Abstract
Introduction: Recent research has increasingly acknowledged the thalamus’s role in the development of neuropsychological deficits, which were previously considered to be primarily related to cortical processes. Among these deficits, neglect is of particular importance in stroke survivors, as it is a predictor of poor functional outcome. This review aimed to clarify the relationship between stroke lateralization and location within the thalamus and the occurrence of neglect. Methods: In the present study, we performed a systematic review according to the PRISMA guidelines. PubMed, Scopus, CINHAL, and Web of Science were searched for articles published from inception to June 30, 2024. All studies presenting cases of isolated vascular thalamic stroke (hemorrhagic, ischemic) and clinical neglect were included. Study quality was assessed using the Joanna Briggs Institute critical appraisal checklist for case reports, case series, and case-control studies. We divided the thalamus into four parts (anterior, lateral, medial, and posterior) based on the four classical vascular territories and performed a qualitative and a simple descriptive statistical analysis using absolute numbers and percentages of the data collected. Results: A total of 23 articles involving 37 patients were included: 31 cases (84%) with right-sided thalamic stroke and 6 cases (16%) with left-sided thalamic stroke. In the hemorrhagic stroke group (21 cases), there was a clear predominance of localization in the posterior (10 cases; 47%) and entire thalamus (9 cases; 43%), with no cases in the anterior part of the thalamus and only one case (5%) each in the medial and lateral parts. In contrast, ischemic cases were predominantly located in the anterior and lateral parts (6 cases each; 37.5%) with only 3 cases (19%) in the medial part and 1 case (6%) in the posterior part. Conclusion: Thalamic neglect appears to occur more frequently in right-sided thalamic strokes than in left-sided thalamic strokes. However, the exact neuroanatomical correlates differed between hemorrhagic and ischemic groups and the underlying mechanisms remain unclear due to the heterogeneity and paucity of data. Rather than drawing definitive conclusions, this work synthesizes existing literature and underscores the need for prospective studies with standardized assessments and advanced neuroimaging.
Introduction
Thalamic strokes can be challenging to recognize because of the wide variety of symptoms they can cause and the different clinical conditions they can mimic. These include neuropsychological deficits, which are commonly associated with cortical deficits.
While aphasia is now widely recognized as a symptom of thalamic stroke [1, 2], the occurrence of neglect following isolated thalamic stroke appears to be less clear. However, its recognition and understanding in clinical routine are crucial, as neglect after stroke is known to be an independent predictor of poor functional outcome and is associated with lower rates of community discharge [3, 4].
The term “thalamic neglect” was first used by Watson and Heilman in 1979 [5] to describe three cases of neglect induced by right thalamic hemorrhage. Supported by evidence from animal studies [6‒10], they proposed that an activation defect along a cortico-limbic-reticular loop, including the intralaminar nuclei, might account for neglect and other neuropsychological deficits usually associated with right hemisphere cortical dysfunction [5]. Two years later, in 1981, Watson et al. [11] emphasized the possible role of the medial thalamus and, particularly, of the centromedian and parafascicular nuclei. In subsequent years, numerous studies have examined the role of the thalamus in neglect, yielding somewhat inconsistent results. For instance, Karnath and colleagues [12] suggested that the posterior territory, specifically the pulvinar, might be crucial in the etiology of neglect based on findings in hemorrhagic stroke patients. Conversely, Rauch and colleagues [13], in a recent MRI-based study of ischemic stroke patients, found an association of neglect with strokes located within the inferolateral and ventral territories. To disentangle these somewhat contradictory previous results, the present study aimed to systematically review the reported cases of patients with isolated hemorrhagic and ischemic stroke of the thalamus and clinical neglect and to clarify the relationship between stroke lateralization and location within the thalamus and the occurrence of neglect.
Methods
For this systematic literature review, we followed the PRISMA 2020 guidelines [14].
Information Sources and Electronic Search Strategy
We systematically searched the following databases: PubMed, Scopus, CINHAL, and Web of Science for articles published from inception to June 30, 2024. The PubMed-entered search terms were the following: (“Neglect”[tw] OR “hemineglect”[tw] OR “right hemineglect”[tw] OR “left hemineglect”[tw] OR “hemispatial neglect”[tw] OR “spatial neglect”[tw] OR “thalam* neglect”[tw] OR “spatial inattention”[tw] OR “visuospatial neglect”[tw] OR “visual spatial neglect”[tw] OR “unilateral neglect”[tw] OR “unilateral spatial neglect”[tw] OR “USN”[tw] OR “visual neglect”[tw] OR “motor neglect”[tw] OR “sensory neglect”[tw] OR “auditory neglect”[tw]) AND (“Thalamus”[mesh] OR “Thalamus”[tw] OR “Thalam*”[tw] OR “thalam* stroke”[tw] OR “thalam* infarct”[tw] OR “thalam* hemorrhage”[tw] OR “thalam* bleeding”[tw]).
Eligibility Criteria
Inclusion criteria were as follows: studies of adult patients with a vascular (ischemic or hemorrhagic) isolated thalamic lesion, confirmed by CT, MRI, or autopsy, presenting with neglect as a clinical symptom. Due to the small number of published cases, the type of assessment used for neglect was not further specified. Patients with any type of neglect were included in this study.
We excluded all studies in which the lesion was not limited to the thalamus or was not of vascular origin. Furthermore, studies that did not provide sufficient information regarding the lesion location were excluded.
All study designs were deemed suitable for consideration, with the exception of narrative reviews and poster/conference abstracts. Articles written in English, French, and German were deemed eligible for inclusion.
Study Selection
The citations were uploaded into Covidence, and any duplicates were removed. The screening of eligible articles was conducted by one reviewer (A.D.). The titles and abstracts were initially evaluated, and subsequently, the full texts of the remaining publications were assessed. In the event of uncertainty, a second reviewer was consulted (T.N.).
