Hemispatial Neglect, Balance and Eye-Movement Control Paresh Malhotra, Elizabeth Coulthard and Masud Husain

Hemispatial neglect, balance and eye-movement control Paresh Malhotra, Elizabeth Coulthard and Masud Husain

Purpose of review Introduction Disorders of spatial awareness and balance following The neglect syndrome is a common disorder following stroke are common but often under-diagnosed. They lead to stroke, particularly prominent and severe after right- poor outcome and frequently coexist. Here we focus on hemisphere lesions. Patients with hemispatial neglect recent progress in the understanding of the mechanisms often fail to be aware of or acknowledge objects or people underlying these disorders and potential therapeutic on their contralesional side (the left side for patients with advances. right brain damage), attending instead to ipsilesional Recent findings items (the right side for right-hemisphere stroke Right-hemisphere networks are important for both spatial patients). Many such patients have intact visual fields attention and postural awareness. Neglect patients show and can initiate eye or limb movements contralesionally. multiple oculomotor impairments including reduced But despite this they appear selectively to attend to items saccade amplitude and difficulty retaining spatial locations on their ipsilesional side. across saccades. There has been controversy regarding the brain regions associated with neglect, although most From a clinical perspective hemispatial neglect repre- studies show the right inferior parietal lobe to be crucial and sents an important disorder because enduring neglect is new imaging modalities have provided insight into neglect associated with a poor functional outcome following caused by subcortical stroke. The ‘pusher syndrome’ is a stroke [1]. In addition to its clinical importance, neglect poorly understood balance disorder where patients push has also attracted a great deal of interest because of the towards their paretic side, resulting in falls. It may involve insights it provides into the brain networks that underlie impairment of subjective verticality but experimental studies spatial awareness and cognition [2,3], as well as non- have reported diverse findings. Advances in treatment for spatial functions such as the ability to sustain vigilant neglect include the successful use of prism adaptation and attention [4]. Here we review recent studies that have pilot data suggesting noradrenergic stimulation may begun to investigate the relationship and impact of the improve search in selected patients. cognitive deficits in neglect with eye movement and Summary balance control in these patients. In addition, we consider New experimental techniques have provided insight into the recent controversies regarding the anatomical basis of the debilitating disorders of spatial and postural awareness that neglect syndrome, the relationship of these findings to often follow stroke. There are currently no widely used current knowledge about the cortical control of balance therapies for neglect but both new behavioural techniques and eye movements, the clinical impact of neglect on and pharmacological methods are promising. patients and novel treatments for the disorder. More general recent reviews of neglect are also available for Keywords the interested reader [5–8]. parietal lobe, pusher syndrome, stroke, unilateral neglect, vestibular Eye-movement studies Oculomotor studies in neglect have tended to concen- Curr Opin Neurol 19:14–20. ß 2006 Lippincott Williams & Wilkins. trate on lateralized saccadic impairments, but more Institute of Cognitive Neuroscience, UCL, Queen Square, London WC1N 3AR, UK recently there has been interest in a failure to update and Imperial College, Division of Neuroscience, Charing Cross Hospital, London W6 8RF, UK representations of space across saccades, leading to recur- Correspondence to Paresh Malhotra, Institute of Cognitive Neuroscience, UCL, sive search in patients with neglect. Queen Square, London WC1N 3AR, UK Tel: +44 (0)207 679 5499; fax: +44 (0)207 916 8517; e-mail: [email protected] Contralesional saccades in neglect Current Opinion in Neurology 2006, 19:14–20 Patients with hemispatial neglect often spontaneously

Abbreviations fail to explore contralesional space in everyday life, confining their eye movements instead to their ipsi- IPL inferior parietal lobe IPS intraparietal sulcus lesional side. It is vital to understand though that the SPV subjective postural verticality demarcation between the neglected and the non- STG superior temporal gyrus SVV subjective visual verticality neglected ‘field’ of vision need not be neatly through TPJ temporoparietal junction the vertical meridian of the visual field, or the midline of the search display (see Fig. 1). The degree of neglect, as ß 2006 Lippincott Williams & Wilkins 1350-7540 assessed for example by fixation patterns [9], varies

Figure 1 Eye-movement scan path of a right-hemisphere patient stimuli [17]. The authors argued that paying attention to with neglect searching for letter Ts embedded among distractor an item at central fixation leads to shrinkage of the letter Ls effective visual field, worse to the contralesional side. Thus when searching a display with high attentional demands, right-hemisphere patients with neglect may have a constricted effective field of vision, accounting for the generally small-amplitude saccades they make.

