Buccal Hemineglect

Objectives  To determine whether the peripersonal and intrapersonal buccal space can be affected by a hemispheric stroke and to evaluate the clinical signs resulting from buccal neglect.

Methods  A prospective study comparing 2 groups of patients with hemiplegia, 1 with a right hemispheric lesion and the other with a left hemispheric lesion. Patients were selected consecutively on the basis of specific criteria at least 1 month after stroke.

Results  Buccal hemineglect was usually concomitant with other hemineglect phenomena resulting from lesions of the right hemisphere (10 of 12 in right lesions and 1 of 12 in left lesions). Clinical signs associated with this condition consisted of impaired swallowing (retention, defective insalivation, presence of food debris in the left hemibuccal space, loss of saliva from the left side of the mouth, and choking); loss of the ability to perceive salty, sweet, or acid tastes; and impaired buccal representation. These problems were usually incorrectly diagnosed initially. Outcome was usually favorable, but functional disorders persisted in some patients for more than 18 months. The underlying attention and representation mechanisms are discussed with reference to experimental lesions of the postarcuate (area 6) cortex in rhesus monkeys. The area around the mouth may be considered to be, as in monkeys, a peripersonal space, ie, probably of little functional importance. The lesion may involve area 6 or its projections to the thalamus or posterior parietal cortex.

Conclusions  Buccal hemineglect, which is likely to cause social embarrassment, should be considered whenever the oral phase of swallowing is impaired in a context of neglect syndromes. Prophylactic measures and rehabilitation can reduce the impact and complications of the condition (food bolus).

TO OUR knowledge, no studies have been published on neglect affecting half of the mouth, particularly the left half. Similarly, sensory extinction affecting taste does not seem to have been reported, unlike other forms of sensory extinction. Reviews of dysphagia resulting from stroke^1-4 report various mechanisms,⁵ but neglect is not mentioned. Hemineglect in the "pericutaneous buccal space" was described in monkeys with contralateral inferior area 6 (postarcuate cortex) lesions.⁶ Monkeys use the mouth for exploration, grasping, and defense; adult human mouths also have these rudimentary functions. Taste is involved in short-distance orientation in all the higher vertebrates.⁷

Our suspicions that buccal hemineglect occurs in humans⁸ led to this prospective study. We investigated the clinical features and onset of buccal hemineglect in a group of 12 patients who had had a stroke of the right hemisphere. Buccal neglect can have 2 types of functional expression: choking and social handicap as a result of dribbling and regurgitation of food some time after eating.

The reference population comprised 12 right-handed patients with neglect syndrome admitted to the hospital after attending a neurorehabilitation clinic between November 1996 and February 1998. Magnetic resonance imaging had confirmed the existence of an ischemic lesion in the area of the right middle cerebral artery. Patients had no severe facial motor or sensory disorders, no buccofacial apraxia or previous neurological history, no mental impairment, and no sign of pseudobulbar impairment or lacunar syndrome (vomiting reflex present).

All patients were examined after the acute phase, more than 30 days after stroke, by the same person (N.M.), who was unaware of the hypothesis being tested.

Motor deficit was assessed using conventional medical and neurological examination. Scores were based on the Scandinavian Neurological Stroke Scale⁹ for motor strength of the arm, hand, and leg.

Evaluation of anosognosia and unilateral spatial neglect was performed using tests of body neglect¹⁰ and self-representation. Evaluation of unilateral spatial neglect alone consisted of visual and motor tests (Diller¹¹ and Bell¹² tests, copying a scene,¹³ drawing a clock, and a line bisection test¹⁴) and perception tests (intertwined figure,¹⁵ reading and writing tests, and body image [body and facial puzzles]). Scores for unilateral spatial neglect were based on results of the Diller test interpreted according to the criteria determined by Diller.¹¹ Anosognosia was assessed using test procedures described by Bisiach et al¹⁶ and rated as severe (+++), moderate (++), slight (+), or normal (0). The hemiasomatognosia scores were rated as symptom being present or absent.