Data Collection Process
The data were entered into a predefined Excel spreadsheet by one reviewer (A.D.) and cross-checked by another reviewer (T.N.). As the majority of case reports were older and based on low-resolution CT imaging, a detailed localization of the affected nuclei could not be achieved. Therefore, the thalamus was divided into four parts (anterior, lateral, medial, and posterior) based on the four classical vascular territories, as illustrated in a schematic representation adapted from Bogousslavsky [15] (shown in Fig. 1). Accordingly, the anterior part is supplied by the tuberothalamic (also called polar) artery, which arises from the middle third of the posterior communicating artery and irrigates the anterior nuclei group, the ventral amygdalo-fugal pathway, the mammillothalamic tract, the ventral internal medullary lamina, the intralaminar nuclei, the ventral anterior nucleus, the reticular nucleus, the rostral part of the ventral lateral nucleus, and the ventral pole of the mediodorsal nucleus [16, 17]. The medial part is supplied by the paramedian arteries, which arise from the P1 segment of the posterior cerebral artery (PCA) and irrigate the mediodorsal nucleus, the posteromedial part of the ventral lateral nucleus, the dorsal internal medullary lamina, the intralaminar nuclei, as well as the ventromedial pulvinar and the paraventricular lateral dorsal nucleus [16, 17]. The lateral territory is supplied by the thalamogeniculate arteries, which arise from the P2 segment of the PCA and irrigate the ventroposterior complex (VPM, VPL, and VPI), the ventral lateral nucleus, the medial geniculate body, the rostral and lateral parts of the pulvinar, and the lateral dorsal nucleus [16, 17].
Schematic representation of the thalamus, adapted from Bogousslavsky et al. [15]. PCA, posterior cerebral artery; ICA, internal carotid artery; PCoA, posterior communicating artery.
Schematic representation of the thalamus, adapted from Bogousslavsky et al. [15]. PCA, posterior cerebral artery; ICA, internal carotid artery; PCoA, posterior communicating artery.
The posterior territory is supplied by lateral and medial branches of the posterior choroidal artery, which arise from the P2 segment of the PCA. The lateral branches irrigate the lateral geniculate body, the lateral dorsal nucleus, the lateral posterior nucleus, and the inferolateral parts of the pulvinar. The medial branches irrigate the medial geniculate body, the posterior parts of the centromedian and centrolateral nuclei (intralaminar nuclei), as well as the pulvinar [16, 17].
If CT or MR imaging was available in the article, classification into the four territories was performed visually by two reviewers (A.D. and T.N.) according to the aforementioned criteria. In the case of ischemic strokes with available MRI slices, we additionally relied on the vascular map proposed by Antoniello and Fernandes-Torres [18]. In instances where imaging data were unavailable, we instead relied on the written description provided.
Data Extraction
We extracted the following data: study design, year of publication, patient age and sex, brain imaging modality, lesion lateralization and location, neglect characteristics, assessment of neglect, and other neurological symptoms described.
Quality Assessment of the Selected Studies
The quality of the selected studies was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklist for case reports, case series, and case-control studies [19]. Letters and short communications were assessed using the same checklist as for case reports.
Data Presentation and Statistical Analysis
Given the heterogeneity of the data, we conducted a qualitative analysis and a simple descriptive statistical analysis with absolute numbers and percentages of the data collected.
Results
Study Selection
After the literature search, a total of 1,017 articles were identified. After removing duplicates, 547 references remained, which were screened for abstract and title. We selected 69 articles for full-text screening. Thirty-two studies were excluded because of the wrong patient population (i.e., mostly because the lesion was not limited to the thalamus, or it was not a vascular lesion, or the patient did not present with neglect). Seven studies were excluded because of the wrong design (i.e., review or conference abstract) and 7 because the localization was not clearly described. The detailed reason for exclusion is provided for each reference in the online supplementary Data (for all online suppl. material, see https://doi.org/10.1159/000545473). The remaining 23 articles were included, with a total of 37 patients. These studies included 17 case reports, 1 letter to the editor, 1 short communication, 3 case series, and 1 case-control study. The PRISMA flowchart diagram is shown in Figure 2.
Risk of Bias in Studies
We used the Joanna Briggs Institute critical appraisal checklist for case reports, case series, and case-control studies [19] to assess the risk of bias in each study. All 19 case reports clearly described patient demographics (age, sex). Patient medical history was reported in only 11 of 19 cases. The clinical condition was clearly described in 18 of 19 cases, and the assessment methods of the neglect were provided in 17 of 19 cases. Questions 5, 6, and 7 of the checklist were not applicable to most of the case reports because no intervention was performed. Regarding the 3 case series [20‒22], all had clear inclusion criteria and used valid methods to identify and reliably measure the condition. Demographic data were clearly reported in 2 out of 3 case series, and clinical information was clearly provided in 2 out of 3 studies. The detailed quality evaluation is available in the online supplementary Data.
Evidence Synthesis
The main characteristics of the patients are shown in Figure 3. Of the 37 patients, 24 were men (65%) and 11 were women (30%). The age of the patients ranged from 38 to 86 years. In 2 patients, sex and age were unknown. Brain imaging was mostly performed by CT (n = 19; 51%). Eight patients had MRI (22%), 2 (5%) had both CT and MRI, 7 (19%) had either CT or MRI (not specified in the publication), and 1 (3%) had an autopsy. Twenty-one of the patients had a hemorrhagic stroke (57%), and 16 had an ischemic stroke (43%). Visuospatial neglect was the most common type of neglect, accounting for 20 cases (54%). Multimodal neglect was described in 12 patients (32%), whereas pure motor neglect was described in 2 patients (5%) and pure representational neglect was described in 1 patient (3%). The type of neglect was not specified in two cases (5%). Each case is detailed in Table 1.
Patient characteristics: sex distribution, type of stroke, and type of neglect.
