190 Journal ofNeurology, Neurosurgery, and Psychiatry 1992;55:190-194

Association of atrophy of the medial J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.55.3.190 on 1 March 1992. Downloaded from with reduced blood flow in the posterior parietotemporal cortex in patients with a clinical and pathological diagnosis of Alzheimer's disease

K A Jobst, A D Smith, C S Barker, A Wear, E M King, A Smith, P A Anslow, A J Molyneux, B J Shepstone, N Soper, K A Holmes, J R Robinson, R A Hope, C Oppenheimer, K Brockbank, B McDonald

Abstract lateral temporal lobes are related to the degree A combination of medial temporal lobe of cognitive decline,"8 but such findings are atrophy, shown by computed tomo- difficult to reconcile with neuropathological graphy, and reduced blood flow in the studies showing that the most severely affected parietotemporal cortex, shown by single parts of the brain in Alzheimer's disease are in photon emission tomography, was found the medial lobe, mainly the amygdala, the in 86% (44151) of patients with a clinical hippocampal formation and adjacent parahip- diagnosis of senile dementia of the Alz- pocampal gyrus.>'3 Atrophy of these areas can heimer type (SDAT). The same combina- be revealed in life by CT studies in which the tion ofchanges was found in four out of 10 scan angle is adjusted to give a clearer view of patients with other clinical types of the medial temporal lobe, and such studies in dementia and in two out of 18 with no patients with a clinical diagnosis ofAlzheimer's evidence of cognitive deficit. Of the 12 disease have, indeed, shown severe damage to patients who died, 10 fulfilled histopatho- structures in the medial temporal lobe.'""6 logical criteria for Alzheimer's disease, However, atrophy of the medial temporal lobe nine ofthem having a clinical diagnosis of is also found in other conditions-notably, SDAT, and one a clinical diagnosis of epilepsy and hypoxia,'7 schizophrenia,'8 and multi-infarct dementia. All 10 patients amnesia'9-and so is not unique to dementia of Radcliffe Infirmary, with histopathologically diagnosed Alz- the Alzheimer's type. Oxford OX2 6HE University heimer's disease had shown a combina- In none of the CT and SPET studies des- Department of tion ofhippocampal atrophy and reduced cribed above were the two procedures carried Clinical Neurology parietotemporal blood flow in life. In 10 out in the same patients. The purpose of our K A Jobst A Wear patients (nine with SDAT) out of 12 in report is to describe CT studies ofthe temporal E M King whom the hippocampal atrophy was lobe and SPET studies of perfusion of the A Smith more noticeable on one side of the brain cortex in the same group of patients. We University than on the other the parietotemporal wished to see whether there was any relation http://jnnp.bmj.com/ Department of perfusion deficit was also asymmetrical, Pharmacology between atrophy of the medial temporal lobe A D Smith being greater on the side showing more and a perfusion deficit in the parietotemporal Department of hippocampal atrophy. These results sug- cortex and, if so, whether such a relation was Neuroradiology gest that the combination of atrophy of associated with Alzheimer's disease defined C S Barker the hippocampal formation and reduced clinically and, when possible, by histopatho- P A Anslow A J Molyneux bloodflowin the parietotemporal regionis logical criteria. Preliminary findings have been

