Hidden Dysfunctioning in Subacute Stroke

Transcription

1 Hidden Dysfunctioning in Subacute Stroke Assia Jaillard, MD, PhD; Bernadette Naegele, PhD; Sandra Trabucco-Miguel, PhD; Jean François LeBas, MD, PhD; Marc Hommel, MD Background and Purpose Determining cognitive dysfunctioning (CDF) after stroke is an important issue because it influences choices for management in terms of return to previous activities. Because previous research in subacute stroke has shown important variations in CDF rates, we aimed to describe the frequency and neuropsychological profile of CDF in subacute stroke outside dementia. We used a large battery of tests to screen any potentially hidden CDF. Methods Patients with Mini-Mental State Examination scores 23 were prospectively and consecutively included 2 weeks after a first-ever ischemic brain infarct. Stroke features were based on MRI. Four domains were evaluated: instrumental and executive functions, episodic memory, and working memory (WM). Patients were scored using means and compared with education- and age-matched control subjects. Then we attributed Z-scores for each test and each domain. The most relevant cognitive tests characterizing CDF were determined using logistic regression. Results Among 177 patients (mean age, 50.6 years), 91.5% failed in at least one cognitive domain. WM was the most impaired domain (87.6%) with executive functions (64.4%), episodic memory (64.4%), and instrumental functions (24.9%) being relatively preserved. CDF was associated with age, education, depression, neurological deficit, and leukoaraiosis in bivariate analysis. Using logistic regression, WM tests and age predicted CDF (Modified Paced Auditorial Serial Addition Test: OR 0.96 CI 0.93 to 0.98; Owen-spatial-WM: OR 1.07 CI 1.02 to 1.12; age: OR 0.96 CI 0.93 to 0.98). Conclusion CDF appears to be almost constant, although underestimated, in subacute stroke. WM could reflect some hidden dysfunctioning, which may interfere with rehabilitation and return to work. Clinical routine may include WM tests in young patients with mild stroke. (Stroke. 2009;40: ) Key Words: acute stroke brain infarction cerebral infarct cognition cognitive impairment cognitive neurology depression leukoaraiosis MRI neuropsychology stroke in young adults stroke management stroke recovery vascular cognitive impairment young stroke in working memory Because stroke units and thrombolysis have improved stroke outcome and then disability related to neurological deficit, the impact of cognitive dysfunctioning (CDF) on handicap becomes a major issue. 1 Thus, the concept of stroke, as principally founded in motor function or ability in activities of daily living, may be insufficient with respect to changes in cognitive functioning. 2 Subtle CDF may indeed be missed within the hospital context and revealed only when returning to previous socioprofessional activities, leading to social handicap even more disturbing because unrecognized. 3,4 Research on CDF in subacute stroke, outside dementia, has evidenced widespread cognitive deficits, CDF rates varying from 31% to 77%. 5 9 Variations in CDF frequency may reflect some heterogeneity in terms of demographic parameters, stroke characteristics, or methodological approach. Actually, no widely accepted cognitive tests designed for the assessment of CDF after stroke exist. In most stroke studies, CDF was assessed using global tests such as Mini-Mental State Examination (MMSE) and Wechsler Adult Intelligence Scale (WAIS) battery aimed at detecting dementia or Alzheimer disease. 8,10 These tests are not very sensitive to executive and working memory dysfunction, which may account for the heterogeneous reported CDF rates. We aimed to describe prospectively the frequency and neuropsychological profile of CDF in nondemented patients with subacute stroke. In contrast to previous research, we investigated CDF using a comprehensive neuropsychological battery of tests assessing working memory and executive functions and able to reveal some hidden dysfunctioning. Moreover, although most studies on CDF were based on CT, Received January 31, 2009; final revision received February 25, 2009; accepted February 27, From the Institut Fédératif de Recherche (A.J., J.F.L.), Unité d Imagerie par Résonance Magnétique, Pôle Imagerie CHU Grenoble, France; Grenoble Institut des Neurosciences (A.J., B.N., J.F.L.), Inserm U.836, Equipe 5, Neuroimagerie Fonctionnelle et Métabolique, Grenoble, France; Unité Neurovasculaire (B.N., S.T.-M.), Pôle Neurologie-Psychiatrie, CHU Grenoble, France; and Centre d Investigations cliniques (M.H.), INSERM CIC 003 CHU Grenoble, France. Correspondence to Assia Jaillard, MD, PhD, Unité d IRM, IRF 1-Pôle Imagerie, CHU Grenoble, BP Grenoble-Cedex 09, France. Assia.Jaillard@ ujf-grenoble.fr 2009 American Heart Association, Inc. Stroke is available at DOI: /STROKEAHA