Saccade amplitudes to the left may be particularly reduced under some circumstances, as demonstrated by Niemeier and Karnath [18]. These researchers used two interesting variants of a visual search paradigm on right-hemisphere patients with and without neglect. In the first task subjects freely searched for the letter A in a random array of letters. The subjects’ scan paths when performing this task were recorded and used to generate the second ‘replay’ condition. In this second task, participants viewed the same display but had to follow a red Black dots show fixations and white lines denote saccades. Note that neglect of the left side of the array is not strictly down the vertical square, which took exactly the same path as the fixations meridian of the display. In this case the patient found some Ts well to the made by subjects in the previous task, thus replaying it. left of centre in the upper part of the screen. The figure also vividly In the first (free search) condition, neglect patients illustrates how such patients with neglect often refixate locations they have already searched on the right, at the expense of exploring the left of tended to search on the right side of array and ignore the array. Such recursive ipsilesional search may, in some patients, be the left, but there was no major discrepancy between due to a failure to keep track of spatial locations that have already been leftward and rightward saccade amplitude or frequency. explored (from [10]). However, in the replay task, saccades were of reduced amplitude, particularly when leftward, despite the pre- between patients, and even within the same patient scribed eye movements being identical to the first task. depending upon the task and stimulus characteristics This finding suggests the contralesional saccadic deficit [10]. Thus a greater degree of contralesional neglect in neglect may be most prominent for stimulus-driven or is often found when patients are asked to search a dense, reflexive saccades (replay task) than for volitional move- cluttered visual scene compared to a non-cluttered one ments (free search condition). [10]. Moreover, there is often a gradient of neglect, such that the probability of finding a target in contralesional Spatial working memory and saccadic control space falls off gradually with distance, rather than When searching a visual scene, neglect patients not only patients demonstrating an abrupt failure to detect targets fail to explore contralesional space, they often keep at the midline [9,10]. Even in the dark, the resting refixating items on their ipsilesional side [9] (for example, position of gaze of right-hemisphere patients with neglect see Fig. 1). Many of these refixations are associated with a is far to the right of the midline [11]. failure to keep track of spatial locations across saccades, with patients revisiting locations they have previously When asked to shift their gaze contralesionally, neglect visited, unaware that they have looked there before patients demonstrate a variety of impairments, often [10,16,19]. Mannan and colleagues [10] monitored being slower to initiate saccades in that direction or eye movements while patients searched for target letter making small, multistep saccades [12–14]. When search- Ts among distractors. Subjects were asked to click a ing a visual scene, they do not necessarily make fewer response button to indicate whenever they identified a saccades in the contralesional direction [15,16], but their new target T (one they had not previously found). Right- eye movements are generally of small amplitude [15]. hemisphere neglect patients with lesions involving either This behaviour may be related to constriction of the the intraparietal sulcus (IPS) or the inferior frontal gyrus ‘effective field of vision’ when searching a dense array frequently misjudged targets on the right as new when for small targets among distractors. A recent study of they had in fact fixated and clicked on them before. In right-hemisphere-lesioned patients who had recovered neglect patients with IPS damage (see Fig. 2 for anatomy) from neglect and had full visual fields demonstrated that the probability of re-clicking on a target increased with when performing an attentionally demanding task at time – and number of intervening saccades – consistent fixation these individuals often missed stimuli within with a deficit in remembering target locations or in 68 of the fovea bilaterally, but worse to the left. However, remapping and updating representations of target if the task demands at fixation were made easier, they locations across saccades. In contrast, the re-clicking encountered far less difficulty in detecting para-foveal behaviour of neglect patients with inferior frontal gyrus

Figure 2 Cortical brain regions typically associated with neglect Vestibular cortex and relationship to neglect after right middle cerebral artery stroke include the inferior Severalimagingstudiesinhealthyhumanshaveimplicated parietal lobe (IPL), consisting of the angular (Ang) and supra- marginal (SMG) gyrus, the temporoparietal junction (TPJ) and a network of regions activated by vestibular stimulation. the inferior frontal gyrus (IFG) For example, one recent positron emission tomography study using caloric stimulation identiﬁed foci of activity in the inferior parietal lobe (IPL), temporoparietal junction (TPJ), posterior insula and frontal eye ﬁelds, most prominently in the right hemisphere of right-handed individuals (Fig. 2) [25]. This cortical vestibular network has often been considered to play a role in perception of spatial orientation relative to gravity. But a major new study proposes that it plays a more general role in representing the physical laws of motion, by generating an internal model of gravity that calculates the effects of gravity on external objects as well as on the body’s own parts, such as the arm as it reaches to grasp an object [26]. As we discuss later, the areas that form part of the vestibular cortical network are often also damaged in patients with neglect.