a feature of dementia of the Alzheimer on September 24, 2021 by guest. Protected copyright. Department of reported.20 Neuropathology type and that the functional change in the B McDonald parietotemporal region might be related University to the loss ofthe projection neurons in the Patients and methods Department of parahippocampal gyrus that innervate Patients were recruited, after having been Nuclear Medicine, John Radcliffe this region of the neocortex. given a full explanation of the study protocol, Hospital, Oxford from general practitioners and hospital based B J Shepstone services, with some subjects entering them- N Soper CT has shown that many patients with a selves. Ethical approval for the study was K A Holmes clinical of senile dementia of the Littlemore Hospital, diagnosis obtained from both the Central Oxford Oxford Alzheimer type (SDAT) display more Research and the Psychiatric Sector Research J R Robinson appreciable cerebral atrophy than expected for ethics committees. Informed consent was ob- K Brockbank their age.' Functional changes in the brains of tained from those without cognitive deficit, and University demented patients have been found by meth- for those with a significant deficit it was given by Department of Psychiatry, ods that show cerebral blood flow, or uptake of a close relative. Subjects were referred with Warneford Hospital, glucose, oxygen or radioligands: the most varying degrees of mental deterioration and Oxford frequent finding is a functional deficit in the memory loss, many for detailed screening with R A Hope posterior parietal and temporal lobes ofthe cor- a view to making a diagnosis. Diagnoses were C Oppenheimer tex.2" Single photon emission tomography made according to DSM-III-R criteria. Drug Received 11 March 1991 and in revised form 10 June 1991. (SPET) has recently been used to show that the treatment remained unchanged during screen- Accepted 18 July 1991 functional deficits in the posterior parietal and ing. Association ofatrophy of the medial temporal lobe with reduced bloodflow in the posterior parietotemporal cortex 191

Figure CT in the SPET studies were carried out with an J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.55.3.190 on 1 March 1992. Downloaded from temporal lobe orientation intravenous injection of 500 MBq hexamethyl- of a patient who was not propyleneamine oxime (HMPAO) labelled clinically demented (top) as a and a patient with a with technetium-99m, which was given clinical diagnosis of senile bolus and prepared according to the manufac- dementia of the Alzheimer turer's instructions (Ceretec, hexametazime, type (SDA T) (bottom). Sixty four SPET The arrows indicate the Amersham Intemational). thinnest part of the medial images were obtained over 360°, each having an temporal lobe adjacent to acquisition time of 20 s and a resolution of the brain stem where a x 64 Axial and coronal slices were linear measurement was 64 pixels. taken. The left of the brain reconstructed by filtered back projection and is shown on the right, were two pixels in width (13 mm). SPET scans where there is also a scale were assessed blind to the clinical diagnosis on in cm. The non-demented patient was aged 80 at the a scale of zero to three, in which 0 represented time of the scan, while the no perfusion deficit, 1 a minimal to mild deficit, patient with SDA T was 82; the respective right 2 a moderate deficit, and 3 a severe perfusion medial temporal lobe deficit "breaching" the cortex. To localise and thicknesses were 17 5 mm grade the deficits more accurately scans were and 10 mm. assessed in both transverse and coronal planes. Grading was done alongside the CT scan so as to identify perfusion deficits in which intra- cerebral pathology was clearly discernible (such as a cerebral infarct) and accounted for the observed SPET perfusion anomaly. Any CT evidence of appreciable atrophy of the was also noted. SPET perfusion scores in the left, right, or bilateral parietotem- poral regions with a grading of 2 or more were considered to indicate a clear perfusion deficit for the purpose of subsequent analysis. Each subject was assessed with the CAMDEX.21 The cognitive section, CAM- COG, includes sections for testing memory, praxis, language, attention, concentration, orientation, abstract thinking, and calculation and has a maximum score of 107. It also enables the derivation of the mini mental state examin- ation, which is scored out of 30. A score of less than 80 in the CAMCOG and of less than 24 in the MMSE was considered to indicate a sig- nificant cognitive deficit. Neuropathological examination was perfor- A detailed clinical and psychiatric history med on cerebral hemispheres that had been http://jnnp.bmj.com/ was taken from each subject, followed by a suspended in buffered neutral formalin for at full physical examination of all systems, blood least six weeks. Histological sections were tests to exclude identifiable metabolic causes prepared from the frontal, temporal, and of memory loss and dementia, and neuro- parietal neocortex, hippocampus, cerebellum, psychological screening with the Cambridge midbrain, and pons. Sections were stained with mental disorders of the elderly examination haemotoxylin and eosin, congo red, luxol and