Four methods were used to assess buccal neglect (Table 1): (1) examination of the face, lips, tongue, cheeks, pillars, and velum of the palate to test motor performance, praxis (opening and closing of mouth, sticking out one's tongue, repeating [o/õ], adjusting lips to imitate a kiss or a smile, etc), voluntary cough, and surface sensitivity (stimulation of one side and then both sides to check for extinction); (2) exploration of the sense of taste, ie, perception of salty, sweet, and acid tastes (lingual V near the tip of each half of the tongue separately and then simultaneously on both sides of the tongue to identify extinction); (3) exploration of swallowing to check for choking, dribbling, and particularly any loss of saliva through the lips, stasis of the saliva, or food and to pinpoint their location; and (4) a questionnaire about dribbling and swallowing (Are you aware of dribbling? Do you feel saliva building up in your mouth? Or dribbling down your chin? When you swallow, do things sometimes go down the wrong way? Do you see saliva running down your clothing? Does this distress you? Do you sometimes find food in your mouth after meals? How long after? Does this sometimes come out of your nose? Is there any difference between solids and liquids?).

Brain computed tomographic scanning or magnetic resonance imaging was used to assess brain lesions. Radiovideoscopic examination during swallowing was performed in 6 patients.

Difficulties in swallowing, dribbling, buccofacial sensitivity, and taste were investigated in the same way in 12 patients with a cerebrovascular lesion of the left hemisphere. These patients met the same selection criteria, and their language impairment did not prevent them from giving reliable responses in the tests performed.

The χ² test was used to compare groups for their main features.

The test group consisted of 5 women and 7 men (mean age, 51 years; range, 30-75 years). All patients had had a right middle cerebral artery stroke with superficial (n = 1), deep (n = 4), or superficial and deep (n = 7) infarction. There was an accompanying parietal lesion in all patients with superficial infarction. Hemiplegia was serious or moderate in 9 patients and absent or slight in 3. Visuospatial neglect was severe or moderate in 8 patients and slight or absent in 4. Anosognosia was severe or moderate in 9 patients and absent in 3. Disorders of the body image were severe in 8 patients and slight or absent in the other 4. Dysarthria observed in 4 patients was not explained by a single motor disturbance and was too slight to impair intelligibility. The main signs are shown in Table 2.

The control group consisted of 7 women and 5 men (mean age, 46 years; range, 21-56 years) with left middle cerebral artery stroke with superficial (n = 4), deep (n = 3), or superficial and deep (n = 5) infarction. The basal ganglia were impaired in 8 patients and the parietal area was impaired in 7. Hemiplegia was severe or moderate in 9 patients and slight or absent in the other 3. No patients had visuospatial neglect. There was 1 patient with severe or moderate anosognosis, which was present but slight in 3 patients and absent in 8. Disturbance of body image was severe in no patients, slight in 2, and absent in 12. No difficulties in swallowing, no dribbling and no extinction of taste were reported in this group (Table 2).

The test and control groups differed significantly in visuospatial hemineglect (χ² = 17.14; P<.001), anosognosis (χ² = 12.05; P =.007), retention (χ² = 12.00; P =.007), sensory loss (χ² = 8.47; P =.04), and loss of taste (χ² = 14.67; P =.005). No differences were found for motor deficit (χ² = 1.73; P =.64), dribbling (χ² = 7.0; P =.07), or aspiration (χ² = 4.8; P =.09).

Buccal hemineglect (generally on the left side) is characterized by an inability to detect the presence of food or saliva in the left half of the buccal cavity or to initiate chewing when food is on the left side or swallowing when saliva or food in the left side of the mouth needs to be swallowed. This is a possible result of lesions of the right hemisphere. In patients with right-sided lesions, buccal hemineglect was usually accompanied by other neglect phenomena. None of our patients had buccal hemineglect alone, but this might be attributable to a recruitment bias. However, in 1 patient, the signs of buccal hemineglect progressively dissociated from signs of neglect, the buccal hemineglect lasting longer than the other phenomena.

Difficulties in swallowing combine stasis of saliva and food debris, which might be intact and only partially impregnated with saliva; saliva and food are found in the left side of the mouth space (crevices and buccogingival grooves) between meals. These food residues might give rise to difficulties in swallowing, which might be delayed. Sudden swallowing of these items can cause aspiration.

Retention of liquids or solids might give rise to delayed swallowing or spillage through the lips. Dribbling often occurs as a result of saliva escaping through the left side of the lips. The patient is unaware of dribbling onto the chin and ignores it, although sensitivity of the chin is normal. Nonspecific dribbling of this type might also occur in facial paralysis but can be distinguished from it by the fact that the patient with neglect is unaware of it and that it occurs intermittently throughout the day. Liquids rarely cause choking, and nasal regurgitation does not occur. In our patients, radiovideoscopic explorations confirmed the prominence of problems during the oral phase of swallowing (Figure 1). Patients often remain unaware of these difficulties, which usually can be corrected using verbal encouragement.

