Diffusion-Weighted MRI: A New Diagnostic Technique in Stroke

Sources: Lutsep HL, et al. Clinical utility of diffusion-weighted MRI in the assessment of ischemic stroke. Ann Neurol 1997;41:574-580; Koroshetz WJ, Gonalez G. D-W MRI: An ECG for "brain attack?" (Editorial). Ann Neurol 1997;41:565-566; Baird AE, et al. Enlargement of human cerebral ischemic lesion volumes measured by D-W MRI. Ann Neurol 1997;41:581-589; Zivin JA. Diffusion-weighted MRI for diagnosis and treatment of ischemic stroke (Editorial). Ann Neurol 1997;41:567-568.

The increasing use of thrombolytic therapy in acute ischemic stroke (NINDS rt-PA stroke study group. N Engl J Med 1995;333:1581-1587) has made necessary a technique to identify and localize ischemic brain injury at the time when clinical stroke symptoms are present but the CT scan and conventional MRI are still negative.

Diffusion-weighted (D-W) MRI is a new diagnostic imaging technique that can detect changes in the apparent diffusion coefficient (ADC) of water within minutes after the onset of ischemia in tissue destined for infarction. In animal models, a decrease in ADC occurs within 30 minutes after arterial occlusion; in human studies, D-W MRI has detected such early ischemic changes at the earliest scan times achieved, namely within two hours of the onset of clinical stroke (Gonzalez G, et al. Stroke 1997;28:2).

Two groups, Lutsep et al and Baird et al, have used this technique to identify and characterize developing ischemic lesions in their early phases. Lutsep et al compared D-W MRI with T2W MRI in patients with cerebral ischemia in whom they assessed changes in ADC and T2W images over the course of one year. They obtained the first scan at a mean time of 10.4 days after onset of symptoms. In four patients scanned within eight hours of stroke onset, D-W MRI was positive when T2W MRI was negative. T2W MRI became positive at 8-24 hours after onset. Therefore, D-W MRI is of a clinical value mainly in the first 8-12 hours after stroke when D-W MRI can detect areas with low ADC before their T2W signal becomes abnormal. Koroshetz and Gonzalez, in their accompanying editorial, have proposed that ischemic strokes be staged according to their D-W and T2W MRI characteristics. (See Table.)


Evolution of Ischemic Stroke on MRI

Stage Time After Onset MRI finding

ADC T2W Signal

No injury 0 normal normal

Hyperacute 8-12 hours low normal

Acute 12 hours - 1 week low high

Subacute 1 week + high high

Chronic 1 month + normal high

Baird et al performed 44 D-W and T2W MRIs in 28 patients between two and 53 hours after stroke onset. Acute stroke volumes measured by D-W MRI, as well as regions of hypoperfusion measured in 13 patients by perfusion MRI, were compared with lesion volumes measured by T2W MRI at seven days or later. There was a significant correlation between the acute abnormality on D-W MRI and the final infarct volume seen on T2W MRI, but they were not congruent. In 12 of 28 (43%) patients, the size of the initial lesion increased by 20% or more. In some of the patients whose lesions enlarged, there was evidence of a penumbra on the initial perfusion and D-W MRI, namely the region of hypoperfusion was larger than the region of low ADC. In addition, enlargement of the cerebral ischemic lesion volumes occurred later than expected, at six, 12, or 24 hours or more after onset. These results may mean that ischemic damage may be reversible for longer than the 3-6 hour therapeutic window of opportunity currently accepted by clinicians.


In an editorial commenting on this report, Zivin reminds us that much more basic and clinical work needs to be done before the findings of D-W MRI can be accepted as reliable measurements of the extent of ischemic damage. There is no evidence at present that D-W MRI estimates of stroke severity are any better than the clinical neurological examination. While these statements are true, there is no doubt that the widespread use of D-W MRI will soon provide the correlations required to define the sensitivity and specificity of this "high-tech" MRI.

Clinicians who up to now have had to make therapeutic decisions in patients with acute, CT-negative strokes, on the basis of their clinical findings will welcome the corroborative information that D-W MRI provides. —jjc