(CAMDEX).2' Screening specifically included fast blue cresyl violet and with methenamine on September 24, 2021 by guest. Protected copyright. CT and SPET of the brain. silver22 and Cross's modification of the CT scans were taken in the axial plane with Palmgren stain23 in order to demonstrate argy- 4 mm contiguous sections through the pos- rophylic plaques and neurofibrillary tangles, terior fossa and 8 mm contiguous sections respectively. Alzheimer's disease was diag- through the cerebrum parallel to the orbito- nosed if the diagnostic criteria proposed by meatal line. After this procedure and starting Khachaturian were fulfilled,24 taking into with a lateral topogram sections were then account any clinical history of dementia. taken passing through the front of the hard palate and top of the mastoids in 4 mm parallel sections through the temporal lobes; this scan Results plane is approximately 20 degrees caudad to the Clinical diagnosis orbitomeatal line. Left and right medial tem- Fifty one of the 79 patients fulfilled the DSM- poral lobe width (combined hippocampal for- III-R criteria for SDAT and could be further mation and parahippocampal gyrus) was then categorised, as shown in the table, into those measured from the film, using calipers, at the who were thought to have SDAT alone and narrowest point adjacent to the brain stem those with other central nervous system disease (figure 1). Measurements were made blind to as well. Ten patients had dementia not thought the clinical diagnosis. All scans were evaluated to be primarily SDAT, including five who by a neuroradiologist for evidence of any initially presented with dementia but who had intracerebral pathology. significant psychiatric symptoms consistent 192 Jobst, Smith, Barker, Wear, King, Smith, et al

Table Resultsfor patients in different diagnostic categories. Values are numbers ofpatients J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.55.3.190 on 1 March 1992. Downloaded from

Patients with Total no Patients with MTL parietotemporal Patients with both Clinical diagnosis ofpatients atrophy on CT deficit on SPET CT and SPET changes Clinically demented SDAT: Alone 23 23 23 23 With ischaemia 19 18 15 15 With other CNS disease 9* 8 6 6 Multi-infarct 3 3 3 3 Progressive supranuclear palsy 2 1 0 0 Pseudodementia 5 2 1 1 Not clinically demented Lesion absent 15 4 2 2 Cerebrovascular accident 3 it 3t it MTL = medical temporal lobe; patients were considered to have atrophy of the medial temporal lobe on CT if the thickness of one or both lobes was < 11 5 mm. A change in the SPET scan of the parietotemporal region on either side indicative of a moderate to severe reduced blood flow was defined by the scoring procedure described in methods. SDAT = senile dementia of Alzheimer's type. *Two with Parkinson's disease, two motoneurone disease, one Lewy body dementia, one dementia, one hypothyroid, two unknown. tCerebrovascular accident could account for atrophy of the medial temporal lobe. tTwo with a cerebrovascular accident and one with previous surgery that could account for abnormal SPET scans.

