Radiat Med (2007) 25:553–557 DOI 10.1007/s11604-007-0180-0 CASE REPORT Reversible widespread ischemia after early reperfusion detected by initial diffusion-weighted magnetic resonance imaging Yasuyuki Ezaki · K. Nakashima · K. Kamada M. Kaminogo Received: May 23, 2007 / Accepted: July 25, 2007 © Japan Radiological Society 2007 Abstract We herein report the time-related dynamic changes according to the diffusion-weighted image (DWI) findings after a cerebral ischemic attack in a 63year-old woman. At 2 weeks after undergoing a lower limb amputation due to diabetic atherosclerosis, she experienced a sudden loss of consciousness and right hemiparesis. Magnetic resonance image revealed left frontal and parietal areas with an increased signal on the DWI, and magnetic resonance angiography (MRA) showed the left middle cerebral artery to be occluded at the superior M2 branch. However, on the next day the lesion on DWI, except for the gray matter, was observed to have almost completely resolved, and MRA showed complete recanalization of the left superior M2 branch with diminished clinical symptoms. Although a few cases of reversible DWI-identified lesions have been described in the literature, the occurrence of large, reversed DWI lesions in the middle cerebral artery territory with severe apparent diffusion coefficient decreases, as seen in our case, are exceedingly rare. Key words Diffusion weighted image · Infarction · Reversible Y. Ezaki1 (*) · K. Kamada · M. Kaminogo Department of Neurosurgery, Sasebo City General Hospital, Sasebo, Japan K. Nakashima Department of Radiology, Sasebo City General Hospital, Sasebo, Japan Present address: Spinal Disorders Center, Fujieda Heisei Memorial Hospital, 123-1 Mizukami, Fujieda 426-8662, Japan Tel. +81-54-643-1230; Fax +81-54-643-1289 e-mail: ezaki-nsu@umin.ac.jp 1 Introduction Although it is widely accepted that diffusion-weighted imaging (DWI) is useful for identifying the ischemic lesion prone to infarction, reversible ischemic lesions with moderately decreased apparent diffusion coefficient (ADC) values have recently been reported.1 However, few reports describe reversed lesions of a large territory with severely decreased ADC values. In this article, we describe a patient who exhibited large hyperintensities on DWI at the onset but which had almost vanished by the next day. Case report A 63-year-old female patient was referred to our orthopedic surgery service for the amputation of diabetic foot ulceration. At 2 weeks after lower limb amputation surgery, she was found unresponsive in bed at 2:00 p.m. Although she had been alert until noon, we could not precisely confirm the onset time. A neurological examination revealed severe right hemiparesis with aphasia [NIH Stroke Scale (NIHSS) score 25]. Electrocardiography indicated atrial fibrillation. Her blood glucose level at onset was 70 mg/dl, which immediately recovered to 150 mg/dl after a glucose injection; the low level was thought to be due to a hypoglycemic attack. This procedure did not, however, restore her consciousness. Computed tomography (CT) obtained at 2:45 p.m. revealed neither infarction, including early CT signs, nor hemorrhage (Fig. 1A,B). The magnetic resonance imaging (MRI) study (1.5-T MRI unit; Siemens, Erlangen, Germany) at 3:30 p.m. demonstrated hyperintensity on DWI (b value 1000 s/mm2), and the ADC had 554 Radiat Med (2007) 25:553–557 A C E G B D F H I Fig. 1. Computed tomography (CT) scans obtained at the onset did not reveal infarction, including early CT signs (A, B). Magnetic resonance imaging (MRI) scans, obtained within a few hours after the onset of symptoms, shows abnormalities involving the left frontal and parietal regions, which were hyperintense on diffusionweighted imaging (DWI) (C, D) and hypointense on the apparent diffusion coefficient (ADC) map (E, F). FLAIR images show no hyperintensity in a similar region; however, arterial hyperintensity is observed in the left sylvian fissure (G, H). Magnetic resonance angiography (MRA) showed occlusion (arrow) of the distal middle cerebral artery (MCA) at the superior M2 branch (I) decreased on the ADC map in the left frontal and parietal regions (Fig. 1C–H). MRA showed an occlusion of the distal middle cerebral artery (MCA) at the superior M2 branch (Fig. 1I). We could not use an intravenous tissue plasminogen activator (t-PA) because we could not precisely identify the time of onset. Instead, she received heparin therapy. The next day, the patient remained confused, but the right hemiparesis had gradually diminished. DWI demonstrated resolution of the initial hyperintense signal changes in a large part of the white matter (WM) lesion; hyperintense signals were observed in the gray matter (GM) and in a small part of the WM of the left parietal lobe (Fig. 2A–F). Complete recanalization of the