Abstract
Objective: To determine the frequency and etiologic significance of multiple acute ischemic lesions in stroke.
Background: Although patients may have more than one stroke during the course of their lives, acute ischemic stroke is usually thought of as a single event. Using diffusion-weighted imaging (DWI), an MRI technique that detects ischemic injury within minutes after onset, we have often observed multiple acute ischemic lesions.
Methods: The MRI scans of 59 consecutively studied patients were reviewed to determine the frequency and etiologic significance of multiple acute ischemic lesions on DWI.
Results: Multiple acute ischemic lesions were present in 10 (17%) of 59 patients. The lesions usually occurred within one major circulation (anterior or posterior), but in two patients (3%), lesions occurred in both cerebral hemispheres or in the anterior and the posterior circulations. The lesions often were small and resulted from presumed multiple emboli or the break-up of an embolus. Two patients had internal carotid artery occlusive disease and four had a cardiac or aortic source. In the other four patients the source was not determined. Lesions larger than 1 cm in diameter progressed to infarction, but some smaller lesions were not seen on follow-up T2-weighted imaging.
Conclusions: Multiple acute stroke lesions on DWI are common and could be caused by multiple emboli or the breakup of an embolus. In some cases it might become possible to make early inferences concerning the stroke mechanism that could be of use for immediately directing the clinical work-up and treatment of the patient.
Although patients may have more than one stroke during the course of their lives,1-6 acute ischemic stroke is typically thought of as a single event. Previous estimates of the occurrence of multiple acute ischemic lesions have been few,3 because conventional imaging shows minimal changes during the first 24 hours, and it can be difficult to distinguish between acute and chronic infarcts. In one study the incidence of simultaneous acute infarcts in one hemisphere was 2%.7 The occurrence of multiple acute lesions suggests embolism from the heart, the aorta, or major extracranial or intracranial vessels, or a coagulation or systemic disturbance, and generally leads to intensified investigations to find the cause.3 Multiple acute infarcts in more than one major circulation (both cerebral hemispheres or anterior and posterior circulations) strongly favors a proximal embolic source or systemic cause.3,8 The presence of multiple lesions in one circulation suggests an ipsilateral large-vessel source or a cardiac embolic source.1,7,8 Patients with multiple acute lesions may be at higher risk of recurrent stroke, and it is important that therapy be instituted as early as possible.
Diffusion-weighted imaging (DWI) is an MRI technique that is of particular interest in the investigation of acute ischemic stroke. Using DWI, ischemic lesions can be detected as early as the first hour after the onset of symptoms,9-12 in contrast to the traditional imaging modalities (CT and conventional MRI), on which definitive lesions are best seen at 24 hours or later. Additional advantages of DWI are that very small ischemic areas can be detected because of the high signal-to-noise ratio11 and that it is possible to differentiate between acute and chronic ischemic.11,13,14 Correlative perfusion imaging allows the state of the relative blood flow to be ascertained and can be obtained in the same imaging study as MR resonance angiography.
Using MR DWI we have often observed that more than one acute DWI lesion may be present. We now report a preliminary analysis of the frequency and causes of multiple acute ischemic lesions as detected by DWI.
Patients and methods.
Patient recruitment.
Patients had been studied with DWI at the Beth Israel Deaconess Medical Center in Boston. The DWI scans of the first 59 consecutive patients in our database (evaluable studies stored on the first 10 compact discs in our archives) were reviewed to determine the frequency of multiple acute ischemic lesions. All MR scans had been performed within 72 hours of the onset of symptoms, most within 24 hours or earlier.
Neuroimaging.
All studies were performed on a 1.5-tesla MR whole-body system (Siemens AG, Erlangen, Germany)—the Vision system or its prototype. DWI was performed using single-shot echo planar imaging, as described previously.10,11 Perfusion imaging was performed after the manual injection of 15 to 20 mL of gadolinium-diethylenetriaminepentaacetic acid (DTPA). Perfusion maps of the relative mean transit time were generated.12 T1-weighted, proton density–weighted, and T2-weighted images were obtained using a spin echo technique.
Definition.
