Is faciobrachial dystonic seizure a misnomer? A case report of atypical leg-predominant tonic-dystonic attacks in LGI1-associated encephalitis

Abstract

Background

Antibody‑mediated encephalitis due to leucine‑rich glioma‑inactivated 1 (LGI1) is a frequent cause of immune‑responsive paroxysmal motor phenomena in older adults. The disorder is commonly associated with brief, stereotyped involuntary movements traditionally labeled faciobrachial dystonic seizures (FBDS). However, the “faciobrachial” descriptor may underrepresent the true phenotypic spectrum. We report a case of LGI1‑associated autoimmune encephalitis presenting with leg‑predominant tonic-dystonic attacks, expanding the recognized clinical presentation and highlighting variability in semiology.

Case presentation

A 62‑year‑old man presented following multiple falls caused by sudden, repetitive involuntary movements primarily affecting the left lower extremity. Episodes were short‑lived, painless, and occurred dozens of times daily, often provoked by initiation of voluntary movement and preceded by a brief sensory warning. Consciousness was consistently retained, although some attacks spread to involve the ipsilateral arm and face. Neurological examination, cognitive assessment, brain MRI, and EEG were unremarkable. Cerebrospinal fluid analysis demonstrated mildly elevated protein without pleocytosis. Antiseizure therapy provided no benefit. High‑dose intravenous corticosteroids resulted in minimal improvement, whereas plasma exchange led to complete symptom resolution. Serum autoimmune testing was positive for LGI1-IgG antibody and a low-titer P/Q-type calcium channel antibody, suggesting an autoimmune encephalitis— most likely LGI1 encephalitis. FDG-PET/CT showed no malignancies or hypermetabolic abnormalities.

Conclusions

This case highlights that LGI1‑associated paroxysmal events may present as leg‑predominant tonic-dystonic attacks, rather than the more typical faciobrachial distribution. Clinicians should maintain a high index of suspicion for autoimmune causes in late‑onset paroxysmal movement disorders, as prompt immunotherapy can prevent falls, reduce morbidity, and significantly alter disease course. Recognition of serum‑positive, imaging‑negative LGI1 autoimmunity is particularly important, as such presentations may lack common cognitive or limbic features and risk diagnostic delay.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-026-05018-w.

Background

Autoimmune encephalitis is increasingly recognized as a common, treatable cause of subacute neurological syndromes, and LGI1-antibody–associated disease represents one of the most frequently encountered entities in this category, particularly in older men. While LGI1 encephalitis is often framed as a limbic disorder with memory impairment and behavioral change, presentations can be clinically heterogeneous, and motor manifestations may precede, overshadow, or even occur without prominent cognitive symptoms. Early recognition is important because timely immunotherapy is associated with improved outcomes and may limit subsequent neurological morbidity [1–4].

Among the most distinctive motor manifestations in LGI1 autoimmunity are brief, frequent, stereotyped paroxysmal events historically termed faciobrachial dystonic seizures (FBDS). The label reflects both their characteristic distribution (face and upper limb) and their seizure-like abruptness and repetition. However, preserved awareness and frequent absence of ictal EEG correlates have fueled ongoing debate about whether these episodes are best conceptualized as epileptic seizures, movement disorders, or a borderland phenomenon. This controversy parallels earlier discussions in which paroxysmal dyskinesias were variably interpreted as seizure equivalents before a movement disorder framework became more widely accepted [5–8].

In addition to mechanistic ambiguity, the anatomic framing embedded in the term “faciobrachial” may be overly restrictive. Existing descriptions emphasize facial and brachial involvement, yet systematic reviews suggest broader variability in motor semiology across LGI1 encephalitis. Importantly, lower-extremity–predominant paroxysmal hyperkinesias are not commonly highlighted and may lead clinicians toward alternative diagnoses—such as paroxysmal kinesigenic dyskinesia (PKD), focal epilepsy, structural etiologies, or functional movement disorder—especially when cognition is intact and MRI/CSF findings are non-inflammatory. In older adults, late onset and high attack frequency should prompt consideration of symptomatic (secondary) causes rather than primary genetic PKD, and autoimmune encephalitis is an important, treatable consideration within that differential [3, 7, 8].

