Middle Cerebral Artery btrokes Causing Homonymous Hefianopia: Positron Emission Tomography Motohiro Kiyosawa, MD,*t Thomas M. Bosley, MD,*$ Michael Kushner, MD," Dara Jamieson, MD,* Abass Alavi, MD,f and Martin Reivich, MD* Eight patients were evaluated with "F-fluorodeoxyglucose positron emission tomography between 3 and 30 days after isolated stroke involving the middle cerebral artery territory that caused homonymous hemianopia. Diffuse hypometabolism was present throughout the damaged cerebral hemisphere, even in cortical areas not obviously ischemic by clinical examination or neuroimaging. Glucose metabolism in primary and association visual cortex of the damaged hemisphere was decreased by more than 47% ( p < 0.01). Metabolism in the undamaged hemisphere was less profoundly affected, but significant decrements were found in calcarine (40%;p < 0.01) and lateral occipital cortex (35%;
p < 0.05). Kiyosawa M, Bosley TM, Kushner M, Jamieson D, Alavi A, Reivich M. Middle cerebral artery strokes causing homonymous hemianopia: positron emission tomography. Ann Neurol 1990;28:180-183
Previous reports of the cortical metabolic effects of strokes causing homonymous hemianopia in humans have shown that ischemia in the calcarine cortex results in reduced glucose metabolism in areas of cortex appropriate for visuai field loss, and that ischemia optic radiation typically produces more modest hypometabolism in the calcarine region than does direct cortical ischemia El, 21. These reports involved small numbers of patients studied metabolically at variable times after o n e or more episodes of cerebral ischemia. Analysis of metabolic data focused on side-to-side asymmetries in visual cortex metabolism (rather than absolute meta-
From the "Cerebrovascular Research Center and ?Department of Nuclear Medicine, University of Pennsylvania, and the $NeuroOphthalmology Service, Wills Eye Hospital, Philadelphia, PA. Received Aug 3, 1989,and in revised form Feb 22, 1990. Accepted for publication Feb 26, 1990. Address correspondence ro Dr Bosley, Neuro-Ophthalmology Service, Wills Eye Hospital, 9th and Walnut Streets, Philadelphia, PA 19107.
bolic rates) and did not address metabolism in other brain areas. T h e afferent visual system offers an ideal opportunity to analyze metabolic changes in cortex denervated by damage to subcortical white matter tracts. T h e purpose of this study was to evaluate more carefully the subacute effects on visual cortex metabolism of isolated stroke that involves the middle cerebral artery (MCA) territory, causing homonymous hemianopia d u e to damage to the optic radiations.
Methods Patient Selection Records were reviewed of all patients with cerebral ischemia who had studies of local cerebral glucose metabolism (1CMRglu) performed between 1981 and 1988 in the Cerebrovascular Research Center of the University of Pennsylvania. From this group were identified every patient who had: (1) an isolated stroke involving the MCA territory diagnosed by neurological symptoms, physical examination, and computed tomography (CT) andor magnetic resonance imaging (MRI); (2) good visual acuity bilaterally, with a dense homonymous hemianopia contralateral to the side of the stroke, as determined by confrontation visual field or Goldman kinetic perimetry performed as part of a complete neurological examination at the time of ICMRglu study; (3) a ICMRglu study performed between 3 and 30 days after ictus (subacute phase); (4) no evidence by neuroimaging of direct damage to visual cortex in either hemisphere (decreased attenuation or contrast enhancement o n CT; bright signal on MRI T2-weighted images); and ( 5 ) no evidence by history, examination, or neuroimaging of a second cerebral ischemic lesion or of other neurological or ophthalmological disease. Table 1 details clinical data on the 8 patients (7 men and 1 woman aged 5 5 to 78 years; mean age, 65 years) who met these criteria. All 8 patients were alert and cooperative at the time of the ICMRglu study.
Positron Emission Tomography All patients had imaging of ICMRglu using 18F-fluorodeoxyglucose and positron emission tomography (PET) between 3 and 30 days after ictus. The technique has been fully described previously 131. Quantitative metabolic data were derived using the operational equation of Sokoloff and colleagues [41. The measured value of the lumped constant (0.542) was used in these calculations {5}. Images of local cerebral glucose consumption were obtained using a modified version of the PETT V {b].The image characteristics of this system, the scanning protocol, and the reconstruction procedures and system calibration have been described previously {7). The procedures used to generate the radiopharmaceuticals employed in this study have also been described previously [S]. Metabolic information from stroke patients was compared to positron emission tomography (PET) data from 11 normal volunteers of similar age and sex who were studied in the same fashion and who had no neurological or ophthalmolog-
180 Copyright 0 1990 by the American Neurological Association
Table I . Clinical Data Patient No.
