Isolated truncal contrapulsion as a rare presentation of acute thalamic infarct.

Infarcts involving the thalamus can yield many deficits, including sensory syndromes, altered consciousness, and cognitive disturbances, depending on the thalamic vascular territory involved. Isolated truncal contrapulsion due to pure thalamic infarct has been rarely reported. Truncal lateropulsion is a compelling sensation of being pulled toward one side that cannot be explained by weakness or limb ataxia. It is commonly reported in lateral medullary infarcts. It may occur with lesions that involve the peripheral vestibular system, brainstem, cerebellum, basal ganglia, ponto-mesencephalic, and thalamic lesions. We hereby report a 64-year-old woman who presented with truncal contrapulsion as the sole manifestation of an acute right lateral thalamic infarct.

Towards subject-level cerebral infarction classification of CT scans using convolutional networks.

Automatic evaluation of 3D volumes is a topic of importance in order to speed up clinical decision making. We describe a method to classify computed tomography scans on volume level for the presence of non-acute cerebral infarction. This is not a trivial task, as the lesions are often similar to other areas in the brain regarding shape and intensity. A three stage architecture is used for classification: 1) A cranial cavity segmentation network is developed, trained and applied. 2) Region proposals are generated 3) Connected regions are classified using a multi-resolution, densely connected 3D convolutional network. Mean area under curve values for subject level classification are 0.95 for the unstratified test set, 0.88 for stratification by patient age and 0.93 for stratification by CT scanner model. We use a partly segmented dataset of 555 scans of which 186 scans are used in the unstratified test set. Furthermore we examine possible dataset bias for scanner model and patient age parameters. We show a successful application of the proposed three-stage model for full volume classification. In contrast to black-box approaches, the convolutional network's decision can be further assessed by examination of intermediate segmentation results.

Classification of Covert Brain Infarct Subtype and Risk of Death and Vascular Events.

Background and Purpose- To test the hypothesis that covert brain infarcts (CBIs) are more likely to be located in noneloquent brain areas compared with clinical strokes and that CBI etiological subtypes carry a differential risk of vascular events compared with people without CBI. Methods- We used brain magnetic resonance imaging from 1290 stroke-free participants in the NOMAS (Northern Manhattan Study) to evaluate for CBI. We classified CBI as cardioembolic (ie, known atrial fibrillation), large artery atherosclerosis (extracranial and intracranial), penetrating artery disease, and cryptogenic (no apparent cause). CBI localized in the nonmotor areas of the right hemisphere were considered noneloquent. We then evaluated risk of events by CBI subtype with adjusted Cox proportional models. Results- At the time of magnetic resonance imaging, 236 participants (18%) had CBI (144 [61%] distal cryptogenic, 29 [12%] distal cardioembolic, 26 [11%] large artery atherosclerosis, and 37 [16%] penetrating artery disease). Smaller (per mm, odds ratio, 0.8 [0.8-0.9]) and nonbrain stem infarcts (odds ratio, 0.2 [0.1-0.6]) were more likely to be covert. During the follow-up period (10.4±3.1 years), 398 (31%) died (162 [13%] of vascular death) and 117 (9%) had a stroke (99 [85%]) were ischemic. Risks of events varied by CBI subtype, with the highest risk of stroke (hazard ratio, 2.2 [1.3-3.7]) and vascular death (hazard ratio, 2.24 [1.29-3.88]) noted in participants with intracranial large artery atherosclerosis-related CBI. Conclusions- CBI can be classified into subtypes that have differential outcomes. Certain CBI subtypes such as those related to intracranial large artery atherosclerosis have a high risk of adverse vascular outcomes and could warrant consideration of treatment trials.

The relationship between oxidized low-density lipoprotein and related ratio and acute cerebral infarction.