with a diagnosis of pseudodementia (table). patients with SDAT and four out of 10 of the Eighteen patients were thought not to have a other demented patients had a clear blood flow cognitive deficit consistent with a diagnosis of deficit (moderate to severe) on one or both sides dementia, of whom two had CT evidence of a (table). None of the demented group showed cerebrovascular accident and one had had CT evidence of appreciable parietal lobe atro- neurosurgery in the parietal region. phy, focal ischaemia, or surgery that would Cognitive testing gave average scores for the have accounted for the changes seen on SPET. demented group (n = 61) including those with Five out ofthe 18 non-demented patients had a pseuodementia of 51 (SD 28) (range 0-96) for clear parietotemporal perfusion deficit, but the CAMCOG and 15 (8) (0-29) for the MMSE; deficit could be accounted for in three of these for the non-demented group (n 18) the cor- by identifiable causes: two of the three had CT responding scores were 98 (6) (77-105) and evidence ofinfarction in the parietal region and 28-5 (1-5) (25-30). Two patients were classified one had had surgery in the parietal lobe. as demented by DSM-III-R criteria, even Although the location ofthe perfusion deficit in though they had cognitive test scores above the the posterior cortex varied slightly between normally accepted cut off values, because patients, it always included one or more of the estimates of their premorbid intelligence following areas: the superior parietal lobe, the quotient (IQ) were extremely high; one ofthese angular and supramarginal gyri, and the patients had a clinical diagnosis of pseudo- superior and middle temporal gyri. dementia, the other a diagnosis of Alzheimer's disease and Parkinson's disease. Comparison of CT and SPET changes in the http://jnnp.bmj.com/ same patient Computed tomography The occcurrence of reduced blood flow in The mean of the average thickness of left and parietotemporal regions as shown by SPET right medial temporal lobes, measured at the was invariably associated with atrophy of the thinnest point, in the 51 patients with SDAT medial temporal lobe as shown by CT: thus, all was 8-5 (2.8) mm (2-75-15 mm). The median 50 patients who had clear parietotemporal value was different < 0-0001, significantly (p perfusion deficits that could not be accounted on September 24, 2021 by guest. Protected copyright. Mann-Whitney) from that of the 18 non- for by infarction or surgery also displayed demented patients, whose mean medial tem- atrophy ofone or both medial temporal lobes as poral lobe thickness was 14-7 (3 0) mm (range shown by CT. On the other hand, atrophy of 10-18-5). On this basis we defined a medial the medial temporal lobe was not invariably temporal lobe thickness of 11 5 mm or less as associated with reduced parietotemporal blood indicative of atrophy (figure). With this flow, although such an association was par- criterion 49 out of the 51 patients with SDAT ticularly striking in patients with a clinical had atrophy ofthe medial temporal lobe on one diagnosis of SDAT (table). All 23 patients who or both sides, as did six out of 10 of the other were considered to have SDAT without other demented patients. However, only five out of 18 disease of the central nervous system displayed non-demented patients had medial temporal a combination ofmedial temporal lobe atrophy lobe atrophy; one of these five patients had and clear parietotemporal perfusion deficit; CT evidence ofcerebral infarction extending to the scans from one of these patients and a non- temporal lobe that could have accounted for the demented patient are shown in the figure. atrophy (table). Overall, only five ofthe 49 patients with SDAT who had atrophy of the medial temporal lobe SPET did not show a marked perfusion deficit in the With the scale outlined above for scoring the parietotemporal region (table); however, all five SPET scans for the degree of perfusion deficit ofthese patients had a SPET perfusion score of in the parietotemporal region, 44 out of the 51 1 either on one or on both sides, which with our Association ofatrophy of the medial temporal lobe with reduced bloodflow in the posterior parietotemporal cortex 193