2 2474 Stroke July Patients admitted to the Stroke Unit 1194 patients with first-ever ischemic stroke 879 eligible patients 211 Selected Patients for Neuropsychological examination 177 Patients with complete procedure Exclusion related to MRI 402 Without stroke 209 Hemorrhagic strokes 60 With hypersignal >10 mm in diameter 74 Language barrier 102 Aphasia 139 Severe Neurological deficit; Neglect; Hemianopia; Medical or Psychiatric illness 668 patients with MMSE <23 9 Refusal 13 Inability to complete the neuropsychological examination 9 Delay >31 days 3 Other Excluded patients Figure. Diagram of the study s stroke population. MRI was performed in this study because of its sensitivity and specificity for stroke features. Methods Participants This prospective case control study included consecutive patients admitted to the stroke unit from November 2000 to May 2005 with a first-ever acute ischemic stroke and a control group of 81 healthy subjects matched for age and education level. Patients with first-ever ischemic stroke demonstrated on MRI and neuropsychological evaluation completed at Day 15 ( 15) were eligible (Figure). Participants were excluded if age 18 years, MMSE score 23 of 30, or in case of current illness, neurological disease or psychiatric disorder. Psychiatric disorder was considered when patients had a prior diagnosed psychosis requiring treatment with a depression with Beck score 27. When necessary, the diagnosis was confirmed by a psychiatrist. Neurological deficit was assessed by the stroke neurologist using the National Institutes of Health Stroke Scale (NIHSS), which was recorded at admission and at Day 15. Patients presenting at Day 15 with a score 2 for aphasia, hemianopia, and neglect were excluded. A score of 1 allowed selecting patients, including additional specific tests during the neuropsychological examination. Subjects who could not perform the tests in relation with apraxia, illiteracy, or decreased visual acuity were excluded. Control subjects were volunteers living in the same community as patients with stroke. Recruited in the vicinity by word of mouth (n 70) and in a healthy retirement home (n 11), they filled a brief structured questionnaire before inclusion focused on health status. The Medical Ethics Committee of the institution approved the study. We obtained signed informed consent from all subjects. Medical data were recorded by a stroke neurologist (A.J., M.H.). Stroke features included neurological deficit, risk factors, Trial of Org in Acute Stroke Treatment classification (TOAST), 11 and antidepressant medication. Motor handedness, MMSE, and depression using the Beck Depression Inventory were recorded. MRI, including T2, fluid-attenuated inversion recovery, and diffusion-weighted imaging sequences (Phillips 1.5 T; axial slices; thickness 5 mm) was performed within the first days. Stroke diagnosis, side, arterial territory, 12 imaging TOAST criteria, 11 and leukoaraiosis 13 were assessed by 2 stroke neurologists or radiologists unaware of the clinical data. Patients with clinical stroke not confirmed by MRI or with prior stroke, defined as a hypersignal 10 mm in diameter, were excluded. Leukoaraiosis was assessed using Scheltens criteria. 13 This scale evaluates occipital and frontal caps; lateral ventricles bands (score 0 to 6) and white matter hyperintensities in frontal, parietal, temporal, and occipital lobes (score 0 to 24); basal ganglia (score 0 to 30); and infratentorial structures (score 0 to 24). The total score was recorded. Neuropsychological Assessment All participants underwent a battery of tests measuring cognitive domains. In addition, Edinburgh handedness score, MMSE, and depression using the Beck Depression Inventory were recorded (Supplemental Appendix). The following 4 domains were thoroughly assessed: (1) episodic memory: the Grober and Buschke procedure including 5 scores: immediate cued recall, free recall, recognition, delayed cued recall, and free recall of 16 words. Visual memory was measured with the Pattern Recognition Memory test from the Cambridge