This study aimed to study the value of oxidized low-density lipoprotein (OX-LDL) and related ratio in the diagnosis of acute cerebral infarction and the classification of acute cerebral infarction.Of the 129 patients enrolled in this study, 94 patients with acute cerebral infarction were assigned to the case group, and 35 healthy subjects were enrolled as control group (n = 35). And then the case group were divided into large-artery atherosclerosis (LAA) group (n = 61) and small-artery occlusion (SAO) group (n = 33) according to the TOAST classification standard. Plasma OX-LDL levels were determined by enzyme-linked immunosorbent assay. OX-LDL/total cholesterol (OX-LDL/TC), OX-LDL/high-density lipoprotein (OX-LDL/HDL), OX-LDL/LDL were calculated.There were significant differences in OX-LDL, OX-LDL/TC, OX-LDL/HDL, and OX-LDL/LDL in patients with acute cerebral infarction and those in control group (P < .001). The area under the receiver-operating characteristic curve of OX-LDL and related ratio was >0.7 (P < .001). There was a slight positive correlation between OX-LDL/TC and National Institutes of Health Stroke Scale score at admission (r = 0.265, P = .039) in the LAA group.OX-LDL, OX-LDL/TC, OX-LDL/HDL, and OX-LDL/LDL were closely related to acute cerebral infarction, especially with large atherosclerotic cerebral infarction. OX-LDL/TC can reflect the severity of acute cerebral infarction for LAA, but it cannot predict the short-term prognosis of acute cerebral infarction.

Retrospective analysis of prognosis and risk factors of patients with stroke by TOAST.

To determine differences in 90-day mortality and identify risk factors among different etiological classifications of ischemic stroke using the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) classification.Our retrospective analysis included 538 ischemic stroke patients. The cause of stroke was categorized according to the TOAST criteria, and 90-day mortality rates were obtained through the patient follow-up. Age, sex, previous medical history, and clinical features were used in the analysis of potential risk factors.There were 38 deaths during the 90-day follow-up period. Patients in the undetermined cause subgroups experienced significantly higher mortality rate than those in subgroups with small artery occlusion and large artery atherosclerosis. Factors independently associated with 90-day mortality for patients with the large artery atherosclerosis stroke subtype were age (95% confidence interval [CI], 1.010-1.192, P = .028), history of hypertension (95% CI, 3.030-99.136, P = .001), high blood glucose (95% CI, 1.273-2.354, P < .001), high cholesterol (95% CI, 0.017-0.462, P = .004), high uric acid (95% CI, 2.360-64.389, P = .003), and National Institute of Health Stroke Scale(95% CI, 1.076-1.312, P = .001). Age (95% CI, 1.012-1.358, P = .034) and high cholesterol (95% CI, 0.011-0.496, P = .007) were independently associated with 90-day mortality for patients with the small artery occlusion subtype of stroke.Our analysis identified that certain risk factors and 90-day mortality differ significantly among different stroke subtypes, as classified by the TOAST criteria. These risk factors must be considered carefully to provide the best clinical management of these patients and thus reduce mortality.

Silent cerebral infarct definitions and full-scale IQ loss in children with sickle cell anemia.

OBJECTIVE: To evaluate whether application of the adult definition of silent cerebral infarct (SCI) (T2-weighted hyperintensity ≥5 mm with corresponding T1-weighted hypointensity on MRI) is associated with full-scale IQ (FSIQ) loss in children with sickle cell anemia (SCA), and if so, whether this loss is greater than that of the reference pediatric definition of SCI (T2-weighted hyperintensity ≥3 mm in children on MRI; change in FSIQ -5.2 points; p = 0.017; 95% confidence interval [CI] -9.48 to -0.93). METHODS: Among children with SCA screened for SCI in the Silent Cerebral Infarct Transfusion trial, ages 5-14 years, a total of 150 participants (107 with SCIs and 43 without SCIs) were administered the Wechsler Abbreviated Scale of Intelligence. A multivariable linear regression was used to model FSIQ in this population, with varying definitions of SCI independently substituted for the SCI covariate. RESULTS: The adult definition of SCI applied to 27% of the pediatric participants with SCIs and was not associated with a statistically significant change in FSIQ (unstandardized coefficient -3.9 points; p = 0.114; 95% CI -8.75 to 0.95), with predicted mean FSIQ of 92.1 and 96.0, respectively, for those with and without the adult definition of SCI. CONCLUSIONS: The adult definition of SCI may be too restrictive and was not associated with significant FSIQ decline in children with SCA. Based on these findings, we find no utility in applying the adult definition of SCI to children with SCA and recommend maintaining the current pediatric definition of SCI in this population.

Isolated INO as a presentation of midbrain paramedian area lacunar infarction in patients with diabetes.

