Diffusion-Negative Stroke

Abstract & Commentary

Synopsis: The occurrence of normal diffusion-weighted imaging findings in the setting of impending infarction emphasizes the need for timely perfusion assessment in hyperacute stroke patients, especially if therapeutic intervention is contemplated.

Source: Wang PY, et al. Diffusion-negative stroke: A report of two cases. Am J Neuroradiol 1999;20:1876-1880.

Diffusion-weighted mr imaging has been shown to be useful for the early diagnosis of stroke in humans. In general, it is considered highly sensitive for the detection of early changes associated with hyperacute stroke (< 6 hours since symptom onset), and is felt to be much more sensitive to these early changes than conventional T2-weighted MR images or CT. In this paper, two patients with negative diffusion-weighted studies obtained within four hours of symptom onset who later developed imaging evidence of infarction are presented. These patients represent a subset of a group of 10 patients studied at this institution within 24 hours of symptom onset, three of whom had negative findings on initial diffusion-weighted images (DWI).

The first patient, a 67-year-old man with a history of transient ischemic attack (TIA), atrial fibrillation, and myocardial infarction (MI), presented with Wernicke’s aphasia and mild right arm apraxia. He was clinically diagnosed with acute embolic stroke in the left temporal region, and an initial CT scan was negative. Conventional MR and DWIs were normal. Perfusion imaging was not performed, as the patient refused gadolinium. The patient was treated with intravenous heparin, and a follow-up MR scan obtained four days later showed clear-cut evidence of a new left temporal infarct. The second patient, a 65-year-old man with multiple cardiovascular risk factors, presented with two hours of expressive aphasia, slight left facial droop, and minimal left pronator drift. Findings on initial conventional and diffusion imaging were normal, though perfusion images showed a large region of markedly increased time-to-peak in the right major coronary artery (MCA) distribution. This latter information was not available for clinical decision making, as it required off-line post-processing. The patient deteriorated clinically, and follow-up MR imaging 13 hours later demonstrated normal T2-weighted images and interval development of an abnormality on the DWIs. Further follow-up four days later demonstrated hemorrhagic infarction in the right posterior MCA territory.

Both patients were symptomatic at the time of imaging but had normal findings on T2-weighted and isotropic DWIs, as well as average diffusion coefficient maps. Region of interest measurements showed the diffusion constant to be 4-9% lower in the eventual region of infarction as compared with the contralateral hemisphere, but this difference was not statistically significant, nor could this be visually detected by an observer. Wang and colleagues felt that there were three potential explanations for the lack of diffusion changes in the acute phase in these patients. First, it was considered that cerebral blood flow might have been at an intermediate level below the threshold for neuronal dysfunction but above that for reduced diffusion. A second possibility, applicable only to case 1, was that reperfusion had occurred, restoring the diffusion constant to normal but not preventing eventual delayed infarction. Third, is the possibility that a second ischemic event might have occurred, but this was not felt to be consistent with the clinical courses of these patients. Overall, Wang et al felt that the occurrence of normal DWI findings in the setting of impending infarction emphasizes the need for timely perfusion assessment in hyperacute stroke patients, especially if therapeutic intervention is contemplated.

Comment by Nancy J. Fischbein, MD

In the relatively brief period that DWI has been clinically available, it has revolutionized the evaluation of patients with acute stroke. Lesions that might have previously gone unrecognized on CT or conventional MR images are now obvious as bright regions on DWI. However, we must all recognize that this emerging technology is only partly understood, and may have some limitations in the assessment of patients with hyperacute stroke.

The paper by Wang et al cautions us that a negative finding on a DWI obtained within the first four hours after the onset of symptoms may not indicate that the patient is experiencing merely a TIA rather than a stroke, but simply that for whatever reason there has not been a reduction in diffusion of such a level that we are able to visually appreciate it. Persistence or even worsening of deficits on clinical examination would be expected in ongoing ischemia/infarction, so results, of course, must always be interpreted in a clinical context. In some cases early changes may be subtle, as illustrated in a companion article in the American Journal of Neuroradiology demonstrating two cases of hyperacute ischemic stroke "missed" by DWI but where, in fact, lesions are visible on retrospective review.1 The combination of perfusion MR imaging with diffusion-weighted MR should be a powerful tool in the assessment of the acute stroke patient, but perfusion imaging continues to require significant post-processing, and the most reliable parameters to evaluate (time to peak, mean transit time, cerebral blood volume, etc.) have certainly not been agreed upon in the literature.

An additional caution is that not every entity presenting with acute neurological deficits and apparent reduction in diffusion on DWIs represents cerebral infarction, as a reduction in apparent diffusion coefficients has been demonstrated in other settings such as viral encephalitis,2 Creutzfeldt-Jakob disease,3 and cerebral abscesses.4 Findings on conventional anatomic images and clinical examination must, therefore, always be incorporated into the overall interpretation. Furthermore, normal DWI in patients with strokelike deficits should stimulate a search for a nonischemic cause of symptoms, as some of these patients may prove to have migraine, post-ictal paresis or paralysis, or brain tumors.5


1. Lefkowitz D, et al. Hyperacute ischemic stroke missed by diffusion-weighted imaging. Am J Neuroradiol 1999;20:1871-1875.

2. Tsuchiya K, et al. Diffusion-weighted MR imaging of encephalitis. Am J Roentgenol 1999;173:1097-1099.

3. Bahn MM, Parchi P. Abnormal diffusion-weighted magnetic resonance images in Creutzfeldt-Jakob disease. Arch Neurol 1999;56:577-583.

4. Desprechins B, et al. Use of diffusion-weighted MR imaging in differential diagnosis between intracerebral necrotic tumors and cerebral abscesses. Am J Neuroradiol 1999;20:1252-1257.

5. Ay H, et al. Normal diffusion-weighted MRI during stroke-like deficits [see comments]. Neurology 1999; 52(9):1784-1792.