“Self-Fulfilling Prophecy” or Recognition Requires a Concept of Perception
To the Editor:
The NINDS rt-PA Stroke Trial did not prospectively assess quantity and quality of ischemic brain parenchyma as detected by baseline CT.1 In this trial, no CT reading panel graded CT scans on the basis of predefined definitions. Now the trialists hypothesized that there would be disagreement among them about the presence of early CT changes.2 They randomly selected 70 baseline scans from the trial, and 16 of them reread the scans in a 1-day session. Among the investigators was 1 neuroradiologist who served as the “gold standard.” He was able to predict the lesion location at 24 hours in 96% (95% CI 92% to 100%) of the scans based on the information provided by the baseline CT. In comparison, the 16 raters of the NINDS group were much less sensitive (78%) and specific (57%) in detecting any early CT change or in identifying changes involving >33% of the MCA territory. With this poor performance, it is not at all surprising that the agreement beyond chance among these raters was somewhat low and heterogeneous. Moreover, the authors did not design and power their study to detect an effect of the method of film viewing, the effect of image quality, or an effect of the raters’ different experiences and training on agreement. To our surprise, however, they conclude from their data that all these factors do not affect the raters’ agreement.
Can we generalize this experience? Correct interpretation of CT is a problem not only for the experienced stroke specialists of the NINDS rt-PA Stroke Trial Group. Emergency physicians in another study had an error rate in stroke detection by CT twice that of neurologists and radiologists, and only 17% of emergency physicians and 40% of neurologists achieved a 100% sensitivity for identification of intracranial hemorrhage.3 Missing the ischemic edema in its early stage on CT could result from technical limitations or the lack of a perceptual concept.4 Moreover, we do not agree with Dr Grotta’s assertion that the technique is the main problem. At least the group’s neuroradiologist was able to predict ischemic tissue damage from the baseline CT with high specificity, confirming the experience of others.5 6 7 The problem is the feature of findings in acute stroke and a lack of training to interpret these findings. Obviously, even experts in neurology are puzzled when anatomic information gets lost on CT because of gray matter hypoattenuation. The old prejudice that CT remains negative within the first 24 to 48 hours after ischemic stroke is still prevalent in review articles,8 although many studies have reported positive findings even within the first 3 to 6 hours of stroke onset.9 The specificity of positive CT findings for irreversible lesions is still widely ignored. Moreover, the diagnostic significance of a normal CT in a patient with stroke is underestimated by many stroke physicians and explained by a low sensitivity of the method. A normal CT in acute stroke means, however, that the patient will soon recover or the ischemic edema is delayed and appears after 6 hours, which occurs in about one third of the patients.7 This observation is really important for treatment strategies and the discussion about the therapeutic time window.10
Key to recognizing early ischemic changes on CT is understanding the concept of how stroke pathophysiology may appear on CT. Dr Grotta and his 20 coauthors like to remain vague in this regard: “The pathological significance of such findings, in particular hypoattenuation of the x-ray signal, is uncertain, but it probably represents increased tissue water content caused by early cytotoxic edema, which may or may not signal irreversible injury” (page 1529).2 Their uncertainty is reflected by the categories applied in this study: The authors differentiated between “loss of gray-white distinction” and “hypodensity,” although both mean x-ray hypoattenuation due to tissue water uptake. Unfortunately, the authors do not explain how they interpret the decreasing x-ray attenuation after arterial occlusion and do not discuss why CT enabled Dr Patel to predict irreversible ischemic damage with high specificity within 3 hours of stroke onset.2 The authors discuss abnormalities on diffusion-weighted MRI that may be used instead to predict the irreversible damage, ignoring the relatively small database and a recent report on spontaneous reversibility of disturbed diffusion in TIA patients.11
Maybe white spots on MRI can be more easily detected than subtle changes of the gray scale on CT. Agreement among physicians seems not a problem exclusively with regard to radiological findings. We would very much like to know to what extent there is agreement among the NINDS investigators about the clinical categories “large- and small-vessel occlusion” on which they based their important conclusion that rt-PA is beneficial even in patients without large vessel occlusion.
We fully agree with the authors that improved methods of recognizing early CT changes are needed not only for the NINDS rt-PA Stroke Trial Group. These methods should include formal training in reading CT for all physicians dealing with acute stroke and who cannot rely on a 24-hour neuroradiological service. Moreover, our scientific journals should further foster the comprehension of imaging by taking care of optimal image quality and interpretation. It should not happen again that an article about the need for an improvement in CT interpretation is accompanied by a CT printed upside down12 or that this journal publishes a CT with an easily visible ischemic lesion as a normal CT.13
- Copyright © 2000 by American Heart Association
The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333:1581–1587.
