To the Editor:
Fujioka et al1 reported 4 cases of transient internal carotid artery–middle cerebral artery occlusion; serial MRI revealed hyperintensity on T1-weighted (T1W) imaging in the caudoputamen in all patients and in the cerebral cortex in 2 patients. In the companion article,2 Fujioka et al reproduced the MRI finding in rats by 15-minute middle cerebral artery occlusion but not by 60-minute occlusion. Histological examination revealed that this specific ischemic change on MRI corresponded to selective neuronal death and gliosis with preservation of the macroscopic structure of the brain.
I have previously reported3 10 cases with similar MRI finding; these patients presented not only with sudden hemispheric stroke followed by rapid improvement but also with hemichorea-hemiballism. A biopsy from the hyperintense putamen in one of my patients revealed a fragment of gliotic brain tissue with abundant gemistocytes. It is interesting that in both my studies and those of Fujioka et al similar conclusions were obtained.
First, the MRI finding corresponded with an incomplete infarction. This was demonstrated not only by the relative preservation of the macroscopic structure of the brain in both studies but also by the presence of patchy lesions intermixing with relatively normal brain tissue in the biopsy specimen of my patient.
Second, Fujioka’s study confirmed my hypothesis that the MRI finding was related more to vascular compromise than to petechial hemorrhage or hyperglycemia, as had been proposed in patients with hemichorea-hemiballism.
Third, while Fujioka et al demonstrated the delayed appearance of the ischemic hyperintensity of the T1W MRI, the onset of hemichorea-hemiballism in some of my patients was also delayed. Both findings suggested a progressive course existing in an incomplete infarction.
Some additional findings in my study are worth mentioning.
First, 1H MR spectroscopy on the biopsy specimen of my patient demonstrated an increase in lactic acid and a decrease in creatine and N-acetylaspartate, which suggests the presence of anaerobic glycolysis, energy depletion, and neuronal dysfunction. These findings were consistent with the presence of an ischemic injury.
Second, my study demonstrated that after years of follow-up in 2 patients, T2-weighted MRI revealed slit-shaped cystic lesions in lateral part of the putamina, consistent with the presence of watershed infarction.
Third, some of my patients presented with hemichorea-hemiballism from the onset and without preceding attacks of hemispheric stroke, which suggests that incomplete infarction alone was sufficient to produce the MRI signal change, with or without hemiparesis.
Fourth, in one of my patients the hyperintense lesion extended down to midbrain level, a location presumably remote from the site of vascular compromise. This finding suggested that the MRI signal was related to changes along the striatonigral fibers and not limited to the striatum.
Although some biochemical changes affecting the magnetic field might be responsible for the MRI signal change, in my article,3 I proposed that the hyperintensity on T1W MRI could be due to the presence of abundant gemistocytes locating along the axons and persisting for years. Shortening of T1 relaxation time could result from the protein hydration layer inside the cytoplasm of swollen gemistocytes, as in a case with gemistocytic astrocytoma.4 Gemistocytes are swollen reactive astrocytes that usually appear during acute injury; after that, they gradually shrink in size. However, gemistocytes are also found in some chronic diseases, which suggests the presence of a long-lasting pathological reaction.
It would be interesting to know whether the reactive astrocytes found in the rat striatum by Fujioka et al belong to the type of gemistocyte, and if the appearance and disappearance of this specific type of reactive astrocytosis correlate with the appearance and disappearance of the ischemic hyperintensity of the T1W MRI.
- Copyright © 2000 by American Heart Association
Fujioka M, Taoka T, Hiramatsu KI, Sakaguchi S, Sakaki T. Delayed ischemic hyperintensity on T1-weighted MRI in the caudoputamen and cerebral cortex of humans after spectacular shrinking deficit. Stroke.. 1999;30:1038–1042.
Fujioka M, Taoka T, Matsuo Y, Hiramatsu KI, Sakaki T. Novel brain ischemic change on MRI: delayed ischemic hyperintensity on T1-weighted images and selective neuronal death in the caudoputamen of rats after brief focal ischemia. Stroke.. 1999;30:1043–1046.
