Infarction of the artery of Percheron (AOP), a rare anatomical variant of the thalamoperforating artery, accounts for approximately 4–35% of thalamic stroke cases and 0.1–0.6% of ischemic stroke events.1,2 In most cases, the paramedian part of thalamus is supplied by the thalamoperforating arteries of the thalamus, which usually originate from the ipsilateral posterior cerebral artery (PCA). However, perforating branches may arise from the AOP, a single thalamoperforating artery from the unilateral P1 segment of the PCA irrigating the bilateral paramedian thalamus, with variable contribution to the anterior thalamus and midbrain.3 Therefore, AOP occlusion is mostly characterized by bilateral paramedian thalamic infarctions.
The aim of this study is to summarize the findings of clinical manifestation, radiological patterns, management, and prognosis of AOP infarction in clinical settings.
We retrospectively reviewed patients with acute bilateral thalamic infarction from January 2009 to August 2020 in Taipei Veterans General Hospital, a tertiary medical center in Taiwan. We defined acute infarction as diffusion-weighted imaging sequence hyperintensity on magnetic resonance imaging (MRI) with corresponding decrease in the apparent diffusion coefficient.3
We defined the inclusion criteria for AOP infarction as the presence of acute bilateral paramedian thalamic infarction with or without involvement of anterior thalamus and rostral midbrain, which is consistent with past studies.3–5 The thalamus was divided by both its long and short axes into four quarters, considering it imperfectly oval. We then assigned the medial-anterior quarter as the paramedian part and the lateral-anterior quarter as the anterior part (Fig. 1). We excluded patients with (1) no acute infarction in the bilateral paramedian thalamus, (2) thalamic infarction extending beyond the bilateral paramedian and anterior portion, (3) preexisting or active noncerebrovascular brain parenchymal diseases that might interfere with radiological studies of acute stroke, and (4) an overt diagnosis other than AOP infarction.
We recorded the patients’ demographic characteristics and clinical presentations. We reviewed the patients’ diagnostic findings, including ischemic territories in brain imaging and stenotic patterns on magnetic resonance angiography (MRA). We defined arterial stenosis as >50% stenosis of the large arteries on MRA. Following the classification system from past research,3 we categorized the brain MRIs of all patients into four distinct infarct patterns: (1) bilateral paramedian thalamus with rostral midbrain, (2) bilateral paramedian thalamus without midbrain, (3) bilateral paramedian and anterior thalamus with rostral midbrain, and (4) bilateral paramedian and anterior thalamus without midbrain. The presumed stroke etiologies were in accordance with the trial of ORG 10172 in acute stroke treatment (TOAST) classification system, which denotes five subtypes of ischemic stroke: (1) atherosclerosis, (2) small vessel occlusion, (3) cardioembolism, (4) stroke of other determined etiologies, and (5) stroke of other undetermined etiologies.6
We collected data regarding treatment and prognosis, which included stroke onset to diagnosis time, summary of treatment approach, and graded general prognosis. The stroke onset to diagnosis time was defined as the time interval between the first neurological symptom onset/last time normal and to acquisition of the first brain MRI adequate for AOP infarction diagnosis. We adopted binominal prognosis grading, which included both favorable and unfavorable prognoses, based on a review of medical charts and stroke databases. We defined a favorable prognosis as a modified Rankin Scale (mRS) ≤2 and an unfavorable prognosis as an mRS ≥3 at discharge.7,8
During the study period, we identified 20 patients with bilateral thalamic infarction. We excluded 10 patients because one had no available MRI, four no bilateral paramedian thalamic infarction, two thalamic infarctions beyond the paramedian and anterior part of the thalamus, two presumed diagnosis of basilar tip syndrome (BTS), and one preexisting medulloblastoma with previous stenting and embolization of multiple posterior circulation aneurysms. Finally, 10 patients (six males and four females, onset age 42–100 years, median 82.5 years) were enrolled in the study (Table 1). Hypertension (50%) was the most common comorbidity. For clinical presentation, altered mental status (AMS) was most common in nine patients.
Table 1 -
Summary of demographics, clinical presentations and imaging findings in all study subjects
||n or %
|Age, y, median (25–75th percentile)
| Diabetes mellitus
| Loss of PLR
|MRI infarct patterns
| P + M
| P + A + M
| P + A
|MRA arterial stenosis
| VA (N/U/B)
| PCA (N/U/B)
| BA (N/O)
|Median time to diagnosis in hours (25–75th percentile)
|Presumed etiology (1/2/3/4/5)b
|Prognosis on discharge (favorable/unfavorable)
A = anterior thalamus; AMS = altered mental status; B = bilateral arterial occlusion; BA = basilar artery; M = midbrain; MRA = magnetic resonance angiography; MRI = magnetic resonance imaging; N = no occlusion; O = occlusion; P = paramedian thalamus; PCA = posterior cerebral artery; perc. = percentile; PLR = pupillary light reflex; U = unilateral arterial occlusion; VA = vertebral artery; VGP = vertical gaze palsy.
aData missing for one patient (case 8).
bFive presumed etiologies are represented by numbers: 1, atherosclerosis; 2, small vessel occlusion; 3, cardioembolism; 4, other determined etiologies; and 5, other undetermined etiologies.
After reviewing MRA, bilateral occlusion in the vertebral artery (VA) was seen in two patients, and unilateral VA dissection was seen in one. For PCA, five patients had unilateral occlusion, and one had bilateral occlusion. No patient had basilar artery (BA) occlusion, but BA intramural hematoma and flap were seen in one patient each. Among the four AOP infarct patterns, bilateral paramedian thalamus with rostral midbrain, bilateral paramedian thalamus without midbrain, and bilateral paramedian and anterior thalamus without midbrain were the three most common patterns.