classification represented a mild perfusion to show reduced perfusion or uptake of tracer, J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.55.3.190 on 1 March 1992. Downloaded from deficit. Two patients classified as having SDAT particularly in the posterior parietotemporal as well as other disease of the central nervous region, of patients with a clinical diagnosis of system showed neither atrophy of the medial Alzheimer's disease.65 2530 temporal lobe nor a clear perfusion deficit in the The striking feature of the results was the parietotemporal lobe, but both had unusual common occurrence ofthese two changes in the clinical features in addition to SDAT, which same patient. Both the CT and SPET changes will be described in detail in a later report. Out often occurred in the same patient ifthe patient ofsix demented patients considered not to have was clinically demented, particularly with SDAT but who had atrophy of the medial dementia of Alzheimer's type. Eighty six per- temporal lobe, four also showed a parietotem- cent (44/51) of patients with dementia of poral perfusion deficit. The only one of these Alzheimer's type had a combination of medial patients who died was found at necropsy to temporal lobe atrophy and reduced parieto- have Alzheimer's disease (see below). temporal blood flow, while only four out of the In subjects in whom there was medial tem- 10 demented patients who did not have a poral lobe atrophy-that is, thickness of clinical diagnosis of Alzheimer's disease 11-5 mm or less on one or both sides-the left showed this combination, one of whom was and right sides were compared to see whether found to have Alzheimer's disease at necropsy there was any relation between asymmetrical (see below). Such a combination ofchanges was medial temporal lobe atrophy and perfusion rare in patients who were not demented, occur- changes shown by SPET. Asymmetry of the ring in only two out of 18 patients when we atrophy was arbitrarily defined as a difference in excluded the patient in whom the changes thickness of 3 mm or more between the two might be accounted for by focal ischaemic sides. With this criterion, 12 of the 60 subjects damage in the medial temporal lobe. with medial temporal lobe atrophy showed A combination of medial temporal lobe such an asymmetry (six on each side). In 10 atrophy and reduced parietotemporal blood (nine with SDAT) of these 12 patients SPET flow in the same patient may thus be indicative scans showed that the parietotemporal blood ofAlzheimer's disease if due account is taken of flow ipsilateral to the medial temporal lobe clinical and CT evidence of ischaemic damage atrophy was lower than on the contralateral as a possible causative factor. Consistent with side. The remaining two subjects showed sym- this suggestion is the finding that, of the 12 metrical perfusion deficits in the parieto- patients who died, the 10 with a histological temporal regions. diagnosis of Alzheimer's disease all displayed the combination of CT evidence of atrophy of Histopathology the medial temporal lobe with SPET evidence Ten of the 51 patients with a clinical diagnosis of a parietotemporal perfusion deficit. It is of SDAT died: nine fulfilled the histopatho- noteworthy that the patient who died in the logical criteria ofKhachaturian for Alzheimer's group classified clinically as having multi- disease24 and the other was considered to have infarct dementia in fact had a histological normal pressure hydrocephalus. All 10 ofthese diagnosis of Alzheimer's disease and also dis- patients showed the combination of medial played the combination of CT and SPET temporal lobe atrophy and parietotemporal changes. We hope to follow all our patients perfusion deficit. Two patients out of the 10 through to necropsy to establish the histo- http://jnnp.bmj.com/ demented patients without SDAT died. One of logical diagnoses and, clearly, further com- these patients had a clinical diagnosis of multi- bined CT and SPET studies should be carried infarct dementia but had a histopathological out on patients with forms of dementia distinct diagnosis of Alzheimer's disease; this patient from Alzheimer's disease. also showed medial temporal lobe atrophy and A causal relation between the two changes a parietotemporal perfusion deficit in life, as may be indicated by the finding that in 10 out of

well as CT evidence of low density in the 12 patients in whom the atrophy of the medial on September 24, 2021 by guest. Protected copyright. periventricular white matter. The other patient temporal lobe was more severe on one side than had a clinical diagnosis of progressive supra- on the other, there was an identical asymmetry nuclear palsy confirmed at necropsy, and did in the parietotemporal perfusion deficit. We not show the combination of medial temporal suggest that the reduced parietotemporal blood lobe atrophy and parietotemporal perfusion flow might reflect pathological change in the deficit. Thus, so far, all 10 demented patients projection neurones in the parahippocampal with histologically confirmed Alzheimer's dis- gyrus, which in monkeys3 32 innervate inferior ease showed a combination of the two changes. parietal areas equivalent to the angular gyrus in man.33 Neurons that are pathologically affected in Alzheimer's disease are predominantly pro- Discussion jection neurones that interconnect association Our CT findings of atrophy of the medial and limbic areas of the cortex,'334 pyramidal temporal lobe are in agreement with several neurones of the hippocampal formation and recent studies on demented patients who are parahippocampal gyrus being particularly af- presumed to have Alzheimer's disease,'5 16 al- fected. '13 Medial temporal lobe atrophy in though we have used a relatively simple dementia might follow the loss of these projec- measurement that can be made directly from tion neurones, whereas in other conditions with the film. Likewise, the results of our SPET such atrophy (see introduction) it might be due studies agree with those of many published to the loss of different neurQnes. Our finding reports in which various tracers have been used suggests that, at least in the parietal lobe, one 194 Jobst, Smith, Barker, Wear, King, Smith, et al

