J7ournal ofNeurology, Neurosurgery, and Psychiatry 1996;60:91-93 91 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.1.91 on 1 January 1996. Downloaded from SHORT REPORT

Executive dysfunction in early Alzheimer's disease

G Binetti, E Magni, A Padovani, S F Cappa, A Bianchetti, M Trabucchi

Abstract have been related to dysfunction. Twenty five patients with probable mild It has been suggested that in addition to more Alzheimer's disease were assessed for direct involvement with selected aspects of deficits in executive functioning and the memory, such as remote memory, confabula- impact ofthese deficits on performance in tion, and memory for temporal sequences, other neuropsychological domains. The the frontal lobes play a strategic part as a Wisconsin card sorting test, the release "cognitive mediation" system in the encoding from proactive interference paradigm, and retrieval of memories.4 Frontal lobe the verbal fluency test, and the Stroop test involvement in the clinical heterogeneity were adopted to classify patients with of Alzheimer's disease, however, has not (AD+) and without (AD -) executive been fully investigated and it is still unclear deficits. Seven of the patients showed an whether frontal lobe related neuropsychologi- impairment in executive function (AD+), cal impairments are integral to Alzheimer's defined as a performance below the cut disease in the early phase, and whether they off score in at least two of these tests. are associated with a differential pattern of There were no significant differences neuropsychological impairment. The aim of in clinical assessments, demographic this study was to evaluate executive function features, or other cognitive functions in patients with mild Alzheimer's disease and between patients. to investigate the relation between executive Executive dysfunction may be an early dysfunction and performance in other neuro- additional feature in a subgroup of psychological domains such as language, patients with mild Alzheimer's disease. memory, , abstract reasoning, and Impairment on frontal lobe tests does not visuospatial abilities. For this purpose, the seem to be related to the severity or dura- Wisconsin card sorting test,5 the release tion of disease, or to a different pattern of from proactive interference,6 the verbal fluency

impairment in other cognitive domains. test,7 and Stroop test8 were adopted to classify http://jnnp.bmj.com/ patients with (AD+) and without (AD-)

_J Neurol Neurosurg Psychiatry 1996;60:91-93) executive deficits.

Keywords: Alzheimer's disease; executive function Subjects and methods All patients included in this study were Alzheimer's disease is characterised by impair- evaluated in the Alzheimer's disease unit at ment of multiple memory related systems. In S Couore Fatebenefratelli Hospital in Brescia, on September 26, 2021 by guest. Protected copyright. Alzheimer's Disease addition patients with Alzheimer's disease Italy, and were drawn from a consecutive Unit, Istituto S Cuore have disturbances in several other cognitive series of patients included in a longitudinal Fatebenefratelli, Brescia, Italy and domains such as language, visuospatial skills, programme on mild Alzheimer's disease Geriatric Research praxis, and abstraction.' In addition to clinical, (MADI project). Group, Brescia, Italy epidemiological, biological, and neuroimaging Twenty five consecutive patients, affected G Binetti E Magni findings, the evidence of distinct profiles of by probable Alzheimer's disease according to A Padovani cognitive deficits has led some authors to sug- currently accepted clinical criteria,9 and with A Bianchetti gest the existence of heterogeneous subtypes a questionable or mild form of (clin- M Trabucchi of patients with Alzheimer's disease.' 2 For ical dementia rating (CDR) 0-5 or 1)10 were Department of example, there is evidence that in individual included in the study. All patients had CT or Neurology, University of Brescia, Italy patients the presenting symptom may not be MRI to exclude other possible aetiologies of S F Cappa the usual memory loss but one or more of a dementia. Correspondence to: number of deficits involving other cognitive Twenty five normal elderly subjects, either Dr Giuliano Binetti, par- correlated volunteers or spouses of patients, Alzheimer's Disease Unit, functions. As there is growing evidence, Istituto S Cuore ticularly from imaging studies, suggesting a were also included in the study. History of Fatebenefratelli, Via cortex disabilities, Pilastroni, 4, 25123 Brescia, considerable involvement of the frontal alcoholism, drug abuse, learning Italy. in Alzheimer's disease,3 study of executive and serious neurological or psychiatric ill- Received 25 August 1994 dsyfunction is essential to the understanding nesses were considered as exclusion criteria. and in final revised form of the disease. Written informed consent was obtained from 30 August 1995 of heterogeneity Accepted 1 September 1995 Various aspects of memory and learning all subjects (patients and controls), or care- 92 Binetti, Magni, Padovani, Cappa, Bianzchetti, Trabucchi

