STROKE FREQUENTLY LEADS to a serious and general

Transcription

1 1499 Cognitive Recovery After Stroke: A 2-Year Follow-Up Jacqueline B. Hochstenbach, PhD, Rob den Otter, MSc, Theo W. Mulder, PhD ABSTRACT. Hochstenbach JB, den Otter R, Mulder TW. Cognitive recovery after stroke: a 2-year follow-up. Arch Phys Med Rehabil 2003;84: Objectives: To determine (1) whether long-term improvement of cognitive function takes place after stroke and (2) which clinical factors influence cognitive recovery. Design: Cohort study with patients who were assessed at 2.3 and 27.7 months after stroke. Setting: Home-based stroke patients. Participants: From a group of 229 stroke patients, 92 were approached to participate. Sixty-five (43 men, 22 women; mean age, 56.4y) agreed, and they were neuropsychologically assessed at 72.2 days after stroke. A group of 33 controls (12 men, 21 women; mean age, 52.4y) was used as a reference sample. Interventions: Not applicable. Main Outcome Measures: Orientation, memory, attention, visuospatial, visuoconstructive, language, and arithmetic abilities were assessed with an extensive neuropsychologic test battery. Results: Significant improvements across time were noted for all cognitive domains. The biggest improvement was found in the attentional domain; the least, in the memory domain. In addition, a small subset of patients accounted for the significant improvement in all cognitive domains; most patients showed no improvement or declined. Factors influencing recovery were side of the stroke and incidence of lowered consciousness on admission. Patients with right-side brain damage performed better than those with left-side brain damage and showed more improvement over time. Patients with lowered consciousness on admission performed worse than patients without lowered consciousness. No significant effect was found for gender, type of stroke, cortical versus subcortical lesions, having 1 stroke or multiple strokes, or the interval between the stroke and the neuropsychologic assessment. Conclusion: There was room for improvement in all cognitive domains, although this improvement was gained by only a small number of patients. Hence, most patients must cope with serious permanent cognitive decline after stroke. Key Words: Cognition; Neuropsychology; Recovery of function; Rehabilitation; Stroke by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation From the Centre for Brain Damage Aftercare, Department of Rehabilitation, University Hospital & University of Groningen, Groningen (Hochstenbach); Sint Maartenskliniek-Research, Nijmegen (den Otter); and Institute of Human Movement Sciences, University of Groningen, Groningen (Mulder), The Netherlands. Presented in part at the 3rd World Congress in Neurological Rehabilitation, April 2 6, 2002, Venice, Italy. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Theo Mulder, PhD, Institute of Human Movement Sciences, Univ of Groningen, P.O. Box 196, 9700 AD Groningen, The Netherlands, T.Mulder@ppsw.ug.nl /03/ $30.00/0 doi: /s (03) STROKE FREQUENTLY LEADS to a serious and general cognitive decline, 1-4 but little is known about the extent of recovery over time. Although a number of studies have investigated recovery after stroke, most of them have focused on functional recovery Among the studies concerned with the recovery of cognitive functions, dementia and the improvement of specific cognitive impairments (eg, neglect or aphasia) have been focal research topics. Furthermore, these studies have focused on a relatively short period after stroke, usually 3 to 6 months. Hence, there is little information concerning the recovery of general cognitive functioning in a longer period after stroke. Only 1 study 24 has addressed this topic, and results showed improvement in 19 of the 151 patients who were tested 3 months and again 1 year after stroke. There is a need for more accurate and quantitative information about the natural history of recovery of general cognitive dysfunctioning after stroke, as well as for more insight into factors that might influence this recovery. Such information and insight not only will enrich our theoretical understanding of recovery processes, but will also provide us with clinically relevant information. 25 Indeed, knowing the cognitive domains in which recovery is more pronounced may help clinicians formulate more adequate treatment strategies. 26 Furthermore, successful reintegration into the physical, vocational, and social roles in which patients were previously engaged often depends on factors other than the degree of motor or sensory deficit. 