In this context, it is interesting to note that several significant asymmetries in the vestibulo-ocular reflex have A more recent study [35] has suggested a special association with the mid-portion of the superior temporal gyrus (STG), but this is controver- been identified in stroke patients with posterior cortical sial. Grey areas denote regions that are activated by vestibular stimu- lesions, some of them associated with hemispatial neglect lation in healthy individuals [25]. Note their proximity to the regions [27,28]. For example, the vestibulo-ocular reflex slow- typically damaged in neglect. Different regions within the intraparietal sulcus (IPS), which separates the inferior parietal lobe from the superior phase time constant is reduced contralesionally in right- parietal lobe (SPL), are often activated in neuroimaging studies of hemisphere patients with neglect, with this impairment saccadic eye movements and spatial working memory. J, sulcus of being associated with right TPJ lesions [27]. It has been Jensen; MFG, middle frontal gyrus; SF, sylvian fissure; STS, superior temporal sulcus. proposed that the TPJ may be responsible for a slow integrative process, extracting inertial semicircular signals to establish a head-centred spatial reference frame [27]. lesions was high from the very beginning of the search, Lesions to such a representation might also play a role in suggesting that these frontal patients may be displaying a degraded representations of contralesional space in the ‘perseverative’ type of behaviour, failing to stop them- neglect syndrome. By contrast, lesions of the posterior selves returning to previously inspected target locations insular cortex may lead to deficits in perceiving visual and clicking on them. verticality [29].

It is important to appreciate that such recursive searching Pusher syndrome of ipsilesional space, at the expense of exploring contra- Perception of verticality – visual and postural – has lesional regions, does not occur in all patients with become an extremely controversial issue in one particular neglect. Similarly, deficits in spatial working memory disorder of balance that has received a great deal of are also not universal in the neglect syndrome. Patients interest lately: the ‘pusher syndrome’. Stroke patients with right lateral parietal involvement appear to be particu- with this disorder actively push away from the ipsilesional larly vulnerable [10,20]. Moreover, the high-resolution side and have a tendency to fall towards their paretic, anatomical analysis of Mannan and her colleagues [10] contralesional side (the left side for right-hemisphere indicates that damage specifically to the IPS may be patients). This phenomenon is more prominent when critically associated with spatial working-memory impair- patients are upright compared to when lying down. ments. These findings correlate reasonably well with Although initially associated with the neglect syndrome functional imaging studies which demonstrate saccade and anosognosia (denial of a contralesional deficit), sub- [21] and spatial-memory related [22] activity in this region. sequent investigation has shown that the pusher phenomenon is doubly dissociable from both disorders, but may Balance control and neglect coexist with them in many individuals. But what is the Disorders of balance are extremely common in stroke- mechanism underlying this behaviour? Could it be due to rehabilitation settings [23], and they frequently coexist an alteration of the sense of verticality? with spatial neglect in the same patient [24]. However, both their pathophysiology and the precise nature of any Karnath and his associates [30] studied both subjective relationship with the neglect syndrome remain unclear. visual verticality (SVV) and subjective postural verticality

(SPV), testing the latter with eyes open or closed. They Recent controversies on the anatomy of reported that their five right-hemisphere pusher patients neglect showed no deficit in making SVV judgements, but they Until recently, the majority of patients with left neglect were impaired in SPV, and this only when vision were considered to have damage to the right IPL or TPJ, was occluded. Curiously, they found that the SPV was with a smaller proportion having damage to right inferior rotated clockwise, to the ipsilesional side, so the right- frontal gyrus or subcortical regions (Fig. 2) [34]. However, hemisphere pusher patients considered themselves to the advent of new scanning modalities and lesion map- be upright when in fact they were tilted approximately ping tools has led to a profusion of studies attempting to 188 to the right of true vertical. Control stroke patients determine the critical lesion loci associated with the who did not show the pusher phenomenon, including syndrome. patients with neglect, demonstrated no alteration of either SVV or SPV. The findings in the pusher patients Cortical regions seem perplexing, given that they actually pushed actively In 2001, using a combination of computed tomography towards the left in everyday