum is more indicative of functional impairment in NMDAR and that cerebellar synaptic dysfunction arising from this impairment could potentially serve as a pivotal contributor to the neurological deficiencies observed in individuals suffering from anti-NMDAR encephalitis.

The cerebellum is believed to participate in motor regulation, and with the advancements in neuroimaging studies, the role of the cerebellum in the regulation of higher cognition has gradually gained attention (Schmahmann et al., 2019). Schmahmann and colleagues found that patients with cerebellar diseases experience impaired executive function, spatial-visual cognition, language ability, and emotion regulation and introduced the clinical concept of “cerebellar cognitive-emotional syndrome,” offering clinical support for the cerebellum’s involvement in advanced cognitive processes (Schmahmann and Sherman, 1998). Based on functional heterogeneity, Lobules I–V comprise the anterior lobe of the cerebellum, the posterior lobe comprises Lobules VI–IX, Lobule VII is segmented into cerebellum Crus I and II, and VIIb. Lobule X belongs to the flocculonodular lobe (Schmahmann et al., 1999). Motor function is primarily associated with the anterior lobe of the cerebellum, whereas cognitive and emotional functions are associated with the posterior lobe (Schmahmann et al., 2009).

Although relevant studies have indicated that the structure and function of the cerebellum in anti-NMDAR encephalitis are abnormal, there has been limited research on functional differentiation in cerebellar subregions using functional MRI (fMRI) data specific to this condition. This gap may have overlooked the potential significance of functional differences within these subregions. The present study endeavors to explore the impact of abnormalities in the FC of cerebellar subregions affect cognitive impairment in patients with anti-NMDAR encephalitis. FC analysis exploits the temporal synchronization of blood oxygen level-dependent signals to reveal functional interactions between different brain areas (Etkin et al., 2019; Li et al., 2020). Resting-state fMRI (rs-fMRI) offers an advantage for patients with impaired cognitive functions as it does not necessitate specific tasks. Even in the absence of evident abnormalities on conventional MRI, rs-fMRI can be used to detect abnormal functional brain activity. We aimed to use rs-fMRI to analyze cerebellar subregions and brain FC alterations in patients with anti-NMDAR encephalitis, exploring the relationships between these changes and neurological dysfunction, along with clinical parameters.

Materials and Methods

Participants

We enrolled 23 patients who were diagnosed with anti-NMDAR encephalitis and were being treated at the Neurology Department of Guangdong Second Provincial General Hospital, forming the anti-NMDAR encephalitis group. A healthy controls (HCs) group, comprising 23 age, sex, and education-matched volunteers with no history of neurological or psychiatric disorders, served as the comparison group. The study was approved by the Ethics Committee of Guangdong Second Provincial General Hospital. Prior to participation, all participants were thoroughly informed about the study’s aims and provided written informed consent.

For inclusion in the anti-NMDAR encephalitis group, participants were required to fulfill one or more of the diagnostic criteria specified for the condition. These criteria encompassed psychobehavioral abnormality or cognitive impairment, epilepsy, dyskinesia, speech disorders, autonomic nervous system dysfunction or insufficient ventilation, decreased level of consciousness, positive cerebrospinal fluid (and serum) tests for anti-NMDA receptor antibodies, right-handedness, and age from 16 to 60 years (Graus et al., 2016). The exclusion criteria were as follows: Participants with low-titer (1:10) positive serum or negative cerebrospinal fluid anti-NMDA receptor antibodies; inability to complete neuropsychological assessment tests; MRI suggesting brain structure abnormalities or MRI contraindications or poor image quality; diseases affecting neuropsychological examination, including severe physical, visual, and hearing disabilities; aphasia. The control group included the following criteria: negative NMDAR antibody test, no mental or neurological diseases, and age from 16 to 60 years.

