The Time Concept in Ischemic Stroke: Misleading
To the Editor:
Wang and colleagues studied cerebral blood flow (CBF), the apparent diffusion coefficient (ADC), and brain tissue sodium concentration ([Na+]) in the experimental animal after occlusion of the middle cerebral artery and of both common carotid arteries.1 They observed a steady increase in [Na+] in the most severe ischemic areas with CBF ≤40 mL · 100 g−1 · min−1 and an ADC ≤520 μm2/5. They suggest the use of [Na+] in addition to MR diffusion and perfusion imaging to exactly assess the time of stroke onset, which they regard as essential before initiating thrombolytic treatment.
The authors overlook the fact that the clinical onset of stroke in humans is not identical with the fall of CBF below thresholds that induce edema and irreversible damage. Otherwise, the reversibility of stroke symptoms would be impossible, and the penumbra concept would be invalid. The [Na+] method may allow determination of the time point at which hypoperfusion has reached a critical level in certain areas of ischemic brain tissue. This time point is presumably not identical with the time of stroke onset. The authors did not show that [Na+] steadily increases in areas with less-severe ischemia, which may, however, be responsible for functional impairment. This reflects the obvious weakness of the time concept in ischemic stroke: the clinical symptoms may represent a volume of ischemic brain tissue with functional impairment of viable tissue, already irreversibly damaged tissue, and in most instances a mixture of both. Ischemic functional impairment can last over hours without visible brain tissue damage.2 This explains why about one third of CT scans remained normal in large stroke studies during the first 6 hours of stroke onset.3 4
The proposed [Na+] method, though impractical, appears to be useful in identifying the partition of ischemic tissue with developing edema after very severe ischemia. It is well established that [Na+] highly correlates with tissue water content after arterial occlusion.5 6 There are more practical measures, however, to determine the water content of ischemic brain tissue under clinical conditions: x-ray attenuation directly correlates with the specific gravity of tissue.7 The linear decline of x-ray attenuation could be used to determine the time point of edema onset that requires a period of very severe ischemia.8 Moreover, the relatively simple CT method allows determination of the extent of edema within an arterial territory at risk. It has been shown that this extent is associated with the risk of secondary hemorrhage after treatment with tissue plasminogen activator (tPA).9 I do not agree with the authors that any study could prove that the risk of cerebral hemorrhage increases with time after tPA administration. It is more likely that the extent of very severe ischemia determines this risk.10 The slogan “Time is brain” is useful for discussion with laypersons but misleading. I think it is more correct to say that “perfusion is brain,” and time is an important cofactor. The ticking clock is not needed. It may keep away a promising treatment from patients who really need it.11 The clinical status of the patient tells us that the brain is at risk, and CT (and probably diffusion-weighted MRI) depicts the volume of tissue that is already damaged. Do we really need more information for the decision to treat a patient with thrombolytics, neuroprotective agents, or decompressive surgery?
- Copyright © 2000 by American Heart Association
Wang Y, Hu W, Perez-Trepichio A, Ng T, Furlan A, Majors A, Jones S. Brain-tissue sodium is a ticking clock telling time after arterial occlusion in rat focal cerebral ischemia. Stroke.. 2000;31:1386–1392.
von Kummer R, Holle R, Rosin L, Forsting M, Hacke W. Does arterial recanalization improve outcome in carotid territory stroke? Stroke.. 1995;26:581–587.
Schuir FJ, Hossmann KA. Experimental brain infarcts in cats, II: ischemic brain edema. Stroke.. 1980;11:593–601.
Gotoh O, Asano T, Koide T, Takakura K. Ischemic brain edema following occlusion of the middle cerebral artery in the rat, I: the time courses of the brain water, sodium, and potassium contents and blood-brain-barrier permeability to 125I-albumin. Stroke.. 1985;16:101–109.
Unger E, Littlefield J, Gado M. Water content and water structure in CT and MR signal changes: possible influence in detection of early stroke. AJNR Am J Neuroradiol.. 1988;9:687–691.
von Kummer R, Weber J. Brain and vascular imaging in acute ischemic stroke: the potential of computed tomography. Neurology.. 1997;49:S52–S55.
Larrue V, von Kummer R, del Zoppo G, Bluhmki E. Hemorrhagic transformation in acute ischemic stroke: potential contributing factors in the European Cooperative Acute Stroke Study. Stroke.. 1997;28:957–960.
del Zoppo G, von Kummer R, Hamann G. Ischemic damage of brain microvessels: inherent risks for thrombolytic treatment in stroke. J Neurol Neurochir Psychiatry.. 1998;65:1–9.
Baron JC, von Kummer R, del Zoppo GJ. Treatment of acute stroke: challenging the concept of a rigid and universal time window. Stroke.. 1995;26:2219–2221.
We appreciate the comments of Dr von Kummer concerning our recent publication.R1 His comments concern several issues: (1) the direct application of experimental results to the clinical situation, (2) confusion about what we were actually doing in terms of using [Na+] (determining functional tissue versus determining unsalvageable tissue), (3) that there are more practical ways to measure time (using CT to determine water content), (4) that time after onset is not the best measure of pathological progression in ischemic stroke (perfusion plus time), and (5) that current techniques are satisfactory for stroke diagnosis. In addition to the obvious internal inconsistencies between points 3 and 4, they and the other issues do deserve serious consideration.
