Diffusion-Weighted Imaging II:Is There a New Gold Standardin Acute Stroke Imaging?
Abstracts & Commentary
Sources: Albers GW, et al. Yield of diffusion-weighted MRI for detection of potentially relevant findings in stroke patients. Neurology 2000;54:1562-1567; Lansberg MG, et al. Comparison of diffusion-weighted MRI and CT in acute stroke. Neurology 2000;54:1557-1561; Hacke W, Warach S. Diffusion-weighted MRI as an evolving standard of care in acute stroke. Neurology 2000;54:1548-1549; Powers WJ. Testing a test: A report card for DWI in acute stroke. Neurology 2000;54:1549-1551.
Ct scanning is the current standard of care for evaluation of acute ischemic stroke. The manuscripts by Lansberg and associates and Albers and colleagues with accompanying pro and con editorials by Hacke and Warach and Powers debate whether magnetic resonance imaging (MRI) with diffusion-weighted imaging (DWI) should supplant computerized tomography (CT) as the study of choice for acute stroke.
Lansberg et al studied 19 consecutive acute stroke patients, evaluated with both CT and DWI. DWI was more accurate for identifying acute infarction, showing evidence of acute infarction in all cases, compared with 42% or 63% on CT, depending on the reader. DWI was more sensitive for identification of middle cerebral artery (MCA) involvement of more than 33% (14% or 43% on CT compared to 57% or 86% for MRI, again depending on the reader). MCA involvement of more than 33% on CT is a key contraindication to intravenous thrombolysis as shown in the ECASS II trial. Lesion volume on DWI rather than CT was found to most closely correlate with the ultimate infarct size.
In a series of 40 patients, Albers et al describe the use of DWI in extending knowledge beyond that of a clinical impression and conventional MRI alone. Using DWI, Albers et al demonstrate that the clinical localization was incorrect in 30% of patients and identified the wrong vascular territory in 18% of cases. DWI may assist in determinations of stroke pathogenesis. Acute lesions in multiple vascular territories, implying an embolic etiology, were found in 13% of patients. DWI may also help differentiate between acute and chronic lesions. Twenty percent of lesions thought to be acute on conventional MRI were actually old.
In the "pro" editorial, Hacke and Warach interpret these data in favor of MRI over CT. As they observe, MRI with DWI/PWI and magnetic resonance angiography (MRA) is superior to achieve detailed anatomic localization, define the extent of acute stroke, determine a vascular anatomy, and draw informed conclusions about appropriate therapy. One concern is the MRI is not sufficient to rule out acute intracerebral hemorrhage prior to anticoagulation or thrombolysitc therapy but, according to Hacke and Warach, MRI is sufficient to distinguish between hyperacute intracerebral hemorrhage and ischemia.
In the "con" editorial, Powers notes that MRI techniques have not been subject to the kind of rigorous experimental design or validated outcomes data needed to accept MRI as a new imaging standard. As Powers observes, the Lansberg et al study is limited by "incorporation bias," using late DWI as a standard (the test itself) to validate the accuracy of early DWI imaging. Neither does Lansberg prove that DWI end points correlate with patient outcomes, the true measure of clinical benefit. Similarly, the Albers et al study does not demonstrate that the additional "clinical pearls" gained from DWI ultimately achieved diagnostic accuracy that could not be gained by other means: e.g., further clinical evaluation, follow-up imaging, or testing such as echocardiography or Doppler ultrasonography.
Commentary
Initial interpretations of DWI data suggested that a diffusion abnormality was an all-or-nothing ischemic event. Data now indicate that these lesions may vary significantly in time, course, and extent. As the works of Li and Kidwell indicate, early DWI changes may or may not prove to correlate with irreversible ischemic damage. Resumption of blood flow at an early time interval after a stroke may allow resolution of DWI abnormalities seen in the hyperacute stage. DWI abnormalites may also transiently vanish and then reappear at a later time interval. This may be due to ongoing excitotoxic injury or late apoptosis. In addition, an ultimately normal DWI does not ensure normal tissue. Histological data suggest that, even with a normal DWI, neuronal damage may still exist.
DWI is a crucial extension of our diagnostic acumen, as the works of Lansberg et al and Albers et al clearly show. There is little doubt that it vastly improves sensitivity over CT. It is also probably sufficient to rule out hemorrhage prior to thrombolysis. However, as Powers argues, data from randomized trials incorporating DWI into acute stroke management are needed before a new gold standard can be declared. —alan z. segal & ayeesha kamal (Dr. Kamal is Chief Resident of Neurology at New York Presbyterian Hospital.)
References
1. Li F, et al. Transient and permanent resolution of ischemic lesions on diffusion-weighted imaging after brief periods of focal ischemia in rats: Correlation with histopathology. Stroke 2000;31:946-954.
2. Kidwell CS, et al. Thrombolytic reversal of acute human cerebral ischemic injury shown by diffusion/perfusion magnetic resonance imaging.Ann Neurol 2000;47:462-469.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content