The clinical features and pathogenesis of the pure midbrain infarction need to be described. Of 4257 stroke patients who were on the stroke registry between January 2000 and December 2015, 25 patients with pure midbrain infarctions, as demonstrated on diffusion-weighted magnetic resonance imaging, were enrolled. We analyzed the clinical features, MRI findings, and etiologic mechanisms of the infarctions. According to the distribution of each penetrating artery of the midbrain, we classified all the infarctions into paramedian (PM) area (13 patients), lateral area (10 patients), and PM and lateral areas (2 patients). In the lacunar infarction of the PM area group (8 patients), 6 patients had diabetes mellitus (DM), and 5 patients showed isolated anterior internuclear ophthalmoplegia (INO)-the ratio was much higher than that of the non-PM-lacunar infarction group. Patients with DM were more likely to develop lacunar infarction of the midbrain PM area than that of non-diabetics. Data suggest that isolated INO is a presentation of midbrain PM area lacunar infarction in patients with DM.

Intracranial Blood Flow Changes in Patients with High-Grade Severe Carotid Artery Stenosis After Stenting.

OBJECTIVE: We investigated whether the cerebral hemodynamic changes of pre- and poststenting in patients with severe carotid stenosis differ by stenosis grades. METHODS: We prospectively recruited patients who underwent carotid artery stenting (CAS) after acute ischemic stroke from June 2014 to December 2015. We compared the mean relative cerebral blood flow (rCBF) changes (measured by whole-brain computed tomography perfusion) pre- and poststenting in patients with high-grade severe stenosis (HGSS) (90%-99%) versus patients with low-grade severe stenosis (LGSS) (70%-89%). RESULTS: Among 24 patients included in the study (mean age, 66.2 ± 7.2 years; 91.7% men), 62.5% (15/24) were in the HGSS group and 37.5% (9/24) were in the LGSS group. In the HGSS group, rCBF increased in territories of the anterior cerebral artery (P = 0.021), middle cerebral artery (P < 0.001), posterior cerebral artery (P = 0.001), and basil ganglia (P = 0.003) after stenting. Of the patients with HGSS, 53.3% (8/15) had collateral flow through anterior communicating artery (AcomA) prestenting. After stenting, all the AcomA collaterals in HGSS reverted to normal (P = 0.002). CONCLUSIONS: The improvement of brain perfusion combined with the normalization of collateral flow through the circle of Willis after CAS was observed only in patients with HGSS.

Proposed methodology and classification of Infarct in New Territory (INT) after endovascular stroke treatment.

While the overall complication rates for endovascular treatment for acute stroke has been extremely low in recent trials, it is important to separate out and accurately document complications. One of these complications that is usually related to the endovascular intervention is Infarct in New Territory (INT). We propose a standardized methodology for documenting INT after the procedure. This new classification takes into account variations in vascular anatomy and location of the occlusion. In addition, given that after the recent trials, vascular imaging (eg, CT angiography (CTA)) is now the standard of care in the work up of acute ischemic stroke, this classification utilizes the information on the pre-procedure non-invasive vascular imaging, the angiography images from end of procedure and the location of lesions on the follow-up scan.

Infarct in a New Territory After Treatment Administration in the ESCAPE Randomized Controlled Trial (Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion With Emphasis on Minimizing CT to Recanalization Times).

BACKGROUND AND PURPOSE: Infarct in a new previously unaffected territory (INT) is a potential complication of endovascular treatment. We applied a recently proposed methodology to identify and classify INTs in the ESCAPE randomized controlled trial (Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion With Emphasis on Minimizing CT to Recanalization Times). METHODS: The core laboratory identified INTs on 24-hour follow-up imaging, blinded to treatment allocation, after assessing all baseline imaging. INTs were classified into 3 types (I-III) and 2 subtypes (A/B) based on size and if catheter manipulation was likely performed across the vessel territory ostium. Logistic regression was used to understand the effect of multiple a priori identified variables on INT occurrence. Ordinal logistic regression was used to analyze the effect of INTs on modified Rankin Scale shift at 90 days. RESULTS: From 308 patients included, 14 INTs (4.5% overall; 2.8% on follow-up noncontrast computed tomography, 11.7% on follow-up magnetic resonance imaging) were identified (5.0% in endovascular treatment arm versus 4.0% in control arm [P=0.7]). The use of intravenous alteplase was associated with a 68% reduction in the odds of INT occurrence (3.0% with versus 9.1% without; odds ratio, 0.32; 95% confidence interval, 0.11-0.96; adjusted for age, sex, and treatment type). No other variables were associated with INTs. INT occurrence was associated with reduced probability of good clinical outcome (common odds ratio, 0.25; 95% confidence interval, 0.09-0.74; adjusted for age, type of treatment, and follow-up scan). CONCLUSIONS: INTs are uncommon, detected more frequently on follow-up magnetic resonance imaging, and affect clinical outcome. In experienced centers, endovascular treatment is likely not causal, whereas intravenous alteplase may be therapeutic. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT01778335.