life. Moreover, in natural and magnetic resonance data that had originally been circumstances such pushing is obvious with vision of the obtained for clinical purposes, Karnath et al. [35] con- environment permitted, not simply when the eyes are cluded that the mid-portion of the right superior temporal closed. The investigators proposed that the pusher gyrus (STG), well anterior to the TPJ, was the critical phenomenon arose from a conflict or mismatch between lesion location associated with neglect. One particularly visual and postural vertical. However, the results of two contentious aspect of the study involved the removal of other studies challenge this view. all patients with visual-field deficits from the analysis. However, the same group of researchers has more First, Perennou and his colleagues [31] reported that the recently come to similar conclusions using a larger popu- three right-hemisphere pusher patients they studied lation that included patients with field deficits [36]. demonstrated a tilt of the pelvic girdle to the left com- Other investigators have also recently used clinical scans pared to control stroke patients and healthy individuals. for lesion analysis, but none of them has demonstrated a This effect was worse without vision, but still clearly special association between neglect and STG damage evident even when vision was permitted. The authors [37,38,39,40,41]. concluded that there was in fact a tilt of SPV to the left in their pusher patients. More recently, Saj et al. [32] Mort and his colleagues [42] employed a different examined the SVV, lying and sitting, in 17 right-hemi- approach, using painstaking mapping methods and dedi- sphere patients. Participants used a luminous rod in the cated high-resolution magnetic resonance imaging pro- dark to indicate verticality. Patients without the pusher tocols to delineate the lesions of 35 right-hemisphere syndrome demonstrated an anticlockwise (leftward) tilt stroke patients. Lesions were demarcated on every of SVV, and this was significantly worse in neglect (1 mm) slice of each individual’s scan, so they could be patients without pushing behaviour (just over 68 to the compared with high fidelity between patient groups. The left of true vertical). However, the pusher patients, most subsequent analyses (on normalized and un-normalized of whom also had left-sided neglect, showed completely brain scans) showed that damage to the angular gyrus of the reverse effect: their SVV was tilted clockwise by 78 (to the IPL (Fig. 2) was necessary for neglect to occur after the right). Importantly, this tilt was modulated by pos- right middle cerebral artery stroke. Although the right ture, with pusher patients showing less rightward tilt STG was involved in 50% of middle cerebral artery when lying than sitting. neglect patients, damage there was neither necessary nor sufficient (without concomitant IPL involvement) Clearly, the findings of these investigations of the pusher to produce neglect. Given the common vascular supply syndrome raise far more questions than they answer. to both IPL and STG, it is perhaps not surprising that There seems to be no consensus on the issue of either both these regions should be damaged in many patients; SVV or SPV tilt in this group of patients. Nor is there a but the findings of Mort et al. [42] show that the critical clear explanation for how both these factors might lead relationship is between the IPL and the neglect syn- to the pusher phenomenon. Methodological differences drome. The debate on this issue has sometimes been between the studies are likely to be critical in accounting extremely lively (see the correspondence in [43,44]). for the different patterns of behavioural results that have been obtained. Emerging anatomical studies suggest inter- Mort and colleagues [42] also examined right-hemisphere estingly that lesions of the posterior thalamus, rather than patients with posterior cerebral artery stroke and reported the vestibular cortex, are required for the pusher syndrome that large lesions in this territory extending from the to be apparent [33]. The time is ripe for a detailed occipital to medial temporal lobe can also be associated investigation of this important disorder of balance that is with neglect. When the lesions of non-neglect posterior commonly associated also with the neglect syndrome. cerebral artery patients were subtracted from those with

