mise à jour du
28 octobre 2006
Yawning in acute anterior circulation stroke
OC Singer, MC Humpich, H Lanfermann, T Neumann-Haefelin
Department of Neurology & Institute for Neuroradiology
J.W. Goethe-University Frankfurt am Main, Germany
Yawning and stroke
Editorial Commentary O. Walusinski


Pathological yawning can be a clinical sign in disorders affecting the brain stem. Here we describe seven patients with pathological yawning due to acute middle cerebral artery (MCA)-stroke, indicating that pathological yawning does also occur in supratentorial stroke. We hypothesize that excessive yawning is a consequence of lesions in cortical or subcortical areas, which physiologically control diencephalic yawning centers.
Pathological yawning has been described in various neurological disorders, including migraine, epilepsy, basal ganglia disorders, multiple sclerosis or brain tumors.(1) In patients with focal brain lesions, infratentorial lesions dominated and pathological yawning has been linked to a disturbance of the ascendant activatory reticular system (MRS). While various neurotransmitters (dopamine, acetylcholine, serotonin, GARA) or hormones (oxytocine, ACTH) as well as nitric oxide have been proven to modulate yawning, the exact anatomical structures involved in yawning are still not well characterized. However, there is experimental evidence that the hypothalamus (especially the paraventricular nucleus, PVN) plays a pivotal role n the elicitation of yawning. A current hypothesis claims, that activation of oxytocinergic neurons of the PVN projecting to extrahypothalamic regions including the hippocampus, pans and the medulla oblongata elicits yawning. (2, 3) Lately, the involvement of anatomical structures of the lower brain stem in human yawning has been emphasized by a case report of two patients presenting with pathological yawning due to acute brain stem ischaemia. (4) The authors speculated that the brain lesions, located at the pontomesencephalic junction, liberated the control of a putative yawning center caudal to the described lesions. However, little is known about the involvement of cortical-subcortical brain areas in the control of yawning. We hypothesized that if neocortical structures are involved in the control of spontaneous yawning, it is likely to find pathological yawning also in acute supratentorial stroke.
During an observation period of 6 months patients referred to our neurological department with the clinical suspicion of acute middle cerebral artery (MCA) stroke (symptom onset <12h) were observed immediately after arrival (i.e. during taking the patient's history or shortly thereafter) for an abnormal yawning frequency. This was arbitrarily defined as more than 3 yawns during 15 minutes. Healthy individuals yawn about 20 times per day, although the frequency differs substantially according to age, circadian rhythms and between individuals (range O-28 per day). (5, 6) However, more than 3 yawns per 15 minutes appears to be a reasonable cut-off between physiological and excesslve" yawning. The restriction to patients with a defined symptom onset shorter than 12 h prior to admission was chosen to minimize secondary reasons for an increased yawning frequency as tiredness associated with the sudden disturbance of the daily routine or space-occupying cerebral edema compromising blood flow in other vascular territories than the primarily affected. Basic clinical data, the time-point of examination, as well as the National Institutes of Health Stroke Scale (NIHSS) scores at admission and modified Rankin Scale (MRS) scores at discharge were assessed. All patients received computed tomography immediately after arrival. Values are given as mean±SD unless otherwise stated.
During the observation period, we registered seven acute stroke patients (three males, mean age 73*12 years, time interval between symptom onset and examination 5h±3.6h) with an abnormal high yawning frequency (mean 7.3±5.3 in 15 mm). Five patients arrived at the hospital during daytime (between 9.40 am. and 5.00 p.m.), while two patients were examined during evening hours (7.00 p.m. and 9.20 p.m.). Six patients experienced ischaemic and one patient a haemorrhagtc stroke. Five patients had left hemispheric and two patients right hemispheric strokes. Average NIHSS-score at admission was 17*4 (mean±SD). 4 patients had a mild disturbance of consciousness at arrival (patients being not alert but arousable by minimal stimulation, scored I on the LOC item of the NIHSS). There was no association between the level of consciousness and the yawning frequency. All patients presented with signs of cortical dysfunction (aphasia n=5, neglect n=3, gaze palsy n=6). Outcome at hospital discharge was unfavourable in 6 of 7 patients (MRS 5or6).