Morphological classification of penetrating artery pontine infarcts and association with risk factors and prognosis: The SPS3 trial.

BACKGROUND: Pontine infarcts are common and often attributed to small vessel disease ("small deep infarcts") or basilar branch atherosclerosis ("wedge shaped"). A well-described morphological differentiation using magnetic resonance images has not been reported. Furthermore, whether risk factors and outcomes differ by morphology, or whether infarct morphology should guide secondary prevention strategy, is not well characterized. METHODS: All participants in the Secondary Prevention of Small Subcortical Strokes Study with magnetic resonance imaging -proven pontine infarcts were included. Infarcts were classified as well-circumscribed small deep (small deep infarct, i.e. lacunar), paramedian, atypical paramedian, or other based on diffusion-weighted imaging, T2/fluid-attenuated inversion recovery, and T1-magnetic resonance images. Inter-rater reliability was high (90% agreement, Cohen's kappa = 0.84). Clinical and radiologic features independently associated with small deep infarct versus paramedian infarcts were identified (multivariable logistic regression). Differences in stroke risk and death were assessed using Cox proportional hazards. RESULTS: Of the 3020 patients enrolled, 644 had pontine infarcts; 619 images were available: 302(49%) small deep infarct, 245 (40%) paramedian wedge, 35 (6%) atypical paramedian, and 37 (6%) other. Among vascular risk factors, only smoking (OR 2.1, 95% CI 1.3-3.3) was independently associated with small deep infarct versus paramedian infarcts; on neuroimaging, old lacunes on T1/fluid-attenuated inversion recovery (OR 1.8, 1.3-2.6) and intracranial stenosis (any location) ≥50% (OR 0.62, 0.41-0.96). Small deep infarct versus paramedian was not predictive of either recurrent stroke or death, and there was no interaction with assigned treatment. CONCLUSIONS: Pontine infarcts can be reliably classified based on morphology using clinical magnetic resonance images. Few risk factors differed between small deep infarct and paramedian infarcts with no differences in recurrent stroke or mortality. There was no difference in response to different antiplatelet or blood pressure treatment strategies between these two groups. REGISTRATION: http://www.clinicaltrials.gov/NCT00059306.

MRI Detection of Cerebral Infarction in Subarachnoid Hemorrhage.

OBJECTIVE: To investigate magnetic resonance imaging (MRI) detection of cerebral infarction (CI) in patients presenting with subarachnoid hemorrhage (SAH). BACKGROUND: CI is a well-known complication of SAH that is typically detected on computed tomography (CT). MRI has improved sensitivity for acute CI over CT, particularly with multiple, small, or asymptomatic lesions. METHODS: With IRB approval, 400 consecutive SAH patients admitted to our institution from August 2006 to March 2011 were retrospectively reviewed. Traumatic SAH and secondary SAH were excluded. Data were collected on demographics, cause of SAH, Hunt Hess and World Federation of Neurosurgical Societies grades, and neuroimaging results. MRIs were categorized by CI pattern as single cortical (SC), single deep (SD), multiple cortical (MC), multiple deep (MD), and multiple cortical and deep (MCD). RESULTS: Among 123 (30.8 %) SAH patients who underwent MRIs during their hospitalization, 64 (52 %) demonstrated acute CI. The mean time from hospital admission to MRI was 5.7 days (range 0-29 days). Among the 64 patients with MRI infarcts, MRI CI pattern was as follows: MC in 20 (31 %), MCD in 18 (28 %), SC in 16 (25 %), SD in 3 (5 %), MD in 2 (3 %), and 5 (8 %) did not have images available for review. Most infarcts detected on MRI (39/64 or 61 %) were not visible on CT. CONCLUSIONS: The use of MRI increases the detection of CI in SAH. Unlike CT studies, MRI-detected CI in SAH tends to involve multiple vascular territories. Studies that rely on CT may underestimate the burden of CI after SAH.

Stroke Subtypes and Topographic Locations Associated with Neurological Deterioration in Acute Isolated Pontine Infarction.