givers where appropriate. Patients and con- version of WCST5 with 64 cards was adopted, J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.1.91 on 1 January 1996. Downloaded from trols were assessed on the geriatric computing an index of perserveration (IP; scale (GDS)" to exclude subjects with severe number of errors/number of perseveration). depressive symptoms (GDS score > 20). The The verbal fluency for letters (PFL) was mini mental state examination (MMSE)"2 was assessed by recording the number of words given to characterise the global cognitive level. produced in the course of one minute for each letter.7 The Stroop colour word test' is an NEUROPSYCHOLOGICAL ASSESSMENT executive task for the assessment of the The following neuropsychological tests were ability to change perceptual sets to conform to given to both groups (for details of tests see changing task requirement. The interference Binetti et al). score (IS) was calculated by subtracting the Short term memory was assessed by audi- predicted colour word score from the raw tory-verbal forward digit span and visuospatial colour word scores. The higher the resultant span (Corsi's block tapping test). Episodic score, the less susceptible the patient is to memory was tested with the logical memory interference. The release from proactive inter- test (recall of a short story) and with the 20 ference (RPI) paradigm was evaluated using minutes delayed recall of the Rey complex the Wickens'6 modification of the Peterson figure. A 30 item version of the Boston naming and Peterson'3 distractor technique according test provided a measure of naming and seman- to Freedman and Cermak.'4 The stimuli, con- tic memory. Remote memory was assessed sisting of five trials of words triads grouped in with the Q60 questionnaire. The Street test four blocks, were presented on a 23 x 18 cm and the copy of the Rey figure were used to monitor. After the presentation of each triad assess visuospatial abilities. The Raven's subjects were asked to count backwards from coloured progressive matrices test, the a random number as rapidly as possible, to attentional matrices test, and the token test preclude rehearsal of the triad. After counting were also included in the neuropsychological for 10 seconds, the patient was asked to recall battery. the three words he had just read. Each block "Executive" function was evaluated by the of five trials consisted of the presentation of Wisconsin card sorting test (WCST), PFL four triads of words from the same semantic verbal fluency test,7 Stroop test,' and the category, followed by a fifth triad of words release from proactive interference paradigm.6 from a different semantic category (shift con- The WCST is a problem solving task dition (SC)). Each block consisted of high that measures the ability to identify abstract lexical frequency words from four categories categories and shift cognitive set. A shortened taken from Battig and Montague Connecticut category norms." The categories were, in sequence, animals, fruits, colours, and vegeta- Table I Perfonnance ofpatients with (AD+) anid without (AD-) bles. The SC effect was calculated by sub- deficits tracting the fourth trial score from the score of AD- AD+ each fifth trial throughout the blocks. (n = 18) (n = 7) No ofpatients As an arbitrary cut off value to identify (mcan (SD)) (meani (SD)) below the cut off (0) patients with prominent executive dysfunc-