Neuropsychological Test Automated Battery; (2) general and instrumental functions. Mattis scale, verbal automatisms, visuospatial and visuoconstructive functions were assessed with the Incomplete Letters and Position Discrimination tests from VOSP and language with a 80 picture naming test (LEXIS); (3) executive functions: the Luria alternating sequences task, the Modified Wisconsin Card Sorting Test, the semantic and phonemic fluency tests, the Hanoï Tower, the Stroop tests, and the Trail- Making Tests A and B; (4) working memory: the backward digit span (WAIS-III) and the Modified Paced Auditorial Serial Addition Test (M-PASAT) 14 assessed auditory working memory. The subject was asked to listen to 60 numbers listed one by one, every 4 seconds, and at the same time add the 2 last numbers heard, answering orally. The score, ranging from 0 to 60, corresponded to the number of correct answers. Spatial working memory (SWM) was assessed using the Owen test 15 from the SWM battery of the Cambridge Neuropsychological Test Automated Battery. The subject had to find a token in each of a number of 4 to 8 boxes. Once a token had been taken from a box, this box would not be used again. Memory and strategy errors were scored. The forward digit span from the WAIS-III digit span subtest and the spatial span from the Corsi Blocks test assessed short-term memory. In addition, primary attentional capacities were assessed using the Big/little Cir Test and the Reaction Time from the Attention Battery of the Cambridge Neuropsychological Test Automated Battery. Control subjects underwent the same neuropsychological evaluation except instrumental functions and attentional abilities. Raw performances and Z-scores were recorded for each test. To calculate Z-scores, we used norms derived from published data or working groups such as the GREFEX or the GREMEM, so they were based on large samples of healthy populations and adjusted for age and education. Diagnosis of CDF On the basis of Z-scores, participants were classified into 3 groups for each domain. Normal group consisted of patients who scored in the normal range for all tests, mild impairment defined patients cutoff Z-score between 1 and 2 on at least one test, and patients with moderate impairment with cutoff Z-scores beyond 2 onat least one test. Global CDF, defined as any impairment in at least one cognitive domain, was rated 1 (mild) or 2 (moderate) according to the score of the most impaired domain. Statistical Analysis To compare patients and control subjects characteristics and performances, bivariate analyses were performed with 2 test, analysis of variance, and independent-sample t test with Levene s test for equality of variances or nonparametric appropriate tests, according to the Gaussian distribution of the variables with a 5% level of significance. Bivariate comparisons were completed by a correlation analysis between tests adjusted for multiple comparisons using the Sidak test. To isolate the most important factors characterizing CDF in the stroke population, compared with the control population, we used a logistic regression model. The status (stroke versus control) was the dependent variable, whereas the demographic and cognitive data were entered in the model to determine predictors of stroke-related CDF with 95% jackknife estimation of CIs. Data were analyzed using STATA 10.1 (STATA Corp, College Station, Texas).

3 Jaillard et al Hidden Dysfunctioning 2475 Table 1. Demographic Characteristics of Patients With Stroke and Control Subjects Patients Control Subjects Demographics (N 177) (N 81) P Mean age (SD) 50.6 (16.1) 51.9 (17.8) 0.55 Males/females, n (%) 112 (62.3)/65 (36.7) 47 (58.0)/34 (42.0) 0.49 Primary and secondary/university level, N (%) 104 (58.7)/73 (41.3) 50 (61.7)/31 (38.3) 0.89 Mean Edinburgh handedness score (SD)* 92.1 (33.8) 93.3 (25.8) 0.89 Mean Beck Depression Index score (SD)* 8.4 (7.3) 7.4 (4.5) 0.84 MMSE, mean (SD)* 28.5 (2.26) (1.58) 0.01 *Mann Whitney test. Table 2. Clinical and Imaging Features for the Stroke Population (N 177) Features N (%) Vascular territory N 177 Anterior circulation 129 (73.4) Anterior cerebral 6 (3.4) Middle cerebral 108 (61.0) Anterior choroidal 16 (9.0) Posterior circulation 47 (26.6) Posterior cerebral 19 (10.7) Vertebrobasilar 28 (15.8) Territory Deep N 60 (33.9) Superficial N 88 (49.7) Deep and superficial N 29 (16.4) Leukoaraiosis Scheltens Scale, mean (SD) 2.52 (2.84) Infarct size: 15 mm/ 15 mm 50 (28.2)/127 (71.8) Clinical Cortical or cerebellar/lacunar syndrome 131 (74.6)/45 (25.4) NIHSS score at Day 1, mean (SD) 5.7 (4.5) NIHSS score at Day 15, mean (SD) 1.5 (2.0) Delay stroke neuropsychological 14.6 (8.4) examination, mean days (SD) NIHSS indicates National Institutes of Health Stroke Scale. Results We included prospectively 177 consecutive patients within 1 month of a first-ever infarct admitted to the stroke unit from November 2000 to May 2005 and 81 control subjects. Demographic characteristics of patients and control subjects are reported in Table 1. There were no significant differences in age, gender, education, handedness, or Beck Depression Inventory. MMSE was lower in patients than in control subjects, although the mean scores were within the normal range. Neurological, imaging, and stroke features are reported in Tables 2 and 3. The raw scores and Z rates of cognitive tests are reported in Table 4. Patients performed significantly worse than control subjects in most tests except for the spatial span, cued episodic memory tasks, phonemic fluency, and Modified Wisconsin Card Sorting Test. The M-PASAT was the most impaired test. The Big/little Cir Test was normal in all patients. Better performances were observed for the nonparetic than the paretic hand on the Reaction Time test (P ). This test was not correlated with other cognitive tests in multiple correlation analysis. Leukoaraiosis correlated significantly with tests assessing free recall, executive function, and working memory. Correlation coefficients between tests performances, demographics, and neurological measures are reported in the Supplemental Table I, available online at CDF reached 91.5%, working memory being the most impaired domain (87.6%; Table 5). In bivariate analysis, CDF was associated with age (P 0.003), level of education (P 0.003), depression (P 0.019), neurological deficit at Day 15 (P 0.028), and leukoaraiosis (P 0.023), but not Table 3. Risk Factors and Trial of Org in Acute Stroke Treatment Classification of Stroke Subtypes (177 Patients) Risk Factors of Atherosclerosis Hypertension 78 (44.1) Diabetes mellitus 10 (5.7) Admission glycemia: 5.5 mmol/l/ 8 mmol/l 12 (6.8)/2 (1.1) Low-density lipoprotein cholesterol 1.2 g/l 73 (41.2) Daily smoking 57 (32.9) Alcohol 25 g/d 18 (10.3) Homocysteine 15 mmol/l 10 (5.7) TOAST subtype 52 (29.4) Large artery atherosclerosis Cardioembolisms 41 (23.2) Atrial fibrillation 15 (8.5) Patent foramen ovale (N 165 TOE) 23 (13.9) Other: cardiac valve, intracardiac thrombus 4 (2.3) Small vessel occlusion 28 (15.8) Other determined etiology 39 (22.0) Extracranial artery dissection 13 (7.3) Oral contraceptive and migraine 10 (5.6) Prothrombotic conditions: factor V Leiden, 9 (5.1) G20210A prothrombin gene variant, ATIII deficit, polyglobulia, thrombocythemia, myelodysplasia, drepanocytosis Drugs (cannabis, heroin) 3 (1.7) Other: pulmonary arteriovenous fistula, 4 (2.3) antibody antiphospholipids, Marfan Undetermined etiology 17 (9.6) TOE indicates trans-oesasophageal echography.

4 2476 Stroke July 2009 Table 4. Neuropsychological Tests in Patients With Stroke (N 177) and Control Subjects (N 81) Neuropsychological Tests Z-Score 1, N (%) Z-Score 2, N (%) Patients, Mean (SD) Control Subjects, Mean (SD) P* Memory: Short-term and episodic memory Digit span (NP, KS) 40 (22.6) 23 (13.0) 5.4 (1.0) 5.6 (0.8) Spatial span (NP, KS) 40 (22.6) 13 (7.3) 5.2 (1.1) 5.0 (0.8) 0.75 RL/RI-16 (NP, KS) Total cued recall 12 (7) 30 (17.5) 45.4 (5.1) 47.5 (0.9) Total free recall 53 (31.0) 33 (18.6) 29.1 (8.0) 32.0 (6.8) Delayed cued recall 14 (8.2) 16 (9.4) 15.4 (1.6) 15.9 (0.3) Delayed free recall 36 (20.3) 48 (27.1) (3.8) 13.7 (2.0) Recognition 11 (6.4) 8 (4.7) 15.8 (0.7) 16.0 (0.0) Pattern recognition 17 (14.0) 10 (8.3) 88.5 (10.3) ND... General /Instrumental functions Mattis scale 5 (1.9) 14 (5.4) (7.5) ND... Verbal automatisms (NP, MN) 12 (7.0) 10 (5.8) (16.1) (11.8) WAIS Similitudes (t test) 12 (8.1) 4 (2.7) 11.3 (3.3) 13.7 (1.6) Lexis Naming test (NP, MN) 10 (6.1) 3 (1.8) 77.8 (6.0) 79.3 (1.3) VOSP (N 137) 12 (6.8) 13 (7.3) 19.3 (1.1) ND... Incomplete letters Position Discrimination 4 (2.3) 12 (6.8) 9.1 (1.6) Executive functions Phonemic fluency (t test) 34 (19.8) 20 (11.6) 18.6 (8.1) 20.0 (5.4) Semantic fluency (t test) 45 (26.2) 21 (12.2) 27.4 (10.7) (8.3) Trail Making A (NP, KS) 10 (5.6) 10 (5.6) 47.5 (32.9) 40.8 (22.5) Trail Making B (NP, KS) 21 (11.9) 45 (25.4) (82.7) (67.0) M-WCST (NP, MN) Categories 