Computed tomography revealed in 5 of 6 patients signs of cerebral infarction of more than 1/3 of the MCA territory. No patient had CT signs of additional infarctions in other than the MCA-territory. Only one patient (No 2) had CT-morphological evidence of relevant space occupying cerebral edema at presentation.
The observation of pathological yawning in seven patients with acute anterior circulation stroke provides strong evidence that excessive yawning can be a sign of supratentorial lesions affecting the middle cerebral artery (MCA) territory.
The interpretation of this finding in the light of the known neuroanatomy of yawning is not straightforward. Both the hypothalamus and the brain stem, which include regions critical for yawning, are not supplied by the MCA. Currently, in particular the paraventricular nucleus (PVN) is believed to be the dominant diencephalic relay station, which sends oxytocinergic neurons to brain stem structures involved in yawning. One possible explanation for our finding of pathological yawning in supratentorial stroke could be that the lesions release the PVN from (presumably existing) neocortical control mechanisms, leading to an increase in the activity of the PVN. Whether this phenomenon is due to a sudden reduction of an inhibitory input from cortical structures resulting in a disinhibition of the PVN or due to an increase of excitatory input i.e. via anoxic depolarization of penumbral tissue remains speculative.
Due to the small sample size and often large lesions of our patients we did not attempt an exact topographic lesion analysis. However, there is little doubt that circumscribed neocortical dysfunctions can lead to excessive yawning, as several case reports on yawning after epileptic seizures (predominantly temporal lobe seizures) confirm. (7, 8) Of further interest is a case report of a woman experiencing excessive (spontaneous) yawning months prior to the development of a bulbar/pseudobulbar palsy due to amyotrophic lateral sclerosis. (9) The author speculated, that her pathological yawning was due to a dysfunction of the upper motor neurones loosing their inhibitory influence on the brain stem and lower motor neurones.
There is evidence from functional MRI studies, that neocortical areas are involved in the well known phenomenon of contagious (visually mediated) yawning: Platek et al. (10) describe substantial activations in the posterior cingulate cortex (Brodman area (BA) 31) and precuneus (BA 23) bilaterally as well as in the thalamus and parahippocampal gyrus (BA 30) of both hemispheres. In addition, Schürmann et al. (11) found significant bilateral activation of the anterior part of the superior temporal sulcus (STS) and in the posterior part of the right STS being involved in the visual processing of observed yawning. More importantly, they found a negative association between the subjective yawning susceptibility and the BOLD response in the left periamygdalar region, an area mostly supplied from the anterior choroidal artery. (12) Given the known tight hippocampal-hypothalamic neuroanatomical connections, and the fact that five of our patients had left hemispheric strokes partially involving temporal structures, case reports on yawning after temproal lobe seizures, it is tempting to speculate, that a dysfunction of hippocampal/periamygdalar structures may be linked with excessive yawning. However, the fMRI data refer to the specific phenomenon of visually mediated contagious yawning; thus, their relevance for spontaneous yawning remains speculative.
Our pilot study has several shortcomings: First, the cut-off for excessive yawning (3/15mm) is somewhat arbitrary, since healthy individuals may also yawn in clusters of similar frequency. However, we believe that frequent yawning in the setting of hospital admission and physical examination is inappropriate, especially since the majority of the patients arrived during daytime hours. Second, we did not keep a screening log, making estimates about the incidence of excessive yawning in acute stroke impossible. Finally, we did not attempt a qualitative or phenonienologic description of the yawns, i.e. if appearing solitary or in sequences.
We conclude that pathological yawning can occur in supratentorial stroke. Our observations suggest that neocortical brain areas have a regulatory effect on diencephalic and brain stem yawning centers. Further imaging studies will need to clarify the exact neuroanatomical structures involved in the higher control of (excessive) spontaneous yawning.
Daquin G, Micallef J, Blin O. Yawning. Sleep Med Rev 2001;5(4):299-312.
Goessler UR, Hein G, Sadick H, Maurer JT, Hormann K, Verse T. [Physiology, role and neuropharmacology of yawning]. Laryngorhinootologie 2005;84(5):345-351.
Argiolas A, Melis MR. The neuropharmacology of yawning. Eur J Pharmacol1998;343(1):1-16.