Characteristics of published cases with thalamic stroke and neglect
Author and year . | Patient . | Brain imaging . | Type of stroke . | Hemisphere . | Localization . | Neglect characteristic . | Neglect assessment . | Other neurological symptoms described . |
---|---|---|---|---|---|---|---|---|
Watson et al. [5] (1979) | Case 1: 84 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Map localization cities, drawing, cancelation task, line bisection, extinction testing | Left hemiparesis, anosognosia, prosopagnosia, flattened affect, memory disturbance |
Case 2: 78 Y, woman | CT | Hemorrhagic | Right | Entire | Multimodal | Drawing, extinction testing, clinical observation, cancelation task | Left sensorimotor hemiparesis, anosognosia, flattened affect | |
Case 3: 69 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Clock drawing, map localization cities | Left hemiparesis, sensibility disturbance, hypokinesia, decreased attention, memory disturbance, flattened affect, anosognosia, left hemianopia | |
Watson et al. [11] (1981) | 62 Y, man | CT | Ischemic | Right | Medial (previous right parietal superior infarction without evidence of neglect on extensive testing) | Multimodal | Extinction testing, line crossing out, line bisection, map localization cities | Lethargy, hypokinesia, hemiparesis left, decreased perception of pain left, memory disturbance, flattened affect |
Schott et al. [23] (1981) | 54 Y, woman | CT | Hemorrhagic | Right | Posterior | Multimodal (predominant motor) | According to clinical definition from Castaigne 1970, extinction test, line bisection test, cancelation test, drawing, city map | Initially decreased consciousness and oculomotor disorders, left facial paresis, mild motor deficit |
Laplane et al. [24] (1982) | Case 1: 64 Y, woman | Autopsy | Ischemic | Right | Lateral | Motor | According to clinical definition from Castaigne 1970 | Mild motor deficit left, hypermetria left lower extremity |
Graff-Radford et al. [25] (1984) | Patient 1: 67 Y, man | CT | Ischemic | Left | Anterior | Not specified | Not specified | Dysarthria, aphasia, right homonymous hemianopia, right hemiparesis, impaired proprioception right, memory disturbance, disorientation |
Hirose et al. [26] (1985) | Patient 1: 83 Y, woman | CT | Hemorrhagic | Right | Entire | Multimodal (spatial, sensory) | Extinction testing, clock drawing, line bisection | Mild left hemiparesis, sensory deficits, right gaze preference, upward gaze limitation, Horner syndrome |
Patient 4: 73 Y, woman | CT | Hemorrhagic | Right | Entire | Visuospatial | Line bisection | Sensory deficit, right gaze preference, moderate left hemiparesis | |
Patient 6: 66 Y, woman | CT | Hemorrhagic | Right | Posterior | Visuospatial | Line bisection | Sensory deficit, right gaze preference, downward gaze deviation | |
Bogousslavsky et al. [27] (1986) | Case 3: 68 Y, man | CT | Ischemic | Right | Anterior | Multimodal | Extinction testing/clinically, writing test | Fatigue, transient blurring of vision, persisting left-sided weakness, left-sided extinction, hypesthesia, constructional apraxia, visual > verbal memory impairment, “frontal” syndrome |
Graveleau et al. [28] (1986) | 71 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal (visual, auditory, motor) | Target cancelation, picture description, reading, locating cities/monument on a map, extinction testing | Left hemiplegia and left hemihypesthesia, anosognosia, disorientation, constructional apraxia |
Waxman et al. [29] (1986) | 61 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Extinction testing, clock drawing, copying, writing, line crossing test | Hypokinesia, apathy, memory impairment, left sensorimotor deficit |
Ferro et al. [30] (1987) | Case 2: 57 Y, man | CT | Hemorrhagic | Left | Posterior | Visuospatial | Writing, drawing, line cancelation test | Global aphasia, right hemianopia. Right hemiparesis and hemihypesthesia, left conjugate gaze deviation. Perseveration, memory disturbance, ideomotor apraxia |
Rafal et al. [31] (1987) | Patient VM: 65 Y, man | CT | Hemorrhagic | Left | Entire (+intraventricular bleeding) | Visuospatial | Not clearly described, at follow-up: letter cancelation task | Comatose, right hemiplegia, hemianesthesia, ocular skew deviation, psychomotor retardation |
Weisz et al. [32] (1995) | Patient 1: 54 Y, man | CT | Ischemic | Right | Posterior | Multimodal | Activity of daily life, behavioral inattention test | Left spastic hemiplegia, hemisensory loss, hemianopsia |
Ebert et al. [33] (1999) | 65 Y, man | MRI | Ischemic | Right | Anterior | Visuospatial | Observation of behavior: collision, lack of exploration, lack of eyes movement to the left, left-sided visual extinction, drawing, and copying test (line bisection + line cancelation normal) | Quadrantanopia lower left, mild left facial paresis, mild left hemiparesis, mild right Horner syndrome, anomic aphasia, dysarthria, deficit of working and short-term memory, impaired encoding of both verbal and figural materials. Reduced attention, impaired planning |
Manabe et al. [34] (1999) | 43 Y, man | MRI | Hemorrhagic | Left | Medial | Motor (without visuospatial) | Motor neglect: clinical observation; visuospatial: complex image description, tridimensional constructive apraxia, clock and room drawing | mild right hemiparesis |
Barrett et al. [35] (2000) | 52 Y, woman | MRI | Ischemic | Left | Anterior | Visuospatial only in extrapersonal space | Line bisection task with laser pointer in near and far extrapersonal space, line cancelation test, extinction testing | Sudden onset of dizziness and confusion. Memory and concentration disturbance. Mild motor impersistence and bilateral defective response inhibition of lateral eye movements to peripheral stimuli on both sides. Reduced verbal fluency |
Karussis et al. [20] (2000) | Case 13: 77 Y, man | CT | Ischemic | Right | Lateral | Multimodal (visuospatial, tactile) | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, anosognosia |
Case 14: 38 Y, man | CT | Ischemic | Right | Lateral | Multimodal (visuospatial, tactile) | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, moderate sensory deficits, anosognosia | |
