In a retrospective study of 8,482 monitored epilepsy patients, researchers at the Epilepsy Center of West China Hospital identified thirteen individuals with pilomotor seizures, finding that these rare episodes predominantly originate in the temporal lobe and display distinctive rhythmic delta electroencephalogram activity. The investigation, which screened patients undergoing video-electroencephalogram monitoring between January 2021 and October 2024, evaluated clinical presentations, neurophysiological patterns, multimodal neuroimaging features, and underlying etiologies of this underrecognized semiology. Across the cohort, pilomotor seizures were consistently accompanied by other autonomic or psychic symptoms, and most patients achieved favorable clinical responses when managed with anti-seizure medications or targeted surgical resection.

Piloerection, commonly recognized as goosebumps, is a physiological phenomenon characterized by the contraction of arrector pili muscles upon exposure to cold temperatures or emotional stimuli. Although voluntary control over piloerection has been recorded in a very small number of cases, the response is primarily mediated by the autonomic nervous system, with the hypothalamus, limbic system, and frontal cortex playing key roles. In the International League Against Epilepsy 2025 position paper on the updated classification of epileptic seizures, piloerection is formally defined as an autonomic semiological feature of epileptic seizures. Clinicians frequently misdiagnose pilomotor seizures as manifestations of anxiety or depression, particularly when episodes present alongside emotional symptoms such as fear or other autonomic disturbances.

To clarify the clinical features and epileptogenic zones of pilomotor seizures, researchers retrospectively reviewed medical records from 8,482 individuals diagnosed with epilepsy meeting 2025 classification criteria. While seventeen patients initially presented with ictal piloerection, four were excluded due to incomplete clinical data, yielding a definitive cohort of thirteen right-handed patients comprising six males and seven females. The mean age of the cohort was 36.7 plus or minus 10.9 years, with a mean age at seizure onset of 32.2 plus or minus 12.7 years. Prior to admission at the specialized center, ten patients had been clinically diagnosed with focal epilepsy of unknown etiology, whereas the remaining three had temporal lobe epilepsy. Ten of the thirteen patients experienced active seizures during synchronized video-electroencephalogram monitoring.

In all thirteen cases, piloerection manifested as an integral component alongside other clinical presentations rather than as an isolated event. Each patient exhibited autonomic manifestations, including tachycardia, epigastric distress, nausea or vomiting, and diaphoresis, while several patients reported disturbances such as olfactory hallucinations, déjà vu, limb paresthesia, automatisms, altered consciousness, or secondary progression to generalized tonic-clonic seizures. Cognitive problems were particularly documented in Case 7, characterized by memory impairment, an inability to recognize her way home, and a lack of recall for recent events. Synchronized clinical and video documentation confirmed that piloerection consistently presented in close association with these broader autonomic and psychic features across the cohort.

Neurophysiological evaluation detected interictal epileptiform discharges in eleven of the thirteen patients, representing 84.6 percent of the cohort, predominantly localized to the temporal region in eight cases. Within these temporal interictal patterns, four patients demonstrated right temporal discharges, one displayed left temporal discharges, and three showed bilateral temporal involvement, while three other cases exhibited discharges in the midline, left frontotemporal region, or bilateral temporoparietal areas. Ictal recordings obtained in ten patients during monitoring revealed that seven cases had seizures of temporal origin, consisting of three left-sided, three right-sided, and one bilateral onset. The remaining three patients exhibited extra-temporal seizure origins, including two cases in the left frontotemporal region and one case in the left frontal lobe.

Among nine patients demonstrating ictal electroencephalogram evidence of temporal seizure onset, seven displayed rhythmic delta activity characterized by 1 to 4 Hertz frequencies. This distinctive slow-wave rhythm contrasted with the typical theta activity observed in classic mesial temporal lobe epilepsy, suggesting a unique pathophysiological mechanism. Scalp recordings in Case 4 demonstrated 2 to 3 Hertz rhythmic slow-wave activity, a pattern also documented in Cases 1, 3, 5, 7, and 8. In Case 12, ictal activity began as rhythmic 7 to 8 Hertz discharges before evolving into 3 to 5 Hertz activity, whereas Case 8 displayed 2 to 3 Hertz spike-and-wave complexes in the right hemisphere during non-convulsive seizures, and Case 6 originated from the frontocentral region. The duration of recorded ictal discharges ranged from 10 seconds to 5 minutes.

The chronological onset of piloerection relative to electroencephalogram discharges was identified in six patients during monitoring. In two individuals, piloerection began 1 to 4 seconds before electroencephalogram onset, while four patients showed piloerection occurrence at 2 to 12 seconds after electroencephalogram onset. Unilateral onset of piloerection was documented in nine cases, providing key lateralizing clues. Among these nine patients, two showed strictly ipsilateral propagation, while seven exhibited contralateral spread. Furthermore, the initial side of piloerection was contralateral to the epileptogenic zone in six cases and ipsilateral in only one case, showing that recording the originating side and spreading characteristics offers valuable lateralizing data for the epileptogenic zone despite the complex network nature of focal epilepsy.