Shan DE, Ho DM, Chang C, Pan HC, Teng MM. Hemichorea-hemiballism: an explanation for MR signal changes. AJNR Am J Neuroradiol.. 1998;19:863–870.
Abe K, Hasegawa H, Kobayashi Y, Fujimura H, Yorifuji S, Bitoh S. A gemistocytic astrocytoma demonstrated high intensity on MR images: protein hydration layer. Neuroradiology.. 1990;32:166–167.
We appreciate Dr Shan’s comments regarding our recent articles.R1 R2 As correctly pointed out, the “delayed ischemic hyperintensity (DIH) on T1W MRI” histologically corresponds to the incomplete infarct of selective neuronal death and reactive astrocyte proliferation after mild ischemia.R2
However, we suggest that this shortening of T1 relaxation time results at least partly from induced manganese superoxide dismutase (Mn-SOD) in mitochondria of the reactive astrocyte and may represent long-lasting oxidative stress in the incomplete infarct after mild ischemia.
Hyperintense basal ganglia on T1W MRI has been reported to occur in patients with or after various pathological conditions, including chronic hepatic encephalopathy,R3 long-term total parenteral nutrition,R4 hyperglycemia,R5 post–cardiac arrest encephalopathy,R6 R7 hypoglycemic coma,R8 and mild focal ischemia.R1 R9 This interesting change can lead to choreaballism.R5 R10 The exact mechanism of the T1 hyperintensity in these cases remains controversial. Possible causative factors of this T1 hyperintensity involve the followingsR1 R2 R8 : (1) factors immobilizing water molecules (ie, macromolecular hydration effect or surface relaxation mechanism eg, protein and calcification), (2) lipid, (3) flow-related enhancement,R11 and (4) paramagnetic substance (eg, methemoglobin in hemorrhagic tissue, free radicals, molecular oxygen, melanin, and metals such as iron, manganese, copper).
Therefore, the T1 hyperintensity in brain ischemia tends to be simplistically considered a hemorrhagic transformation. Indeed, we also reported that MRI revealed symmetrical changes suggestive of minor hemorrhages, which CT scans could not detect, in the basal ganglia, thalami, and/or substantia nigra in the patients after cardiac arrest.R6 R7 The lesions appeared hyperintense on both T1W and T2W MRI in the late stage after heart arrest, and then could be considered to be petechial hemorrhages.
However, in patients after hypoglycemia or brief focal brain ischemia (both of which lead to relatively mild energy failure in the brain compared with cardiac arrest), MRI showed persistent hyperintensity/hypointensity on T1W/T2W MRI, respectively, in the basal ganglia, cerebral cortex, hippocampus, and/or substantia nigra from a week after each insult.R1 R8 In the patients after brief hemispheric ischemia,R1 ischemic change of T1 hyperintensity subsided with time. These changes on repeated MRI and CT scans seemed to clearly differ from edema, infarct, hemorrhage and calcification. The T1 hyperintensity seemed to be caused by an unknown common mechanism that was related to neuronal death.
We tried to reproduce the ischemic change of hyperintensity on T1W and hypointensity on T2W MRI in the rat.R2 This MRI change appeared in the rat striatum at day 7 but not day 3 after 15 minutes’ middle cerebral artery occlusion (MCAO). This DIH histologically corresponded to selective neuronal death and glial proliferation without infarct or hemorrhage.
Certainly, there may be a possibility that astrocyte proliferation per se or ultrastructural changes in astrocyte cytoplasm (eg, proliferation of mitochondria, rough endoplasmic reticulum, and vacuoles) shortened the T1 and T2 relaxation times via surface relaxation mechanism.R10 R12 However, glial reactions appear from the early stage after ischemia.R13 R14 In our previous study,R2 abundant GFAP-positive astrocytes existed in the rat striatum 3 days after 15-minute MCAO. These astrocytes had common features of “reactive astrocyte” characterized by hypertrophy with enlarged and extended processes and increases in intermediate filaments.R13 At this time (3 days after ischemia), MRI did not demonstrate T1 hyperintensity in the striatum. Therefore, we think that other factors, such as paramagnetic effect, are strongly related to the DIH rather than surface effect caused by those subtle structural changes of brain tissue.