cortex and the amygdala in Alzheimer's disease. J Neurol ofthe or J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.55.3.190 on 1 March 1992. Downloaded from consequence damage loss ofprojection Neurosurg Psychiatry 1990;53:161-5. neurones in the parahippocampal gyrus might 14 LeMay M. CT changes in dementing diseases: a review. be reduced American Journal ofNeuroradiology 1986;7:841-53. metabolism and blood flow in the 15 Kido DK, Caine ED, LeMay M, et al. Temporal lobe target area. As the disease progresses the loss of atrophy in patients with Alzheimer's disease: a CT study. from the American Journal ofNeuroradiology 1989;10:551-5. input hippocampal region might in 16 de Leon MJ, George AE, Stylopoulos LA, Smith G, Miller turn lead to transneuronal pathological changes DC. Early marker for Alzheimer's disease: the atrophic in the hippocampus. Lancet 1989;ii:672-3. parietal lobe and consequent loss of 17 Esiri MM, Oppenheimer DR. Diagnostic neuropathology. tissue, which seems to be more severe in the Oxford: Blackwell Scientific Publications, 1989. later of Alzheimer's 18 Roberts GW. Schizophrenia-the cellular biology of a stages disease.9"35 functional psychosis. Trends Neurosci 1990;13:207-1 1. 19 Press GA, Amaral DG, Squire LR. Hippocampal ab- We thank normalities in amnesic patients revealed by high-reso- the patients, their carers, and the lution magnetic resonance imaging. Nature 1989;341: clinicians in the Oxford region who referred 54-7. their 20 Jobst KA, Anslow PA, Barker S, et al. Diagnostic value of a patients for this study. Sue Stannard for combination oftemporal lobe CT scans and SPET studies coordinating many aspects of the study; and in dementia: definition of a subgroup of patients with Drs M structural and metabolic changes. Neurobiol Aging 1990; Esiri, J Marshall, and K McPherson for 11:262. advice. BMcD held a Wellcome Research 21 Roth M, Huppert FA, Tym E, et al. CAMDEX: the Cambridge examination for mental disorders of the elderly. Training Fellowship during part of this work. Cambridge: Cambridge University Press, 1988. The research was supported by a grant from the 22 Lamy C, Duyckaerts C, Delaere P, et al. Comparison of seven staining methods for senile plaques and Bristol-Myers Squibb Company. neurofibrillary tangles in a prospective series of 15 elderly patients. Neuropathol Appl Neurobiol 1989;15:563-78. 23 Cross RB. Demonstration of neurofibrillary tangles in paraffin section-a quick and simple method using Palmogren's technique. Med Lab Sci 1982;39:67-9. 24 Khachaturian ZS. Diagnosis of Alzheimer's disease. Arch 1 DeCarli C, Kaye JA, Horwitz B, et al. Critical analysis ofthe Neurol 1985;42:1097-105. use of computer-assisted transverse axial tomography to 25 Gemmell HG, Sharp PF, Evans NT, Besson JAO, Lyall D, study human brain in agingand dementia ofthe Alzheimer Smith FW. Single photon emission tomography with 123I_ type. Neurology 1990;40:872-83. isopropylamphetamine in Alzheimer's disease and multi- 2 Hoffman JM, Guze BH, Baxter LR, et al. < F-18 > Fluoro- infarct dementia. Lancet 1984;ii:1348. deoxyglucose (FDG) and positron emission tomography 26 Neary D, Snowden J, Shields R, Burjan AWI, Northen B, (PET) in aging and dementia-a decade of studies. Eur MacDermott N, et al. Single photon emission tomography Neurol 1989;29:16-24. using "'Tc-HM-PAO in the investigation of dementia. 