27,28 Therefore, the main goal of our study was to determine whether long-term improvement of cognitive function takes place after stroke. In addition, a number of specific factors were examined to determine whether they influence this recovery. Although there are abundant studies concerned with the prognosis of functional recovery in terms of activities of daily life (for reviews, see Jongbloed 29 and Kwakkel et al 30 ), little is known about correlates of improvement in cognitive function. Against this background, the influence of factors such as gender, type of stroke, side of stroke, location of stroke, and level of consciousness on admission to the emergency department were studied. Given the paucity of pertinent research in this area, we did not formulate specific hypotheses for this exploratory study. METHODS Participants The 65 patients in our study were a sample from a series of 229 patients enrolled in a study about neuropsychologic deficits after stroke. 3 Of the 92 patients originally approached, 65 gave informed consent for reassessment; the other 27 patients declined participation. At the time of their stroke, the patients were between 18 and 70 years old, with the diagnosis of a stroke confirmed by computed tomography of the brain. The age limit was set at 70 years to avoid the compromising effect of age on cognitive functioning. Those affected by any other major physical diseases (life-threatening, neurologic, or disabling) or psychiatric disorders (under psychiatric treatment in the last 10y) were excluded. At the time of the neuropsychologic reassessment, all patients were living at home. There were 43 men and 22 women, and they had a mean age standard deviation (SD) of years (range,

2 1500 RECOVERY OF COGNITION AFTER STROKE, Hochstenbach Level Table 1: Dutch Educational System Years of Education 1 6 (elementary school) (university) 21 73y) and a mean educational level of (see table 1). There were 35 patients with left-hemisphere strokes, 27 with right-hemisphere strokes, and 3 with a lesion in both hemispheres. The stroke types could be divided into 57 infarctions and 8 hemorrhages. There were 24 patients with subcortical lesions, 38 with cortical lesions, and 2 with both cortical and subcortical lesions; these data were missing for 1 patient. Five patients had had a previous stroke. The neuropsychologic reassessment of these 65 patients took place at a mean of months (range, 19 49mo) after stroke onset. The results of this reassessment were compared with the results of the first neuropsychologic assessment, which took place at a mean of months (range, mo) after stroke onset. Because not all neuropsychologic tests have norms adequately stratified for age, a group of healthy controls was tested once and used as a reference sample. This group consisted of 12 men and 21 women, with a mean age of 52.4 years (range, 25 73y) and a mean level of education of years (see Hochstenbach et al 3 ). Neuropsychologic Examination Orientation, memory, attention and concentration, visuospatial and visuoconstructive functions, language, and arithmetic skills were assessed. These functions were selected because they represent important aspects of cognitive functioning (for a detailed description of the tests used, see Hochstenbach 3 ). The following tests were used at both testing times. Orientation. Orientation was determined by questions about time, length of time, place, and person. The score on this test was a composite score. Memory. Memory performance was evaluated with the Dutch version of the Auditory-Verbal Learning Test 31 (AVLT), the Wechsler Adult Intelligence Scale II 32,33 (WAIS-II) digit span subtest, the Rivermead Behavioural Memory Test, 34,35 and recounting a story. Attention. Attention was assessed through the Trail-Making Tests A and B 36 (TMT-A, TMT-B), the WAIS-II digit symbol subtest, a letter cancellation task, and using a photograph from the subtest picture scanning of the Behavioural Inattention Task 37 (BIT). Visuospatial and visuoconstructive. Visuospatial and visuoconstructive functions were assessed by the WAIS-II block design subtest both with and without a time limit, the Bobertag, 38,39 a structured clock test, a clock drawing task, 40 the copying task of the BIT, and Money s road map test for left-right orientation. 41 Language and arithmetic. Language and arithmetic skills were evaluated by means of the similarities subtest from the WAIS-II, a writing task, and 4 subtests from the Dutch Aphasia Society 42 word comprehension, sentence comprehension, naming, and verbal fluency (in the category animals ). Arithmetic skills were assessed with different aspects of handling money recognizing, counting, and arithmetic. Statistical Analyses To investigate the possibility of a selection bias, Student t tests were used to determine whether there were differences between the group that had a second neuropsychologic assessment (n 65) and the group that was assessed only once (n 164). Analyses were performed for the following variables: age, gender, education, various health aspects (eg, heart disease, hypertension, diabetes mellitus), type of stroke (infarction, hemorrhage), the side of the stroke (left, right), the location of the lesion (cortical, subcortical), single or multiple strokes, lowered consciousness on admission or not, and the interval between stroke and the first assessment. Further