The duration from stroke onset to diagnosis ranged from 4.9 to 552.0 hours (median 39.70 hours), after excluding one patient (case 8) without available records.
All patients received antiplatelet therapy. No patient died due to direct results or sequelae of cerebral infarction. The most common presumed etiology was atherosclerosis (60%). “Unfavorable prognosis” and “favorable prognosis” were obtained for eight and two patients, respectively.
Our study found that AMS was the most common clinical presentation (90%) and was present in almost all cases. Although atherosclerosis was the main presumed etiology (60%) for AOP stroke, atherosclerotic changes in VA (20%) and BA (0%) were markedly rarer than in the P1 segment of the PCA (60%). Therefore, infarcts were mostly restricted to the thalamus.
The age distribution in our study was slightly higher than that the few existing case series of AOP infarction.3,5,9 Seven of the nine patients (78%) with complete records had a confirmed diagnosis of AOP infarction more than 24 hours after stroke onset. The low mortality (0%) was similar to results from the few published case series of AOP infarction. However, 80% of patients had an unfavorable prognosis (mRS ≥3) at discharge after the acute stroke in our study. This indicates that residual symptoms of AOP infarction often persist and affect daily life activity despite medical therapy and rehabilitation.
We retrospectively examined the excluded patients with bilateral thalamic infarction and found that the presumed stroke mechanism could be cardioembolism, cerebral venous thrombosis, BTS, and atherosclerosis of other main trunks. This indicates that various stroke mechanisms can lead to bilateral thalamic infarction, highlighting the importance of recognizing territories irrigated by the AOP. However, we also noticed two patients with bilateral paramedian thalamic infarction who were eventually excluded after a comprehensive review. The final diagnosis was BTS due to compatible findings on MRA and the involvement of other territories typically irrigated by the BA. Past studies have suggested that BTS due to distal BA occlusion often involves the cerebellum, pons, or occipital lobe in addition to the bilateral thalamus.10,11 Therefore, it is important not to solely rely on the infarct pattern in the thalamus but also in other posterior circulation territories.
This study has limitations. We adopted the “four-quarter method” to objectively define the paramedian and anterior part of the thalamus. This method might not perfectly identify AOP ischemic areas in all patients due to the highly variable vascular supply of the thalamus and different observers’ selection of MRI slices. However, it was still supported by general compatibility with the “interpeduncular profundus territory” and “tuberothalamic territory” in a widely accepted research-use computed tomography template proposed by Graff-Radford et al12 and the fact that the infarct territories of all 37 patients in the largest case series for AOP stroke to date, as recognized in the superimposed graph of every MRI ischemic area, were restricted to these two zones.3 Finally, cases with unilateral thalamic infarction, which were excluded from our study, may occasionally occur in AOP infarction. This may lead to potential missing cases during patient enrollment.
This study was supported by grants from the Ministry of Science and Technology, Taiwan (MOST 109-2314-B-075-052-MY2, MOST 110-2321-B-001-011-, 110-2634-F-A49-005-, 111-2321-B-001-010-), Taipei Veterans General Hospital (V111C-216, V110C-059, VGHUST110-G1-5-1), National Health Research Institutes, Taiwan (NHRI-11A1-CG-CO-05-2225-1), and the Brain Research Center, National Yang-Ming Chiao Tung University from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan. The interpretation and conclusions contained herein do not represent those of the funding agencies.
1. Sandvig A, Lundberg S, Neuwirth J. Artery of Percheron infarction: a case report. J Med Case Rep. 2017;11:221.
2. Lin PC, Lee CW, Liu HM, Xiao FR. Acute infarction in the artery of Percheron distribution during cerebral angiography: a case report and literature review. J Radiol Case Rep. 2018;12:1–9.
3. Lazzaro NA, Wright B, Castillo M, Fischbein NJ, Glastonbury CM, Hildenbrand PG, et al. Artery of percheron infarction: imaging patterns and clinical spectrum. AJNR Am J Neuroradiol. 2010;31:1283–9.
4. Arauz A, Patiño-Rodríguez HM, Vargas-González JC, Arguelles-Morales N, Silos H, Ruiz-Franco A, et al. Clinical spectrum of artery of Percheron infarct: clinical-radiological correlations. J Stroke Cerebrovasc Dis. 2014;23:1083–8.
5. Xu Z, Sun L, Duan Y, Zhang J, Zhang M, Cai X. Assessment of Percheron infarction in images and clinical findings. J Neurol Sci. 2017;383:87–92.
6. Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in acute stroke treatment. Stroke. 1993;24:35–41.
7. Rankin J. Cerebral vascular accidents in patients over the age of 60. III. Diagnosis and treatment. Scott Med J. 1957;2:254–68.
8. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19:604–7.
9. Agarwal S, Chancellor B, Howard J. Clinical-radiographic correlates of artery of Percheron infarcts in a case series of 6 patients. J Clin Neurosci. 2019;61:266–8.
10. Rodriguez EG, Lee JA. Bilateral thalamic infarcts due to occlusion of the artery of Percheron and discussion of the differential diagnosis of bilateral thalamic lesions. J Radiol Case Rep. 2013;7:7–14.
11. Jiménez Caballero PE. Bilateral paramedian thalamic artery infarcts: report of 10 cases. J Stroke Cerebrovasc Dis. 2010;19:283–9.
12. Graff-Radford NR, Damasio H, Yamada T, Eslinger PJ, Damasio AR. Nonhaemorrhagic thalamic infarction. Clinical, neuropsychological and electrophysiological findings in four anatomical groups defined by computerized tomography. Brain. 1985;108 (Pt 2):485–516.