3 McGeer EG, Peppard RP, McGeer PL, et al. Fluorodeoxy- J Neurol Neurosurg Psychiatry 1987;50:1 101-9. glucose-18 positron emission tomography studies in 27 Perani D, Di Piero V, Valler G, et al. Technetium-99m- presumed Alzheimer cases, including 13 serial scans. Can HMPAO-SPECT study of regional cerebral perfusion in JNeurol Sci 1990;17:1-11. early Alzheimer's disease. J Nucl Med 1988;29:1507-14. 4 Perani D, Di Piero V, Valler G, et al. Technetium-99m- 28 Gemmell HG, Sharp PF, Smith FW, et al. Cerebral blood HMPAO-SPECT study of regional cerebral perfusion in flow measured by SPECT as a diagnostic tool in the study early Alzheimer's disease. J NuclMed 1988;29:1507-14. of dementia. Psychiatry Res 1989;29:327-9. 5 Prohovnik I, Mayeux R, Sackeim HA, et al. Cerebral 29 Leys D, Steinling M, Petit H, et al. Single photon emission perfusion as a diagnostic marker of early Alzheimer's tomography with HMPAO Tc99 in Alzheimer's disease. disease. Neurology 1988;38:931-7. Rev Neurol (Paris) 1989;145:443-50. 6 Burns A, Philpot MP, Costa DC, Ell PJ, Levy R. The 30 Hellman RS, Tikofsky RS, Collier BD, et al. Alzheimer investigation of Alzheimer's disease with single photon disease-quantitative analysis of I-123-iodoamphetamine emission tomography. J Neurol Neurosurg Psychiatry SPECT brain imaging. Radiology 1989;172:183-8. 1989;52:248-53. 31 Mesulam MM, van Hoesen GW, Pandya DN, et al. Limbic 7 Hunter R, McLuskie R, Wyper D, et al. The pattem of and sensory connections of the inferior parietal lobule function-related regional cerebral blood flow investigated (area PG) in the rhesus monkey: a study with a new by single photon emission tomography with Tc-99M- method for horseradish peroxidase histochemistry. Brain HMPAO in patients with presenile Alzheimer's disease Res 1977;136:393-414. and Korsakoffs psychosis. Psychol Med 1989;19:847-55. 32 Cavada C, Goldman-Rakic PS. Posterior parietal cortex in 8 Montaldi D, Brooks DN, McColl JH, Patterson J, Barron E, Rhesus monkey. 1. Parcellation of areas based on distinc- McCulloch J. Measurements of regional cerebral blood tive limbic and sensory corticocortical connections. flow and cognitive performance in Alzheimer's disease. J Comp Neurol 1989;287:393-421. http://jnnp.bmj.com/ J Neurol Neurosurg Psychiatry 1990;53:33-8. 33 Eidelberg D, Galaburda AM. Inferior parietal lobule. 9 Brun A, Englund E. Regional pattern of degeneration in Divergent architectonic asymmetries in the human brain. Alzheimer's disease: neuronal loss and histopathological Arch Neurol 1984;41:843-52. grading. Histopathology 1981-5:549-64. 34 Pearson RCA, Esiri MM, Hiorns RW, et al. Anatomical 10 Wilcock GK, Esiri MM. Plaques, tangles and dementia. A correlates ofthe distribution ofthe pathological changes in quantitative study. J Neurol Sci 1982-56:343-56. the neocortex in Alzheimer's disease. Proc Nat Acad Sci 11 Hyman BT, van Hoesen GW, Damasio AR, et al. USA 1985;82:4531-4. Alzheimer's disease: Cell-specific pathology isolates the 35 Najerahim A, Bowen DM. Biochemical measurements in hippocampal formation. Science 1984;225:1168-79. Alzheimer's disease reveal a necessity for improved 12 Ball MJ, Hachinski V,Fox A, Kirshen AJ, Fisman M, Blume neuroimaging techniques to study metabolism. Biochem J

W, et al. A new definition of Alzheimer's disease: a 1988;251:305-8. on September 24, 2021 by guest. Protected copyright. hippocampal dementia. Lancet 1985;i:14-6. 36 Bowen DM, Najlerahim A, Proctor AW, et al. Circum- 13 Esiri MM, Pearson RCA, Steele JE, Bowen DM, Powell scribed changes ofthe in neuropsychiatric TPS. A quantitative study of the neurofibrillary tangles disorders of later life. Proc Nat Acad Sci USA 1989;86: and the choline acetyltransferase activity in the cerebral 9504-8.