26 (10.1) 20 (7.8) 5.7 (0.9) 5.5 (0.9) 0.85 Perseverations 27 (10.5) 31 (12.0) 2.9 (3.6) 2.7 (2.6) 0.69 Maintenance errors 4 (1.6) 4 (1.6) 0.86 (1.3) 0.68 (1.2) 0.55 Hanoï Tower 3 pieces (LN) 10 (8.5) 20 (17.1) 11.7 (14.8) 7.1 (8.1) Stroop test inks (N 27) 6 (10.7) 2 (3.4) (29.8) ND... Luria sequences ND... Gestual sequences (4.0) Graphic sequences (3.8) Working memory Backward digit span (NP, MN) 48 (27.3) 40 (22.7) 3.86 (1.41) 4.4 (0.82) M-PASAT (NP, MN) 18 (10.2) 107 (60.5) 33.8 (21.3) 47.2 (11.4) CANTAB SWM test Memory errors (NP, MN) 42 (24.3) 41 (23.1) 30.7 (24.2) 19.0 (16.4) Strategy errors (NP, MN) 49 (28.3) 36 (20.8) 31.3 (7.9) 27.6 (7.2) Reaction time (N 72) Paretic hand, msec 8 (11.1) 9 (12.5) 55 (76.2) (67.9) Non-paretic hand, msec 4 (5.6) 6 (8.3) 62 (86.1) (60.3) Mean refers to raw scores. *P for the comparison between patients and control subjects scores. Comparison with t test for parametric tests and with Mann Whitney (MN) or Kolmogorov-Smirnov (KS) tests for nonparametric tests (NP): comparison; LN indicates comparison with the napierian logarithm transformation; ND, not done; M-WCST, Modified Wisconsin Card Sorting Test; SWM, spatial working memory. with gender (P 0.12) or with delay between stroke onset and neuropsychological examination (P 0.39). Furthermore, delay did not correlate with any cognitive test. Antidepressant therapy, which was associated with depression (P 0.011), was not associated with CDF (P 0.86). The logistic regression resulted in finding the M-PASAT (OR 0.96 per point; P ; 95% CI 0.93 to 0.98), Owen-SWM test ( strategy errors: OR 1.07 per error; P 0.005; 95% CI 1.02 to 1.12), and age (OR 0.96 per year; P ; 95% CI 0.93 to 0.98) as significant pre-

5 Jaillard et al Hidden Dysfunctioning 2477 Table 5. Frequencies of Domain-Specific Cognitive Deficits and CDF in 177 Patients With Stroke Domain At Least One Test Z 1, N (%) At Least One Test Z 2, N (%) Instrumental functions 27 (15.3) 17 (9.6) Episodic memory 49 (27.7) 65 (36.7) Executive functions 38 (21.5) 76 (42.9) Working memory* 32 (18.1) 123 (69.5) CDF (any domain) 32 (18.1) 130 (73.4) *Including spans. dictors of CDF. Note that gender, level of education, depression, neurological deficit, and leukoaraiosis were not kept by the model. Discussion The 177 patients achieved the neuropsychological battery and had lower performances than matched control subjects in most neuropsychological tests. As a group, 91.5% (95% CI 0.86 to 0.95) of patients with stroke had difficulties in at least one cognitive domain and 73.4% (95% CI 0.67 to 0.80) had severe difficulties. Studies conducted on subacute stroke have reported rates of cognitive impairment varying from 49% up to 80%, 6 8,16 19 suggesting that our CDF rates were higher than others. The neurological deficit, age, methodological bias, or other confounders could explain such discrepancies. The mean National Institutes of Health Stroke Scale score was at Day 1 and improved to at Day 15, whereas the mean MMSE (28.4) was close to control subjects MMSE (29.15), corresponding to strokes with good outcome. In the previously cited studies, patients were older, presented with more severe stroke, and had lower MMSE than ours. Because age, stroke severity, and MMSE are associated with CDF, 7,20,21 better cognitive functioning could be expected in our population. Patients are usually compared with a matched control group, resulting in large CIs due to small samples. In our study, the use of standardized cutoff scores, based on age and level of education derived from large samples, may have increased the sensitivity to CDF. However, even studies using standardized cutoff scores have reported lower CDF rates than ours. 22 Regarding neuropsychological tests, CDF is usually assessed using MMSE, the WAIS battery, or some standard executive functions tests designed for degenerative diseases. 8,10 When additional episodic memory 22 and executive function 8,22 tests were added, increased CDF was reported. In our study, although executive and episodic memory functions account for 64.4% of CDF, working memory dysfunction reached 87.6% (Table 5). Therefore, using a comprehensive battery including several executive functions and specific working memory tests may have revealed hidden cognitive dysfunctioning. The neuropsychological profile of CDF, using multivariate analysis, was characterized by the M-PASAT and Owen- SWM tests after adjustment for age, although gender, educational level, and depression were not predictive of CDF. Performing M-PASAT is highly demanding of auditory working memory, because subjects have to retain relevant auditory information (the last 2 numbers heard) and simultaneously manipulate this information (adding these 2 numbers). 