neglect, the parahippocampal region, an area known to generally be picked up as commonly as dysphasia at have strong reciprocal connections to the parietal lobe, the bedside, often because clinicians do not test for it. was identified as a critical location. Further studies will be required to examine whether neglect in such posterior Prognosis and treatment of neglect cerebral artery territory patients is due to diaschisis or Recent studies continue to emphasize the poor prognosis hypoperfusion in distant parietal regions. associated with the presence of enduring neglect. For example, Gillen and colleagues [50] report that right- Subcortical lesions and magnetic resonance perfusion hemisphere patients with neglect stay longer in hospital, studies make slower progress with rehabilitation and have It has been known for some time that pure subcortical particular problems with grooming, bathing and walking. strokes can also be associated with neglect, particularly Even when matched with non-neglect patients who had after right-hemisphere damage. Lesion mapping has equally poor functional status on admission to a rehabi- implicated the putamen, pulvinar and, to a lesser extent, litation unit, individuals with neglect still had longer the caudate nucleus [45]. Hillis and colleagues [46] have admissions and slower progress. A detailed prospective also used the combination of diffusion-weighted and investigation of the predictive value of early neuropsy- perfusion-weighted imaging – which is sensitive to chological examination in 228 acute stroke patients also poorly perfused (but potentially reversible) brain tissue found neglect and perceptual impairments to be the most – to identify areas of cortical hypoperfusion in patients important independent prognostic indicators of final out- with subcortical infarcts [46]. In their group of patients come [1]. However, the majority of patients with acute with right-hemisphere stroke, all of whom were neglect no longer demonstrated neglect at the time of examined and scanned within 24 h of symptom onset, their follow-up examination (6–10 months post-stroke). the seven who had neglect all had areas of cortical Thus although the presence of acute neglect appears to hypoperfusion and the seven without neglect showed predict long-term functional outcome in a very strong no such hypoperfusion. The authors concluded that way, long-term disability in this group may not be directly many, if not all, cases of neglect that appeared to be related to persistent neglect, but other factors that may due to a solely subcortical infarct could be accounted for coexist with neglect. Future studies may need to examine by hypoperfusion in right cortical middle cerebral artery this issue more closely. territories. But exactly which cortical regions? Perhaps the two most interesting advances in treatment A subsequent study by the same group has recently are the use of prism adaptation and, more recently, a pilot reported that left egocentric neglect was most strongly proof-of-concept study involving a noradrenergic drug. associated with hypoperfusion of the right angular gyrus, The beneficial effects of short bursts of prism adaptation but that allocentric neglect was most strongly associated on standard clinical tests and everyday behaviour can last with hypoperfusion of the right STG [47]. There was no as long as 5 weeks [51] and have been shown to affect association between the subcortical infarct location and visuo-verbal and haptic tasks as well as multiple sensory egocentric or allocentric neglect. But clearly the debate modalities [52–54]. Ocular scanning behaviour also on the critical cortical location associated with neglect shows significant differences following prism adaptation looks set to continue, even from a subcortical perspective. such that there is an improvement in oculomotor bias and One important clinical implication of perfusion magnetic neglect dyslexia, with further left-sided exploration and resonance studies is the finding by Hillis et al. [48] that an increased amplitude of the first contralesional saccade left neglect is a better indicator of changes in brain [55]. Morris et al. [56] could not find a change in spatial perfusion in right-hemisphere stroke patients than the attention, as measured on a serial search task, after prism National Institute of Health Stroke Scale (NIHSS). adaptation. Because there was no evidence for a change in Stroke scales typically do not give much weight to deficits the gradient of (right-to-left) search performance after that are more typically associated with right-hemisphere prism adaptation, these investigators felt that prism adap- lesions; by contrast, language deficits following left-hemi- tation was unlikely to be mediated by an adaptive redis- sphere stroke are often easily detected by clinicians and tribution of spatial attention. The exact mechanisms therefore also figure prominently in many stroke scales. underlying its beneficial effects remain to be elucidated. One recent retrospective study from Canada reported that right-hemisphere patients with stroke were 45% less Previous investigations of drug treatment for neglect likely to receive thrombolysis acutely than their left- have largely focused on dopaminergic modulation and hemisphere counterparts [49]. Interestingly, patients produced mixed results. A new placebo-controlled cross- with profound hemispatial neglect who also generally over study [57] has reported the effects of guanfacine, an present with severe weakness were likely to receive a-2 noradrenergic agonist that most likely exerts its thromobolytic treatment, but the authors rightly beneficial effects on cognitive function via post-synaptic expressed concerns that less severe neglect may not receptors in the prefrontal cortex [58]. The results show