Grotta JC, Chiu D, Lu M, Patel S, Levine SR, Tilley BC, Brott TG, Haley EC, Lyden PD, Kothari R, Frankel M, Lewandowski CA, Libman R, Kwiatkowski T, Broderick JP, Marler JR, Corrigan J, Huff S, Mitsias P, Talati S, Tanne D, and the NINDS rt-PA Stroke Trial Study Group. Agreement and variability in the interpretation of early CT changes in stroke patients qualifying for intravenous rtPA therapy. Stroke. 1999;30:1528–1533.
Schriger D, Kalafut M, Starkman S, Krueger M, Saver J. Cranial computed tomography interpretation in acute stroke: physician accuracy in determining eligibility for thrombolytic therapy. JAMA. 1998;279:1293–1297.
Berlin L. Malpractice issues in radiology: perceptual errors. AJR Am J Roentgenol. 1996;167:587–590.
von Kummer R, Nolte PN, Schnittger H, Thron A, Ringelstein EB. Detectability of hemispheric ischemic infarction by computed tomography within 6 hours after stroke. Neuroradiology. 1996;38:31–33.
Moulin T, Cattin F, Crépin-Leblond T, Tatu L, Chavor D, Piotin M, Viel JF, Rumbach L, Bonneville JF. Early CT signs in acute middle cerebral artery infarction: predictive value for subsequent infarct locations and outcome. Neurology. 1996;47:366–375.
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.
Gilman S. Imaging of the brain. N Engl J Med. 1998;338:812–820.
von Kummer R, Patel S. Neuroimaging in acute stroke. J Stroke Cerebrovasc Dis. 1999;8:127–138.
Baron JC, von Kummer R, del Zoppo GJ. Treatment of acute ischemic stroke: challenging the concept of a rigid and universal time-window. Stroke. 1995;26:2219–2221.
Kidwell CS, Alger JR, Di Salle F, Starkman S, Villablanca P, Bentson J, Saver JL. Diffusion MRI in patients with transient ischemic attacks. Stroke. 1999;30:1174–1180.
Bonn D. Better training needed to interpret CT scans. Lancet. 1998;351:1259.
Firlik AD, Kaufmann AM, Wechsler LR, Firlik KS, Fukui MB, Yonas H. Quantitative cerebral blood flow determinations in acute ischemic stroke: relationship to computed tomography and angiography. Stroke. 1997;28:2208–2213.
Our study should not be interpreted as discounting the importance of CT in early stroke diagnosis and the pioneering work of the ECASS group in general and Drs von Kummer and Hacke in particular in drawing attention to early changes of ischemia on CT. However, we stand by our findings that disagreement about the presence and extent of such findings is likely in everyday practice.
Our study was not powered to detect the effect of method of film viewing, image quality, and different training and experience of raters. However, as stated, we found no consistent differences in performance between specialties, nor did we see improved agreement later in our study when newer CT scans had been introduced. Nevertheless, we agree with Hacke and von Kummer that these are important factors in CT interpretation. The variability in these factors in everyday practice, in which CT scans may be interpreted by general radiologists on their home computer using telemetry techniques or experienced neurovascular radiologists viewing the films on the CT console, only emphasizes the fact that variability in interpretation is likely.
We also agree with Hacke and von Kummer that “loss of gray-white distinction” and “hypodensity” are continuums of the same process, namely, hypoattentuation due to tissue water uptake. In fact this is already clearly stated in the paper. Nevertheless, partly as a response to concerns about identifying very subtle changes, clinicians commonly subdivide early CT changes into microcategories, such as “insular ribbon sign,” etc. However, even if all of these are collapsed into a single category of “any change occurring in more than one third of the MCA territory,” only fair agreement was found.
Although we mention MRI in our paper, we did not advocate MRI over CT, since there are even less data evaluating variability in MRI interpretation and correlating early MRI changes with response to therapy.
In conclusion, the ideal test, or how to best interpret existing tests such as CT, has not yet been determined. CT is currently the gold standard, but our data suggest that we need more precision in its interpretation. Most importantly, clinicians should recognize the pitfalls inherent in overreliance on subtle CT changes in acute stroke patients, particularly since at present there are still no data to support exclusion of patients on the basis of early ischemic changes on CT who would otherwise qualify for thrombolytic therapy within 3 hours.