analyses were performed by using general linear models. First, analyses were performed for each cognitive domain to determine whether there was a significant difference between assessment 1 and assessment 2. To establish whether the improvement shown in this analysis represented a general improvement shared by all subjects, the following equation was used: C (x i1 x i2 )/2SD controls, where C is the change across time, x i1 is the test score of subject i at time 1, x i2 is the test score of subject i at time 2, and 2SD is 2 standard deviations of the scores of 33 healthy controls. The criterion of 2SD, a relatively rigorous and statistically relevant criterion, 3 was chosen to ensure that improvement was not an artifact of the measurements (eg, test-retest effect, small control group). Improvement was defined as C greater than or equal to 1, no change as 1 C 1, and deterioration as C less than or equal to 1 and was corrected for the orientation of the test scores. Multivariate analyses were performed to examine the influence of different variables on the changes between the 2 assessments (within-subjects factor: time). The between-subjects factors were gender, type of stroke (infarction, hemorrhage), the side of the stroke (left, right), the location of the lesion (cortical, subcortical), single or multiple strokes, the interval between stroke and the first assessment, and lowered consciousness on admission or not. Stages of lowered consciousness ranged from stupor to somnolent and were determined by a neurologist when the patient was admitted to the emergency department. The patient was considered conscious when he/she was, for instance, able to answer a question or to obey simple commands. For status of the patient, level of consciousness was reported as being lowered or not, which we extracted for research purposes. RESULTS No significant differences were established between the group that was assessed twice and the group that was assessed only once, except for a single variable: the former comprised fewer patients who had had a previous stroke ( , P.036). Therefore, it seems fair to conclude that no selection bias existed. In table 2, the results of the multivariate analyses indicate a significant improvement across time for all cognitive domains. The more specific results of the univariate analyses show that, in the memory domain, working memory and recognition of previously learned material improved. In the attention domain, the biggest improvement was seen in the TMT-B, indicating improvement in the ability to work under time pressure as well as in the ability to adequately shift between mental sets. This was reflected in a decrease in time taken and errors made. Neglect, too, as measured by the omissions on the letter can-

3 RECOVERY OF COGNITION AFTER STROKE, Hochstenbach 1501 Table 2: Significant Results of General Linear Model for the Various Cognitive Domains, With Univariate Analyses to Indicate Significant Differences Domain F df P Univariate F df P Means Assessment 1 Means Assessment 2 Memory , ,56 AVLT immediate recall AVLT recognition WAIS digit span backward Attention , ,54 TMT-B correct TMT-B time Letter cancellation OL Visuospatial and visuoconstructive , ,54 WAIS-II block design WTL WAIS-II block design NTL Clock drawing Money s road map time Money s road map defaults Language and arithmetic , ,58 Naming Sentence comprehension Writing WAIS-II similarities Counting money Orientation , Abbreviations: NTL, no time limit; OL, omissions left; WTL, with time limit. cellation task, diminished. Visuospatial and visuoconstructive abilities improved as well: not only did the construction of 2-dimensional and 3-dimensional figures (both with and without a time limit) improve, but also the left-right orientation was enhanced. Language abilities showed improvement in the more basic aspects of language, such as word fluency, sentence comprehension, and writing, as well as in the more abstract use of language. Although in the more complex aspect of handling money arithmetic patients did not show any significant improvement, they did better in more basic aspects, such as counting money. Orientation is a composite score of questions about time, length of time, place, and person; the results in table 2 show a significant improvement across time. At this juncture, a relevant concern is the extent to which all subjects share this improvement. To examine this, factor C (change) was determined for all tests that were univariately significant (see table 2) according to the equation stated in the Methods section. As a result, table 3 depicts the percentage of patients who showed improvement, no change, or deterioration. The results indicated that a large number of subjects with C greater than or equal to 1 could be found in the attentional domain, as reflected by the better scores on the TMT and fewer omissions in the letter cancellation task. The same was true for the language domain, although only a small percentage of patients showed