Case 15: 55 Y, man | CT | Hemorrhagic | Right | Lateral | Not specified | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, anosognosia | |
Ghika-Schmid et al. [22] (2000) | Patient L8: unknown age and sex | CT and MRI | Ischemic | Left | Anterior | Visuospatial | Line cancelation and bisection test | Memory disturbance, anosognosia, topographic disorientation, word finding difficulties, semantic paraphasia, dysarthria, hypophonia, naming difficulties, reduced verbal fluency, executive dysfunction, apathy |
Patient R2: unknown age and sex | CT and MRI | Ischemic | Right | Anterior | Visuospatial | Line cancelation and bisection test | Disoriented, confabulation, memory disturbance, anosognosia, word finding disorder, reduced verbal fluency and naming difficulties, emotional and behavioral disinhibition, distractibility | |
Ortigue et al. [36] (2001) | 86 Y, man | MRI | Ischemic | Right | Medial | Representational neglect (without perceptual neglect) | Line bisection test, cancelation test, landscape copying test, clock copying test, word and sentence reading, visualization of space | Dysarthria, moderate left hemiparesis without hemisensory loss, impairment in figural fluency |
Pack et al. [37] (2002) | 73 Y, man | MRI | Ischemic | Right | Lateral | Visuospatial | Not described | Mild hemiparesis, dyspraxia, dysmetria, diminished sensation to light touch and proprioception, alien hand syndrome |
De Witte et al. [38] (2008) | 70 Y, man | MRI | Ischemic | Right | Medial | Visuospatial | BORB subtest “length match task,” “size match task,” “position of gap match task” | Left hemiparesis, mild left facial nerve paresis, dysarthria, aphasia, left hemianopia, concentration deficits, short-term memory disturbance, acalculia, apraxia, MMSE 18/30 |
Chen et al. [39] (2014) | 55 Y, woman | CT | Hemorrhagic | Right | Entire (+intraventricular bleeding and ventricular enlargement) | Visuospatial | Line cancelation test, letter cancelation test, star cancelation test, line bisection test, copy and continuation of graphic sequence test | Pusher syndrome, left hemiplegia, decreased response to pain left, MMSE test 22/30 |
Sebastian et al. [21] (2014) | Patient 11: 68 Y, man | MRI | Ischemic | Right | Lateral (+ severe right PCA stenosis with hypoperfusion) | Visuospatial | Perceptual tasks, perceptual motor tasks (line bisection, clock drawing, copying), oral reading and spelling, motor extinction test | Not clearly described |
Patient 12: 61 Y, man | MRI | Ischemic | Right | Lateral (+ severe right PCA stenosis and right M1 stenosis with hypoperfusion) | Visuospatial | Perceptual tasks, perceptual motor tasks (line bisection, clock drawing, copying), oral reading and spelling, motor extinction test | Not clearly described | |
Karnath et al. [12] (2022) | 7 patients: median 74 Y (61–79), 3 women, 4 men | MRI or CT | Hemorrhagic | Right | Posterior | Visuospatial | Clinical features, letter cancelation test, Bell test, baking tray task, copying task | 86% paresis of contralesional side, no somatosensory deficit |
Author and year . | Patient . | Brain imaging . | Type of stroke . | Hemisphere . | Localization . | Neglect characteristic . | Neglect assessment . | Other neurological symptoms described . |
---|---|---|---|---|---|---|---|---|
Watson et al. [5] (1979) | Case 1: 84 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Map localization cities, drawing, cancelation task, line bisection, extinction testing | Left hemiparesis, anosognosia, prosopagnosia, flattened affect, memory disturbance |
Case 2: 78 Y, woman | CT | Hemorrhagic | Right | Entire | Multimodal | Drawing, extinction testing, clinical observation, cancelation task | Left sensorimotor hemiparesis, anosognosia, flattened affect | |
Case 3: 69 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Clock drawing, map localization cities | Left hemiparesis, sensibility disturbance, hypokinesia, decreased attention, memory disturbance, flattened affect, anosognosia, left hemianopia | |
Watson et al. [11] (1981) | 62 Y, man | CT | Ischemic | Right | Medial (previous right parietal superior infarction without evidence of neglect on extensive testing) | Multimodal | Extinction testing, line crossing out, line bisection, map localization cities | Lethargy, hypokinesia, hemiparesis left, decreased perception of pain left, memory disturbance, flattened affect |
Schott et al. [23] (1981) | 54 Y, woman | CT | Hemorrhagic | Right | Posterior | Multimodal (predominant motor) | According to clinical definition from Castaigne 1970, extinction test, line bisection test, cancelation test, drawing, city map | Initially decreased consciousness and oculomotor disorders, left facial paresis, mild motor deficit |
Laplane et al. [24] (1982) | Case 1: 64 Y, woman | Autopsy | Ischemic | Right | Lateral | Motor | According to clinical definition from Castaigne 1970 | Mild motor deficit left, hypermetria left lower extremity |
Graff-Radford et al. [25] (1984) | Patient 1: 67 Y, man | CT | Ischemic | Left | Anterior | Not specified | Not specified | Dysarthria, aphasia, right homonymous hemianopia, right hemiparesis, impaired proprioception right, memory disturbance, disorientation |
Hirose et al. [26] (1985) | Patient 1: 83 Y, woman | CT | Hemorrhagic | Right | Entire | Multimodal (spatial, sensory) | Extinction testing, clock drawing, line bisection | Mild left hemiparesis, sensory deficits, right gaze preference, upward gaze limitation, Horner syndrome |
Patient 4: 73 Y, woman | CT | Hemorrhagic | Right | Entire | Visuospatial | Line bisection | Sensory deficit, right gaze preference, moderate left hemiparesis | |
Patient 6: 66 Y, woman | CT | Hemorrhagic | Right | Posterior | Visuospatial | Line bisection | Sensory deficit, right gaze preference, downward gaze deviation | |
Bogousslavsky et al. [27] (1986) | Case 3: 68 Y, man | CT | Ischemic | Right | Anterior | Multimodal | Extinction testing/clinically, writing test | Fatigue, transient blurring of vision, persisting left-sided weakness, left-sided extinction, hypesthesia, constructional apraxia, visual > verbal memory impairment, “frontal” syndrome |