Age (yr)/Sex
5 51M 631M 6OIM 631M 7 8lM 6liM 63lM 74lF
Days to PET
Visual Field
Infarct on CT”
Neurological Findingsb
4 7 14 30 14
RHH RHH RHH RHH LHH LSQ RHH RHH
Fr, Te Th, IC Te, BG Normal Fr, Par Post Par Par Fr, Par
Hemiparesis, alexia, agraphia Hemiparesis, mild aphasia, alexia Hemiparesis, global aphasia Hemiparesis, hypesthesia Hemiparesis Hemineglect Hypesthesia, anomia Hemiplegia, global aphasia
17
15 14
“Infarction on CT was always contralateral to homonymous hemianopia. ‘Neurological abnormalities were ipsilateral to homonymous hemianopia. PET = positron emission tomography; CT = computed tomography; RHH = right homonymous hemianopia; LHH = left homonymous hemianopia; LSQ = left superior quadrantanopsia; Fr = frontal lobe; Te = temporal lobe; Th = thalamus; Par = parietal lobe; IC = internal capsule; BG = basal ganglia; Post = posterior.
ical complaints and normal findings on neurological examinations. All subjects were instructed to look at the ceiling of the dimly lit PET room without a specific target. Ears were not plugged, but ambient noise was minimal.
Dada Analysis Statistical analysis was performed on data from 10 noncon-
tiguous cortical and subcortical regions in each hemisphere, including areas both with and without known visual function {9]. Metabolic information from the damaged cerebral hemisphere of stroke patients was analyzed separately from that of the undamaged hemisphere. For control subjects, metabolic data were selected randomly from the right or left hemisphere, and mean 1CMRglu for each region of interest (ROI) was calculated using these values. Statistical comparison between patients and control subjects was performed by Student’s t test using the Bonferoni correction for repeated tests.
Results Table 2 contains aggregate metabolic data for patients with MCA infarction and for normal control subjects. In the damaged hemisphere of the infarction group all brain regions in the MCA territory had values of lCMRglu that were significantly reduced when compared to control values. Glucose metabolism was also decreased in the posterior calcarine cortex ( - 50%; p 5 O.OOl), anterior calcarine cortex (-47%; p 5 O.OOl), peristriate cortex (-49%; p 5 O.Ol), and lateral occipital cortex ( - 50%; p 5 0.01) of the damaged hemisphere despite the fact that visual cortex was not directly involved in the ischemic process. Glucose metabolism was diffusely decreased in the undamaged cerebral hemisphere, although to a lesser extent than in the damaged hemisphere. Statistically
significant decreases in 1CMRglu were found in motor cortex and the inferior parietal cortex while reductions in other MCA regions did not reach statistical significance. Glucose metabolic rates were also decreased in the posterior calcarine cortex ( - 41 %; p 5 O.Ol), anterior calcarine cortex (-40%; p 5 0.01), peristriate cortex (35%; not significant), and lateral occipital cortex (36%; p 5 0.05) of the undamaged hemisphere. lCMRglu in the cerebellar ROI contralateral to the damaged hemisphere was more affected ( - 3 1 than metabolism in the ipsilateral cerebellar ROI (- 24%) was, although these changes did not reach statistical significance.
s)
Discussion We retrospectively analyzed cerebral metabolic data from patients who received PET scans after isolated MCA strokes causing contralateral homonymous hemianopia, in order to determine the effects of optic radiation ischemia on visual cortex metabolism. Some potential sources of error must be considered. We selected PET studies performed between 3 and 30 days after cerebra1 ischemia because the lumped constant and rate constants used in calculating regional cerebral glucose metabolism may be altered by acute cerebral ischemia [lo]. Ischemia involving the visual cortex itself or the contralateral hemisphere may not have been detected. In addition, hypertension, disseminated atherosclerotic disease, and general health are factors not easily controlled for that could cause widespread cerebral hypometabolism independent of ischemia in the stroke group 1111. Glucose metabolism in the damaged MCA territory was decreased 40% to 60% from control levels, simi-
Brief Communication: Kiyosawa et al: PET of Homonymous Hemianopia
181
Table 2. Regiond Metabolic Data” MCA Infarction (n
=
8)
Regions of Interest
Damaged
Undamaged
Control(n
Frontal eye fields Motor area Auditory association Inferior parietal Lateral occipital Peristriate Anterior calcarine Posterior calcarine Cerebellurn Thalamus Whole brain
1.62(0.77)b 1.66(0.79)b 1.93(0.88)‘ 1.03(0.71)b 1.64(0.65)b 1.79(0.63)b 2.5l(0.75)” 2.14(0.75)d 2.79(0.40) 2.08(0.57)’ 3.41(0.68)
2.20(0.65) 2.36(0.73)’ 2.39(0.78) 1.49(0.67)‘ 2.12(0.64)‘ 2.29(0.83) 2.84(0.78)b 2.53(0.91)b 2.53(1.2 7) 2.49(0.66)
3.78(1.28) 3.64(0.90) 3.59(0.90) 2.71(0.86) 3.28(0.86) 3.51(0.95) 4.74(0.83) 4.27(0.97) 3.66(0.99) 3.65(1.03) 4.98(1.08)
“All values are mean (standard deviation) in mg of plucos&OO “p 5 0.01. ‘p 5 0.05. dp s 0.001. MCA
=
gin
=
7)
of braidmin.