14 The failure to perform the maintenance and operations of the M-PASAT, which require the activation of the prefrontal cortex and posterior parietal cortex, 23 was somewhat expected because these areas are supplied by the middle cerebral artery, which was involved in 61% of our patients. The Owen-SWM test assessed the capacity to maintain a strategy when searching the correct spatial sequence is assumed to explore spatial working memory in relation to the right frontoparietal regions, 24 which are also supplied by the middle cerebral artery. Note that the forward digit and spatial spans were relatively preserved in our patients, because they rather engage storage than processing working memory 25 and could have been more sensitive to the stroke side. 26 Age and level of education were strongly associated with CDF using bivariate analysis in agreement with most previous studies. 27,28 Neurological deficit was also associated with CDF. Because outcome has often been related to cognitive impairment and stroke severity, the association between National Institutes of Health Stroke Scale and CDF at the same time (Day 15) in this study was expected. 18,29 The association between depression and CDF is consistent with prior studies, 30,31 because depression is common after stroke and has been directly related to cognitive impairment It has been argued that CDF could reflect depression or fatigue. 35 In our study, depression and fatigue as a Beck Depression Inventory item were associated with CDF in bivariate analysis. However, they were not predictors of CDF characterized by working memory dysfunctioning. Whether depression could be a confounder has been addressed using logistic regression to model working memory. Neither depression nor fatigue was predictive of working memory dysfunction, suggesting they are measuring another dimension such as mood or patient s self-evaluation. Leukoaraiosis is a consistent predictor of CDF and has been more specifically related to working memory and executive dysfunction. 36,37 Although leukoaraiosis was associated with CDF and correlated with cognitive tests in this study, it was not a predictor, presumably because of the low frequency of leukoaraiosis. Again, this could be explained by the low age and high MMSE of our patients. 38 The dominant finding of our study is that CDF is almost constant within the first month after stroke and related to working memory deficit with some executive and memory dysfunction, whereas instrumental functions were relatively spared. One of the strengths of this study derives from the use of MRI in all patients, which allow more reliable diagnosis of recent strokes, silent strokes, and leukoaraiosis. However, some limitations must be considered. Our population does not reflect CDF of a broadly based stroke population when considering the young age, high MMSE, and low National Institutes of Health Stroke Scale scores. Most ineligible patients presented with major neurological or cognitive deficits, dementia, MMSE 23, or recurrent stroke, all conditions inappropriate for attending the whole neuropsychological examination. However, because ineligible pa-

6 2478 Stroke July 2009 tients were supposed to present with more obvious CDF, they would not have modified our findings. The delay variations between stroke and neuropsychological assessment could have interfered with CDF, because patients are recovering within the first month. To limit the delay interference, we attempted to schedule examinations 2 weeks after stroke. Actually, no correlation between delay and cognitive tests was observed. It could be argued that the battery of neuropsychological tests was too large because some tests were redundant and time-consuming for patients and examiners. However, standard tests may fail to identify CDF, which, in the context of a prospective study, is only captured by a more comprehensive assessment. Moreover, such a large battery could allow selecting the more relevant tests to detect stroke-related CDF. The level of attention has not been specifically addressed in this study and, thus, could have interfered with cognitive performances, because attention is required for the performance of all cognitive tasks. Stricto sensu, attention refers to vigilance, but more recently, the concept has been extended to attentional or executive control, referring to the range of capacities involved in modulating and controlling attention. 