Cattaneo L, Cucurachi L, Chierici E, Pavesi G. Pathological yawning as a presenting symptom of brain stem ischaemia in two patients. J Neurol Neurosurg Psychiatry 2006;77(1):98-100.
Walusinski O. Yawning: unsuspected avenue for a better understanding of arousal and interoception. Med Hypotheses 2006;67(1):6-14.
Baenninger R, Binkley S, Baenninger M. Field observations of yawning and activity in humans. Physiol Behav 1996;59(3):421-425.
Muchnik S, Finkielman S, Semeniuk G, de Aguirre MI. [Yawning and temporal lobe epilepsy]. Medicina (B Aires) 2003;63(2):137-139.
Yankovsky AE, Andermann F, Dubeau F. Post-ictal forceful yawning in a patient with nondominant hemisphere epilepsy. Epileptic Disord 2006;8(1):65-69.
Williams DR. The yawning reflex: an upper motor neuron sign in amyotrophic lateral sclerosis. Neurology 2000;55(10):1592-1593.
Platek SM, Mohamed FB, Gallup GG, Jr. Contagious yawning and the brain. Brain Res Cogn Brain Res 2005;23(2-3):448-452.
Schurmann M, Hesse MD, Stephan KE, Saarela M, Zilles K, Hari R, et al. Yearning to yawn: the neural basis of contagious yawning. Neuroimage 2005;24(4):1260-1264.
Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of the human brain: cerebral hemispheres. Neurology 1998;50(6):1699-1708.

Journal of Neurology Neurosurgery & Psychiatry 2007;78(11):1166
Editorial Commentary
Can stroke localisation be used to map out the neural network for yawning behaviour ?
Olivier Walusinski (pdf)
Since the 19th century, cases of pathological yawning have occasionally been published in medical journals. In this issue of JNNP, O Singer et al present the first study to focus specifically on yawning during acute stroke affecting the middle cerebral artery (MCA) territory.
None of the seven patients suffering from abnormal repetitive yawning had diencephalic lesions. Classically, yawning is thought to originate in archaic brain structures common to all vertebrates. It appears to be a powerful muscular stretch which recruits specific control systems, particularly the paraventricular nucleus of the hypothalamus (PVN), the locus coeruleus, and the reticular activating system; these structures explain its ability to increase arousal. A persistent vestige of the past, yawning has survived evolution with little variation (1). O Singer et al suggest that neocortical brain areas have an inhibitory effect on the PVN, and that in certain MCA strokes, this region is liberated, provoking repetitive yawning. This hypothesis merits discussion.
According to J Lapresle (2), palatal myoclonus is the human homologue of a primitive respiratory reflex in gill-breathing vertebrates, submerged but not lost, reappearing when the inhibitory system is damaged by lesions to the dentato-olivary pathway.
We coined the term "parakinesia brachialis oscitans" (3) to describe cases of hemiplegia where the onset of yawning coincides with involuntary raising of the paralysed arm. We argued that a lesion in the internal capsule affecting an inhibitory pathway liberates certain subcortical structures that coordinate the massive inspiration of yawning and the motor control associated with quadrupedal locomotion.
In these examples, the loss of cortical inhibition following stroke releases a hidden function, phylogenetically more primitive. O Singer et al do not provide evidence of such an event.
Face-scratching, nose-face rubbing, yawning, and sighs are automatisms frequently reported after epileptic seizures. These behaviours are also considered a characteristic pattern in healthy subjects upon waking. Movement speed and repetition are the factors that vary, based on whether the context is physiological (sleep, arousal) or pathological (epileptic seizure, stroke). These behaviours are related to the brainstem and diencephalic activation that occurs when the cortex is disconnected from these areas (where the "central pattern generators" are located) by an epileptic discharge or a stroke.(4)
Adaptive behaviours depend on interactions between neural networks at various levels, requiring continuous mutual feedback. Yawning is an exterior manifestation of the tonic stimulation of the cortex by subcortical structures, particularly when the brainstem does not receive appropriate feedback from the cortex.
I agree with the conclusion reached by O Singer et al: further studies are necessary to determine the exact neuroanatomical structures involved in repetitive yawning during stroke and the pathophysiological role of this behaviour.