that patients with chronic neglect without extensive 10 Mannan S, Mort D, Hodgson T, et al. Revisiting previously searched locations in visual neglect: role of right parietal and frontal lesions in misjudging old prefrontal cortical involvement can benefit from guan- locations as new. J Cogn Neurosci 2005; 17:340–354. Using eye-movement tracking during visual search, these authors found that many facine, with improvements in the extent of search into neglect patients refixate ipsilesional targets, often mistakenly considering them to contralesional space and extended ability to maintain be newly found. The authors propose that patients with intraparietal damage have a deficit in keeping track of locations over time and across saccades that contributes attention on task. The authors suggest that targeting to their neglect. intact prefrontal areas with guanfacine may improve 11 Fruhmann-Berger M, Karnath H-O. Spontaneous eye and head position in the ability to sustain attention and thereby improve patients with spatial neglect. J Neurol 2005; 252:1194–2000. performance on visual search. However, the results are 12 Walker R, Findlay JM. Saccadic eye movement programming in unilateral preliminary and the study is a proof-of-concept investi- neglect. Neuropsychologia 1996; 34:493–508. gation in a small number of patients, rather than a large 13 Behrmann M, Ghiselli-Crippa T, Dimatteo I. Impaired initiation but not execu- tion of contralesional saccades in hemispatial neglect. Behav Neurol 2001; randomized controlled trial. Further experimental inves- 13:39–60. tigation is required before advocating use of this drug in 14 Girotti F, Casazza M, Musicco M, Avanzini G. Oculomotor disorders in cortical clinical practice. lesions in man: the role of unilateral neglect. Neuropsychologia 1983; 21: 543–553. 15 Niemeier M, Karnath HO. Exploratory saccades show no direction-specific Conclusion deficit in neglect. Neurology 2000; 54:515–518. The brain regions that are frequently damaged in patients 16 Husain M, Mannan S, Hodgson T, et al. Impaired spatial working memory with neglect often involve areas that normally participate across saccades contributes to abnormal search in parietal neglect. Brain 2001; 124:941–952. in eye-movement control or are activated by vestibular 17 Russell C, Malhotra P, Husain M. Attention modulates the visual field in healthy stimulation. Studies of eye movement and balance con- observers and parietal patients. Neuroreport 2004; 15:2189–2193. trol in patients with neglect have begun to provide 18 Niemeier M, Karnath HO. Stimulus-driven and voluntary saccades are coded further insight into the nature of the condition and its in different coordinate systems. Curr Biol 2003; 13:585–589. relationship to commonly associated clinical disorders, 19 Sprenger A, Kompf D, Heide W. Visual search in patients with left visual hemineglect. Prog. Brain Res 2002; 140:395–416. such as the pusher syndrome. Treatments for the con- 20 Pisella L, Berberovic N, Mattingley JB. Impaired working memory for location dition are in their infancy but experimental investigations but not for colour or shape in visual neglect: a comparison of parietal and non- parietal lesions. Cortex 2004; 40:379–390. of both behavioural techniques (such as prism adaptation) Study using a change-detection paradigm to assess the working memory deficit in and neuropharmacological modulators (such as noradren- neglect patients, demonstrating that in patients with parietal lesions working memory for location was selectively impaired when compared to memory for ergic drugs) may lead to useful clinical interventions. colour and shape. 21 Koyama M, Hasegawa I, Osada T, et al. Functional magnetic resonance Acknowledgements imaging of macaque monkeys performing visually guided saccade tasks: This work is funded by a Wellcome Trust grant to M.H. We are comparison of cortical eye fields with humans. Neuron 2004; 41: 795–807. extremely grateful to the patients who have participated in our studies. 22 Todd JJ, Marois R. Capacity limit of visual short-term memory in human posterior parietal cortex. Nature 2004; 428:751–754. 23 Perennou DA, Bronstein AM. Balance disorders and vertigo after stoke: References and recommended reading assessment and rehabilitation. In: Barnes M, Dobkin B, Bogousslavksy J, Papers of particular interest, published within the annual period of review, have editors. Recovery after stroke. Cambridge: Cambridge University Press; been highlighted as: 2005. pp. 320–396. of special interest of outstanding interest 24 Wee JYM, Hopman WM. Stroke impairment predictors of discharge function, Additional references related to this topic can also be found in the Current length of stay, and discharge destination in stroke rehabilitation. Am J Phys World Literature section in this issue (pp. 91–92). Med Rehab 2005; 84:604–612. Analysis of a prospective study of several hundred stroke patients, which inves- 1 Nys GM, van Zandvoort MJ, de Kort PL, et al. The