39,40 All patients had normal performances in on the Big/little Cir Test, assessing primary attention, and few were impaired in the Reaction Time, Trail-Making Tests A and B, and Modified Wisconsin Card Sorting Test, assessing processing speed and executive attentional abilities. Therefore, it is questionable that attentional deficit could account for a major part in the CDF and working memory impairment we observed. A final issue is whether CDF after stroke could be related to mild cognitive impairment (MCI), a recent concept proposed to identify a transitional state between normal aging and dementia related to Alzheimer disease. 40 Whereas episodic memory is reported to be the most impaired domain in MCI, working memory and executive functions are also often altered. 40 However, executive dysfunctioning in MCI has been related to leukoaraiosis, which is common in elderly persons with MCI. 41 Because the diagnosis of MCI is excluded by strokes, CDF and MCI differ in their pathogenesis and clinical significance. A remaining issue is whether CDF, similar to MCI, could be a risk factor for dementia. Finally, we showed high CDF rates in patients with stroke with low residual neurological deficit. This is compelling because hidden cognitive difficulties may interfere with recovery. Therefore, identifying such difficulties could help in clinical practice for the management of rehabilitation and return to work. Working memory assessment should be included in the routine evaluation of young patients with mild stroke. Source of Funding Funding for this study provided by Ministère de la Santé (PHRC 1997 CIRCE to A.J.). None. Disclosures References 1. Gorelick PB, Bowler JV. Advances in vascular cognitive impairment Stroke. 2008;39: Carlsson GE, Moller A, Blomstrand C. Consequences of mild stroke in persons 75 years a 1-year follow-up. Cerebrovasc Dis. 2003;16: Hommel M, Trabucco-Miguel S, Joray S, Naegele B, Gonnet N, Jaillard A. Social dysfunctioning after mild to moderate first-ever stroke at vocational age. J Neurol Neurosurg Psychiatry Nov 14 [Epub ahead of print]. 4. Hommel M, Trabucco-Miguel S, Naegele B, Gonnet N, Jaillard A. Cognitive determinants of social functioning after a first-ever mild to moderate stroke at vocational age. J Neurol Neurosurg Psychiatry Apr 8 [Epub ahead of print]. 5. Sachdev PS, Brodaty H, Valenzuela MJ, Lorentz L, Looi JC, Wen W, Zagami AS. The neuropsychological profile of vascular cognitive impairment in stroke and TIA patients. Neurology. 2004;62: Riepe MW, Riss S, Bittner D, Huber R. Screening for cognitive impairment in patients with acute stroke. Dement Geriatr Cogn Disord. 2004;17: Rasquin SM, Lodder J, Verhey FR. Predictors of reversible mild cognitive impairment after stroke: a 2-year follow-up study. J Neurol Sci. 2005; : Nys GM, van Zandvoort MJ, de Kort PL, van der Worp HB, Jansen BP, Algra A, de Haan EH, Kappelle LJ. The prognostic value of domainspecific cognitive abilities in acute first-ever stroke. Neurology. 2005;64: Nys GM, van Zandvoort MJ, de Kort PL, Jansen BP, de Haan EH, Kappelle LJ. Cognitive disorders in acute stroke: prevalence and clinical determinants. Cerebrovasc Dis. 2007;23: Blake H, McKinney M, Treece K, Lee E, Lincoln NB. An evaluation of screening measures for cognitive impairment after stroke. Age Ageing. 2002;31: Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org in Acute Stroke Treatment. Stroke. 1993;24: Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of the human brain: cerebral hemispheres. Neurology. 1998;50: Scheltens P, Barkhof F, Leys D, Pruvo JP, Nauta JJ, Vermersch P, Steinling M, Valk J. A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J Neurol Sci. 1993;114: Naegele B, Mazza S, Grenoble. CD. Test d attention soutenue: Pasat modifié. Adaptation française. Marseille:Solal; Owen AM, Downes JJ, Sahakian BJ, Polkey CE, Robbins TW. Planning and spatial working memory following frontal lobe lesions in man. Neuropsychologia. 1990;28: Hoffmann M. Higher cortical function deficits after stroke: an analysis of 1000 patients from a dedicated cognitive stroke registry. Neurorehabil Neural Repair. 