Motor disorders can occur in the absence of facial or lingual paralysis, and mastication is asymmetrical. Food is transferred preferentially to the right half of the mouth. We also noticed that the teeth were not brushed on the left side of the mouth, whereas the face was washed normally on both sides. This suggests that this is a specific form of neglect.

We found an extinction phenomenon affecting touch and taste. Stimulation of the cheek, lips, and vellum of the palate on one side at a time was perceived on both sides, but during simultaneous stimulation it was perceived only on 1 side (6 of 12 patients). Similarly, salty, sweet, and acid tastes were all perceived separately in the left and right sides of the mouth, whereas simultaneous stimulation was perceived only in the right side. These signs are similar to the sensory losses reported for the limbs.^17-19 Errors about the side being stimulated may be interpreted as a form of alloesthesia or "allogueusia." The loss of touch and taste are probably responsible for food being forgotten.

Body representation tests involving drawings or puzzles may reveal major anomalies of the representation of the mouth, which may be completely or half missing. The mouth may also be missing on profile drawings (Figure 2). Although these disorders have not been specifically reported, they can be seen in figures included in articles about patients with hemineglect,^20-24 particularly those who report changes in the drawings produced by artists with right-sided brain lesions (Figure 3). These defects are not overlooked as a result of left-sided visual neglect because there is a selective absence of representation within a space that is otherwise well depicted and because these defects are repeated. The left half of the mouth is represented asymmetrically during the recovery phase. Facial puzzles may be severely disrupted (Figure 4). We did not find any mistakes in representations of the mouth by patients with lesions of the dominant hemisphere, particularly in those with aphasia, or generally in those with impaired verbal expression or who are unable to consume food by the oral route. Failure to represent the mouth should not therefore be interpreted symbolically, but as the expression of a defect in body image of the same nature as the familiar defects in representing the limbs. In contrast, symbolic representations are exaggerated in the works of famous artists. However, the mouth is missing in some "primitive" art or in children's drawings.

Signs of hemineglect can be improved by vestibular stimulation^25,26; when this procedure was carried out for 4 patients in our group, it resulted in temporary abolition of all signs of buccal neglect.

The outcome was generally favorable, and the signs disappeared in 2 to 8 months. Some patients continued to have abnormalities for more than 18 months. The objectives of rehabilitation were to improve swallowing, to prevent aspiration of food, to protect the airways, and to allow the swallowing of saliva. Exercises designed to make the patient aware of dribbling or the regurgitation of residues were given in addition to the usual rehabilitation care.⁵

Abnormal behavior (in terms of chewing and handling the food bolus inside the mouth, in front of the V of the tongue, and then in passing it back over the V of the tongue) results in poor mastication and insufficient salivation. This can lead to stasis of food in the mouth and aspiration of particles not impregnated with saliva into the airways. The time taken to swallow food might be prolonged because it has not been adequately salivated and has insufficient viscosity. The transfer to the pharynx is no longer synchronized with laryngeal ascension and occlusion. As a result, the food bolus enters the oropharynx when the pharynx reopens and the esophageal sphincter is closed. So food enters the respiratory tract. The amount of time before a food bolus is noticed in the left side of the mouth is longer; liquids flow into the pharynx before their presence in the mouth has been detected.

Analysis of disorders reported on the basis of the main theoretical models of unilateral spatial neglect is unclear for at least 2 reasons: (1) the specific features of the mouth as an anatomical/functional space and a represented space and (2) concomitant sensory extinction, particularly of the sense of taste.

The buccal cavity is a single median cavity with contralateral cerebral cortex areas for each half. Like other segments of the body, the mouth is part of the concept of body image, as a container and as its contents.²⁷ It is a space that is sometimes peripersonal (when the mouth is open or filled) and sometimes internal (when the mouth is closed). Two features distinguish the mouth from other segments: the purpose of this space in functional terms (digestive, communicative, expressive, etc) and the specificity of the sensory supply, in which sensory or taste inputs are particularly important and in which there is no visual input.

