Small Centrum Ovale Infarcts on Diffusion-Weighted Magnetic Resonance Imaging
Background and Purpose— A small centrum ovale infarct (SCOI), caused by occlusion of the white matter medullary arteries, is often equated with a lacunar infarct. We sought to clarify the clinical characteristics of a SCOI visualized by diffusion-weighted MRI (DWI) compared with those of a small basal ganglia infarct (SBGI).
Methods— Patients with a SCOI (SCOI group; n=38) or SBGI (SBGI group; n=68) ≤15 mm in diameter on conventional MRI and DWI were selected from 582 consecutive patients with acute ischemic stroke. Sex, age, neurological symptoms, vascular risk factors, emboligenic heart disease, arterial occlusive disease in the ipsilateral carotid system, and recurrent stroke within the initial 30 days were compared between the 2 groups.
Results— Only 47% of SCOIs but 87% of SBGIs could be identified with the use of conventional MRI, whereas DWI could detect them all. Age, sex, and vascular risk factors were not significantly different between the 2 groups. The SCOI group had more frequently an abrupt onset of symptoms (63% versus 26%; P=0.0002), emboligenic heart diseases (34% versus 12%; P=0.0054), occlusive carotid and/or middle cerebral artery diseases (53% versus 19%; P=0.0004), and recurrent stroke (13% versus 1%; P=0.0216) but less frequently a classic lacunar syndrome (50% versus 81%; P=0.0009) than the SBGI group. On a multivariate analysis, both arterial and heart diseases were independently associated with the SCOI group.
Conclusions— Symptomatic SCOIs detected by DWI may be associated with large-vessel and heart diseases and should be distinguished from lacunar infarcts.
The subcortical white matter of the cerebral hemispheres receives blood supply via 2 different vascular systems, deep and superficial penetrating arteries, both originating from the middle cerebral artery (MCA).1–3⇓⇓ The deep penetrating arteries arise directly from the MCA trunk and irrigate the basal ganglia, internal capsule, corona radiata, and caudate head. The superficial penetrating arteries, namely the white matter medullary arteries, arise from the cortical branches of the MCA and supply blood flow to the centrum ovale. A small basal ganglia infarct (SBGI) is often noted as a lacunar infarct, which is caused predominantly by in situ arteriopathy.4–6⇓⇓ A small centrum ovale infarct (SCOI) has a size and shape similar to those of a SBGI on brain imaging studies. However, the pathogenesis of a SCOI remains unresolved.
Diffusion-weighted MRI (DWI) is a powerful tool for detecting acute ischemic lesions and can clearly discriminate fresh infarcts from old ones.7,8⇓ In this respect, it is far more sensitive and specific than CT and conventional MRI.
The aim of the present study was to clarify the clinical characteristics of a SCOI detected by DWI and to compare them with those of a SBGI.
Subjects and Methods
Patients with a SCOI (SCOI group) or SBGI (SBGI group) were selected from 582 consecutive patients with acute ischemic stroke or transient ischemic attack (TIA) who were admitted to our hospital within 7 days of symptom onset between October 1996 and September 2000. We defined a SCOI as an isolated infarct on MRI with a maximal diameter of ≤15 mm located in the territory of the white matter medullary artery arising from the cortical branches of the MCA, according to the templates of Bogousslavsky and Regli.1 A SBGI was defined as a singular infarct on MRI ≤15 mm in diameter localized in the territory of the deep penetrating artery of the MCA, including putamen, globus pallidus, internal capsule, and caudate head. We excluded patients with an infarct in the thalamus, brain stem, and subcortical white matter of the anterior and posterior cerebral artery territories. To locate the SCOI within the territory of the white matter medullary arteries as accurately as possible, patients with an infarct situated deeper in the corona radiata (ie, in a more paraventricular location, closer to the deep penetrating artery territory) were also excluded from the present study.
The following clinical characteristics were compared between the 2 patient groups: (1) patient sex and age; (2) mode of symptom onset; (3) National Institutes of Health Stroke Scale (NIHSS) score on admission; (4) frequency of a classic lacunar syndrome; (5) vascular risk factors including hypertension, diabetes mellitus, hypercholesterolemia, and cigarette smoking; (6) emboligenic heart diseases; (7) arterial occlusive diseases in the ipsilateral carotid system; and (8) recurrence of ischemic stroke or TIA within 30 days of symptom onset.