Neuropsychological scale assessment

Disease severity was evaluated using the modified Rankin Scale (mRS) before fMRI acquisition. Participants completed the following scales: the California Verbal Learning Test (CVLT) to assess episodic language learning and memory, the Symbol Digit Modalities Test (SDMT) to evaluate attention and information processing speed, the Brief Visuospatial Memory Test (BVMT) to test spatial memory, and the Mini-Mental State Examination (MMSE) to measure overall cognitive function.

fMRI data obtaining

At the Guangdong Second Provincial General Hospital, both patients and healthy volunteers underwent MRI scanning using a 3.0 T MRI scanner. To reduce the noise effects, each participant’s head was pressed against a foam pad to limit head movement. During the acquisition of the MRI data, the participants were asked to keep their eyes closed to prevent them from falling asleep or becoming distracted by their thoughts. rs-fMRI datasets were acquired using a gradient echo planar imaging sequence with the parameters set as follows: each participant’s scanning involved acquiring 33 slices at 240 time points; echo time, 30 ms; repetition time, 2,000 ms; flip angle, 90°; matrix size, 64 × 61; field-of-view, 224 × 224 mm².

Data pretreatment

Data pretreatment was conducted in MATLAB utilizing DPARSF V4.3 (accessible at http://rfmri.org/DPARSF). The specific steps were as follows: removing the first 10 time points of image data; slice timing correction; conducting head movement correction (excluding participants with head rotation of more than 2.5° or 3D translation of more than 2.5 mm); spatial normalization (resampling to a standard space size of 3 × 3 × 3 mm); image segmentation; spatial smoothing with a Gaussian kernel of 4 mm full width at half maximum (Hopfinger et al., 2000); removing linear drift; performing regression analysis to eliminate the effect for cerebrospinal fluid, white matter signals, global brain signal, or Friston-24 parameters of head motion on the results. Finally, all images were temporally band-pass filtered (0.01–0.1 Hz).

Analyzing functional connectivity in cerebellar subregions

We used a seed-based analysis of resting-state FC. The DPARBI Viewer used a probabilistic atlas of the human cerebellum (https://www.diedrichsenlab.org/imaging/suit.htm; Fig. 1), which revealed 28 cerebellar subregions that served as regions of interest (ROIs). According to the literature, cerebellar Crus I was subdivided into two subregions, a and b. Left cerebellar Crus Ia (Montreal Neurological Institute [MNI] coordinates: −12, −78, and −28) and right cerebellar Crus Ia (MNI coordinates: 12, −78, and −28), left cerebellar Crus Ib (MNI coordinates: −32, −76, and −34) and right cerebellar Crus Ib (MNI coordinates: 34, −80, and −36) (Alalade et al., 2011). Spherical seed-point ROIs with a 6 mm radius were created using the DPABI software, and we generated FC maps by correlating the average time series from each ROI with the time series of all other brain voxels using Pearson correlation analysis. For further statistical analyses using Fisher’s r-to-z transformation, the correlation coefficients (r values) were converted to z values.

FIG. 1.

A probabilistic atlas of the human cerebellum.

Statistical analysis

Statistical analysis was conducted using SPSS software (version 26.0; SPSS Inc., Chicago, IL, United). General clinical data and cognitive scale scores were analyzed for both groups. Measured data were compared between groups using the independent samples t test, and count data were analyzed using the chi-square test; both are two-tailed tests. The difference was considered statistically significant at p < 0.05. The independent sample t test for FC based on ROIs was performed using the statistical module of the DPABI software, with age, gender, years of education, and head movement parameters as covariates to minimize the effects of these factors, and with a two-tailed Gaussian Random Field and defined as statistically significant brain regions with a corrected voxel level of p < 0.005 and cluster level of p < 0.05. The FC values of the brain regions with statistically significant differences were extracted, and the patients’ clinical variables and neuropsychological measurements, such as disease course, mRS, CVLT, SDMT, BVMT, and MMSE scores, were evaluated using Pearson correlation analysis.

Results

Demographic and clinical data

Table 1 presents the demographic and clinical data of the participants. Due to excessive head movements during the imaging procedure, four patients with anti-NMDAR encephalitis and two HCs were excluded from further analysis. Ultimately, the study comprised 19 patients with anti-NMDAR encephalitis and 21 HCs. No statistically significant differences were found between the groups regarding age, sex, or years of education (p > 0.05). However, the patient group exhibited lower cognitive assessment scores than the HCs group did, showing obvious differences in CVLT (p < 0.05), SDMT (p < 0.0001), BVMT (p < 0.01), and MMSE (p < 0.01) scores.