prognostic value of domain- tigated the contribution of multiple impairments on outcome measures. The admission balance score and presence of body neglect both predicted discharge specific cognitive abilities in acute first-ever stroke. Neurology 2005; 64: 821–827. destination. A thorough investigation of the predictive value of neuropsychological examination 25 Dieterich M, Bense S, Lutz S, et al. Dominance for vestibular cortical function in acute stroke, assessing long-term cognitive and functional outcomes. Neglect in the non-dominant hemisphere. Cereb Cortex 2003; 13:994–1007. and perceptual impairments were the most important independent prognostic 26 Indovina I, Maffei V, Bosco G, et al. Representation of visual gravitational indicators of functional outcome. motion in the human vestibular cortex. Science 2005; 308:416–419. 2 Corbetta M, Shulman GL. Control of goal-directed and stimulus-driven 27 Ventre-Dominey J, Nighoghossian N, Denise P. Evidence for interacting attention in the brain. Nat Rev Neurosci 2002; 3:201–215. cortical control of vestibular function and spatial representation in man. 3 Behrmann M, Geng JJ, Shomstein S. Parietal cortex and attention. Curr Opin Neuropsychologia 2003; 41:1884–1898. Neurobiol 2004; 14:212–217. 28 Doricchi F, Siegler I, Iaria G, Berthoz A. Vestibulo-ocular and optokinetic 4 Husain M, Rorden C. Non-spatially lateralized mechanisms in hemispatial impairments in left unilateral neglect. Neuropsychologia 2002; 40:2084– neglect. Nat Rev Neurosci 2003; 4:26–36. 2099. 5 Driver J, Vuilleumier P, Husain M. Spatial neglect and extinction. In: Gazzaniga 29 Dieterich M, Brandt T. Vestibular system: anatomy and functional magnetic MS, editor. The cognitive neurosciences III. Cambridge, Mass: MIT Press; resonance imaging. Neuroimaging Clin N Am 2001; 11:263–273. 2004. pp. 589–606. 30 Karnath HO, Ferber S, Dichgans J. The origin of contraversive pushing: 6 Parton A, Malhotra P, Husain M. Hemispatial neglect. J Neurol Neurosurg evidence for a second graviceptive system in humans. Neurology 2000; Psychiatr 2004; 75:13–21. 55:1298–1304. 7 Karnath H-O, Milner AD, Vallar G. The cognitive and neural bases of spatial 31 Perennou DA, Amblard B, Laassel el M, et al. Understanding the pusher neglect. Oxford: Oxford University Press; 2002. behavior of some stroke patients with spatial deficits: a pilot study. Arch Phys 8 Kerkhoff G. Spatial hemineglect in humans. Prog Neurobiol 2001; 63:1–27. Med Rehabil 2002; 83:570–575. 9 Behrmann M, Watt S, Black SE, Barton JJ. Impaired visual search in patients 32 Saj A, Honore J, Coello Y, Rousseaux M. The visual vertical in the pusher with unilateral neglect: an oculographic analysis. Neuropsychologia 1997; syndrome Influence of hemispace and body position. J Neurol 2005; 252: 35:1445–1458. 885–891.

33 Karnath HO, Johannsen L, Broetz D, Kuker W. Posterior thalamic hemorrhage 48 Hillis AE, Wityk RJ, Barker PB, et al. Change in perfusion in acute non- induces ‘‘pusher syndrome’’. Neurology 2005; 64:1014–1019. dominant hemisphere stroke may be better estimated by tests of hemispatial neglect than by the National Institutes of Health Stroke Scale. 34 Vallar G. Extrapersonal visual unilateral spatial neglect and its neuroanatomy. Stroke 2003; 34:2392–2396. Neuroimage 2001; 14:S52–S58. 35 Karnath HO, Ferber S, Himmelbach M. Spatial awareness is a function of the 49 Di Legge S, Fang J, Saposnik G, Hachinski V. The impact of lesion side on temporal not the posterior parietal lobe. Nature 2001; 411:950–953. acute stroke treatment. Neurology 2005; 65:81–86. These investigators show that patients with right-hemisphere stroke are far less 36 Karnath HO, Fruhmann Berger M, Kuker W, Rorden C. The anatomy of spatial likely to receive thrombolysis than those with left-hemisphere stroke, although the neglect based on voxelwise statistical analysis: a study of 140 patients. Cereb presence of neglect (when it was detected) conferred a 2-fold increase in the Cortex 2004; 14:1164–1172. chance of receiving the treatment. Like [48], it shows that the assessment of Part of an ongoing debate concerning the anatomy of spatial neglect (see also neglect might be an important part of the examination when screening a patient for [35,42]).The investigators assessed a large number of computed tomography and thrombolysis. magnetic resonance imaging scans obtained for clinical purposes and concluded that the right STG was the key cortical region associated with spatial neglect. 50 Gillen R, Tennen H, McKee T. Unilateral spatial neglect: relation to rehabilita- 37 Doricchi F, Tomaiuolo F. The anatomy of neglect without hemianopia: a key tion outcomes in patients with right hemisphere stroke. Arch Phys Med role for parietal-frontal disconnection? Neuroreport 2003; 14:2239–2243. Rehabil 2005; 86:763–767. 38 Buxbaum LJ, Ferraro MK, Veramonti T, et al. Hemispatial neglect: subtypes, 51 Frassinetti F, Angeli V, Meneghello F, et al. Long-lasting amelioration of neuroanatomy, and disability. Neurology 2004; 62:749–756. visuospatial neglect by prism adaptation. Brain 2002; 125:608–623.