Demographic data: tion, we adopted the criterion of a score 1SD http://jnnp.bmj.com/ Age (y) 72-1 (9 5) 73 7 (6 9) Education (y) 7-4 (4-1) 5-7 (1-6) below the controls' mean on at least two out MJF 3/15 F7 - of the four measures of frontal lobe function CDR (0 5/1) 4/14 3/4 - MMSE 21 5 (2 2) 21 4 (2 4) (IP, PFL, IS, SC). Duration of illness (months) 30 7 (18 9) 36.8 (24-1) Geriatric depression scale 12-0 (7 6) 10 4 (6-8) Tests: Verbal fluency (PFL)t 17 5 (7-9) 10 0 (9-4) 3 (12) Results Wisconsin card sorting test: Index of perseveration 0-3 (0-14) 0-5 (0 17)* 7 (28) According to the described cut off, we Stroop test interference - 31 (12-7) - 3-8 (9-4) 5 (29) identified two subgroups of patients with on September 26, 2021 by guest. Protected copyright. Release from proactive interference 1 5 (3 1) -0-8 (30)* 12 (48) (block 5-block 4) Alzheimer's disease: seven patients (AD+) were considered to show executive dysfunc- *p < 0 05. tScores are not corrected for age and education. tion and 18 patients (AD -) not. Table 1 CDR = Clinical dementia rating. shows the number of patients below the cut off for each test. The AD+ and AD - groups were com- Table 2 Performance ofpatients with Alzheimer's disease with (AD+) and without parable in terms of age, education, sex, and (AD -) frontal lobe dysfunction on tests of memory, language, attention, abstract thinking, and visuospatial abilities depressive synptoms (table 1). There was no difference between AD+ and AD - patients AD- AD+ of illness. (n = 18) (n = 7) for MMSE score and duration Test (mean (SD)) (mean (SD)) As shown in table 2, we did not find Digit span 5-3 (1 1) 4.7 (0-7) a significant difference between AD+ and Corsi test 4.7 (0 9) 4-4 (0-9) AD- patients on neuropsychological tests Logical memory test 2 5 (3-5) 2 6 (2 0) Rey figure 20 minute recall 1.5 (3-4) 0-9 (2 4) of memory, language, attention, abstract Token test 28 0 (3-1) 29.0 (4 0) thinking, or visuospatial abilities. Boston naming test* 13 7 (5-1) 12 8 (8 2) Attentional matrices 30 3 (8 7) 28 5 (8 4) Raven 36 PM colour 18 8 (5 2) 17 5 (3 5) Rey figure copy 24 2 (8 5) 19 7 (9-7) Street test 4-4 (2 0) 5 0 (2 0) Discussion Q60 212 (5 1) 14 9 (8-3) Although the deterioration of cognitive *A 30 items version was used. function is considered to be widespread in Executive dysfunction in early, Alzheimer's disease 93 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.60.1.91 on 1 January 1996. Downloaded from Alzheimer's disease,' frontal lobe involvement nitive domains, and suggest the possible in the clinical presentation of the disease has involvement of the frontal lobe in early stages not been investigated in detail. Thus it is still of the disease. unclear whether executive dysfunction is an integral component of early Alzheimer's dis- We thank the two anonymous reviewers for their helpful ease or rather characterises later stages of the suggestions. disease. This study showed that executive dys- 1 Binetti G, Magni E, Padovani A, Cappa SF, Bianchetti A, Trabucchi M. Neuropsychological heterogeneity in mild function may be an early manifestation of Alzheimer's disease. Denmentia 1993;4:321-6. Alzheimer's disease. In fact seven out of 25 2 Mayeaux R, Stern Y, Spanton S. Heterogeneity in dementia of Alzheimer's type: evidence of subgroups. NeurologU patients were found to be impaired on tests 1985;35:453-61. measuring executive function, supporting 3 Grady CL, Haxby B, Horwitz M, Sundaram M, Berg G, Shapiro M, et al. Longitudinal study of the early previous findings in Alzheimer's disease.'6 17 neuropsychological and cerebral metabolic changes in To evaluate whether the presence of dementia of the Alzheimer's type. _7 Clin Exp Aenropsvyhol execu- 19881O:576-96. tive dysfunction could be related to variables 4 Warrington EK, Weiskrantz L. : a disconnection such as age, education and duration of the ill- ? Neuropsychologia 1980;20:233-48. 