First, his comments concern human stroke, and while our experimental laboratory results might have bearing on this situation, this will only be after much additional work. The editorial commentR2 accompanying our work mentions that “considerable work remains to confirm this hypothesis in humans.” We clearly state that “this scheme must be validated in the clinical setting of acute stroke.” Perhaps our attempt to aim an experimental design at a clinical problem was misplaced, but this cannot be determined unless this scheme is attempted in the “real” world of clinical stroke.
Second, an important point that Dr von Kummer seems to have missed is that we chose the most severely ischemic regions without regard to functional status or the extent of marginal ischemia for use in the sodium clock method, as shown in our Figure 2A.R1 The choice of these regions was based on CBF and/or ADC thresholds of 40 mL · 100 g−1 · min−1 and 520 μm2/s, respectively. The ischemic threshold in the rat of 35 to 40 mL · 100 g−1 · min−1 corresponds to 10 to 15 mL · 100 g−1 · min−1 in nonhuman primates.R3 R4 For the sodium clock to work, just one region of severe ischemia and/or ADC decrease large enough to be imaged needs to be consistently present. If this region exists, then so does the pathological progression to irreversibility characterized by the eventual increase in blood-brain barrier permeability and increased intensity in contrast-enhanced CT and T2-weighted MRI. If this region of severe ischemia is present, the clock will tick, whether for the increase of [Na+] or the eventual loss of that tissue to encephalomalacia. This region corresponds to the proposed first side of the therapeutic window that “may already be closed when the patient first presents.”R5 Thus, the basis of the method is to choose an area in which [Na+] is increasing. Perhaps in the future information could be extracted, using [Na+] or the rate of [Na+] increase, from other regions that might be recruited or salvaged, but for now, all we have shown is that in the most severely ischemic regions in experimental ischemia, the rate of increase is consistent and provides a “ticking clock.”
Part of Dr von Kummer’s objections could be based on the possibility that the most severe ischemia does not occur at the onset of clinical symptoms in human stroke, whereas in experimental stroke caused by arterial occlusion it definitely does. If this is true, the sodium clock will not start ticking until a certain threshold of perfusion is exceeded. In this regard, the [Na+] threshold proposed by Thulborn et alR6 might be a more relevant concept that avoids the issue of timing; or, alternatively, time from the point of critical ischemia, not clinical symptoms, could be determined by the sodium clock.
Third, Dr von Kummer proposes using “the linear decline of x-ray attenuation … to determine the time point of edema onset,” even though he states that “‘time is brain’ is a comment for laymen.” Although this method to determine time after onset has been presented,R7 it has not to our knowledge been prospectively validated or related to any other variable. The experimental data supporting this CT measure of edema is briefly presented as an unpublished observation, in what seems to be ex vivo tissue,R7 and several studies seem to challenge the presumption that CT observation of edema is useful in early acute stroke. A study comparing CT and pathological examination of 13 human brains concludes that low-density areas can be caused by factors other than high fluid content and can be obscured by minerals, fat, or blood.R8 A more recent studyR9 indicates that “there is a considerable lack of agreement, even among experienced clinicians, in recognizing and quantifying early CT changes” in the first few hours after stroke onset. These difficulties might be related to the small increase in water content of 2.5% from normal in the most severely ischemic regions, 4 hours after onset, with correspondingly small CT density changes of about 6.5 Hounsfield units.R10 It is useful to note in this discussion that low-attenuation changes on CT scans are not unequivocally an indication of permanent ischemic injury.R11 This method of measuring water content with CT might be complicated by the same factors present in a study by von Kummer et alR12 that finds in “8 patients examined by CT within 180 minutes of the stroke, no low density could be identified, even in retrospect with the knowledge of the findings on follow-up.” Further adding to the questionable effectiveness of CT at detection of edema in early stroke is the acknowledgment that the large body of data supporting its usefulness was obtained in a cohort with 86% of the population studied 3 hours after onset.R13 In this cohort, before 2 hours “the increase in tissue water is too small to cause a visible decrease in x-ray attenuation.”R13 However, this hyperacute period is the crucial time for thrombolysis.
Fourth, we agree that the critical factor in determining the permanence of neurological injury is the relationship between perfusion and time. We also agree that the concept of a sodium clock is a preliminary proposition that requires substantiation. The relationship between accrual of “time” on the sodium clock to “real-time” versus neurological injury has to be determined. However, although the value of the sodium clock as a prognosticator is also undetermined, it may very well prove to be a more useful indicator of reversible versus permanent neurological injury than other methods currently used.
This point addresses the much broader issue of whether time is the appropriate variable to consider, and Dr von Kummer uses our manuscript as a platform to reiterate a point that he has made before. We are aware of Dr von Kummer’s excellent and original work in this regard and acknowledge that he presents a reasonable hypothesis that there are other more sensitive ways to segregate appropriate patients for thrombolysis. However, as of now, it is still a hypothesis.