Walusinski O, Deputte BL. Le bâillement : phylogenèse, éthologie, nosogénie. Rev Neurol (Paris). 2004;160(11):1011-21.
Lapresle J. Palatal myoclonus. Adv Neurol. 1986;43:265-73.
Walusinski O, Quoirin E, Neau JP. La parakinesie brachiale oscitante. Rev Neurol (Paris). 2005;161(2):193-200.
Meletti S, Cantalupo G, Stanzani-Maserati M, Rubboli G, Tassinari A. The expression of interictal, preictal, and postictal facial-wiping behavior in temporal lobe epilepsy: a neuro-ethological analysis and interpretation. Epilepsy Behav. 2003;4(6):635-43.
Différentes localisations d'AVC peuvent-elles aider à concevoir la neuro-anatomie d'un comportement comme le bâillement ?
Depuis le XIX° siècle, quelques observations de bâillements pathologiques sont publiées dans la littérature médicale. Dans ce numéro du JNNP, O. Singer et al. proposent, pour la première fois, un travail dont le but assigné est l'étude des bâillements au cours d' AVC affectant l'artère sylvienne.
Aucun des sept patients souffrant de bâillements anormalement fréquents et répétés n'a de lésion diencéphalique. Classiquement, l'origine du bâillement est située au niveau du noyau paraventriculaire de l'hypothalamus (PVN) avec des projections vers le Locus Coeruleus et la réticulée ascendante du tronc cérébral. C'est par l'implication de ces structures qu'est expliqué son effet physiologique de stimulation de l'éveil. Le bâillement apparaît comme un comportement aux allures de vestige ubiquitaire et ancestral, persistant sans variation évolutive notable, O. Singer et al. suggèrent que des aires néocorticales auraient une fonction inhibitrice sur le PVN, levées par certains AVC, et permettant alors l'extériorisation de bâillements répétés. Cette proposition mérite d'être discutée.
J. Lapresle explique que la myoclonie du voile du palais est l'homologue humain d'un réflexe branchiale des vertébrés aquatiques, enfoui mais non disparu au cours de l'évolution, et réapparaissant lors d'un lésion de la voie dentatoolivaire.
Nous avons nommé « parakinésie brachiale oscitante » l'apparition , au cours d'une hémiplégie, d'un mouvement portant le bras paralysé vers la bouche, simultanément à un bâillement. Nous avons proposé de l'expliquer par l'effet d'une lésion, au niveau de la capsule interne, d'une voie inhibitrice libérant une structure sous-corticale, coordonnant l'inspiration ample et la motricité du membre supérieur, lors de la course chez les quadrupèdes.
Dans ces exemples, la perte de l'effet inhibiteur d'origine corticale, secondaire à l'AVC libère une fonction cachée, constamment inhibée et phylogénétiquement plus ancienne. O. Singer et al. ne proposent pas d'explication comparable, extériorisant une fonction archaïque normalement inhibée.
Se gratter le visage, se frotter le nez, bâiller, soupirer sont des gestes d'allure automatique fréquemment notés après des crises d'épilepsie, par exemple. Ces comportements se voient également aussitôt après l'éveil, chez des sujets sains. C'est la rapidité et la répétition de ces gestes, suivant qu'ils apparaissent de façon physiologique (sommeil, éveil) ou pathologiques (épilepsie, AVC) qui se modifient. Ils peuvent être rapportés à l'activation de centres de mouvements coordonnés, automatiques, situés au niveau du tronc cérébral, quand la crise épileptique ou l'AVC déconnecte le tronc cérébral du cortex.
Les comportements adaptés dépendent d'une interaction harmonieuse entre différents circuits neuronaux cortico sous-corticaux engageant une dialogue bidirectionnel continu. L'apparition de ces comportements répétitifs extériorisent l'activité stimulante des structures sous-corticales quand elles ne reçoivent pas de feed-back adapté des aires corticales.
Nous nous associons à la conclusion de O. Singer et al.: d'autres études seront nécessaires pour clarifier les structures neuroanatomiques mises en jeu lors des bâillements répétés survenant lors des AVC et pour en expliquer le rôle physiopathologique.