Graveleau et al. [28] (1986) | 71 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal (visual, auditory, motor) | Target cancelation, picture description, reading, locating cities/monument on a map, extinction testing | Left hemiplegia and left hemihypesthesia, anosognosia, disorientation, constructional apraxia |
Waxman et al. [29] (1986) | 61 Y, man | CT | Hemorrhagic | Right | Entire | Multimodal | Extinction testing, clock drawing, copying, writing, line crossing test | Hypokinesia, apathy, memory impairment, left sensorimotor deficit |
Ferro et al. [30] (1987) | Case 2: 57 Y, man | CT | Hemorrhagic | Left | Posterior | Visuospatial | Writing, drawing, line cancelation test | Global aphasia, right hemianopia. Right hemiparesis and hemihypesthesia, left conjugate gaze deviation. Perseveration, memory disturbance, ideomotor apraxia |
Rafal et al. [31] (1987) | Patient VM: 65 Y, man | CT | Hemorrhagic | Left | Entire (+intraventricular bleeding) | Visuospatial | Not clearly described, at follow-up: letter cancelation task | Comatose, right hemiplegia, hemianesthesia, ocular skew deviation, psychomotor retardation |
Weisz et al. [32] (1995) | Patient 1: 54 Y, man | CT | Ischemic | Right | Posterior | Multimodal | Activity of daily life, behavioral inattention test | Left spastic hemiplegia, hemisensory loss, hemianopsia |
Ebert et al. [33] (1999) | 65 Y, man | MRI | Ischemic | Right | Anterior | Visuospatial | Observation of behavior: collision, lack of exploration, lack of eyes movement to the left, left-sided visual extinction, drawing, and copying test (line bisection + line cancelation normal) | Quadrantanopia lower left, mild left facial paresis, mild left hemiparesis, mild right Horner syndrome, anomic aphasia, dysarthria, deficit of working and short-term memory, impaired encoding of both verbal and figural materials. Reduced attention, impaired planning |
Manabe et al. [34] (1999) | 43 Y, man | MRI | Hemorrhagic | Left | Medial | Motor (without visuospatial) | Motor neglect: clinical observation; visuospatial: complex image description, tridimensional constructive apraxia, clock and room drawing | mild right hemiparesis |
Barrett et al. [35] (2000) | 52 Y, woman | MRI | Ischemic | Left | Anterior | Visuospatial only in extrapersonal space | Line bisection task with laser pointer in near and far extrapersonal space, line cancelation test, extinction testing | Sudden onset of dizziness and confusion. Memory and concentration disturbance. Mild motor impersistence and bilateral defective response inhibition of lateral eye movements to peripheral stimuli on both sides. Reduced verbal fluency |
Karussis et al. [20] (2000) | Case 13: 77 Y, man | CT | Ischemic | Right | Lateral | Multimodal (visuospatial, tactile) | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, anosognosia |
Case 14: 38 Y, man | CT | Ischemic | Right | Lateral | Multimodal (visuospatial, tactile) | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, moderate sensory deficits, anosognosia | |
Case 15: 55 Y, man | CT | Hemorrhagic | Right | Lateral | Not specified | Line bisection, line cancelation, flower drawing, cube copying, extinction testing | Mild motor deficits, anosognosia | |
Ghika-Schmid et al. [22] (2000) | Patient L8: unknown age and sex | CT and MRI | Ischemic | Left | Anterior | Visuospatial | Line cancelation and bisection test | Memory disturbance, anosognosia, topographic disorientation, word finding difficulties, semantic paraphasia, dysarthria, hypophonia, naming difficulties, reduced verbal fluency, executive dysfunction, apathy |
Patient R2: unknown age and sex | CT and MRI | Ischemic | Right | Anterior | Visuospatial | Line cancelation and bisection test | Disoriented, confabulation, memory disturbance, anosognosia, word finding disorder, reduced verbal fluency and naming difficulties, emotional and behavioral disinhibition, distractibility | |
Ortigue et al. [36] (2001) | 86 Y, man | MRI | Ischemic | Right | Medial | Representational neglect (without perceptual neglect) | Line bisection test, cancelation test, landscape copying test, clock copying test, word and sentence reading, visualization of space | Dysarthria, moderate left hemiparesis without hemisensory loss, impairment in figural fluency |
Pack et al. [37] (2002) | 73 Y, man | MRI | Ischemic | Right | Lateral | Visuospatial | Not described | Mild hemiparesis, dyspraxia, dysmetria, diminished sensation to light touch and proprioception, alien hand syndrome |
De Witte et al. [38] (2008) | 70 Y, man | MRI | Ischemic | Right | Medial | Visuospatial | BORB subtest “length match task,” “size match task,” “position of gap match task” | Left hemiparesis, mild left facial nerve paresis, dysarthria, aphasia, left hemianopia, concentration deficits, short-term memory disturbance, acalculia, apraxia, MMSE 18/30 |
Chen et al. [39] (2014) | 55 Y, woman | CT | Hemorrhagic | Right | Entire (+intraventricular bleeding and ventricular enlargement) | Visuospatial | Line cancelation test, letter cancelation test, star cancelation test, line bisection test, copy and continuation of graphic sequence test | Pusher syndrome, left hemiplegia, decreased response to pain left, MMSE test 22/30 |
Sebastian et al. [21] (2014) | Patient 11: 68 Y, man | MRI | Ischemic | Right | Lateral (+ severe right PCA stenosis with hypoperfusion) | Visuospatial | Perceptual tasks, perceptual motor tasks (line bisection, clock drawing, copying), oral reading and spelling, motor extinction test | Not clearly described |
Patient 12: 61 Y, man | MRI | Ischemic | Right | Lateral (+ severe right PCA stenosis and right M1 stenosis with hypoperfusion) | Visuospatial | Perceptual tasks, perceptual motor tasks (line bisection, clock drawing, copying), oral reading and spelling, motor extinction test | Not clearly described | |
Karnath et al. [12] (2022) | 7 patients: median 74 Y (61–79), 3 women, 4 men | MRI or CT | Hemorrhagic | Right | Posterior | Visuospatial | Clinical features, letter cancelation test, Bell test, baking tray task, copying task | 86% paresis of contralesional side, no somatosensory deficit |
Localization