All patients underwent cranial magnetic resonance imaging at 3.0 Tesla using T1-weighted, T2-weighted, and fluid-attenuated inversion recovery sequences, evaluated independently by two blinded neuroradiologists. Imaging identified temporal lobe or hippocampal abnormalities in nine patients, while four patients demonstrated normal findings. Abnormalities included temporal lobe nodules in two cases, a cavernous angioma in one case, hippocampal sclerosis in three cases, bilateral temporal changes associated with autoimmune encephalitis in one case, and non-specific hippocampal signal abnormalities in two cases. Seven cases showed concordance between imaging findings and electroencephalogram onset, three had negative imaging, one had negative electroencephalogram findings with a right temporal lesion, and two showed discordance. In Case 7, combined magnetic resonance and positron emission tomography revealed patchy T1 hyperintensity and increased glucose metabolism in bilateral hippocampi and amygdala alongside symmetrically decreased frontal metabolism, confirming autoimmune encephalitis.

A definitive etiology was determined in eight of the thirteen patients, while five had undetermined etiologies despite comprehensive evaluation. Specific etiologies comprised encephalitis in four patients, hippocampal sclerosis in two, cavernous hemangioma in one, and vascular malformation in one. Among the four patients with normal magnetic resonance imaging, Case 9 was diagnosed with viral encephalitis. Additionally, one patient tested positive for LGI1 antibodies, and two had comorbid systemic lupus erythematosus. The authors noted that a literature review of 109 published cases identified thirty patients with confirmed autoimmune encephalitis, with approximately 46 percent of pilomotor seizure cases associated with astrocytoma or limbic encephalitis. Furthermore, while pilomotor seizures occur in about 0.65 percent of patients with refractory temporal lobe epilepsy, their reported frequency rises to between 13.3 and 14 percent in limbic encephalitis cohorts, suggesting a link to immune-mediated neuroinflammatory processes.

The physiological circuitry of piloerection operates through an integrated tri-lineage unit consisting of the arrector pili muscle originating from the mesenchymal layer, the hair follicle from the epithelial layer, and direct sympathetic innervation anchored to the follicle bulge region. Temperature-triggered pathways begin with cutaneous TRPM8 cold receptors transmitting signals via A-delta and C fibers to the spinal dorsal horn, ascending through the spinothalamic tract to the brainstem parabrachial nucleus and hypothalamic preoptic area. Emotion-triggered signals originate in the amygdala and relay to the hypothalamic paraventricular nucleus. Both pathways converge as hypothalamic signals descend to the spinal intermediolateral column, where postganglionic sympathetic fibers release norepinephrine onto alpha-1 adrenergic receptors in arrector pili muscles. Modulated by the prefrontal cortex and anterior cingulate cortex, ictal piloerection may result from pathological hyperactivation of sympathetic-cutaneous circuitry, the thalamus, or the amygdala by epileptogenic networks.

The case series demonstrated that pilomotor seizures predominantly affected adults, with eleven patients experiencing seizure onset in adulthood and two at age 16. The mean age of first seizure was 32.2 years, spanning an overall range from 16 to 57 years, an age distribution that may reflect the predominantly adult population treated at the center. A literature review of 104 cases revealed that among 64 patients with documented age of onset, 56 occurred in adulthood, with a mean onset age of 40.4 years across a range from 1 to 82 years. Although several studies have reported piloerection in underage epilepsy patients, a retrospective study of sixty children with epilepsy presenting with autonomic symptoms identified none with piloerection, suggesting that pilomotor seizures may be associated with neurological development and maturation.

Regarding clinical outcomes, the mean follow-up duration was 18.6 months, ranging from 6 to 36 months. Nine patients achieved significant seizure control with anti-seizure medications, while one patient voluntarily discontinued medication and remained seizure-free during one year of follow-up. In addition, Case 4 underwent left temporal epileptogenic focus resection with selective hippocampectomy and remained seizure-free for nine months postoperatively. Conversely, three patients had poor responses: Case 6 continued experiencing three to four seizures monthly despite four anti-seizure medications, and Case 7 partially improved cognitively following intravenous immunoglobulin and methylprednisolone but continued experiencing one to two monthly seizures. The authors noted limitations including the retrospective single-center design, modest sample size, and lack of advanced imaging or comprehensive autoimmune testing in certain patients, emphasizing that future prospective multicenter studies are warranted to validate these electroclinical findings.

Subject of Research: Clinical and electroencephalographic characteristics of pilomotor seizures in epilepsy

Article Title: Clinical and electroencephalogram characteristics of pilomotor seizures

Article References: Le, Z., Li, B., Zhang, H., Chen, Y., Yan, R., Zhou, D., Chen, J., & Wu, X. (2026). Clinical and electroencephalogram characteristics of pilomotor seizures. Acta Epileptologica, 8(1), Article 19. https://doi.org/10.1186/s42494-026-00260-8

Image Credits: AI Generated

DOI: 10.1186/s42494-026-00260-8

Keywords: pilomotor seizures, piloerection, temporal lobe epilepsy, electroencephalogram, rhythmic delta activity, autoimmune encephalitis, neuroimaging, semiology, Clinical, characteristics, pilomotor, seizures