middle cerebral artery.
lar to findings in previous reports of large MCA strokes Cllf and thalamic strokes 111, 12). Calcarine cortex hypometabolism was almost as profound as cortical hypometabolism in the MCA territory and somewhat more profound than would be expected from previous studies {l, 2) involving smaller patient series and different data analysis. Metabolism of the visual association cortex metabolism in the damaged hemisphere was equally affected, perhaps because of decreased stimulation from the denervated calcarine cortex and from cortical areas directly damaged by ischemia. The visual association cortex might also appear hypometabolic on PET scanning because of volume averaging with cortical areas having decreased metabolism located both lateral (MCA territory) and medial (calcarine cortex). Glucose metabolism in the undamaged hemisphere was decreased by 30 to 45% from levels in the control subjects. A major factor in this contralateral hypometabolism may be transcallosal diaschisis due to damaged callosal connections between homologous regions of the hemispheres [ll, 12). Metabolism in the visual cortex of the undamaged hemisphere was also decreased by 35 to 40% despite the presence of normal visual acuity and apparently normal visual function in the intact visual field. Transcallosal denervation may not completely explain this hypometabolism because callosal connections between calcarine cortices are relatively weak in other mammals {13]. Calcarine hypometabolism may be related to “gaiting” of the lateral geniculate relay of visual information after suppression of activity in the association cortex [14}. Volume averaging with the damaged hemisphere might cause
182 Annals of Neurology Vol 28 No 2 August 1990
spurious calcarine hypometabolism but could explain the similar metabolic defect in undamaged visual association cortex. Callosal connections between visual association areas are powerful 1131, and decreased metabolism in visual association of the undamaged hemisphere may be caused by decreased stimulation from both contralateral visual association cortex and ipsilatera1 calcarine cortex. Eye closure in normal individuals caused only a 14% decrease in metabolism of the primary visual cortex and no change in metabolism of the visual association {15]. MCA ischemia involving the optic radiations in this study clearly had effects on visual cortex metabolism that were more profound and widespread than could be explained by simple sensory deprivation. Quantitatively similar metabolic results were found in the visual cortices bilaterally in a patient who had a left homonymous hemianopia due to progressive multifocal leukoencephalopathy involving only the right cerebral hemisphere, as demonstrated by MRI C161. This disease affects only white matter, and the case confirms the metabolic effects of subcortical denervation. Cortical glucose metabolism is clearly not the only determinant of cortical function because hemispheric white matter damage due to both MCA stroke and progressive multifocal leukoencephalopathy caused contralateral cortical hypometabolism without striking clinical deficits. This research was supported by the U.S. Public Health Service program project grant NS-14867-09. Dr Kiyosawa was supported by Fight For Sight Fellowship E-PD-100.
