The first description of involuntary arm elevation during yawning in hemiplegic patients was reported in 1844 by John Abercrombie. In 2005, Walusinski et al. coined the term “parakinesia brachialis oscitans” (PBO) to describe involuntary elevation of a paretic upper limb triggered by yawning. ^{Reference Walusinski, Neau and Bogousslavsky1} Despite being recognized for nearly two centuries, this phenomenon remains poorly understood and likely underrecognized in clinical practice.

PBO most commonly affects the paretic upper limb and has been reported in association with ischemic and hemorrhagic stroke, as well as non-vascular disorders including amyotrophic lateral sclerosis, tumefactive demyelination and brainstem tuberculoma. ^{Reference de Lima, Munhoz, Becker and Teive2–Reference Tsai, Lee and Chen5} We describe three patients with ischemic stroke who developed PBO during the course of their disease.

Case 1. A 20-year-old woman with a 2-year and 5-month history of oral contraceptive use and no other relevant medical history presented with a sudden-onset frontal headache followed by left-sided weakness resulting in a fall and inability to walk. Neuroimaging demonstrated a right internal carotid artery dissection with ischemic stroke involving the right internal capsule. One week after stroke onset, she developed involuntary elevation of the paretic arm triggered by yawning, consistent with PBO (Video 1).

Case 2. A 45-year-old woman with hypertension (post-preeclampsia), hypothyroidism and dyslipidemia presented with subacute dysarthria and right hemiparesis (NIHSS 10) following chest pain and headache. Brain magnetic resonance imaging (MRI) demonstrated a left hemipontine infarction with hemorrhagic transformation. By day 4 after stroke onset, she developed PBO. Further workup revealed a patent foramen ovale, considered incidental given the presence of uncontrolled vascular risk factors.

Case 3. A 49-year-old woman with hypothyroidism, recently diagnosed hypertension and paternal history of stroke presented with left central facial palsy and hemiparesis. Examination confirmed left hemiparesis with ipsilateral hyperreflexia and hemiparetic gait. Brain MRI revealed a right frontal subcortical white matter ischemic lesion with diffusion restriction and no hemorrhagic transformation. During the second week after stroke onset, she developed PBO. Extensive etiologic workup was unrevealing, and the stroke was classified as cryptogenic.

The main characteristics of the patients are summarized in Table 1.

Table 1.

Clinical characteristics of patients with parakinesia brachialis oscitans

Table 1. Long description

Table 1 summarizes the clinical characteristics of three patients with parakinesia brachialis oscitans. All three patients were female and had cerebral infarction.

ESUS = embolic stroke of undetermined source; HTN = hypertension; TOAST = Trial of ORG 10172 in Acute Stroke Treatment; T2DM = type 2 diabetes mellitus.

The term “parakinesia” refers to an abnormal involuntary movement accompanying or substituting a voluntary movement, while “brachialis” refers to the upper limb and “oscitans” derives from the Latin word for yawning. ^{Reference Walusinski, Neau and Bogousslavsky1} The movements are typically stereotyped, involving shoulder abduction with internal rotation, elbow flexion and finger extension. They begin during the inspiratory phase of yawning and end with the rapid return of the limb to its resting position. ^{Reference Lim and Neo3–Reference Chowdhury, Datta, Biswas and Biswas6}

The onset of PBO varies widely, ranging from one day to several months after stroke and often occurring during the spastic phase. ^{Reference Lim and Neo3–Reference Chowdhury, Datta, Biswas and Biswas6} These movements usually resolve as motor function in the affected limb improves, most often within 6 months. ^{Reference de Lima, Munhoz, Becker and Teive2,Reference Salavisa, Mohamed, Allen-Philbey, Stennett, Campion and Schmierer4} Lesions most frequently involve the internal capsule and pontomedullary region, although supratentorial locations such as the centrum semiovale and frontal subcortex have also been reported. ^{Reference Lim and Neo3}

Pathophysiology likely involves disruption of corticospinal and corticopontocerebellar pathways with preservation of spinocerebellar circuits. Interruption of the corticopontocerebellar pathway allows the release of subcortical structures from cortical inhibition, while preservation of spinocerebellar pathways allows phylogenetically primitive structures to modulate movement in response to yawning. ^{Reference Chowdhury, Datta, Biswas and Biswas6} One proposed mechanism involves activation of an “emotional motor system,” in which descending pathways distinct from the pyramidal tract coordinate bulbar reticular activity during yawning and activate reticulospinal pathways controlling the upper limb. ^{Reference Lim and Neo3–Reference Chowdhury, Datta, Biswas and Biswas6} Another hypothesis involves a proprioceptive loop in which strong respiratory muscle contraction during yawning generates signals transmitted through the ventral spinocerebellar tract to cerebellar structures and the lateral reticular nucleus, which then activates spinal motor neurons. A limitation of this theory is that it does not explain the lower limb movements observed in some case reports. ^{Reference de Lima, Munhoz, Becker and Teive2,Reference Lim and Neo3,Reference Chowdhury, Datta, Biswas and Biswas6}

Distinguishing the phenomenon of PBO from an epileptic seizure prevents the unnecessary treatment of a non-existent epilepsy, as this phenomenon lacks an electrographic correlate and is strongly associated with the act of yawning. It also avoids exposing the patient to antiseizure medications and their potential adverse effects. Likewise, it should be differentiated from a voluntary movement, as misinterpretation could lead to an overestimation of motor recovery by incorrectly suggesting preserved voluntary motor pathways.

Our cases broaden the clinical spectrum of PBO. The first demonstrates its occurrence in a young patient with carotid artery dissection. The second highlights its presence in pontine infarction with multiple vascular risk factors. The third shows that supratentorial subcortical lesions may also produce this phenomenon. Together, these observations emphasize the heterogeneity of PBO and the importance of recognizing this sign during neurological examination.

The analysis of these three clinical cases of women with PBO, each associated with different comorbidities, distinct lesion locations, and varying etiologies of cerebral infarction, is relevant because it highlights the diversity of clinical contexts in which this phenomenon may occur. To date, no common underlying factor has been identified to explain its occurrence. This variability suggests that its emergence is not attributable to a single cause, but rather to the interaction of multiple pathophysiological mechanisms. Furthermore, its description reinforces recognition of this phenomenon as a specific neurological sign with important diagnostic implications, as it allows differentiation from focal epileptic seizures and voluntary movements, thereby preventing inappropriate therapeutic interventions and misinterpretation of neurological recovery.

Although rarely reported, PBO may be underrecognized in clinical practice. Increased awareness may improve understanding of post-stroke motor network reorganization and provide insights into preserved extrapyramidal pathways involved in motor recovery.