an improvement in the more abstract use of language, as measured by WAIS-II similarities. The results on the visuospatial and visuoconstructive tasks were not so clearcut: subjects showed improvement in some tasks and no improvement in others. In Money s road map test, there was a decrease in time but an increase in the incidence of defaults. An improvement in clock drawing was seen in 37.5% of the patients, but little change was found on the WAIS-II block design. Orientation was an area of improvement for 21.9% of the subjects. The results also showed that only a small percentage of subjects improved in memory tasks, as measured by the AVLT and the WAIS-II digit span. To gain more insight into factors that may influence the observed changes between the 2 moments of assessment, further multivariate analyses for each cognitive domain were performed. The factors included were gender, type of stroke (infarction, hemorrhage), the side of the stroke (left, right), lesion location (cortical, subcortical), instance of single or multiple strokes, consciousness (having lowered consciousness on admission or not), and whether there was a difference in recovery between patients who were assessed early after stroke and those who had their first assessment later. This factor is important because it could be that patients who were assessed earlier after stroke had more room for improvement than patients who were assessed later. Therefore, the interval between the stroke and the moment of assessment 1 was divided into 4 categories: 3 to 37 days (n 14), 37 to 70 days (n 15), 70 to 95 days (n 24), and more than 95 days (n 12). The results of the analyses, summarized in table 4, show that 2 factors had a significant main effect: side of the stroke and consciousness. For the other factors, no significant effect was found. In exploring these results, we saw that the most pronounced effect was caused by the side of the stroke, which generated a significant main effect for memory (F 9, , P.012), language and arithmetic (F 9, , P.003), and orientation (F 1, , P.03). Further scrutiny of the main effects revealed that patients with right-hemisphere damage performed better in all cognitive domains than patients with left-hemisphere damage. In the memory domain, the performance of patients with right-sided lesions was, on average, 27% better than that of those with left-sided lesions; in the

4 1502 RECOVERY OF COGNITION AFTER STROKE, Hochstenbach Table 3: Percentage of Patients Who Changed on the Neuropsychologic Tests, Which Showed Significance From Assessment 1 to Assessment 2 (see table 2) Neuropsychologic Test 2SD Deterioration No Change Improvement AVLT immediate recall (0 75) AVLT recognition (0 15) WAIS digit span backward (0 12) TMT-B correct* (0 24) TMT-B time (s) Letter cancellation OL WAIS-II block design WTL (0 26) WAIS-II block design NTL (0 26) Clock drawing (0 10) Money s road map time Money s road map defaults (0 32) Naming (0 18) Sentence comprehension (0 45) Writing (0 1) WAIS-II similarities (0 26) Counting money (0 10) Orientation (0 4) NOTE. Number in parentheses indicates range of possible scores. * Correct is the number of correct connections. language domain, the performance was 11% better; and in the orientation domain, the results were as much as 78% better. In addition, there was also a significant time by side interaction in the attentional domain (F 9, , P.026). It appears that the group with left-sided brain damage improved, on average, by only 8% in attention tasks, whereas the group with rightsided brain damage improved by 28%. There was a main effect of consciousness for orientation (F 1, , P.022), showing a performance improvement of 228.2% for patients who were fully conscious on admission. Also, significant time by consciousness interactions in the memory domain (F 9, , P.045) and in the visuospatial and constructive domain (F 8, , P.009) were obtained. The group that was fully conscious on admission registered an improvement of 5.1% in the memory domain, whereas the other group showed an improvement of 22.1%. In the visuospatial and constructive domain, conscious subjects showed an improvement of 8.8%, whereas the group with lowered consciousness registered an improvement of 33.8%. DISCUSSION The main goal of our study was to determine whether longterm recovery of cognitive functioning after stroke takes place. An additional goal was to obtain some insight into the factors that might influence this recovery. Our results suggest that long-term improvement in generalized cognitive function does indeed take place after stroke, although it is an improvement that is not shared by most stroke patients. These results concur with those of Desmond et al, 24 who found a similar long-term recovery in 19 of 151 patients, with significant improvement in memory, orientation, visuospatial function, and attention, but not in language or abstract reasoning. Kotila et al 43 demonstrated an improvement between the first assessment that took place 3 months after stroke and the second assessment 12 months after stroke for domains such as visuospatial functioning, memory, and intelligence. Furthermore, improvement was noted for combined dyslexia, dysgraphia, and dyscalculia. They found little improvement, however, for speech and language deficits. Although similar, the results of our study are not completely in line with those of Desmond 24 and Kotila. 