OBJECTIVE: This study investigated predictors of neurological deterioration (ND) in acute isolated pontine infarction. METHODS: Two hundred fifty-nine patients with acute isolated pontine infarctions identified using diffusion-weighted imaging were retrospectively analyzed. The patients were divided according to the presence/absence of ND, defined as increased (≥2 units) National Institutes of Health Stroke Scale scores 5 days after onset. Pontine infarctions comprised 3 stroke subtypes: vertebrobasilar large-artery disease, basilar artery branch disease (BABD), and small-artery disease (SAD), according to basilar artery atherosclerosis severity and lesion extent of the transverse axial plane. Topographic locations of longitudinal pontine infarctions in the axial plane were divided into upper, middle, lower, and whole. RESULTS: Of the 259 patients (male : female = 136:123, 68.84 ± 10.24), only 27.4% exhibited ND. The prevalence was significantly increased in females, whereas smoking was significantly decreased in patients with ND. BABD and lower pontine infarctions were significantly more frequent in patients with ND (70.4% and 43.7%, respectively) than in patients without ND (51.6% and 30.3%, respectively). SAD and upper pontine infarctions were significantly less frequent in patients with ND (16.9% and 7.0%, respectively) than in patients without ND (30.3% and 23.4%, respectively). BABD and lower pontine infarctions were positively related to ND. CONCLUSIONS: This is the first study to demonstrate that BABD and lower pons lesions are predictors of ND in acute isolated pontine infarction. These findings indicate the potential importance of early identification of stroke subtypes and topographic locations in the prevention of ND in patients with suspected pontine infarction.

Infarct volume-based subgroup selection in acute ischemic stroke trials.

BACKGROUND AND PURPOSE: We investigated whether hyperintensities with a diameter of at least 3, 3.5, and 4 cm and visible on at least 3 slices on diffusion-weighted imaging enables patient selection with an infarct volume of ≥15 mL. METHODS: Consecutive acute stroke patients were screened for the AXIS2 trial and examined according to a standardized magnetic resonance imaging protocol in 65 sites. Diffusion-weighted lesion diameters were measured and compared with volumetric assessments. RESULTS: Out of 238 patients, 86.2% (N=206) had infarct diameter of at least 3 cm. Volumetric assessments showed infarct volume of ≥15 mL in 157 patients. A cut-off value of 3 cm led to 96.8% sensitivity and 33.3% specificity for predicting lesion volume of ≥15 mL. Analogously, a 3.5 cm cut-off led to 96.8% sensitivity and 50.6% specificity and a 4 cm cut-off led to 91.7% sensitivity and 61.7% specificity. CONCLUSIONS: Lesion diameter measures may enable multicentric patient recruitment with a prespecified minimal infarct volume. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00927836.

Noninvasive Qureshi Grading Scheme Predicts 90-Day mRS in Patients with Acute Ischemic Stroke.

BACKGROUND: The Qureshi grading scheme is an effective classification system for evaluating the severity of acute arterial occlusion. However, this scheme is of limited utility because it is based on invasive angiography. In this study, we assessed whether a relationship between a noninvasive Qureshi score, based on magnetic resonance angiography (MRA) or computed tomography angiography (CTA), and 90-day functional outcome could be observed in patients with acute ischemic stroke. METHODS: A stroke neurologist evaluated all patients with acute ischemic stroke who presented to the emergency room within 12 hour of symptom onset. Two neurologists independently assessed the noninvasive Qureshi score from initial MRA or CTA. We assessed the relationship between the noninvasive Qureshi grading scheme and clinical outcome on day 90. RESULTS: Of a total 125 patients, 75 underwent MRA and 50 underwent CTA. Interobserver reliability showed good agreement (κ = .62). The noninvasive Qureshi score for MRA or CTA and that for CTA alone were directly associated with a good 90-day functional outcome (odds ratio, .672; P = .016 and odds ratio, .511; P = .042). CONCLUSIONS: The noninvasive Qureshi scheme using MRA or CTA provides meaningful information about long-term functional outcomes in patients with acute ischemic stroke.

Multiple variant type thalamic infarcts: pure and combined types.