The localization of the thalamic strokes in the neglect cases is shown in Figure 4. Overall, there was a clear dominance of the right hemisphere in the occurrence of neglect in isolated hemorrhagic or ischemic thalamic strokes (31 cases; 84%). We found only 6 cases (16%) with isolated left thalamic stroke in which neglect was present. Overall, there was no clear predominance of one of the four vascular territories, regardless of the type of stroke or hemispheric lateralization: we identified 6 cases (16%) in the anterior part of the thalamus, 7 (19%) in the lateral part, 4 (11%) in the medial part, 11 (30%) in the posterior part, and 9 (24%) in the entire thalamus. However, a clear predominance appeared according to the type of stroke. In hemorrhagic strokes, there was a clear predominance of localization in the posterior (10 cases; 47%) and entire thalamus (9 cases; 43%), with no cases in the anterior part of the thalamus and only 1 case (5%) in each the medial and lateral parts. In contrast, in ischemic strokes, there was a predominance of localization in the anterior and lateral parts (6 cases each; 37.5%). We found only 3 cases with reported neglect and isolated thalamic stroke in the medial part of the thalamus (19%), and one case in the posterior part (6%).
Localization of the thalamic stroke with lateralization and localization overall, as well as separately for ischemic and hemorrhagic cases.
Localization of the thalamic stroke with lateralization and localization overall, as well as separately for ischemic and hemorrhagic cases.
In order to obtain a more precise localization, we also considered only the cases with ischemic strokes that had been diagnosed by MRI. Nine patients meeting these criteria were identified in seven articles [21, 22, 33, 35‒38]. Of these, seven had right-sided infarctions and two had left-sided infarctions. Again, a trend was observed for the anterior and lateral parts of the thalamus. The nuclei that may have been affected in the case by Barrett et al. [35] were the anterior, the ventral anterior, the anterior dorsal median nuclei, and possibly the anterior intralaminar nuclei. Ebert et al. [33] reported a lesion involving only the right anterolateral nuclei (lateropolaris, fasciculosus, ventrooralis externus and internus, and the ventral part of reticularis). In the case series by Ghika-Schmid and Bogousslavsky [22], 2 patients presented with visuospatial neglect due to an isolated anterior thalamic infarct. In the case-study by Pack et al. [37], the DWI imaging revealed a right lateral thalamic infarction. However, there were no MRI slices or detailed descriptions of the possibly affected nuclei. In the case series by Sebastian et al. [21], 2 patients with lesions within the territory of the right inferolateral (thalamogeniculate) artery presented with left hemispatial neglect. However, both patients had cortical hypoperfusion due to intracranial stenosis in the inferior temporal/fusiform cortex. De Witte [38] described a case of visuospatial neglect following an infarction in the median region of the right thalamus. In this patient, an EDC-SPECT study conducted 13 months post-onset showed a significantly decreased perfusion in the right fronto-temporo-parieto-occipital region and a severe hypoperfusion in the thalamus. Ortigue [36] reported a case of pure representational neglect after a right lesion in the dorsomedial nucleus of the thalamus extending into the pulvinar.
Discussion
The present study systematically reviewed the reported cases of patients with isolated hemorrhagic or ischemic thalamic stroke and neglect, with the aim of describing the most common localizations and potential mechanisms. Nowadays, the thalamus is considered to be a central hub for complex information processing and cognitive control, rather than simply relaying sensory information to the cortex. The historical notion of a single thalamic nucleus performing a single function seems to have become obsolete [40‒42]. Jones [43] proposed a classification based on thalamocortical projections, suggesting that parvalbumin-stained “core” cells project densely to layer IV of the cerebral cortex, whereas calbindin-stained “matrix” cells have a more integrative function and project diffusely to upper cortical layers, innervating multiple cortical areas as well as the basal ganglia [40, 42]. Sherman and Guillery [44, 45] proposed that glutamatergic input can be divided into drivers, which represent the main information to be conveyed, and modulators, which can alter the processing of the drivers. All thalamic relays receive modulatory input from the cortex. Depending on the source of the driver input, Sherman and Guillery [45] distinguished between “first-order” nuclei, such as the lateral geniculate nucleus (LGN), which receive subcortical driver input, and “higher order” nuclei, such as the pulvinar and mediodorsal nuclei, which receive cortical driver input [46]. Higher order relays appear to play a role in transthalamic-corticocortical communication [40, 45]. Furthermore, the thalamus has strong connections with the basal ganglia and cerebellum [47]. Glutamatergic driver inputs from the deep nuclei of the cerebellum project to core cells of the ventral thalamus, whereas GABAergic projections from the globus pallidus internus project to matrix cells [42, 48]. Finally, as a part of the reticular activating system, the thalamus also shows strong connections with the brainstem and hypothalamus [49, 50].
Overall, these studies reveal that the thalamus is centrally located in the brain and has strong connections with cortical and subcortical structures. The diversity of thalamocortical projections and the modulatory role of the thalamus suggest that it may play an important role in cognitive functions, such as attention and neglect, which were previously thought to be exclusively cortical deficits [51].
In our review, there was a clear dominance (84%) of the right hemisphere in the occurrence of neglect in isolated hemorrhagic or ischemic thalamic stroke among the 37 patients included. The predominance of neglect in isolated right thalamic stroke is not a surprising result, as the overall prevalence of neglect is higher in right than in left hemisphere strokes [52‒55].