Dr von Kummer states that the authors “regard [the time of stroke onset] essential before initiating thrombolytic treatment.” Actually, in the United States, this is the standard of care,R14 independent of what the authors, or Dr von Kummer, regard as ideal. The time window of 3 hours has been supported by several recent studies.R15 R16 Indeed, time might not be the perfect arbitrator of the effectiveness of tPA and the occurrence of hemorrhage.R17
We certainly agree that there may very well be individual differences in the pattern of evolution of the ischemic process that interact with vulnerability to thrombolysis. In fact, in published studies,R18 R19 we discuss the relevant clinical conditions that alter the evolution of ischemia. For the foreseeable future, though, we feel it is a most compelling goal to develop methods that extend the opportunity of thrombolysis to as many patients as possible, in particular to those in whom the onset is not known. We feel the methods described here are sufficiently promising that the extensive effort required to test them prospectively in humans is warranted.
The current standard of stroke care in the United States is based on onset time. Although this might be a first faltering step in the evolution of truly rational stroke care, it certainly represents an improvement over the period when time after onset was more or less ignored in stroke therapy.R2 This anti-time argument should really be directed at the design of trials for thrombolytic agents. It is important to think of the future, not the present or past.
Dr von Kummer ascribes to our work that “the risk of cerebral hemorrhage increases with time after tPA administration.” However, we merely cited the results of Clark et al,R16 who clearly document that hemorrhage becomes important in limiting the benefit of intravenous tPA between 3 to 5 hours after onset, whereas before 3 hours, the benefits of tPA outweigh the damage from hemorrhage.R14
We disagree with the statement that “the ticking clock is not needed. It may keep away a promising treatment from patients who really need it.” If the [Na+] increase is taken from the region of maximal ischemia, this region could be at risk for hemorrhage after thrombolysis and recirculation, in contrast to a more highly perfused region where the blood-brain barrier is intact. Alternatively, if this method of determining onset time were used in all possible candidates for thrombolytic therapy, it is possible that some strokes would be found to have occurred earlier than the patients were aware, that a region with severely ischemic tissue had been there longer than the clinical symptoms indicated.
Fifth, we do think that we need “more information for the decision to treat,” including perhaps the sodium clock. It is hard to imagine not trying to explore new techniques just because the present methods, such as CT or diffusion-weighted MRI (DWI), seem satisfactory. Although the proposed method for [Na+] determination in humans using MRI is currently impractical for wide application to ischemic stroke, so was the use CT in 1975R20 and DWI in 1989.R21 If a method proves useful, then it will be used, as has happened with CT, MRI, and DWI.
Wang Y, Hu W, Perez-Trepichio AD, Ng TC, Furlan AJ, Majors AW, Jones SC. Brain tissue sodium is a ticking clock telling time after arterial occlusion in rat focal cerebral ischemia. Stroke. 2000;31:1386–1392.
Haley ECJ. Editorial comment on “Brain tissue sodium is a ticking clock telling time after arterial occlusion in rat focal cerebral ischemia.” Stroke. 2000;31:1392.
von Kummer R, Holle R, Rosin L, Forsting M, Hacke W. Does arterial recanalization improve outcome in carotid territory stroke? Stroke. 1995;26:581–587.
von Kummer R, Weber J. Brain and vascular imaging in acute ischemic stroke: the potential of computed tomography. Neurology. 1997;49(suppl 4):S52–S55.
Grotta JC, Chiu D, Lu M, Patel S, Levine SR, Tilley BC, Brott TG, Haley EC Jr, 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. Agreement and variability in the interpretation of early CT changes in stroke patients qualifying for intravenous rtPA therapy. Stroke. 1999;30:1528–1533.
Unger E, Littlefield J, Gado M. Water content and water structure in CT and MR signal changes: possible influence in detection of early stroke. AJNR Am J Neuroradiol. 1988;9:687–691.
Kim DK, Mayberg MR, Eskridge JM, Newell DW, Winn HR. Reversal of acute ischemic hypodense lesions on computed tomography. J Stroke Cerebrovasc Dis. 1993;3:240–243.
von Kummer R, Allen KL, Holle R, Bozzao L, Bastianello S, Manelfe C, Bluhmki E, Ringleb P, Meier DH, Hacke W. Acute stroke: usefulness of early CT findings before thrombolytic therapy. Radiology. 1997;205:327–333.
Clark WM, Albers GW, Madden KP, Hamilton S, for the Thrombolytic Therapy in Acute Ischemic Stroke Study Investigators. The rtPA (alteplase) 0- to 6-hour acute stroke trial, part A (A0276g): results of a double-blind, placebo-controlled, multicenter study. Stroke. 2000;31:811–816.
Clark WM, Wissman S, Albers GW, Jhamandas JH, Madden KP, Hamilton S. Recombinant tissue-type plasminogen activator (alteplase) for ischemic stroke 3 to 5 hours after symptom onset: the ATLANTIS Study: a randomized controlled trial. Alteplase ThromboLysis for Acute Noninterventional Therapy in Ischemic Stroke. JAMA. 1999;282:2019–2026.