Here, we present a patient with serum LGI1-IgG positivity whose initial and predominant manifestation consisted of frequent, movement-triggered involuntary jerks involving the lower extremity, resulting in recurrent falls, with a marked response to immunotherapy. This case necessitates a re-evaluation of the term “faciobrachial” as a definitive descriptor and questions its clinical adequacy, broadens the recognized distribution of LGI1-related paroxysmal motor phenomena and underscores the need to consider autoimmune causes in late-onset, PKD-like presentations with leg-dominant involvement [3, 4, 7, 9].

Case presentation

A 62-year-old man with type 2 diabetes mellitus, hypertension, and hyperlipidemia presented to the emergency department after a fall precipitated by violent jerking of his left leg. He experienced brief, intermittent painless spasms, occurring 30–40 times daily, often causing violent leg jerks that sometimes spread to his left arm and face. Attacks were triggered by voluntary leg movement and preceded by an “aura”, a premonitory sensation of numbness, that the attack was coming. On the day prior to admission, he fell three times due to these jerks, without loss of consciousness.

There were no systemic symptoms such as fever or headache. Neurological examination revealed no cognitive, psychiatric, cranial nerve, or neuromuscular abnormalities. Serum metabolic profile including sodium level was normal at presentation (140 mEq/L; reference range 133–145). Brain and cervical spine MRI showed non-specific basal ganglia and pontine calcifications. An electroencephalogram (EEG) captured during attacks was normal. CSF was negative for autoantibodies but showed mildly elevated protein (63 mg/dL and 78 mg/dL during the first and second admission, respectively) without pleocytosis. Screening for malignancy (scrotal ultrasound, CT chest/abdomen/pelvic) was unremarkable.

Anticonvulsants were ineffective. High-dose intravenous methylprednisolone (1,000 mg daily for 5 days) produced mild improvement, but attacks persisted. He then underwent five sessions of plasma exchange, which led to complete resolution of symptoms.

Five months later, he experienced recurrence of similar attacks, now involving the right face and upper limb, triggered by hand movements such as texting or drinking from a can. Serum sodium at relapse remained within the normal range (137 mEq/L). He again received pulse-dose steroids followed by plasma exchange, resulting in marked improvement, supporting an autoimmune or paraneoplastic etiology.

Serologic testing revealed LGI1-IgG positivity and a low-titer P/Q-type voltage-gated calcium channel antibody (0.06 nmol/L). CSF remained negative for autoantibodies, including LGI1 antibodies. Given the clinical suspicion for LGI1-associated disease, ictal recordings of the EEG were re-reviewed specifically for subtle features, including fronto-central or mid-frontal slow or infraslow deflections preceding clinical events. No definite ictal correlate or subtle changes were identified.

Total body FDG-PET/CT, including the brain, was obtained approximately one month after the patient’s second admission. The study showed no hypermetabolic abnormalities to suggest an underlying malignancy or paraneoplastic process. However, dedicated cerebral metabolic interpretation was not performed.

Given the recurrent attacks, rituximab was initiated about a month after discharge to reduce the further relapses. The patient was administered two doses of 1000 mg at 15 days interval, followed by maintenance therapy every 6 months. At follow-up visits, attack frequency markedly improved. At most recent evaluation, he experienced infrequent (approximately twice weekly), brief, stimulus-triggered attacks with preserved awareness and no cognitive deficits. His LGI-1 titer remained positive. The neuroimmunologist recommended to continue the maintenance therapy for about 1–2 years and check the patient’s LGI1 titer prior to each infusion.

Discussion and conclusions

LGI1-antibody encephalitis is a subtype of autoimmune encephalitis characterized by antibodies targeting leucine-rich glioma-inactivated 1 (LGI1), a secreted neuronal protein critical for synaptic transmission [1, 10]. It is the second most common autoimmune encephalitis, typically affecting men over 50 years of age and presenting as subacute limbic encephalitis [2]. Hallmark features include rapidly progressive memory impairment, behavioral changes, hyponatremia, and movement disorders, originally described as faciobrachial dystonic seizures (FBDS) [7].