The following definition of multiple lesions was used, modified from that of Bogousslavsky7: 1) the patient presented with a single clinical event, 2) the lesions on DWI were acute, 3) the lesions were “topographically distinct” (separated in space or noncontinuous on contiguous slices), and 4) more than one vascular territory was involved. The vascular territories in the anterior circulation are the anterior cerebral artery, middle cerebral artery (superior division), middle cerebral artery (inferior division), middle cerebral artery (lenticulostriate), single penetrating artery in the deep basal ganglia or white matter, anterior choroidal artery, and watershed. In the posterior circulation the vascular territories are the posterior cerebral artery, superior cerebellar artery, anterior inferior cerebellar artery, posterior inferior cerebellar artery, cerebellar watershed, and brainstem.1
Image analysis.
The images were independently reviewed by two physicians experienced in stroke imaging, one neurologist and one neuroradiologist. There was good agreement between the reviewers in all but one case; after the two reviewers conferred, this patient was removed as a case of multiple acute lesions. The readers otherwise agreed on the location and the number of lesions. At the time of the review the reviewers were not aware of the nature of the symptoms reported by the patients other than that they had had a clinical suspicion of a stroke. The DWI and relative mean transit time (providing an index of relative cerebral blood flow) maps were reviewed separately. A lesion was determined to be acute if 1) the apparent diffusion coefficient was reduced or 2) there was a signal hyperintensity on the DWI but not on the corresponding T2-weighted images. In some of the smallest lesions it was not possible to accurately measure the apparent diffusion coefficient.
In patients with multiple acute lesions, the stroke mechanism was classified into the following categories:15,16 1) large artery atherosclerosis (including large-artery thrombosis and artery-to-artery embolism), 2) cardiogenic embolism, 3) small-artery occlusion (lacunar disease), 4) hypercoagulable state or other determined mechanism, and 5) undetermined mechanism. We used the Trial of Org 10172 in Acute Stroke Treatment classification but did not adhere to the “work-up adequate” and “inadequate” components, as it was difficult to determine these retrospectively. The final diagnosis of stroke was made from the clinical record and from the results of T2-weighted imaging.
Results.
Multiple acute lesions on DWI were found in 10 (17%) of 59 patients. The number of lesions per patient ranged from 2 to 20, with a total of 60 lesions among the 10 patients. Lesions were as small as 4 mm in diameter and often located in the cortex. In 5 of the 10 patients, apart from the index clinically relevant lesion, at least some of the lesions were judged to be clinically silent (table).
Location and etiology of multiple lesions detected on MRI
Location of lesions.
In most patients (8/10) the lesions were located within multiple vascular territories of one circulation. Six patients had more than one lesion in the anterior circulation; two patients had multiple lesions in the posterior circulation. In the other two patients there were lesions in both the anterior and posterior circulations (figure); one of these patients also had lesions affecting both anterior circulations.
Figure. Diffusion-weighted imaging study from a 50-year-old patient who presented with the simultaneous onset of acute right inferior myocardial infarction and left-sided weakness, obtained 12 hours after the onset of symptoms. Hyperintense signal areas on DWI indicate ischemic injury. The T2-weighted study at this time was normal. Multiple areas of abnormality on DWI are present over the right cerebral hemisphere in the right anterior cerebral and middle cerebral artery territories (arrows). There are also lesions in the right posterior cerebral artery territory. Infarction was present in all areas larger than 1 cm diameter on follow-up T2-weighted imaging, but some of the smaller lesions were not seen.
Stroke mechanisms.
In the two patients with anterior and posterior circulation lesions the stroke mechanism was cardiogenic embolism. One patient had an acute myocardial infarct, and the other patient had severe coronary artery disease and left-ventricular dysfunction. In the eight patients with lesions confined to the vascular territories of one circulation, the cause was large-artery atherosclerosis in two, coronary angioplasty in one, aortic-arch thrombus in one (lesions all occurred in left cerebral hemisphere), and undetermined in four. Some lesions were located in the watershed zones in four patients. There were no cases of small-artery occlusive disease (lacunar disease). In patients with single lesions the mechanism was large-artery atherosclerosis in 12, cardiac embolism in 10, lacune in five, other in four, and undetermined in 15 (p > 0.05 by chi-square analysis, compared with patients with multiple lesions).
Eight of 10 patients had also been studied with MR perfusion imaging. There were distinct perfusion defects correponding to the DWI lesions in six patients. In the other patients, the DWI lesions were not associated with a perfusion defect. In two patients discrete DWI lesions were located within one zone of hypoperfusion (Middle cerebral artery [MCA] branch hypoperfusion or total MCA territory hypoperfusion). Lesions larger than 1 cm in diameter were detected on follow-up T2-weighted imaging, but some smaller lesions were not always seen. One patient later developed a confluent infarct that encompassed several of the acute lesions that had been present within a single zone of hypoperfusion.