39 Farne A, Buxbaum LJ, Ferraro M, et al. Patterns of spontaneous recovery of 52 Maravita A, McNeil J, Malhotra P, et al. Prism adaptation can improve neglect and associated disorders in acute right brain-damaged patients. J contralesional tactile perception in neglect. Neurology 2003; 60:1829– Neurol Neurosurg Psychiatr 2004; 75:1401–1410. 1831. Study addressing the important issue of spontaneous recovery of neglect and its time course. The right angular gyrus was found to be involved in both the acute 53 Farne A, Rossetti Y, Toniolo S, Ladavas E. Ameliorating neglect with prism genesis of left neglect and in the continuing presence of the syndrome. adaptation: visuo-manual and visuo-verbal measures. Neuropsychologia 40 Ringman JM, Saver JL, Woolson RF, et al. Frequency, risk factors, anatomy, 2002; 40:718–729. and course of unilateral neglect in an acute stroke cohort. Neurology 2004; 54 McIntosh RD, Rossetti Y, Milner AD. Prism adaptation improves chronic visual 63:468–474. and haptic neglect: a single case study. Cortex 2002; 38:309–320. 41 Vandenberghe R, Geeraerts S, Molenberghs P, et al. Attentional responses to unattended stimuli in human parietal cortex. Brain 2005; doi:10.1093/brain/ 55 Angeli V, Benassi MG, Ladavas E. Recovery of oculo-motor bias in neglect awh522. patients after prism adaptation. Neuropsychologia 2004; 42:1223–1234. 42 Mort DJ, Malhotra P, Mannan SK, et al. The anatomy of visual neglect. Brain 56 Morris AP, Kritikos A, Berberovic N, et al. Prism adaptation and spatial 2003; 126:1986–1997. attention: a study of visual search in normals and patients with unilateral 43 Karnath HO, Fruhmann Berger M, Zopf R, Kuker W. Using SPM normalization neglect. Cortex 2004; 40:703–721. for lesion analysis in spatial neglect. Brain 2004; 127:E10. These investigators used a speeded search task to assess the effects of prism adaptation in normal subjects and patients with neglect. Their results suggest that 44 Mort DJ, Malhotra P, Mannan SK, et al. Reply to: Using SPM normalization for prism adaptation may not improve core aspects of the neglect syndrome. lesion analysis in spatial neglect. Brain 2004; 127:E11. 57 Malhotra P, Parton A, Greenwood R, Husain M. Noradrenergic modulation of 45 Karnath HO, Himmelbach M, Rorden C. The subcortical anatomy of human space exploration in visual neglect. Ann Neurol 2005; doi:10.1002/ana. spatial neglect: putamen, caudate nucleus and pulvinar. Brain 2002; 125: 20701. 350–360. Pilot drug study demonstrating that some of the deﬁcits associated with neglect 46 Hillis AE, Wityk RJ, Barker PB, et al. Subcortical aphasia and neglect in acute might be treated pharmacologically through the manipulation of sustained atten- stroke: the role of cortical hypoperfusion. Brain 2002; 125:1094–1104. tion by guanfacine, a noradrenergic agonist. Importantly this treatment might only be successful if prefrontal cortical areas are relatively intact, indicating the 47 Hillis AE, Newhart M, Heidler J, et al. Anatomy of spatial attention: insights importance of lesion delineation and careful patient selection. from perfusion imaging and hemispatial neglect in acute stroke. J Neurosci 2005; 25:3161–3167. 58 Avery RA, Franowicz JS, Studholme C, et al. The alpha-2A-adrenoceptor This interesting new study examined acute patients with hemineglect and sub- agonist, guanfacine, increases regional cerebral blood ﬂow in dorsolateral cortical stroke using perfusion-weighted imaging. The results point to possible prefrontal cortex of monkeys performing a spatial working memory task. anatomical dissociations between potential subtypes of unilateral neglect. Neuropsychopharmacology 2000; 23:240–249.