MRI studies were performed with the use of a Siemens MAGNETOM Vision 1.5-T MR unit with echo-planar capability. We performed conventional MRI studies involving T1-weighted (repetition time [TR]/echo time [TE], 630/14), T2-weighted (TR/TE, 5400/99), and fluid-attenuation inversion recovery (FLAIR) images (TR/TE, 9000/105) within 14 days of symptom onset (mean, 5.8 days; range, 1 to 14 days) in all patients. Forty patients underwent MRI studies within the initial 3 days, 49 between 4 and 7 days after stroke onset, and 17 between 8 and 14 days. DWI studies were simultaneously performed with a multislice, single-shot, spin-echo echo-planar imaging sequence. Diffusion gradients were applied in each of the x, y, and z directions with 2 b values (0 and 1000 s/mm2). To minimize the effects of diffusion anisotropy, an average of the 3 diffusion directions was calculated. MR images were reviewed by a neuroradiologist and a neurologist, both of whom were blinded to clinical data. Infarcts were considered to be symptomatic when they were focally hyperintense on T2-weighted and FLAIR images, isointense or slightly hypointense on T1-weighted images, and located in the vascular region corresponding to the patient’s symptoms. Criteria for the diagnosis of fresh infarcts on DWI included focal hyperintensity, judged not to be due to normal anisotropic diffusion or magnetic susceptibility artifact. The infarct sizes were measured both on axial T2-weighted image and DWI, and the infarcts with a maximal diameter of ≤15 mm were included in the present study.
Mode of neurological symptom onset was classified into 3 categories: abrupt, gradual, and unknown. A classic lacunar syndrome included pure motor hemiparesis, pure sensory stroke, sensorimotor stroke, ataxic hemiparesis, and dysarthria–clumsy hand syndrome, according to previous reports.2,9,10⇓⇓
The vascular risk factors were identified as follows: (1) use of antihypertensive agents for hypertension, with systolic blood pressure ≥160 mm Hg or diastolic blood pressure ≥95 mm Hg on admission for hypertension; (2) use of oral hypoglycemic agents, insulin, or glycosylated hemoglobin >6.4% for diabetes mellitus; (3) use of antihyperlipidemic agents or serum cholesterol level >220 mg/dL for hypercholesterolemia; and (4) any cigarette usage within the 28 days preceding the index stroke for smoking.
To detect emboligenic heart diseases, all patients underwent 12-lead ECG, 24-hour ECG monitoring, and transthoracic echocardiography. Emboligenic heart diseases included atrial fibrillation, mitral stenosis, left ventricular aneurysm, prosthetic heart valves, infective endocarditis, sick sinus syndrome, and dilated cardiomyopathy.
We performed color-flow duplex carotid ultrasonography (Toshiba SSA 270A, Toshiba Inc, or Ultramark 9 HDI, ATL) in all cases. Intracranial arteries were evaluated with conventional cerebral angiography or MR angiography in all patients. The grade of extracranial carotid stenosis was determined according to the criteria used in the North American Symptomatic Carotid Endarterectomy Trial.11 To determine the degree of intracranial stenosis, the vessel being evaluated was measured at its point of maximal narrowing and compared with the normal section of the vessel adjacent to the stenosis. Arterial diseases were considered significant when stenosis >50% or occlusion was evident in the ipsilateral carotid system.
Statistical analyses were performed with the use of a commercially available software package (Stat-View, version 5.0; SAS Institute). We compared the 2 groups with the χ2 test and Student’s t test, and thereafter multivariate logistic regression analysis was performed. Independent variables were sex, age, hypertension, diabetes mellitus, hypercholesterolemia, cigarette smoking, emboligenic heart diseases, and arterial occlusive diseases in the ipsilateral carotid system. A P value of <0.05 was considered statistically significant.
This study protocol followed the principles outlined in the Declaration of Helsinki.
Among the 582 consecutive patients with acute ischemic stroke or TIA, 38 (6.5%) and 68 patients (11.7%) met the criteria for SCOI and SBGI, respectively. Five SCOI and 2 SBGI patients had a clinical diagnosis of TIA. The subjects consisted of 68 men and 38 women, with an age of 67.4±10.1 (mean±SD) years.