5 Berg EA. A Simple objective test for measuring flexibility in ness, the AD+ and AD - groups were com- thinking. _] Gen Psychol 1948;39:15-22. pared. There were no major differences 6 Wikens DD. Encoding categories of words: an empirical approach to meaning. Psychol Rev 1970;77: 1-15. between the groups. 7 Novelli G, Papagno C, Capitani E, Laiacona M, Vallar G, One of our aims was to investigate whether Cappa SF. Tre test clinici di ricerce e produzione lessi- cale: Taratura su soggetti normali. Archivio di Psicologzia, a deficit in executive functioning in mild Neurologia e Psichiatnia 1986;47:477-506. Alzheimer's disease was associated with a 8 Golden CJ. Stroop color and word test. Chicago, Stoeltig Company, 1978. different pattern of neuropsychological 9 McKhann G, Drachman D, Folstein M, Katzman R, Price impairment. No significant cognitive differ- D, Stadlan EM. Clinical diagnosis of Alzheimer's dis- ease: report of the NINCDS-ADRDA Work Group ences were found between the AD+ and AD - under the auspices of Department of Health and Human groups using a complete Services Task force of Alzheimer's disease. Neurology neuropsychological 1984;34:939-44. battery. This finding suggests that executive 10 Hughes CP, Berg L, Danzinger WL, Coben LA, Martin dysfuncton, as defined in this study, has little LA. A new clinical scale for the staging of dementia. Br3r Psychiatry 1982;140:566-72. impact on the performance of tasks measuring 11 Yesavage JA. Brink TL, Rose TL, Andrey M. Geriatric other cognitive domains such as language, depression rating scale: comparison with other self report and psychaitric rating scales. In: Crook T, Ferris S, visuospatial function, and memory. Bartus R, eds. Assessmient in gemiatnrc psychoparnnacology. The present data may reflect the involve- New Canaan, 1983. 12 Folstein MF, Folstein SE, McHugh PR. Mini mental state. ment of the frontal cortex in the early phase of _7 Psychiatr Res 1975;12: 189-98. disease in a subgroup of patients. 13 Peterson LR, Peterson MJ. Short-term retention of individ- ual items. T Exp Psychol 1959;58:193-8. Studies with PET and SPECT have indi- 14 Freedman M, Cermak LS. Semantic encoding deficits in cated that an anterior frontal lobe disease and amnesia. Brainl Cognt 1986;5: involvement, although 108-14. not typical, can be seen in the early phase of 15 Battig WF, Montague WE. Category norms for verbal disease.3 '8 items in 56 categories: a replication and extension of the Connecticut category norms. Joiurnal of experinen?btal The pattern of the impairment seems to be Monographs 1969;80:1-46. clearly differentiated from the clinical picture 16 Kopelman MD. Frontal dysfunction and memory deficits in the alcoholic Korsakoff syndrome and Alzheimer-type in patients with suspected frontotemporal dementia. 1991;114:117-34. dementia,'9 given the lack of the behavioural 17 Bhutani GE, Montaldi D, Brooks DN, McCulloch J. A http://jnnp.bmj.com/ neuropsychological investigation into frontal lobe features of the "frontal lobe syndrome" which involvement in dementia of the Alzheimer type. are often present in this condition. Neuropsychology 1992;6:211-24. 18 Foster ML, Chase TN, Fedio P, Patronas NJ, Brooks RA. Taken together, these findings seem to sup- Alzheimer's disease: focal cortical changes shown by port the existence of a subgroup of positron emission tomography. Neurology 1983;33: patients 961-5. with mild Alzheimer's disease who have 19 Brun A, Englund B, Gustafson L, Passant U, Mann DMA, additional neuropsychological executive Neary D, Snowden JS. Clinical and neuropathological deficits unrelated to impairment of other criteria for frontotemporal dementia. _7 Neurol Neurosurg cog- Psychiatry 1994;57:416-8. on September 26, 2021 by guest. Protected copyright.