occlusion of the superior M2 branch was noted on the MRA (Fig. 2G). Over a course of 2–3 weeks, the patient’s symptoms showed marked alleviation, but her verbal communication was limited to only several words (NIHSS score 8). Hyperintense findings on the DWI and fluid-attenuated inversion-recovery (FLAIR) images were lessened but remained in the GM and in part of the WM on day 30 (Fig. 3). Radiat Med (2007) 25:553–557 Fig. 2. Hyperintensities on DWI (A, B) and hypointensities of the ADC map (C,D) in the left frontal and parietal lobe diminish on the second day and persist only in the gray matter and the area of white matter in the left parietal lobe. Fluid-attenuated inversion recovery (FLAIR) images (E, F) show a slight increase in the signal intensity of the gray matter in the frontal and parietal lobe and resolution of an abnormally arterial high intensity in the left sylvian fissure. Complete recanalization (arrows) of the occlusion of the superior M2 branch was noted on MRA (G) 555 A C E B D F G We subdivided the initial hyperintense lesions on the DWI into three regions, and the ADC value and the hemispheric ratio of these values (% of ADC, lesion/corresponding contralateral area × 100) were calculated for: (A) the irreversible diffusion abnormality in the WM; (B) the reversible diffusion abnormality in the WM; (C) the reversible diffusion abnormality in the GM. Ten uniform regions of interest were manually drawn on ADC maps for each of the three regions, and the average ADCs were thus determined. The ADC of every region at onset were significantly lower in the lesion (region A: 0.44 × 10−3 mm2/s; B: 0.41 × 10−3 mm2/s; C: 0.48 × 10−3 mm2/s) than they were in the corresponding right area. The percent of ADC for each region was 49.6% (region A), 48.0% (region B), 64.8% (region C). Discussion DWI is the most suitable method for imaging acute stroke patients because of its ability to depict ischemic lesions early. Initially, the lesion visualized by DWI in our patient was thought to represent irreversibly infarcted tissue. However, a reversible diffusion abnormality has recently been reported to be well recognized now in association with acute human stroke.1 As a result, ADC reduction is not equivalent to irreversible ischemia. The reversibility of DWI has potentially important clinical implications in the monitoring of stroke treatment. However, in past reports, DWI reversal was observed only in patients who had modestly decreased ADC values or a small infarct lesion. 556 Radiat Med (2007) 25:553–557 Fig. 3. Hyperintensity found on DWI (A, B) and FLAIR images (C, D) was diminished but remained in the gray matter and in part of the white matter on day 30 There is evidence that the ADC thresholds for infarction and salvage exist in animals and in humans.2,3 Dardzinski et al. suggested that ADC values < 0.55 × 10−3 mm2/s at 2 h following ischemia strongly predict an infarction in a rat permanent occlusion stroke model. Oppenheim et al. reviewed 48 patients not treated by thrombolytics who had a posterior wall infarct seen by DWI within 6 h after onset; their analysis indicated an ADC threshold value of 0.748 × 10−3 mm2/s and a percent of ADC threshold value of 91%. In our case, the ADC value and percent of ADC of the reversible DWIidentified lesion in the WM were 0.412 × 10−3 mm2/s and 48%, respectively; these values are significantly lower than those of past reports. As a result, severe initial ADC decreases do not uniformly predict infarction. The final fate of the acute DWI-identified lesions cannot, however, be predicted from the ADC in these lesions alone. In our case, the follow-up MRI 24 h after onset showed spontaneous reperfusion. For lesions with severe decreases in ADC, such as those seen in our case, early reperfusion seems to be the main factor associated with DWI-evidenced reversal.4,5 The absolute ADC threshold for infarction could not be identified because it is likely to change over time. Loh et al. reported on patients who presented with acute cortical stroke and who were treated with tissue plasminogen activator. They noted A C B D that a shorter time interval to thrombolytic therapy was one of the factors associated with reversal on DWI.5 Our other main finding in this case was that the threshold of ADC values differed between the GM and WM—in our case, the significantly lower ADC values in the DWI lesion of the GM and WM (regions A, B, and C) at onset. However, although part of the WM lesion (region B) was reversed the next day, all DWI lesions in the GM (region C) were not reversed. Based on past reports, there is evidence that GM is more vulnerable to ischemia than WM.6,7 We think that the difference in the ADC thresholds between the GM and WM is probably attributable to two factors.8,9 First, there are well documented differences in the