BACKGROUND AND PURPOSE: Variant topographic patterns of thalamic infarction with distinct manifestations have been classified into three territories: anteromedian, central, and posterolateral. The purpose of this study was to determine clinical, etiological, and radiological features of multiple variant thalamic infarcts. METHODS: We reviewed 8400 patients with a first clinical stroke included in the Ege Stroke Registry between 2000 and 2013. Among 80 patients with an acute multiple thalamic infarcts confirmed by MRI, we selected all patients with lesions outside the classical territories and studied their clinical, etiological, and radiological features. RESULTS: Among 8400 patients with first-ever stroke in our registry, 21 patients (26% of all multiple thalamic infarcts) showed infarction outside the classical territories, allowing us to delineate three variant distributions; (i) unilateral multiple variant infarcts [seven patients (9%) in the anteromedian, central, and posterolateral territories] presented with predominantly decreased vigilance (66% with right lesions, 75% with left lesions), cognitive impairment including amnesia (71%), aphasia (57%) in left-sided or bilateral lesions, and executive dysfunction (43%). The most frequent stroke mechanism was cardioembolism (43%). (ii) Bilateral multiple variant infarcts [five patients (6%)], with lesions on the variant territories of the thalamus, resulting in a variety of neurological and neuropsychological signs, consciousness disturbances (80%), sensory-motor deficits (80%). Cardioembolism (60%) was the most frequent etiology. (iii) Combined multiple variant and classical infarcts [nine patients (11%)], characterized by hemihypesthesia (89%) as the most frequent manifestation, followed by hemiataxia (78%), and cognitive deficits. Cardioembolism (56%) and large-artery disease of the vertebrobasilar system (33%) were the main stroke mechanisms. CONCLUSIONS: We described multiple variant topographic patterns of thalamic infarction with distinct manifestations and etiologies. We thought that multiple variant infarcts are the result of variation in thalamic arterial supply or reflect a source of embolism.

Correlation between the youth cerebral infarction in different TOAST classifications and high homocysteine.

Provide the basis for clinical pathogeny diagnosis and discuss effective ways of curing the youth cerebral infarction in different TOAST classifications through analysis of the correlation between the youth infarction in different TOAST classifications and high homocysteine (Hcy). The selected young patients who diagnosed with cerebral infarction and admitted to the hospital in the past 3 years were regarded as the study group, and the other 136 persons picked accepted health examination in our hospital in the corresponding period were regarded as the control group, then the 136 patients were grouped according to TOAST classification; Hcy levels were determined by enzymatic cycling to compare and analyze, and NIHSS scale is applied to score. Compared with the control group, serum Hcy levels in young patients with cerebral infarction were significantly higher (P < 0.05). The statistical significance in comparative difference was found in the serum Hcy level and NIHSS score of large artery atherosclerotic (LAA) subtype from TOAST classification and the other four subtypes (P < 0.05), but such significance was absent among the other four subtypes (SAA, CA, SOE, SUE). In conclusion, the cerebral etiology of high Hcy is often related to the atherosclerosis and tends to be more serious than the other subtypes after onset of LAA, which requires more clinical attention. Early intervention and therapy are highly necessary specific to cause of disease.

Branching patterns determine the size of single subcortical infarctions.

BACKGROUND AND PURPOSE: The size variation of single subcortical infarcts (SSIs) may be because of differences in stroke mechanisms or branching patterns of perforators. We aim to investigate this issue using high-resolution MRI. METHODS: Patients with high-resolution MRI-identified SSI without significant (≥50%) middle cerebral artery stenosis were enrolled. Perforator stems were defined as perforators originating from the middle cerebral artery and perforator branches as linear structures inside the infarcted area, traceable from >2 adjacent slices. The branching index was calculated as the number of perforator branches divided by the number of perforator stems. Clinical and imaging characteristics were compared between large (≥20 mm in diameter) and small SSI groups. RESULTS: Forty-one patients (10 large and 31 small SSIs) were enrolled. Larger SSIs were more closely associated with diabetes mellitus and severe neurological dysfunction but not with the presence of middle cerebral artery plaque. Although there was no difference in the number of perforator stems, the number of perforator branches (4.8±2.2 versus 2.3±1.4; P=0.005) and branching index (2.9±1.0 versus 1.2±0.8; P<0.001) was higher in the large SSI group. SSI diameter showed a significant correlation with the number of perforator branches (r=0.630; P<0.001) and branching index (r=0.750; P<0.001). CONCLUSIONS: SSI diameter seems to be associated with anatomic branching variation rather than the mechanism of stroke. Definition of small vessel disease with lesion diameter criteria may not be appropriate.