References 1. Bosley TM, Rosenquist AC, Kushner M, et al. Ischemic lesions of the occipital cortex and optic radiations: positron emission tomography. Neurology 1985;35:470-484 2. Kiyosawa M, Mizuno K, HatazawaJ, et al. Metabolic imaging in hemianopsia using positron emission tomography with 18Fdeoxyfluoro-glucose. Am J Ophthalmol 1986;101:310-319 3. Reivich M, Kuhl D, Wolf A, et al. The 18-F-fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ Res 1978;44:127-137 4. Sokoloff L, Reivich M, Kennedy C, et al. The "CO-deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure and normal values in the conscious and anesthetized albino rat. J Neurochem 1977;28:897-916 5. Reivich M, Alavi A, Wolf A, et al. Glucose metabolic rate kinetic model parameter determination in humans: the lumped constant and rate constants for [ 18F)fluorodeoxyglucose and El 1C)deoxyglucose. J Cereb Blood Flow Metab 1985;5:179192 6. Ter-pogossian MM, Mullan NA, Hud JT. Design considerations for a positron emission tomograph (PETT V) for imaging of the brain. J Comput Assist Tomogr 1978;2:539-544 7. Kushner MJ, Alavi A, Reivich M, et al. Contralateral cerebellar hypometabolism following cerebral insult. Ann Neurol 1984; 15:425-434 8. Jones SJ, Robinson GD, McIntyre E. Tandem Van de Graaff accelerator production of positron labeled radiopharmaceuticals for routine clinical use. Int J Appl Radiat Isot 1984;35:721-729 9. Kushner MJ, Rosenquist A, Alavi A, et al. Cerebral metabolism and patterned visual stimulation: a positron emission tomographic study of the human visual cortex. Neurology 1988; 38~89-95 10. Hawkins RA, Phelps ME, Huang SC, Kuhl DE. Effect of ischemia on quantification of local cerebral glucose metabolic rate in man. J Cereb Blood Flow Metab 1981;1:37-51 11. Feeney DM, Baron JC. Diaschisis. Stroke 1986;17:817-830 12. Shishido F, Uemura K, Inugami A, et al. Remote effects in MCA territory ischemic infarction: a study of regional cerebral blood flow and oxygen metabolism using positron computed tomography and 1 5 0 labelled gases. Radiat Med 1987;5:36-41 13. Segraves M, Rosenquist AC. The distribution of the cells of origin of callosal projections in cat visual cortex. J Neurosci 1982;2:1079-1089 14. Heilman KM, Valenstein E, Watson RT. Neglect. In: Asbury AK, McKhann GM, McDonald WL,eds. Diseaes of the nervous system. Philadelphia: W.B. Saunders, 19862356-866 15. Kiyosawa M, Bosley TM, Kushner M, et al. Positron emission tomography study of the effect of eye closure and optic nerve damage on human cerebral glucose metabolism. Am J Ophthalmol 1989;108:147-1 52 16. Kiyosawa M, Bosley TM, Alavi A, et al. Positron emission tomography in a patient with progressive multifocal leukoencephalopathy. Neurology 1988;38: 1864- 1867
Vasoactive Peptide Release in the Extracerebral Circulation of Humans During Migraine Headache P. J. Goadsby, MD, PhD," L. Edvinsson, MD, PhD,? and R. Ekman, MDS The innervation of the cranial vessels by the trigeminal nerve, the trigeminovascular system, has recently been the subject of study in view of its possible role in the mediation of some aspects of migraine. Since stimulation of the trigeminal ganglion in humans leads to facial pain and flushing and associated release of powerful neuropeptide vasodilator substances, their focal release into the extracerebral circulation of humans was determined in patients who had either common or classic migraine. Venous blood was sampled from both the external jugular and the cubital fossa ipsilateral to the side of headache. Plasma levels of neuropeptide Y, vasoactive intestinal polypeptide, substance P, and calcitonin gene-related peptide were determined using sensitive radioimmunoassays for each peptide, and values for the cubital fossa and external jugular and a control population were compared. A substantial elevation of the cdcitonin gene-related peptide level in the external jugular but not the cubital fossa blood was seen in both classic and common migraine. The increase seen in cla5sic migraine was greater than that seen with common migraine. The other peptides measured were unaltered. This finding may have importance in the pathophysiology of migraine. Goadsby PJ, Edvinsson L, Ekman R. Vasoactive peptide release in the extracerebral circulation
of humans during migraine headache. A n n Neurol 1990;28:183-187
In spite of considerable debate concerning the pathophysiology of migraine, there is general agreement, and indeed good clinical evidence {l], that rhe cranial vessels play some role in either the pathogenesis or the expression of the migraine syndrome. Recently, considerable literature has characterized the innervation of these vessels as being with both classic and nonclassic
From the *Department of Neurology, The Prince Henry and Prince of Wales Hospitals, and the School of Medicine, University of New South Wales, Sydney, Australia, and the Departments of TInternal Medicine and # Neurochemistry, University of Lund, Lund, Sweden. Received Jan 16, 1990. Accepted for publication Feb 15, 1990. Address correspondence to Dr Goadsby, Department of Neurology, Clinical Sciences Building, The Prince Henry Hospital, Little Bay, NSW 2036 Australia.
Copyright 0 1990 by the American Neurological Association
183