LGI1 interacts with ADAM22 and ADAM23 to stabilize AMPA receptor function at excitatory synapses [10]. Autoantibody binding disrupts this interaction, leading to synaptic dysfunction, particularly in limbic structures. MRI often shows T2/FLAIR hyperintensities in the medial temporal lobes, and EEG may reveal focal slowing or epileptiform discharges [3, 7].

Although most cases exhibit both motor and non-motor symptoms [3, 4], some patients present solely with movement disorders. The “dystonic seizure” terminology arose because FBDS attacks were brief and stereotyped, yet consciousness was typically preserved and EEG findings normal [5, 7]. This ambiguity mirrors earlier debates surrounding paroxysmal dyskinesias, where both epileptic and non-epileptic pathophysiology was considered but now classified as movement disorders [6, 8].

Our case highlights limitations of both the “seizure” terminology and the “faciobrachial” descriptor in fully capturing the spectrum of these events, as the leg-predominant onset adds to the recognized spectrum. Clinically, the episodes resembled paroxysmal kinesigenic dyskinesia (PKD), triggered by movement initiation. While most PKD is genetic and presents in youth, this patient’s late onset and acute course were inconsistent with primary PKD [8]. Secondary (symptomatic) PKD has been reported with multiple sclerosis, pseudohypoparathyroidism with basal ganglia calcification, and autoimmune encephalitis (e.g., CASPR2). Thus, autoimmune or paraneoplastic etiologies should be considered in older adults presenting with new-onset PKD-like episodes, even with leg-predominant involvement.

Serum LGI1-IgG positivity with CSF negativity is well documented and may reflect higher test sensitivity in serum [9]. Such cases often have milder systemic findings and normal MRI, potentially delaying recognition [9]. The patient’s low-titer P/Q-type calcium channel antibody typically associated with Lambert–Eaton syndrome (LEMS) or lung cancer likely represents an incidental or secondary autoimmune response. The patient had no clinical features suggestive of LEMS and FDG-PET/CT scan revealed no evidence of lung cancer. The coexistence of multiple antibodies in autoimmune encephalitis is not uncommon, although the combination of LGI1 and P/Q-type antibodies has not, to our knowledge, been previously reported. This detection is of uncertain clinical significance for our case at this stage.

Importantly, ictal EEG correlates are frequently absent or subtle in FBDS, and targeted re-review may still yield nondiagnostic results, as observed in this case. Similarly, hyponatremia, although common, is not universal, as demonstrated by consistently normal sodium levels in this patient during both initial admission and relapse three months later.

This case further underscores the importance of considering autoimmune etiologies in adult-onset paroxysmal movement disorders and recognizing that LGI1-related tonic-dystonic attacks may involve the legs. Increased awareness of atypical distributions, including lower extremity-predominant presentations, is essential to avoid diagnostic delay. Early immunotherapy remains critical, as illustrated by the marked clinical response in this patient. First-line treatment typically includes high-dose corticosteroids, intravenous immunoglobulin, or plasma exchange, while rituximab and other steroid-sparing agents may reduce relapse risk [4, 7].

Authors’ contributions

MF and MB: wrote the manuscript, designed this case report and collected the clinical data of this case. AG, SK and MB: guided the treatment of this case, analyzed the data and revised the manuscript. All authors have read and approved the manuscript.

Funding

The authors report no sources of funding for this case report.

Data availability

No datasets were generated or analyzed for this case report. The generated video material is included in this manuscript.

Declarations

Ethics approval and consent to participate

Per institutional policy, this case report is exempt from a Research and Institutional Review Committee (RIRC) approval since it involves fewer than three patients.

Consent for publication

The patient in this case report has reviewed the manuscript along with the video and has provided both verbal and written informed consent for publication of clinical details and images/videos provided in this report. The physician speaking to the patient in the background has also provided verbal and written consent to publish. We confirm that we have read the journal’s position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Competing interests

The authors declare no competing interests.