Disturbed Diffusion and X-Ray Hypoattenuation in Acute Stroke
To the Editor:
Barber and colleagues1 are the first to study the important question of whether diffusion-weighted MRI (DWI) is superior to CT in predicting irreversible ischemic tissue damage. They presented the MRI and CT findings obtained from 17 patients within the first 6 hours of symptom onset and used a T2-weighted image at 90 days as gold standard for the final infarct size. According to their table, 15 patients had an infarct on the follow-up MRI. One of 2 patients without infarct on the follow-up MRI was later identified as a nonstroke patient (patient 13). This patient, however, had a positive DWI but a negative CT at baseline, whereas the other had a negative DWI and CT at baseline. I cannot follow the authors saying that “hyperintense lesions on DWI consistent with acute ischemia were seen in all 16 patients with a final diagnosis of stroke, giving a sensitivity and positive predictive value for DWI of 100%.” The predictive values for a ischemic lesion on follow-up imaging in this study are presented in the Table⇓.
Based on these data, it is premature to conclude that “DWI is able to identify the presence of early infarction with greater sensitivity than CT.” Moreover, CT was obtained earlier than MRI in 10 patients, with a difference of >3 hours in 2 patients. Although disturbed diffusion is more easily depicted on DWI than slight hypoattenuation on CT, it is an open question which finding is more specific for irreversible tissue damage.2 3
I conclude from this study that we need more data obtained in a manner similar to that used by Barber et al.1 I would like to encourage the authors to continue their efforts in imaging acute stroke patients with these two modalities. We would like to know whether disturbed diffusion and x-ray hypoattenuation show the same pathophysiology, eg, early ischemic edema. We need to know at which time after symptom onset or at what degree these phenomena are specific for irreversible tissue damage. And, finally, we should carefully study patients with stroke who have a negative early DWI or CT. Stroke in these patients may be different compared to patients with positive early imaging. Irreversible tissue damage may not occur in these patients or may be delayed.
- Copyright © 2000 by American Heart Association
Barber P, Darby D, Desmond P, Gerraty R, Yang Q, Li T, Jolley D, Donnan G, Tress B, Davis S. Identification of major ischemic change: diffusion-weighted imaging versus computed tomography. Stroke.. 1999;30:2059–2065.
Kidwell C, Alger J, Di Salle F, Starkman S, Villablanca P, Bentson J, Saver J. Diffusion MRI in patients with transient ischemic attacks. Stroke.. 1999;30:1174–1180.
von Kummer R, Bourquain H, Manelfe C, Bastianello S, Bozzao L, Meier D. Predictive value of early CT in acute ischemic stroke. Stroke.. 1999;30:250. Abstract.
We appreciate the opportunity address the points raised by Prof von Kummer. On page 2061 of our study,R1 we mistakenly identified patient 12 in the table as patient 13 in the text. Thus, the only patient without a final diagnosis of stroke was patient 12. He was included in this analysis because his presentation resulted in initial treatment as a stroke patient. However, normal acute CT and DWI studies prompted a search for an alternative diagnosis, which resulted in a final diagnosis of a brachial plexopathy. This oversight was in the misidentification of the patient in the body of the text, not in the calculation of the sensitivity and positive predictive value for DWI. We apologize for the confusion that this mistake may have caused.
The second issue is that Prof von Kummer has reinterpreted the results using day 90 T2-weighted images as the gold standard for a final diagnosis of stroke. While T2-weighted imaging at day 90 was used to measure final infarct size, a final diagnosis of stroke was determined on the basis of both standard clinical criteria and imaging results, a distinction noted in both the Methods and Results sections. Therefore, in patient 13, an investigator blinded to the clinical data and the results of earlier imaging studies was unable to detect evidence of a relatively small acute DWI lesion on the day 90 T2-weighted images. Previous studies have found that chronic infarct size may be smaller than DWI lesions in the first days following stroke.R2 R3 R4 Possible explanations for this observation have included chronic cerebral atrophy or reversal of the DWI lesions at the margins of the hyperacute lesion.R3 R4 In contrast, the attending physicians arrived at a final diagnosis with full knowledge of a patients’ clinical history and course and the unblinded interpretation of all investigations. These included magnetic resonance perfusion imaging (PI) and subacute MRI studies at day 3, as well as other more conventional poststroke investigations not reported in this study. Thus, hyperintense lesions consistent with ischemia were indeed seen in the acute DWI studies of all 16 patients with a final diagnosis of stroke. Our results and conclusions remain the same.
We agree that the question of whether DWI or CT is more specific in the early identification of irreversible change is open. This and the other important questions raised by Prof von Kummer require more animal or larger human studies. Furthermore, the full potential of DWI to identify individual patients most likely to benefit from acute interventional stroke therapies, either alone or in combination with other MRI sequences such as PI, needs investigation. To this end we are recruiting and imaging acute hemispheric stroke patients with both CT and DWI before treatment with tissue plasminogen activator.
Barber PA, Darby DG, Desmond PM, Gerraty RP, Yang GY, Li KL, Jolley D, Donnan G, Tress BM, Davis SM. Identification of major ischemic change: diffusion-weighted imaging versus computed tomography. Stroke.. 1999;30:2059–2065.
Barber PA, Darby DG, Desmond PM, Yang Q, Gerraty RP, Jolley D, Donnan GA, Tress BM, Davis SM. Prediction of stroke outcome with echoplanar perfusion- and diffusion-weighted magnetic resonance imaging. Neurology.. 1998;51:418–426.
Sorensen AG, Buonanno FS, Gonzalez RG, Schwamm LH, Lev MH, Huang-Hellinger FR, Reese TG, Weisskoff RM, Davis TL, Suwanwela N, Can U, Moreira JA, Copen WA, Look RB, Finkelstein SP, Rosen BR, Koroshetz WJ. Hyperacute stroke: evaluation with combined multisection diffusion-weighted and hemodynamically weighted echo-planar MR imaging. Radiology.. 1996;199:391–401.
Baird AE, Benfield A, Schlaug G, Siewert B, Lovblad K, Edelman R, Warach S. Enlargement of human cerebral ischemic lesion volumes measured by diffusion-weighted magnetic resonance imaging. Ann Neurol.. 1997;41:581–589.