Since then, we have investigated the chronological changes in the rat striatum from 4 hours to 4 months after 15-min MCAO with regard to MRI, histology, and immunoreactivity to Mn-SOD. (The details, reported at the 25th International Stroke Conference, February 10–12, 2000, in New Orleans, appear in abstract form in the January 2000 issue of Stroke.R15 ) Based on the results, we think that the delayed ischemic hyperintensity on T1W MRI results at least partly from Mn-SOD induction in mitochondria of the reactive astrocytes. This induction of Mn-SOD seems to reflect a long-lasting oxidative stress after mild ischemia.
Fujioka M, Taoka T, Hiramatsu K-I, Sakaguchi S, Sakaki T. Delayed ischemic hyperintensity on T1-weighted MRI in the caudoputamen and cerebral cortex of humans after spectacular shrinking deficit. Stroke.. 1999;30:1038–1042.
Fujioka M, Taoka T, Matsuo Y, Hiramatsu K-I, Sakaki T. Novel brain ischemic change on MRI: delayed ischemic hyperintensity on T1-weighted images and selective neuronal death in the caudoputamen of rats after brief focal ischemia. Stroke.. 1999;30:1043–1046.
Krieger D, Krieger S, Jansen O, Gass P, Theilmann L, Lichtnecker H. Manganese and chronic hepatic encephalopathy. Lancet. 346;1995;270–274.
Mirowitz SA, Westrich TJ, Hirsch JD. Hyperintense basal ganglia on T1-weighted MR images in patients receiving parenteral nutrition. Radiology.. 1991;181:117–120.
Nagai C, Kato T, Katagiri T, Sasaki H. Hyperintense putamen on T1-weighted MR images in a case of chorea with hyperglycemia. AJNR Am J Neuroradiol.. 1995;16:1243–1246.
Fujioka M, Okuchi K, Miyamoto S, Sakaki T, Hiramatsu K-I, Tominaga M, Kamada Y, Iwasaki S. Changes in the basal ganglia and thalamus following reperfusion after complete cerebral ischemia. Neuroradiology.. 1994;36:605–607.
Fujioka M, Sakaki T, Hiramatsu K, Miyamoto S, Iwasaki S, Specific changes in human brain following reperfusion after cardiac arrest. Stroke.. 1994;25:2091–2095.
Fujioka M, Hiramatsu K, Sakaki T, Sakaguchi S, Ishii Y. Specific changes in human brain after hypoglycemic injury. Stroke.. 1997;28:584–587.
Fujioka M, Maeda Y, Okuchi K, Kagoshima T, Taoka T. Secondary change in the substantia nigra induced by incomplete infarct and minor hemorrhage in the basal ganglia due to traumatic middle cerebral arterial dissection. Stroke.. 1999;30:1975–1977.
Shan D-E, Ho DMT, Chang C, Pan H-C, Teng MMH. Hemichorea-hemiballism: an explanation for MR signal changes. AJNR Am J Neuroradiol.. 1998;19:863–870.
Bradley WG, Waluch V. Blood flow: magnetic resonance imaging. Radiology.. 1985;154:443–450.
Inoue E, Hori S, Narumi Y, Fujita M, Kuriyama K, Kadota T, Kuroda C. Portal-systemic encephalopathy: presence of basal ganglia lesions with high signal intensity on MR images. Radiology.. 1991;179:551–555.
Jabs R, Bekar LK, Walz W. Reactive astrogliosis in the injured and postischemic brain. In: Walz W, ed. Cerebral Ischemia: Molecular and Cellular Pathophysiology. Totowa, NJ: Humana Press; 1999:233–249.
Kondo Y. Activated and phagocytic microglia. In: Walz W, ed. Cerebral Ischemia: Molecular and Cellular Pathophysiology. Totowa, NJ: Humana Press; 1999:251–269.
Fujioka M, Taoka T, Matsuo Y, Hiramatsu K-I, Kondo Y, Ogoshi K, Miyasaki A, Sakaki T, Kato K, Siesjö BK. Delayed ischemic hyperintensity on T1 weighted MRI and induced manganese superoxide dismutase in mitochondria after mild focal ischemia: long-lasting oxidative stress hypothesis. Stroke.. 2000;31:341. Abstract P221.