Table 1.

Demographic and Clinical Characteristics of Subjects

Variables	Patients (n = 19)	Healthy control (n = 21)	p value
Age (years)	30.05 ± 13.03	28.42 ± 9.47	0.653
Gender (male/female)	12/7	12/9	0.698
Education (years)	13.47 ± 2.01	14.09 ± 2.79	0.428
Disease duration (months)	12.00 (11.00, 25.00)	/	/
mRS (scores)	0.47 ± 0.51	/	/
CVLT (scores)	45.74 ± 13.52	54.33 ± 8.98	0.030
SDMT (scores)	49.68 ± 12.59	67.78 ± 9.62	0.000
BVMT (scores)	23.21 ± 8.77	29.94 ± 3.87	0.005
MMSE (scores)	26.84 ± 2.79	29.00 ± 1.08	0.005

Normally, distributed data are expressed as mean ± standard deviation, and nonnormally, distributed data are expressed as median and interquartile spacing.

BVMT, Brief Visuospatial Memory Test; CVLT, California Verbal Learning Test; MMSE, Mini-Mental State Examination; mRS, modified Rankin Scale; SDMT, Symbol Digit Modalities Test.

Functional connectivity results

The anti-NMDAR encephalitis group demonstrated reduced FC compared with that of the healthy HCs group. Similarly, decreased FC was noted between the right cerebellar Crus I and the left thalamus. In addition, a lower FC was observed between the left cerebellar Crus Ia and the right medial superior frontal gyrus, in addition to between the left cerebellar Crus Ib and the right middle temporal gyrus. Furthermore, the FC was lower between Crus II in the vermis of the cerebellum and the left putamen, as well as in the opercular part of the inferior frontal gyrus. Our results also indicated decreased FC between the left cerebellar X lobule and the left putamen (Table 2 and Fig. 2).

Table 2.

Alterations in the Cerebellar Subregions Functional Connectivity in the Patients Group Compared with the Healthy Control Group

			Peak MNI coordinate
ROI	Brain regions	Voxel	X	Y	Z	T-value
Right_CrusI	Thalamus_L	17	−39	42	18	−3.00225
Left_Crus Ia	Frontal_Sup_Medial-R	30	3	48	6	−2.99621
Left_Crus Ib	Temporal_Mid_R	75	51	−66	21	−3.00215
Vermis_CrusII	Putamen_L	57	−24	−3	9	−2.99816
	Frontal_Inf_Oper_L	36	−42	−3	33	−3.00491
Left_X	Putamen_L	35	−15	9	3	−3.00201

Inf, inferior; L, left; MNI, Montreal Neurological Institute; R, right; ROI, region of interest; Sup, superior.

FIG. 2.

FC alterations of cerebellum subregion in patients compared with HCs. Brain regions that had decreased FC with the right cerebellar Crus I (A). Brain regions that had decreased FC with the left cerebellar Crus Ia (B). Brain regions that had decreased FC with the left cerebellar Crus Ib (C). Brain regions that had decreased FC with the Crus II in the vermis of the cerebellum (D). Brain regions that had decreased FC with the left cerebellar X lobule (E). FC, functional connectivity; HCs, healthy controls.

Correlation analysis

Changes in connectivity values between cerebellar subregions were extracted, and their correlation with clinical manifestation and cognitive function scale scores was analyzed to investigate the relationship between FC changes and neurological dysfunction. Since the distribution of the data on disease duration was not normal, its correlation was analyzed using Spearman’s correlation analysis; the rest of the data were analyzed for correlation using Pearson’s correlation analysis. The results of the correlation analysis correlation analysis showed that the reduced FC between the right cerebellar Crus I and the left thalamus was negatively correlated with disease duration (r = −0.504, p = 0.028) while showing positive correlations with attention and information processing speed (SDMT; r = 0.613, p = 0.005) and language learning and memory (CVLT; r = 0.576, p = 0.010). In addition, decreased FC between the left cerebellar X lobe and the left putamen showed a positive correlation with language learning and memory (CVLT; r = 0.698, p = 0.006; Figs. 3 and 4).

FIG. 3.

The FC between the right cerebellar Crus I and the left thalamus was negatively correlated with the disease duration (r = −0.504, p = 0.028) while showing positive correlations with SDMT (r = 0.613, p = 0.005) and CVLT (r = 0.576, p = 0.010). CVLT, California Verbal Learning Test; FC, functional connectivity; SDMT, Symbol Digit Modalities Test.

FIG. 4.

The FC between the left cerebellar X lobe and the left putamen showed a positive correlation with CVLT (r = 0.698, p = 0.006). CVLT, California Verbal Learning Test; FC, functional connectivity.

Discussion

We examined FC changes between the cerebellar subregions and cerebral cortex in patients with anti-NMDAR encephalitis compared with those in HCs using MRI data, along with seed-based whole-brain FC. The results showed that the patients exhibited abnormal functional connections between the cerebellum and specific brain regions, as evidenced by a reduced FC between the right cerebellar Crus I and the left thalamus, which correlated negatively with disease duration and positively with attention, information processing speed, language learning, and memory. Furthermore, the FC between the left cerebellar Crus Ia and the medial superior frontal gyrus was significantly decreased, as well as that between the left cerebellar Crus Ib and the right medial temporal gyrus. The cerebellar vermis Crus II exhibited significantly reduced FC with the opercular part of the inferior frontal gyrus (Frontal_Inf_Oper_L) and the left putamen; the reduced FC between the left cerebellar lobule X and the left putamen was correlated with language learning and memory performance. These findings indicate that abnormal cerebellar subregion–cerebral connections participate in cognitive dysfunction in patients with anti-NMDAR encephalitis.

The study found that the FC between the right cerebellar Crus I and the left thalamus changed, accompanied by a more pronounced decrease with increasing duration of the patient’s disease. The thalamus is a small bilateral mesencephalon structure that is connected to various parts of the central nervous system. It serves as a repeater station and integration hub for the cognitive, motor, and sensory pathways, facilitating communication between the cerebral cortex, subcortical regions, and cerebellum. Moreover, the thalamus represents a critical node within two cerebro-cerebellar circuits: the cortico-striato-thalamo-cortical and cortico-ponto-cerebello-thalamo-cortical loops. These cerebral-cerebellar loops establish the anatomical basis for the cerebellum’s involvement in cognitive functions (Badke D’Andrea et al., 2023). The posterior lobe of the cerebellum, which includes Crus I as a significant part, is crucial for higher cognitive and emotional processing in humans. Crus I plays a vital role in diverse cognitive abilities, including memory, language, and visuospatial skills (Van Overwalle et al., 2020). Cerebellar Crus I–thalamus connectivity is reportedly associated with cognitive processing (Fama and Sullivan, 2015), and our findings further support this report by revealing a significant reduction in the FC between these regions in patients with anti-NMDAR encephalitis. The primary manifestation of cognitive dysfunction in these patients is long-term memory impairment (Bach, 2014; Guo et al., 2022). In a task-state fMRI study focusing on situational memory function, patients with anti-NMDAR encephalitis revealed that the worse the memory function of the patients, the more pronounced the activation of the thalamic regions. The overactivation of the thalamus was a compensatory activity of the brain under memory impairment to compensate for the decline in memory function owing to the disease (Wang et al., 2022). In our study, we found that the lower the cerebellar Crus I–thalamus connections, the worse the patient’s memory performance. This finding validates previous reports on the critical role of cerebellar Crus I and thalamus in memory encoding and reveals that reduced communication and interaction between the two further affects memory function in patients with NMDAR encephalitis (Ashida et al., 2019; Shine et al., 2023).

Furthermore, we observed significant changes in the functional connections between the cerebellum and the default mode network (DMN), particularly involving the medial temporal gyrus and the middle superior frontal gyrus. The DMN is linked to memory processing, abstract thought, and deep introspection, showing significant activity during rest or complex cognitive tasks, and is deemed an essential resting brain network in cognitive function (Smallwood et al., 2021). In anti-NMDAR encephalitis, relevant studies have reported cortical atrophy and white matter damage within the DMN areas (Xu et al., 2022), along with significant alteration in DMN–hippocampal connectivity, which are strongly correlated with disease severity and cognitive function (Yang et al., 2024). Our findings further revealed that the FC between the cerebellum and the DMN regions is abnormally diminished. A direct association was not observed between abnormal cerebellar DMN FC and cognitive assessment scores in this study; however, this is not sufficient to eliminate an association. Because of the disruption of FC in the DMN region, which is prevalent in patients with cognitive impairment (Chen et al., 2019; Miraglia et al., 2020; Qin et al., 2021; Wang et al., 2018), we speculate that the FC dysfunction between the cerebellum and DMN may be underrecognized during cognitive impairment, suggesting its extensive investigation in future studies.

Language, as a complex cognitive function, depends on the complex cerebral cortex and the cortical organization of the cerebellum (Yuan et al., 2022). In patients with anti-NMDAR encephalitis, language dysfunction is a prominent clinical manifestation. Neuroimaging studies have showed strong interconnections between the frontal cortical language areas, posterior cerebellar hemisphere, and vermis (Hodge et al., 2010; Stoodley and Schmahmann, 2009). Patients with schizophrenia, who also have reduced NMDAR function, showed a reduced cerebellar vermis gyrus, resulting in poorer language ability and immediate verbal memory (Antonova et al., 2004). In our study, we observed a reduced FC between the cerebellar vermis Crus II and the Frontal_Inf_Oper_L. Relevant studies have shown that the triangular and opercular parts of the inferior frontal gyrus constitute Broca’s area, which summarizes language information and transmits decision information to the motor cortex, a process that is crucial in language processing (Chang et al., 2015). Therefore, we hypothesized that the dysfunction of the functional connection between the vermis of the cerebellum and the inferior frontal gyrus of the left insular cortex contributes to language dysfunction in patients with anti-NMDAR encephalitis.

Our research revealed that several cerebellar subregions, such as the vermis Crus II, and the functional connections between cerebellar Lobule X and the putamen were anomalously reduced, and the weak connectivity between the cerebellar Lobule X and the putamen was associated with language learning and memory. As the crucial input structure of the basal ganglia, the striatum comprises the caudate and putamen and is the core component of reinforcement learning and reward processing (Admon et al., 2017). A long-term study revealed that the cerebellum plays a role in reward-related signaling, particularly during goal-directed learning tasks that require interactions among the cerebral cortex, basal ganglia, and cerebellar cortex. This finding has cemented a new consensus that the cerebellum significantly influences cognitive processes (Kostadinov and Häusser, 2022). Similarly, this indicates that the cerebellum integrates supervised and reinforcement learning to enhance goal-oriented behaviors. fMRI studies, task-based and at rest, showed that multiple representations of cerebellar cognitive and emotional processing, such as language, memory, and emotional processing, simultaneously involve vital subregions of the cerebellum, including the cerebellar Crus I, Crus II, and Lobule X (Schmahmann et al., 2019). However, in our results, how the connection disorder between cerebellar lobule X and the putamen is related to cognitive function remains unclear and requires further exploration.

The research has some limitations. First, the small number of participants made it challenging to ensure result stability and reproducibility. The new patterns of change we observed should be tested on a larger sample size, which can be used to increase the effect size and the reproducibility of our findings. Second, owing to the challenges in obtaining cooperation from patients with anti-NMDAR encephalitis during the acute phase, our study only included individuals in the postacute phase. This reflects the long-range effects of the disorder; however, at different stages of the disorder, brain function abnormalities differ among patients at different phase of the disease. Consequently, our findings may not entirely represent the exact changes in brain function during the acute stage. Finally, the use of neuropsychological scales should be improved, as anti-NMDAR encephalitis is a rare disease, making the number of patients that can be collected limited. This limited our ability to analyze differences in cognitive functioning between groups and failed to strictly follow score criteria for relevant comparisons. Future studies will aim to increase the sample size to enhance the rigor of the results and strictly follow the criteria for analysis. This was a cross-sectional study, and the correlation between radiographical changes and clinical symptoms did not imply causation. Hence, future longitudinal studies should increase the sample size to improve the representativeness and statistical reliability of the findings, and larger samples can better capture individual differences in cerebellar structural and functional changes and focus on observing the trajectory of alterations in cerebellar morphology and function during the disease course, as well as the relationship between the timing of brain changes and clinical manifestations to facilitate a better understanding of the neurophysiological mechanisms of cognitive impairment in anti-NMDAR encephalitis.

Conclusion

In this rs-fMRI study, cerebellar subregions were considered the ROIs to investigate changes in the cerebellar subregions and brain FC in patients with anti-NMDAR encephalitis. Changes in the FC between the cerebellum and the thalamus, the DMN, and Frontal_Inf_Oper_L may significantly contribute to the cognitive deficits in these patients. Findings on the FC in the cerebellar subregions reaffirm and enhance those of previous studies, offering novel insights and evidence to better understand the pathophysiological mechanisms underlying cognitive dysfunction in anti-NMDAR encephalitis.

Footnotes

Acknowledgment

The authors thank all the patients and volunteers for participating in this study.

Authors’ Contributions

X.M. and M.L. conceived and designed the experiments. M.L. and Z.L. performed the experiments. Z.L. and G.X. analyzed the data. M.L., J.L., and J.T. contributed reagents/materials/analysis tools. Z.L. and J.L. worked on the figures. X.M. and Z.L. wrote the article.

Author Disclosure Statement

The authors of this article declare no relationships with any companies, whose products or services may be related to the subject matter of the article.

Funding Information

This work was supported by the National Science Foundation of China (Grant number: 82472035) and the Science and Technology Program of Guangzhou (Grant Number: 202201020385).

References

Admon

, Kaiser

, Dillon

, et al. Dopaminergic enhancement of striatal response to reward in major depression. Am J Psychiatry, 2017; 174(4):378–386; doi: 10.1176/appi.ajp.2016.16010111

Alalade

, Denny

, Potter

, et al. Altered cerebellar-cerebral functional connectivity in geriatric depression. PLoS One, 2011; 6(5):e20035; doi: 10.1371/journal.pone.0020035

Antonova

, Sharma

, Morris

, et al. The relationship between brain structure and neurocognition in schizophrenia: A selective review. Schizophr Res, 2004; 70(2–3):117–145; doi: 10.1016/j.schres.2003.12.002

Armano

, Rossi

, Taglietti

, et al. Long-term potentiation of intrinsic excitability at the mossy fiber-granule cell synapse of rat cerebellum. J Neurosci, 2000; 20(14):5208–5216; doi: 10.1523/jneurosci.20-14-05208.2000

Ashida

, Cerminara

, Edwards

, et al. Sensorimotor, language, and working memory representation within the human cerebellum. Hum Brain Mapp, 2019; 40(16):4732–4747; doi: 10.1002/hbm.24733

Bach

. Long term rehabilitation management and outcome of anti-NMDA receptor encephalitis: Case reports. NeuroRehabilitation, 2014; 35(4):863–875; doi: 10.3233/nre-141176

Badke D’Andrea

, Marek

, Van

, et al. Thalamo-cortical and cerebello-cortical functional connectivity in development. Cerebral Cortex (New York, NY: 1991), 2023; 33(15):9250–9262; doi: 10.1093/cercor/bhad198

Cai

, Liang

, Huang

, et al. Cerebral functional activity and connectivity changes in anti-N-methyl-D-aspartate receptor encephalitis: A resting-state fMRI study. Neuroimage Clin, 2020; 25:102189; doi: 10.1016/j.nicl.2020.102189

Chang

, Raygor

, Berger

. Contemporary model of language organization: An overview for neurosurgeons. J Neurosurg, 2015; 122(2):250–261; doi: 10.3171/2014.10.Jns132647

10.

Chen

, Huang

, Ye

, et al. Disrupted functional and structural connectivity within default mode network contribute to WMH-related cognitive impairment. Neuroimage Clin, 2019; 24:102088; doi: 10.1016/j.nicl.2019.102088

11.

Dalmau

, Graus

. Antibody-Mediated encephalitis. N Engl J Med, 2018; 378(9):840–851; doi: 10.1056/NEJMra1708712

12.

Dalmau

, Tüzün

, Wu

, et al. Paraneoplastic anti-N-methyl-D-aspartate receptor encephalitis associated with ovarian teratoma. Ann Neurol, 2007; 61(1):25–36; doi: 10.1002/ana.21050

13.

Etkin

, Maron-Katz

, Wu

, et al. Using fMRI connectivity to define a treatment-resistant form of post-traumatic stress disorder. Sci Transl Med, 2019; 11(486); doi: 10.1126/scitranslmed.aal3236

14.

Fama

, Sullivan

. Thalamic structures and associated cognitive functions: Relations with age and aging. Neurosci Biobehav Rev, 2015; 54:29–37; doi: 10.1016/j.neubiorev.2015.03.008

15.

Finke

, Kopp

, Pajkert

, et al. Structural hippocampal damage following Anti-N-Methyl-D-aspartate receptor encephalitis. Biol Psychiatry, 2016; 79(9):727–734; doi: 10.1016/j.biopsych.2015.02.024

16.

Graus

, Titulaer

, Balu

, et al. A clinical approach to diagnosis of autoimmune encephalitis. Lancet Neurol, 2016; 15(4):391–404; doi: 10.1016/s1474-4422(15)00401-9

17.

Guo

, Lv

, Zhang

, et al. Gray matter atrophy and corresponding impairments in connectivity in patients with anti-N-methyl-D-aspartate receptor encephalitis. Brain Imaging Behav, 2022; 16(5):2001–2010; doi: 10.1007/s11682-022-00670-5

18.

Hodge

, Makris

, Kennedy

, et al. Cerebellum, language, and cognition in autism and specific language impairment. J Autism Dev Disord, 2010; 40(3):300–316; doi: 10.1007/s10803-009-0872-7

19.

Hopfinger

, Büchel

, Holmes

, et al. A study of analysis parameters that influence the sensitivity of event-related fMRI analyses. Neuroimage, 2000; 11(4):326–333; doi: 10.1006/nimg.2000.0549

20.

Hughes

, Peng

, Gleichman

, et al. Cellular and synaptic mechanisms of anti-NMDA receptor encephalitis. J Neurosci, 2010; 30(17):5866–5875; doi: 10.1523/jneurosci.0167-10.2010

21.

Kostadinov

, Häusser

. Reward signals in the cerebellum: Origins, targets, and functional implications. Neuron, 2022; 110(8):1290–1303; doi: 10.1016/j.neuron.2022.02.015

22.

Lau

, Zukin

. NMDA receptor trafficking in synaptic plasticity and neuropsychiatric disorders. Nat Rev Neurosci, 2007; 8(6):413–426; doi: 10.1038/nrn2153

23.

Lee

, Lee

, Kim

, et al. Disease progression and brain atrophy in NMDAR encephalitis: Associated factor & clinical implication. Ann Clin Transl Neurol, 2022; 9(7):912–924; doi: 10.1002/acn3.51604

24.

, Choi

, Joshi

, et al. Temporal non-local means filtering for studies of intrinsic brain connectivity from individual resting fMRI. Med Image Anal, 2020; 61:101635; doi: 10.1016/j.media.2020.101635

25.

Liang

, Cai

, Zhou

, et al. Voxel-based analysis and multivariate pattern analysis of diffusion tensor imaging study in anti-NMDA receptor encephalitis. Neuroradiology, 2020; 62(2):231–239; doi: 10.1007/s00234-019-02321-x

26.

Mapelli

, Pagani

, Garrido

, et al. Integrated plasticity at inhibitory and excitatory synapses in the cerebellar circuit. Front Cell Neurosci, 2015; 9:169; doi: 10.3389/fncel.2015.00169

27.

McKeon

, Robinson

, Ryan

, et al. Cognitive outcomes following anti-N-methyl-D-aspartate receptor encephalitis: A systematic review. J Clin Exp Neuropsychol, 2018; 40(3):234–252; doi: 10.1080/13803395.2017.1329408

28.