It should be noted, however, that early reports of right hemisphere predominance in neglect may have been influenced by the predominant use of assessment tools – such as cancelation tasks – that are particularly sensitive to egocentric neglect. This methodological bias may lead to an overestimation of neglect prevalence in right ‐hemisphere strokes [56, 57]. Notably, Kleinman et al. [58] demonstrated that allocentric neglect occurs more frequently in left hemisphere strokes, while Hillis et al. [59] reported that egocentric neglect is more commonly observed in right hemisphere strokes.
Further, Hillis et al. [59] elucidated distinct cortical networks underpinning these neglect types, with the dorsal attention network primarily linked to egocentric neglect and ventral stream regions influencing allocentric neglect. Given that thalamic nuclei function as crucial relay centers within these cortical networks [60, 61], the specific pattern of neglect observed may depend on whether a lesion predominantly affects dorsal or ventral pathways.
Although evidence on the lateralization of thalamic contributions to egocentric and allocentric neglect remains limited, Ten Brink et al. [62] reported that egocentric neglect is associated with right thalamic damage, in addition to lesions in the right parietal and temporal lobes. Additionally, the case report by Ortigue et al. [36] described a patient with right thalamic stroke exhibiting strong representational neglect in viewer-centered conditions, whereas allocentric representations remained intact. This suggests that the right thalamus may serve as a crucial relay for egocentric mental imagery.
While our review underscores a predominance of neglect in right thalamic stroke cases, these findings highlight the importance of distinguishing between egocentric and allocentric neglect in clinical assessments. Importantly, allocentric neglect is less frequently measured in the acute stroke setting. In our review, specific tests – such as the Apple Cancelation Test – were not employed to distinguish between allocentric and egocentric neglect. As a result, the observed predominance of right thalamic stroke in neglect cases could partly reflect methodological biases rather than an absolute hemispheric dominance.
Regarding the hemorrhagic thalamic strokes, we found that 43% of cases involved the entire thalamus. Precise localization of hemorrhage can be difficult due to its diffuse and often large nature, and it is thus questionable whether a statement about the affected area can be made in the acute phase. In addition, in a hemorrhagic stroke, swelling may cause compression of nearby structures, particularly of the surrounding white matter tracts, which may account for a part of the symptomatology [63]. Indeed, several studies have highlighted the importance of white matter disconnections in the development of neglect [64, 65].
Furthermore, high prevalence of neglect (47%) was found after posterior thalamic hemorrhage. Within the posterior localization, the pulvinar deserves special attention. As early as the 1980s, electrophysiological studies in monkeys demonstrated the modulatory effects of attention and eye movements on the responses of pulvinar neurons [66, 67]. In addition, pulvinar inactivation in monkeys resulted in spatial neglect syndrome with severe impairment of visually guided behavior toward the contralateral space [68]. As mentioned above, the pulvinar serves as a higher order nucleus, exhibiting extensive connectivity with the cortex and playing a crucial role in cortico-thalamo-cortical pathways [69, 70]. Evidence from animal studies [71, 72], involving monkeys performing visuospatial attention tasks suggests that pulvino-cortical loops can influence information processing across visual and attentional cortical regions. Specifically, research by Saalmann and colleagues [71] demonstrated that the ventral pulvinar synchronizes neuronal activity between visual cortical areas (V4 and TEO) based on the focus of attention, while Fiebelkorn et al. [72] showed the involvement of the dorsal pulvinar, along with the frontal eye fields (FEFs) and lateral intraparietal area, in sustaining spatial attention and serving as fundamental component of the attention network. Karnath and colleagues [12] also emphasized the role of the pulvinar in the development of subcortical neglect and hypothesized that its connection to the superior temporal gyrus may play a role. Another key area in the posterior group and involved in visual attention control is the LGN [69]. For instance, it is known from monkey studies that the spike rate of LGN neurons is increased for attended stimuli [73]. Furthermore, a human fMRI study has provided evidence that visual processing in the LGN is modulated by selective attention [74], suggesting that the LGN may modulate the flow of information from the retina to the visual cortex based on the specific behavioral context [69].
However, in patients with ischemic stroke, the prevalence for neglect was highest in the anterior and lateral groups with 37.5% each, whereas neglect was rarely described after a stroke in the posterior group with only 6%. A possible reason for this difference from the hemorrhagic group is that the pulvinar, a potentially critical area for the development of neglect in the posterior group, is redundantly supplied by both the posterior choroidal artery as well as the thalamogeniculate and paramedian arteries [16]. Thus, in cases of ischemic stroke affecting the posterior thalamus due to an occlusion of the posterior choroidal artery, the pulvinar may still receive blood supply from other arteries. Conversely, in cases of posterior hemorrhage, damage to the pulvinar is inevitable. This distinction may explain the lower incidence of ischemic strokes compared to hemorrhagic strokes in patients with neglect and posterior thalamic damage.
The anterior region of the thalamus, particularly the anterodorsal, anteromedial, and anteroventral nuclei, is known for its extensive connectivity with the limbic system, including the cingulum and retrosplenial area, as well as prefrontal regions [75]. Neglect is a complex cognitive process that involves not only asymmetric attention but also asymmetric orientation and action deficits [76], and includes impairments in other cognitive processes such as spatial working memory [77‒79]. The retrosplenial cortex is known to play an important role in spatial cognition, including spatial navigation, orientation, and spatial memory [80, 81]. Damage to the cingulate cortex can lead to deficits in spatial aiming and motor-intentional functions, contributing to increased neglect symptoms [76]. Thus, it is arguable that a lesion of the anterior thalamus, through its connections with the cingulate and retrosplenial cortex, may contribute to the development of neglect syndrome by impairing spatial aiming and spatial working memory.