All SCOIs and SBGIs were detected by DWI. In the SCOI group, however, conventional MRI studies could identify a symptomatic infarct in only 18 patients (47%). Because the remaining 20 SCOI patients (53%) had multiple or diffuse white matter lesions, conventional MRI studies failed to distinguish a symptomatic infarct from the white matter lesions because of a similar signal intensity. In the SBGI group, the symptomatic infarcts could be identified in 59 patients (87%) by conventional MRI studies.
The demographic and clinical features of the 2 patient groups are shown in Table 1. Twenty-four patients (63%) in the SCOI group had an abrupt onset of neurological symptoms compared with 18 (26%) in the SBGI group (P=0.0002). The neurological presentation was compatible with a classic lacunar syndrome in 55 patients (81%) of the SBGI group but in only 19 (50%) of the SCOI group (P=0.0009). Ten patients (26%) in the SCOI group presented a monoparesis, whereas 3 (4%) in the SBGI group did (P=0.0017). Sex, age, vascular risk factors, and NIHSS score on admission were not significantly different between the 2 groups.
Emboligenic heart diseases were detected in a total of 21 patients (20%). They consisted of 13 patients (34%) in the SCOI group and 8 (12%) in the SBGI group, indicating a higher incidence in the SCOI group than in the SBGI group (P=0.0054). A total of 33 patients (31%) had occlusive carotid and/or MCA disease. It was found in 20 patients (53%) of the SCOI group but in only 13 (19%) of the SBGI group (P=0.0004). Five patients had both heart and arterial occlusive diseases: 3 in the SCOI group and 2 in the SBGI group. Thus, a total of 30 patients (79%) in the SCOI group harbored an emboligenic heart disease and/or an arterial occlusive disease, which was a greater proportion than the 19 patients (28%) in the SBGI group (P<0.0001).
The results of the multivariate logistic regression analysis are presented in Table 2. After adjustment for sex, age, and vascular risk factors, the presence of an emboligenic heart disease (odds ratio [OR], 5.52; 95% CI, 1.61 to 18.9; P=0.0065) and an occlusive carotid and/or MCA disease (OR, 8.13; 95% CI, 2.83 to 23.3; P<0.0001) were independently associated with the presence of a SCOI.
Mean (±SD) number of days of hospitalization was 35±23. We followed up all the patients until 30 days after onset of initial ischemic event. During this period, 5 patients (13%) of the SCOI group and 1 (1%) of the SBGI group developed recurrent stroke or TIA (P=0.0216). All these patients received anticoagulant or antiplatelet therapy soon after the admission. In each of the 5 SCOI patients with recurrent stroke, DWI scans revealed new lesions in the MCA territory ipsilateral to the initial SCOI. Of these patients, 4 had a significant arterial stenosis of the internal carotid artery, and the remaining 1 patient had an atrial fibrillation, which were the same as initial diagnoses at the time of first stroke. The patient in the SBGI group, however, had a new lesion in the contralateral thalamus on DWI and had neither arterial nor heart disease. None of the patients in the present study had a major complication or died until 30 days after onset of initial ischemic event.
The pathogenesis of the SCOI remains controversial. Bogousslavsky and Regli1 suggested that a symptomatic SCOI was associated with hypertension and diabetes mellitus and with lacunar syndromes but not with occlusive carotid disease and emboligenic heart disease, indicating that it is caused mainly by in situ small-vessel disease involving the white matter medullary artery. Similarly, the other 2 studies12,13⇓ reported that patients with SCOI were more likely to have risk factors for small-vessel disease. However, there were some problems in these studies. First, mechanisms other than small-vessel disease may account for a SCOI, because approximately one third of the patients had occlusive carotid disease and/or emboligenic heart disease. Second, patients with multiple or diffuse white matter lesions were usually excluded to avoid false diagnosis of SCOI on CT or conventional MRI scans, although approximately one third of acute stroke patients have been reported to have silent, chronic white matter lesions.14,15⇓ Third, evaluation of the intracranial arteries with conventional cerebral angiography or MR angiography was performed only in selected patients.
In the present study we used DWI, the most sensitive and specific imaging method to detect acute ischemic lesions, and we always evaluated intracranial arteries by conventional cerebral angiography or MR angiography. The SCOI group showed a significantly higher prevalence of occlusive carotid and/or MCA disease and emboligenic heart disease than the SBGI group, although vascular risk factor profiles were similar between the 2 groups. On the multivariate analysis, both arterial occlusive and heart diseases were independently associated with the SCOI. The substudy of the European Carotid Surgery Trial16 suggested that 66% of patients with a SCOI had ipsilateral carotid stenosis of >50% compared with 40% of those with a SBGI. Waterston et al17 studied 10 patients with small deep infarcts in association with occlusive carotid artery disease, of whom 7 patients had only a SCOI. Moreover, Lammie and Wardlaw18 reported, in their pathological study of 12 consecutive autopsy cases with SCOI, that the mechanism of the infarct was definitely or probably cardioembolic in 3 cases, possibly embolic from the heart or aortic arch in 5, and probably embolic from the ipsilateral carotid disease in 2 (ie, 10 of 12 cases had potential sources of emboli). We also found that only one half of the SCOI patients presented a classic lacunar syndrome and that approximately two thirds of them had an abrupt onset. Therefore, we believe that cerebral large-vessel disease and emboligenic heart disease may play an important role in the mechanism of SCOI rather than small-vessel disease. Large-vessel diseases as well as heart diseases can mediate a SCOI by embolic mechanism. The high incidence of abrupt onset in the SCOI group may support this hypothesis. A hemodynamic mechanism also could be considered in some SCOI patients with large-vessel disease, at least in 2 patients with internal carotid artery occlusion.
Other potential sources of emboli, including right-to-left shunt representing patent foramen ovale19 and atherosclerotic disease of the aortic arch,20,21⇓ may have been overlooked in our study because we did not routinely perform transesophageal echocardiography. Further investigations should be necessary to clarify the relation of these emboligenic diseases to SCOIs.
In conclusion, symptomatic SCOIs on DWI studies, unlike SBGIs, were significantly associated with cerebral large-vessel disease and emboligenic heart disease. We therefore should distinguish a SCOI from a lacunar infarct.
This study was supported in part by Research Grants for Cardiovascular Diseases (8C-4, 9A-2, 9A-3, 9A-8) from the Ministry of Health and Welfare of Japan and by Special Coordinating Funds for Promoting Science and Technology (Strategic Promotion System for Brain Science) from the Science and Technology Agency of Japan.
- Received September 20, 2001.
- Revision received November 14, 2001.
- Accepted December 3, 2001.
- ↵Bogousslavsky J, Regli F. Centrum ovale infarcts: subcortical infarction in the superficial territory of the middle cerebral artery. Neurology. 1992; 42: 1992–1998.
- ↵Moody DM, Bell MA, Challa VR. Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study. AJNR Am J Neuroradiol. 1990; 11: 431–439.
- ↵Fisher CM. Lacunes: small, deep cerebral infarcts. Neurology. 1965; 15: 774–784.
- ↵Ghika J, Bogousslavsky J, Regli F. Infarcts in the territory of the deep perforators from the carotid system. Neurology. 1989; 39: 507–512.
- ↵Singer MB, Chong J, Lu D, Schonewille WJ, Tuhrim S, Atlas SW. Diffusion-weighted MRI in acute subcortical infarction. Stroke. 1998; 29: 133–136.
- ↵Fisher CM. Lacunar strokes and infarcts: a review. Neurology. 1982; 32: 871–876.
- ↵Leys D, Mounier-Vehier F, Rondepierre P, Leclerc X, Godefroy O, Marchau M Jr, Scheltens P, Pruvo JP. Small infarcts in the centrum ovale: study of predisposing factors. Cerebrovasc Dis. 1994; 4: 83–87.
- ↵Jørgensen HS, Nakayama H, Raaschou HO, Gam J, Olsen TS. Silent infarction in acute stroke patients: prevalence, localization, risk factors, and clinical significance: the Copenhagen Stroke Study. Stroke. 1994; 25: 97–104.
- ↵Schmidt R, Fazekas F, Kleinert G, Offenbacher H, Gindl K, Payer F, Freidl W, Niederkorn K, Lechner H. Magnetic resonance imaging signal hyperintensities in the deep and subcortical white matter: a comparative study between stroke patients and normal volunteers. Arch Neurol. 1992; 49: 825–827.
- ↵Toyoda K, Yasaka M, Nagata S, Yamaguchi T. Aortogenic embolic stroke: a transesophageal echocardiographic approach. Stroke. 1992; 23: 1056–1061.