The term "representation" is used to designate the processing of information and the result of this processing (retention of information in memory, regulation, and orientation of actions). It is possible to define an egocentric space consisting of 1 (or several) referential(s) for organized motor programs of the various segments of the body. Results of this analysis, therefore, suggest the existence of distinct spatial referentials. Some observations^10,28,29 suggest a dual dissociation between corporal and extracorporal neglect and the space represented (by mental evocation). Different anatomical areas are important for personal, peripersonal, and extrapersonal spaces. Results of the study by Rizzolatti et al⁶ suggest that the perceptual deficit in monkeys with inferior area 6 lesions and secondary thalamic degeneration in the ventrolateral nuclear complex concerns the personal and peripersonal space and the motor side, the organization of the head, the mouth, ie, "cutaneous peribuccal space," and hand and arm movements. The animal could not grasp food with its mouth when it was presented near the face or even in contact with the mouth on the side contralateral to the lesion. The inferior area is connected to the rostral part of the inferior parietal lobule. The properties of the neurons of this sector are similar to those of the inferior area 6. The thalamus, basal nuclei, and internal capsule were altered in 11 of 12 of our patients, and 7 of 12 had lesions of the parietal area (1 isolated).

Extinction is usually considered to be a minor aspect of unilateral spatial neglect.⁹ Some patients with isolated extinction of taste (patient 5 in our group) with no unilateral spatial neglect and others with unilateral spatial neglect without loss, regardless of the form taken, have also been reported^18-30 (patient 8 in our group). To our knowledge, however, extinction of the sense of taste (salty, sweet, and acid), alone or combined with unilateral spatial neglect contralateral to the lesion, and its impact on swallowing have not been described. Taste-afferent fibers join the nucleus of the solitary tract; the axons of second-order neurons proceed to the contralateral nucleus ventralis posterior medialis thalami; the third set of neurons communicates with the cerebral cortex: area 43 (gustative area), areas 40 and 39 (parietal associative area), and area 38.³¹ Specific pathways support salty, sweet, and acid perception. The primitive taste cortex lies close to the somatic representation of the mouth and to the motor centers controlling jaw and tongue movement. They are contiguous with the gyrus supramarginalis. Our findings suggest that the organization of the taste system is similar to that of the other sensorial systems and is indicative of hemispheric lateralization: (1) left and right within the sense of taste and (2) right (especially posterior parietal areas) within the gustative space.

The importance of attention factors vs perception factors has been highlighted. An imbalance between 2 attention factors may lead to channeling attention to the dominant side from the ipsilesional side relative to the lesion¹⁸ and hence to sensory loss.³² This imbalance may make itself felt at various levels: (1) qualitatively, it may affect the orientation of motor behavior in physical space during the processing of the mental representation and (2) quantitatively, with differing degrees of extinction.

There seems to be a specific syndrome of buccal hemineglect with an inability to detect food and saliva in the left half of the mouth, to initiate chewing when food is in the left half of the mouth, or to swallow when the saliva or food on the left side has to be swallowed. The clinical signs of this syndrome include dribbling, choking, retention of food, and sensory loss (particularly of the sense of taste). Distortions of the representation of the body and anosognosia are associated with buccal hemineglect. The disorder is improved by verbal encouragement. The seriousness of buccal hemineglect results from the potentially life-threatening risk of choking on food. The main characteristics of associated clinical signs and buccal hemineglect are similar to those encountered in other types of hemineglect. The outcome of buccal hemineglect is shorter. However, there are some distinct features, particularly the high incidence and the nature of the distortions of the body image in our group of patients. The brain lesions are located in the right hemisphere (parietal area, thalamus, or basal ganglia). Certain theoretical models based on referential or representational hypotheses look promising; they offer a particularly good integration of the anatomical and functional characteristics of the buccal cavity. The space around the mouth could be a specific peripersonal space, as in monkeys. There might be a specific lesion in the inferior area 6 or in its projections to the thalamus or posterior parietal area. Isolated buccal hemineglect would be anticipated, without extinction of taste sensory in patients with left-sided damage of the postarcuate (area 6) cortex. Rehabilitation should make it possible to prevent or reduce the main symptoms.

Accepted for publication April 11, 2000.

We thank M. O. Thisse, MD; L. Le Chapelain, MD; H. de Barmon, OT; D. Bret, OT; V. Ganis, OT; L. Porée, MD; X. Porée, MD; and J. Kang, who all helped with this work. We are grateful to the patients for taking part in this study.

Reprints: Jean-Marie André, MD, Institut Régional de Réadaptation de Nancy, 35, rue Lionnois, 54042 Nancy-Cedex, France (e-mail: jmandre@irr.u-nancy.fr).