2001;15: Kinsella G, Ford B. Acute recovery from patterns in stroke patients: neuropsychological factors. Med J Aust. 1980;2: Lawrence ES, Coshall C, Dundas R, Stewart J, Rudd AG, Howard R, Wolfe CD. Estimates of the prevalence of acute stroke impairments and disability in a multiethnic population. Stroke. 2001;32: Schmidt R, Mechtler L, Kinkel PR, Fazekas F, Kinkel WR, Freidl W. Cognitive impairment after acute supratentorial stroke: a 6-month follow-up clinical and computed tomographic study. Eur Arch Psychiatry Clin Neurosci. 1993;243: Gazzaley A, Sheridan MA, Cooney JW, D Esposito M. Age-related deficits in component processes of working memory. Neuropsychology. 2007;21: Henon H, Durieu I, Guerouaou D, Lebert F, Pasquier F, Leys D. Poststroke dementia: incidence and relationship to prestroke cognitive decline. Neurology. 2001;57: Rasquin SM, Verhey FR, Lousberg R, Winkens I, Lodder J. Vascular cognitive disorders: memory, mental speed and cognitive flexibility after stroke. J Neurol Sci. 2002; : Audoin B, Au Duong MV, Ranjeva JP, Ibarrola D, Malikova I, Confort-Gouny S, Soulier E, Viout P, Ali-Cherif A, Pelletier J, Cozzone PJ. Magnetic resonance study of the influence of tissue damage and cortical reorganization on PASAT performance at the earliest stage of multiple sclerosis. Hum Brain Mapp. 2005;24:

7 Jaillard et al Hidden Dysfunctioning Malhotra P, Jager HR, Parton A, Greenwood R, Playford ED, Brown MM, Driver J, Husain M. Spatial working memory capacity in unilateral neglect. Brain. 2005;128: Cowan N, Morey CC, Chen Z, Bunting M. What do estimates of working memory capacity tell us? In: Osaka N, Logie RH, D Esposito M, eds. The Cognitive Neuroscience of Working Memory. Oxford: Oxford University Press; 2007: Smith EE, Jonides J, Koeppe RA. Dissociating verbal and spatial working memory using PET [published erratum appears in Cereb Cortex. 1998; 8:262]. Cereb Cortex. 1996;6: Newman GC, Bang H, Hussain SI, Toole JF. Association of diabetes, homocysteine, and HDL with cognition and disability after stroke. Neurology. 2007;69: Rasquin SM, Verhey FR, van Oostenbrugge RJ, Lousberg R, Lodder J. Demographic and CT scan features related to cognitive impairment in the first year after stroke. J Neurol Neurosurg Psychiatry. 2004;75: Mok VC, Wong A, Lam WW, Fan YH, Tang WK, Kwok T, Hui AC, Wong KS. Cognitive impairment and functional outcome after stroke associated with small vessel disease. J Neurol Neurosurg Psychiatry. 2004;75: Kauhanen M-L, Korpelainen J, Hiltunen P, Brusin E, Mononen H, Määttä R, Nieminen P, Sotaniemi K, Myllylä V. Poststroke depression correlates with cognitive impairment and neurological deficits. Stroke. 1999;30: Nys GM, van Zandvoort MJ, van der Worp HB, de Haan EH, de Kort PL, Jansen BP, Kappelle LJ. Early cognitive impairment predicts long-term depressive symptoms and quality of life after stroke. J Neurol Sci. 2006;247: Berg A, Palomaki H, Lehtihalmes M, Lonnqvist J, Kaste M. Poststroke depression in acute phase after stroke. Cerebrovasc Dis. 2001;12: Hackett ML, Yapa C, Parag V, Anderson CS. Frequency of depression after stroke: a systematic review of observational studies. Stroke. 2005; 36: Mondal S, Sharma VK, Das S, Goswami U, Gandhi A. Neuro-cognitive functions in patients of major depression. Indian J Physiol Pharmacol. 2007;51: Staub F, Bogousslavsky J. Fatigue after stroke: a major but neglected issue. Cerebrovasc Dis. 2001;12: Jokinen H, Kalska H, Mantyla R, Ylikoski R, Hietanen M, Pohjasvaara T, Kaste M, Erkinjuntti T. White matter hyperintensities as a predictor of neuropsychological deficits post-stroke. J Neurol Neurosurg Psychiatry. 2005;76: Lamar M, Price CC, Libon DJ, Penney DL, Kaplan E, Grossman M, Heilman KM. Alterations in working memory as a function of leukoaraiosis in dementia. Neuropsychologia. 2007;45: Pantoni L, Simoni M, Pracucci G, Schmidt R, Barkhof F, Inzitari D. Visual rating scales for age-related white matter changes (leukoaraiosis): can the heterogeneity be reduced? Stroke. 2002;33: Mesulam M. Attentional networks, confusional states and neglect syndromes. In: Mesulam MM, ed. Principles of Behavioral and Cognitive Neurology. Oxford: Oxford University Press; 2000: Belleville S, Bherer L, Lepage E, Chertkow H, Gauthier S. Task switching capacities in persons with Alzheimer s disease and mild cognitive impairment. Neuropsychologia. 2008;46: Delano-Wood L, Abeles N, Sacco JM, Wierenga CE, Horne NR, Bozoki A. Regional white matter pathology in mild cognitive impairment: differential influence of lesion type on neuropsychological functioning. Stroke. 2008;39: