Metabolic phenotyping of pilomotor seizures in autoimmune encephalitis

Abstract Objectives Ictal piloerection (IP) is an uncommon symptom in focal epilepsy and is associated with autoimmune encephalitis (AE). However, the networks involved in AE‐associated IP are still unclear. To have a better understanding of IP underlying mechanisms, the current study investigated whole‐brain metabolic networks for the analysis of AE‐associated IP. Methods Patients with AE and IP diagnosed at our Institute between 2018 and 2022 were selected. We then investigated the brain regions associated with AE‐associated IP using positron emission tomography (PET). Anatomometabolic changes (interictal 18F fluorodeoxyglucose PET) in AE patients with IP were compared with those of AE patients of similar age without IP (p‐voxel <0.001, uncorrected). Results Sixteen patients showed significant IP. The overall IP prevalence was 4.09% of patients with AE and 12.9% of patients with limbic encephalitis. The most common autoantibodies were against LGI1 (68.8%) followed by GAD65 (6.3%), NMDA (6.3%), GABAb (6.3%), CASPR2 (6.3%), and antibodies recognizing both GAD65 and mGLUR5 (6.3%). Most patients responded well to immunotherapy. Analysis of the imaging results at the voxel level showed that patients with IP had hypermetabolic changes in the right inferior temporal gyrus, suggesting involvement of this brain region in IP. Conclusions Our findings indicate that IP as an uncommon AE‐associated manifestations should be recognized. We observed that the metabolic pattern of IP was conspicuous in the right inferior temporal gyrus.

Ictal piloerection (IP) has been observed in LE, among other neurological disorders, and is usually classified as an autonomic epilepsy. 10,11 It is an uncommon accompaniment of focal seizures, with the result that the epilepsy diagnosis may be missed. It can occur on either one or both sides of the body and is often associated with autonomic manifestations such as sweating, chills, or pallor. 12 Although IP is thought to be related to temperature dysregulation, its cause remains unknown. Piloerection has been associated with stimulation of various brain regions, including the hippocampus, insula, midbrain, amygdala, and medial prefrontal cortex, suggesting that ictal IP may be linked to autonomic control. 13 The use of network approaches in the investigation of the pathophysiology of LE-associated IP may provide insight into the etiology, diagnosis, and management of this disease. 18 F-fluoro-2-deoxy-d-glucose positron emission tomography ( 18 F-FDG-PET) enables in vivo imaging and examination of brain function, and has been shown capable of diagnosing AE. 14 Quantitative analysis of 18 F-FDG-PET data can also be performed and have proved advantageous in the evaluation of a variety of neurological disorders. 15 However, the metabolic activity associated with AE-associated IP is unknown and the neural mechanisms underlying it remain obscure. The current study, therefore, investigated whole-brain metabolic networks for the analysis of AE-associated IP.

| Study participants
The study was approved by the institutional review board of Beijing Tiantan Hospital and written informed consent was obtained from all participants.
The data of patients diagnosed with AE between December 2018 and October 2022 in Beijing Tiantan Hospital were retrospectively evaluated for the incidence of IP. Patients had been diagnosed with AE in accordance with accepted criteria updated in 2016. 16 Cases with IP as a major ictal characteristic were identified.

| EEG and imaging
Conventional long-term video-electroencephalogram (EEG) recording was performed on each patient with standard 10-20 system electrodes. Magnetic resonance imaging (MRI) was performed with a 3T MRI system (Signa HD xt 3 T Volume, GE, GE Healthcare, USA) while 18 F-FDG-PET scans were done using a PET/CT scanner (acquisition parameters see below).

| Functional assessment and evaluation of outcomes
For most patients, the Montreal Cognitive Assessment (MOCA) and the Mini-Mental State Examination (MMSE) were used at the disease peak for assessing cognition. Subjects were assessed for seizure occurrence both at the disease peak and at follow-up.

| 18 F-FDG-PET voxel-wise analysis
Of the 16 patients, two did not benefit from the PET examination and one had no available original MRI data, so they were excluded from the analyses. Statistical analysis of the whole-brain PET imaging data showed that, relative to patients without IP, those with IP had marked clusters of hypermetabolism in the right inferior temporal gyrus (p-voxel <0.001 uncorrected; k = 36 voxels; MNI coordinates of voxel maximum: 42, -4, -46; Figure 2). No brain areas in the group showed significant hypermetabolism. The maps of significantly hypermetabolic voxels for individual IP patients compared to those without IP are provided in Figure S1.

| DISCUSS ION
We reported here 16 patients with AE-associated pilomotor seizures and suggested a specific etiological link between this type of autonomic seizure and AE. We also performed voxel-based group comparison and demonstrated the specific symptom network associated with IP in AE.
Several studies have proposed that pilomotor seizures should be grouped as a specific multi-seizure type associated with AE. 17,18 However, autonomic seizures rarely have IP as the main ictal symptom. The incidence of these seizures in AE is not known and IP is thought to be an uncommon symptom associated with focal epilepsy, having a predicted prevalence of 0.4%-0.65% 11 LGI1 (68.8%) followed by NMDA (6.3%), CASPR2 (6.3%), GABAb It is thus possible that the right inferior temporal gyrus could be involved in the generation of IP. The current study appears to be first to analyze the origin of pilomotor seizures using PET which enables direct and precise identification of metabolic changes. IP is linked to bilateral or unilateral seizures generated in the temporal lobe. Rocamora et al. 11 reported five AE cases that presented with pilomotor seizures, three of which had anti-LGI1 encephalitis, one anti-Hu, and one anti-Ma2. A follow-up study 25

LE-associated seizures accompanied by autoantibodies against
LGI1 respond poorly to ASMs but respond well to immunotherapy. 2

AUTH O R CO NTR I B UTI O N S
QW and YQS concepted, designed, and supervised the study. YQS and XBZ acquired the data. YQS analyzed and interpreted the data, provided statistical analysis, had full access to all of the data in the study, and are responsible for the integrity of the data and the accuracy of the data analysis. YQS drafted the manuscript, QW and LA critically revised the manuscript for important intellectual content.
All authors read and approved the final manuscript.

ACK N OWLED G M ENTS
The authors thank the following colleague who contributed to this study providing FDG-PET data: Chengxu Jiang.

FU N D I N G I N FO R M ATI O N
The study was financially supported by the National Key R&D

CO N FLI C T O F I NTE R E S T S TATE M E NT
None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.