Discussion.
We found that multiple acute lesions on DWI were present in 17% of patients. They usually occurred within one major circulation but in 3% of patients were bihemispheric or involved both the anterior and posterior circulations. Although the lesions were acute, they may or may not have been of the same age. In acute ischemic stroke the apparent diffusion coefficient remains low for up to 2 to 10 days,10,11,13 after which time it normalizes and then becomes high in the chronic phase. Therefore, in the current study the lesions may have occurred simultaneously, before or after the clinically relevant lesion, but generally would have occurred within a few days or up to a week of the clinically relevant lesion. In many cases these extra “satellite” lesions were silent and would have gone unrecognized, but the high signal-to-noise ratio of DWI allowed the detection of small lesions on the cortical edge, which might often be missed on conventional T2-weighted imaging. The finding of multiple acute stroke lesions may have significance regarding the design and development of acute stroke therapies, particularly in relation to implications for therapeutic administration.
The current study was designed to determine the frequency and cause of the presence of multiple lesions by retrospectively reviewing our MRI database. The next step would be to compare the cause and clinical significance with those of single lesions. We retrospectively looked at this in the 49 other patients, but a formal determination of the positive and negative predictive value of these lesions awaits further formal investigation in a prospective study. We were not able to control the investigative work-up of patients and so cannot be sure that all patients underwent exactly the same work-up to determine the frequency of each subtype. In such a determination, it would be necessary to determine the cause of the patient’s stroke independent of the imaging findings.
We speculate that the majority of the lesions were caused by multiple emboli or the break-up of an embolus. In six patients there was cardiac, aortic, or carotid arterial occlusive disease. In carotid artery occlusive disease, embolism has been shown to be a predominant ischemic mechanism of ischemic stroke.17-20 In reports from angiographic and transcranial doppler ultrasonography studies,17-20 emboli often originate from thrombus in freshly occluded arteries, although hypoperfusion may coexist as a cause of infarction. Embolism is a major accepted mechanism of infarction in patients with aortic and cardiac disease. Emboli can be multiple and simultaneous, or a single embolus may break up and cause multiple ischemic brain lesions. Hypercoagulability is another very important, although less common, cause of recurrent infarcts in different vascular territories.3,8 In the four patients in whom no source was found, the mechanism was presumed to be embolism. In studies of stroke subtype classification there is always a sizable group of patients in whom the source cannot be determined, even if intensive investigations have been performed.15,16 It is postulated that strokes in these patients may be caused by emboli arising from cardiac or vascular sources that have since disappeared or are not able to be detected on investigation.15,16,21
Acutely evolving ischemia on DWI can be patchy and may initially appear multifocal, with lesions subsequently becoming confluent.12 This did occur in one of our cases; in that patient there was a single perfusion defect. To avoid classifying patchy ischemia as multiple ischemic lesions and thereby overestimating the frequency of these, we used a modified classification used by Bogousslavsky:7 Lesions had to involve more than one major vascular territory and had to be topographically distinct. We are confident that multiple acute lesions on DWI do represent ischemic injury, because many occurred in a vascular distribution (for example in the watershed zones), and larger lesions (>1 cm diameter) progressed to infarction. However, as yet unrecognized phenomena cannot be completely ruled out.
It is not clear if all small DWI lesions become very small infarcts or if some resolve; many may be too small to be detected on follow-up T2-weighted images and could account for some cases of “MR negative strokes.”22-24 It is also not clear what the long-term clinical impact, if any, is; for example, on cognitive functioning.
Multiple acute stroke lesions on DWI are common and may result from multiple emboli or the break up of an embolus. In some cases, for example in patients with bihemispheric lesions, it might become possible to make early inferences concerning the stroke mechanism that could be of use for immediately directing the clinical work-up and treatment of the patient. This important subject is deserving of future investigation.
Acknowledgments
Acknowledgment
The authors thank Dr. L. Caplan for his helpful comments and reviews of the manuscript.
Footnotes
-
Presented at the 50th annual meeting of the American Academy of Neurology; Minneapolis, MN; April 1998; and The Lancet Conference on Stroke; Montreal, Canada; October 1998.
-
Dr. Baird is supported by a Clinical Scientist Award from the Doris Duke Charitable Foundation.
- Received February 10, 1999.
- Accepted September 28, 1999.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.