Another important part of the anterior group being associated with attention is the intralaminar nuclei. The intralaminar nuclei receive inputs from the brainstem reticular formation and have therefore been included in the “nonspecific ascending reticular activating system” that projects to extensive areas of the cerebral cortex [82]. For instance, a positron emission tomography study showed activation of the intralaminar nuclei when healthy subjects performed an attentionally demanding task [83].
Regarding the ischemic strokes in the lateral group, the results should be interpreted with caution. In our analysis, 2 out of 6 cases exhibited cortical hypoperfusion in the inferior temporal/fusiform cortex due to intracranial stenosis. Remarkably, the severity of neglect in these cases was linked to the extent of cortical hypoperfusion rather than the volume of the thalamic lesion [21]. Hence, it is uncertain whether neglect was due to the thalamic stroke itself or to cortical hypoperfusion. Furthermore, there may be some bias, as the thalamogeniculate territory seems to be more frequently affected in isolated thalamic stroke (39.7% of all isolated thalamic strokes in the study by Schaller-Paule et al. [17], 39.3% in a study by Del Mar Saez de Ocariz [84]), probably related to the hypertensive arteriolopathy that is common in this region. A possible explanation for the development of neglect after lateral thalamic stroke is damage to the thalamic reticular nucleus (TRN). The TRN forms a thin capsule around the lateral thalamic nuclei and plays an important role in attention [85]. The TRN is thought to inhibit the thalamocortical neurons, thereby acting as an attentional filter [86].
In our review, we found only 4 cases in the medial group (11%, three right ischemic and one left hemorrhagic thalamic strokes), which should be interpreted with caution because two [11, 36] may have involved the pulvinar, and the case by Watson et al. [11] had a previous right parietal superior infarction. In addition, a medial thalamic lesion may be associated with oculomotor dysfunction, such as in Parinaud syndrome, if the mesencephalon is affected, for example, by swelling, which may be a cofounder for the occurrence of neglect. Nevertheless, as a part of the dorsal attention network [61], the mediodorsal nucleus should be considered as an important nucleus for the development of neglect. A further role for the mediodorsal nucleus in attentional control comes from monkey studies. It has been shown that the mediodorsal nucleus, which connects inputs from the superior colliculus to the FEFs, has been suggested to have an internal representation of the planned saccade direction, which the FEF then uses for saccade triggering [87].
Our study has several limitations that should be acknowledged. First, most of the included studies were case reports or case series, which inherently carry a higher risk of bias despite our efforts to assess their quality using standardized checklists. Another major limitation is the reliance on older imaging modalities. The majority of the cases were evaluated using CT rather than MRI, and 17 of the 23 included studies were published before the year 2000, when brain imaging technology was significantly less advanced. This limits the precision of lesion localization, particularly in identifying affected thalamic nuclei. Additionally, as evident from the case descriptions, many hemorrhagic stroke cases involved multiple neurological deficits beyond neglect, likely reflecting damage to structures outside the thalamus. The presence of mass effect or extension into neighboring regions complicates the interpretation of correlations between neglect and particular thalamic nuclei, as symptoms may result from injury to other areas. When restricting the analysis to ischemic strokes documented with MRI, a more reliable clinical-anatomical correlation emerged, particularly with infarctions in the anterior and lateral aspects of the thalamus. However, even within this limited subgroup of 9 subjects [21, 22, 33, 35‒38] additional methodological concerns arise. One case [37] lacked specific data on the affected thalamic nuclei, 2 cases [21] exhibited extensive associated cortical hypoperfusion without detailed documentation of other neurological deficits, and another case [38] showed SPECT evidence of decreased perfusion in the entire right hemisphere and thalamus. These findings further challenge the ability to draw firm conclusions about the role of specific thalamic nuclei in the pathogenesis of neglect. Furthermore, the small sample size and the heterogeneity of the data remain significant limitations. Finally, neglect was assessed using different diagnostic tools and at varying time points poststroke, leading to inconsistencies in the reported prevalence and severity.
Despite these limitations, this study provides a valuable and comprehensive reflection on the current state of the literature regarding neglect in thalamic stroke – a topic that has, until now, received relatively little attention in research. While the thalamus plays a central role in cognitive and attentional processes, its specific contribution to neglect syndromes remains poorly understood. By systematically analyzing the available evidence, our study identifies key knowledge gaps and underscores the need for more rigorous investigations, particularly for prospective studies including only ischemic strokes and incorporating advanced neuroimaging techniques such as high-resolution MRI and diffusion tensor imaging to better delineate thalamic structures and their connections. Additionally, the adoption of standardized assessment protocols is essential to enhance the consistency and comparability of findings across studies.
Conclusion
Our findings suggest a right hemisphere dominance in thalamic neglect, similar to the lateralization observed in classical cortical lesions that induce neglect. However, the precise neuroanatomical correlates and underlying mechanisms remain unclear due to the heterogeneity and paucity of available data. Understanding the complexity of the thalamus and its extensive connections with the cortex, as well as the complexity of the neglect syndrome, suggests the presence of multiple thalamo-cortical loops involved in different stages of spatial cognitive processes. Rather than providing definitive conclusions, this work serves as a comprehensive synthesis of existing literature, emphasizing the need for further research, including only ischemic strokes, with prospective designs, standardized neglect assessments, and advanced neuroimaging technique. Addressing these gaps will enhance our understanding of thalamic contributions to spatial attention and inform more precise diagnostic and therapeutic strategies for patients with thalamic strokes.
Statement of Ethics
A statement of ethics is not applicable because this study is based exclusively on published literature.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This work was supported by SNSF Grant No. Z00P3_154714/1 to DC and SNSF Grant No. 32003B_196915 TNy.
Author Contributions
Conceptualization and writing – review and editing, A.D., B.C.K., D.C., and T.N.; methodology, literature search, data extraction, methodological quality assessment, and writing – original draft preparation, A.D. and T.N. All authors have read and agreed to the published version of the manuscript.
Data Availability Statement
All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.