Early major ischemic changes on CT should not preclude use of tPA. In 1997, Prof von Kummer and his colleagues in ECASS greatly advanced stroke neurology by showing us the importance of early ischemic changes (EIC) on computed tomography.1 They devised the so-called one third rule and taught us to estimate whether EIC subsumed more or less than one third of the territory of the middle cerebral artery (MCA). In the NINDS study, we missed this important finding. In our trial, we focused on treating half our patients within 90 minutes, and the remainder within 3 hours; there was only time to review the scans for hemorrhage, perform a history and physical, and obtain informed consent prior to mixing up the drug.2 Following the regulatory approval of intravenous (IV) thrombolysis in the United States, physicians have tried to implement CT reading and to learn to read EIC and use the one third rule.1 Although FDA approval was based primarily on NINDS, the approval language and package insert incorporated the ECASS warning about EIC and the one third rule. Listening to neurologists and emergency medicine physicians debate the importance of EIC, I would guess that no single aspect of hyperacute stroke therapy has produced more physician angst, and excluded more patients from IV thrombolysis, unless it might be the time window. Yet this angst is ultimately needless, and these patients could be treated, for 3 very simple reasons.
First, von Kummer and I agree on some simple definitions of terms that others seem unwilling to accept; once the terminology is clarified, much of the debate evaporates. The term “subtle EIC” should be confined to the very indistinct signs that can be irreproducible until they are mastered: obscuration of the insular ribbon, blurring of the gray-white matter margin, and a relative paucity of sulci over the affected hemisphere. The signs are hard to teach, and hard to read unless clinical information is provided.3 These signs may reflect mild, reversible tissue edema, but this is not certain. Von Kummer teaches that these findings indicate increased water content, likely early ischemic edema, and that such findings may precede more ominous findings of irreversible tissue damage.4 In contrast to the subtle signs, obvious, frank hypodensity— manifest as attenuated signal on CT—reflects more severe injury. Von Kummer and I concur: here lies severe tissue injury, and there is greater likelihood of hemorrhage if this region is reperfused. Such hypodensity is not subtle: if large, do not treat; if small, recheck the onset time; if <3 hours, consider treating, but note that even some of the NINDS investigators hesitate to treat patients with frank hypodensity. The issue, to me, is that the onset time is likely >3 hours, and the patient has obvious evidence of severe tissue injury. We do not yet know for sure, but reperfusion may be hazardous.
Second, subtle EIC has never been shown to correlate with any differential outcomes, either death/disability or hemorrhage. In the original ECASS articles, the outcomes in patients with subtle EIC greater than one third were not statistically significantly different from those with EIC less than one third or those with no EIC. Although von Kummer has been rigorous and consistent in this regard, he is frequently misquoted. The mantra that subtle EIC greater than one third predicts hemorrhage or death has been drummed into review papers, national guidelines, and book chapters. We examined the NINDS data set to try to prove it, but failed.5 Greater than one third is not true: it was never significant in ECASS and there is not even a trend in NINDS. As confirmation, an analysis showing no relationship to outcome in the Australian Streptokinase Trial data set was reported recently.6 There is no relationship between subtle EIC greater than one third of the MCA territory and outcome. I would not go so far as to say that subtle findings indicate an even greater likelihood of response to tPA, but others have, and the data support this proposition as a testable hypothesis.
Finally, we in the stroke community agree on much more than we acknowledge, but the wider medical world observes and frets over our public squabbling. Consider that intravenous tPA yields an absolute risk reduction for stroke disability of 11%, compared with absolute risk reduction for cardiac disability of 2% in TIMI and 1% mortality in GUSTO. We have an effective therapy that is seldom used because—among other problems—emergency medicine physicians witness us criticizing each other more than we support a proven therapy. In proposing alternatives or improvements to IV tPA, we all feel compelled to indict it as unwieldy or inadequate. I believe that first we must use it, inadequate and difficult as IV therapy is, and then build improvements on a successful foundation. It is time to publicize the agreement we all take for granted: IV thrombolytic therapy is the first choice for patients who present within 3 hours, using the NINDS inclusion and exclusion criteria. Imaging modalities—such as ECASS CT criteria, angiography, or MRI perfusion/diffusion mismatch—may be useful in selecting patients beyond the 3-hour window, but all that is experimental. Our goal should be to boost the frequency of IV therapy in the 3-hour window from the current 2% of all strokes to over 12%. This goal is eminently feasible and has been achieved in communities with a unified medical commitment to acute IV therapy. After that, we may begin trials designed to extend the therapeutic window using new imaging modalities. The first step is to identify our profound areas of extensive agreement, isolating our disagreements for the future.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Stroke Association.
Grotta J, Chiu D, Lu M, Patel S, Levine S, Tilley B, Brott T, Haley EC Jr, Lyden P, Kothari R, et al. Agreement and variability in the interpretation of early CT changes in stroke patients qualifying for intravenous rtPA therapy. Stroke. 1999; 30: 1528–1233.
Patel SC, Levine SR, Tilley BC, Grotta JC, Lu M, Frankel M, Haley EC, Brott TG, Broderick JP, Horowitz S, et al, for the National Institute of Neurological Disorders, and Stroke rt-PA Stroke Study Group. Lack of clinical significance of early ischemic changes on computed tomography in acute stroke. JAMA. 2001; 286: 2830–2838.
Gilligan A, Markus R, Read S, Srikanth V, Hirano T, Fitt G, Arends M, Chambers BR, Davis SM, Donnan GA, Australian Streptokinase Trial Investigators. Baseline blood pressure but not early computed tomography changes predicts major hemorrhage after streptokinase in acute ischemic stroke. Stroke. 2002; 33: 2236–2242.