neurochemical response to ischemia of the WM in comparison to the GM. In the GM, the classic cascade commences with energy failure and glutamate release; it is then followed by a voltagegated calcium channel impairment, potassium and calcium influx, and cell death. However, the WM largely lacks glutamate receptors, and the ischemic response is therefore governed by Na/K channel inhibition and a postulated adenosine-mediated autoprotective feedback loop. Second, although the ADC value is used for a quantitative evaluation of edema severity, the localizations of edema fluid is different in the GM and WM.10 Radiat Med (2007) 25:553–557 Cellular swelling in the GM takes place mainly in the astrocytes, whereas in the WM we observed hydropic swelling of the oligodendroglial cell body as well as of the astrocytes. Intracellular water accumulation in the axon also occurred, and periaxonal space enlargement was seen in many myelinated fibers. Because it seemed that hyperintensity on DWI described the different cytotoxic edema in the GM and WM, only ADC values could not account for the different vulnerability to ischemia between the GM and the WM. In our case, the blood glucose level at onset was low. ADC reductions are not unique to ischemia and occur in various pathological conditions, such as seizures, spreading depression, excitotoxic cerebral injuries, and hypoglycemia.11,12 Previous reports regarding prolonged hypoglycemia demonstrated reversible MRI changes on DWI. However, for the following reasons we concluded that the reversibility on DWI in our case was different from the normalization that occurs in hypoglycemia. First, in our case, hypoglycemia was mild, and intravenous glucose supplementation did not improve her clinical condition. Second, we confirmed the MCA occlusion on MRA, and the hyperintensity on DWI was in the territory of the middle cerebral artery. In previous reports, MR signal changes in hypoglycemia were usually localized in the basal ganglia and the temporal and occipital cortex bilaterally. Conclusion Although the conclusions obtained from our case cannot be easily transferred to all stroke patients, an earlier appearance of reperfusion may increase the probability of a DWI-evidenced reversal even if a severe decrease in the ADC value is observed. Although evaluation by MRI is being added to the acute stroke treatment guidelines of t-PA, more sufficient consideration still seems to be required. 557 References 1. Kidwell CS, Saver JL, Mattiello J, Starkman S, Vinuela F, Duckwiler G. Thrombolytic reversal of acute human cerebral ischemic injury shown by diffusion/perfusion magnetic resonance imaging. Ann Neurol 2000;47:462–9. 2. Dardzinski BJ, Sotak CH, Fisher M, Hasegawa Y, Li L, Minematsu K. Apparent diffusion coefficient mapping of experimental focal cerebral ischemia using diffusion-weighted echo-planar imaging. Magn Reson Med 1993;30:318–25. 3. Oppenheim C, Grandin C, Samson Y, Smith A, Duprez T, Marsault C. Is there an apparent diffusion coefficient threshold in predicting tissue viability in hyperacute stroke? Stroke 2001;32:2486–91. 4. Fiehler J, Knudsen K, Kucinski T, Kidwell CS, Alger JR, Thomalla G. Predictors of apparent diffusion coefficient normalization in stroke patients. Stroke 2004;35:514–9. 5. Loh PS, Butcher KS, Parsons MW, MacGregor L, Desmond PM, Tress BM. Apparent diffusion coefficient thresholds do not predict the response to acute stroke thrombolysis. Stroke 2005;36:2626–31. 6. Arakawa S, Wright PM, Koga M, Phan TG, Reutens DC, Lim I, et al. Ischemic thresholds for gray and white matter: a diffusion and perfusion magnetic resonance study. Stroke 2006;37:1211–6. 7. Bristow MS, Simon JE, Brown RA, Eliasziw M, Hill MD, Coutts SB, et al. MR perfusion and diffusion in acute ischemic stroke: human gray and white matter have different thresholds for infarction. J Cereb Blood Flow Metab 2005;25:1280–7. 8. Dohmen C, Kumura E, Rosner G, Heiss WD, Graf R. Adenosine in relation to calcium homeostasis: comparison between gray and white matter ischemia. J Cereb Blood Flow Metab 2001;21:503–10. 9. Stys PK, Ransom BR, Waxman SG, Davis PK. Role of extracellular calcium in anoxic injury of mammalian central white matter. Proc Natl Acad Sci U S A 1990;87:4212–6. 10. Kuroiwa T, Nagaoka T, Ueki M, Yamada I, Miyasaka N, Akimoto H. Different apparent diffusion coefficient: water content correlations of gray and white matter during early ischemia. Stroke 1998;29:859–65. 11. Maekawa S, Aibiki M, Kikuchi K, Kikuchi S, Umakoshi K. Time related changes in reversible MRI findings after prolonged hypoglycemia. Clin Neurol Neurosurg 2006;108: 511–3. 12. Böttcher J, Kunze A, Kurrat C, Schmidt P, Hagemann G, Witte OW, et al. Localized reversible reduction of apparent diffusion coefficient in transient hypoglycemia-induced hemiparesis. Stroke 2005;36:e20–2.