Miraglia

, Vecchio

, Marra

, et al. Small world index in default mode network predicts progression from mild cognitive impairment to dementia. Int J Neural Syst, 2020; 30(2):2050004; doi: 10.1142/s0129065720500045

29.

Qin

, Tang

, Dou

, et al. Default mode network integrity changes contribute to cognitive deficits in subcortical vascular cognitive impairment, no dementia. Brain Imaging Behav, 2021; 15(1):255–265.

30.

Schmahmann

, Doyon

, McDonald

, et al. Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage, 1999; 10(3 Pt 1):233–260.

31.

Schmahmann

, Guell

, Stoodley

, et al. The theory and neuroscience of cerebellar cognition. Annu Rev Neurosci, 2019; 42:337–364; doi: 10.1146/annurev-neuro-070918-050258

32.

Schmahmann

, MacMore

, Vangel

MJN

. Cerebellar stroke without motor deficit: Clinical evidence for motor and non-motor domains within the human cerebellum. Neuroscience, 2009; 162(3):852–861.

33.

Schmahmann

, Sherman

. The cerebellar cognitive affective syndrome. Brain, 1998; 121(Pt 4):561–579; doi: 10.1093/brain/121.4.561

34.

Shine

, Lewis

, Garrett

, et al. The impact of the human thalamus on brain-wide information processing. Nat Rev Neurosci, 2023; 24(7):416–430; doi: 10.1038/s41583-023-00701-0

35.

Smallwood

, Bernhardt

, Leech

, et al. The default mode network in cognition: A topographical perspective. Nat Rev Neurosci, 2021; 22(8):503–513; doi: 10.1038/s41583-021-00474-4

36.

Stoodley

, Schmahmann

. Functional topography in the human cerebellum: A meta-analysis of neuroimaging studies. Neuroimage, 2009; 44(2):489–501; doi: 10.1016/j.neuroimage.2008.08.039

37.

Tanguturi

, Cundiff

, Fuchs

. Anti-N-Methyl d-Aspartate receptor encephalitis and electroconvulsive therapy: Literature review and future directions. Child Adolesc Psychiatr Clin N Am, 2019; 28(1):79–89; doi: 10.1016/j.chc.2018.07.005

38.

Van Overwalle

, Manto

, Cattaneo

, et al. Consensus paper: Cerebellum and social cognition. Cerebellum, 2020; 19(6):833–868; doi: 10.1007/s12311-020-01155-1

39.

Wang

, Pan

, Liu

, et al. Aberrant default mode network in amnestic mild cognitive impairment: A meta-analysis of independent component analysis studies. Neurol Sci, 2018; 39(5):919–931; doi: 10.1007/s10072-018-3306-5

40.

Wang

, Wu

, Ji

, et al. Abnormal brain activation during verbal memory encoding in postacute Anti-N-Methyl-d-Aspartate receptor encephalitis. Brain Connect, 2022; 12(7):660–669; doi: 10.1089/brain.2021.0046

41.

, Guo

, Li

, et al. Progressive cortical and sub-cortical alterations in patients with anti-N-methyl-D-aspartate receptor encephalitis. J Neurol, 2022; 269(1):389–398; doi: 10.1007/s00415-021-10643-1

42.

Yang

, Fu

, Jiang

, et al. Functional connectivity changes of the hippocampal subregions in anti-N-methyl-D-aspartate receptor encephalitis. Brain Imaging Behav, 2024; 18(3):686–697; doi: 10.1007/s11682-024-00852-3

43.

Yuan

, Li

, Du

, et al. The cerebellum and cognition: Further evidence for its role in language control. Cerebral Cortex (New York, NY: 1991), 2022; 33(1):35–49; doi: 10.1093/cercor/bhac051

44.

Zhang

, Duan

, Ye

, et al. Brain MRI characteristics of patients with Anti-N-Methyl-D-aspartate receptor encephalitis and their associations with 2-year clinical outcome. AJNR Am J Neuroradiol, 2018; 39(5):824–829; doi: 10.3174/ajnr.A5593