43 We found improvement in all domains. Furthermore, the domains in which most subjects showed improvement were attention and language, whereas only a few subjects showed improvement in the memory functions. How can such a difference be explained? A possible explanation can be found in the different ways of analyzing the data. Kotila 43 did not perform any Table 4: Effect of Various Factors on the Cognitive Domains Between Assessment 1 and Assessment 2 Factor Memory Attention Visuospatial/ Constructive Language/ Arithmetic Orientation Gender Type (infarct, bleeding) Side (left, right) Location (cortical, subcortical) No. of strokes ( 1) Consciousness* Time of assessment 1 NOTE. The first column indicates whether the factor exerts a significant main effect ( ) or not ( ); the second column indicates a significant interaction ( ) or not ( ). *Having lowered consciousness or not on admission to the hospital. P.05; P.01.

5 RECOVERY OF COGNITION AFTER STROKE, Hochstenbach 1503 statistical analysis to determine whether the differences between the 2 assessments were significant, so those results have only face value. Desmond 24 transformed the raw test scores of the various tests into z scores, which then were averaged to produce summary scores of each domain, thus leading to less specificity. Although Desmond 24 and Kotila 43 did not find improvement in language functions, there are other studies that have shown recovery of language during the first 3 to 6 months after stroke. 6,22,44-46 These results are in concordance with our results. Recently, a study by van Zandvoort et al 47 showed that there is improvement between early assessment and late follow-up, but it also showed that there was a significantly stable test profile between these 2 assessments. Hence, it can be concluded that, although there is room for significant improvement, these improvements are relatively limited. Also, even though we did find these improvements, we should not forget that most patients showed no improvement or even declined, which leaves a vast number of stroke patients with considerable cognitive impairments. Two factors that may influence the changes between the 2 assessments could be isolated: the side of the stroke (with patients with right-sided brain damage performing better and improving more) and level of consciousness on admission. Common sense dictates that patients who had lowered consciousness on admission to the hospital performed worse than those without lowered consciousness. It is also logical that, in the memory domain and in the visuospatial and constructive domain, more subjects in the group with lowered consciousness improved than in the group without lowered consciousness. Because of their lower initial scores, they simply had more room for improvement. However, despite their larger relative improvement, 2 years after stroke their performance still did not equal that of the group without lowered consciousness on admission. In future research, it would be interesting to explore this factor further, especially because use of the Glasgow Coma Scale score in stroke patients is becoming more and more common, thus providing differential scores. The most intriguing factor concerns the time interval from the occurrence of stroke until the first assessment. It has been shown 3 that no significant difference existed between the results of patients who were assessed at different poststroke times. Still, it could be hypothesized that patients who were assessed early had more room for recovery than patients who were assessed much later after stroke. Such a formulation is not only guided by logic but also supported by research findings, which show that functional recovery is most rapid during the first months after stroke This raises some interesting questions, especially in a decade when neural plasticity is a central topic. 51 Is plasticity more feasible in the sensory motor domain than in the neuropsychologic domain? Did we ask the wrong questions, or are we using the wrong methods? Or are the results found in the motor domain artifacts of a Gallian error? It is clear that further research is needed to explore the robustness of the findings of this study. Despite these novel and interesting findings, caution remains necessary. Indeed, the categorization of the tests into the various cognitive domains, as described in the Methods section, remains a bit arbitrary. It is well known that tests often measure various aspects of cognitive functioning. For instance, the time score on Money s road map test may also be seen as a measure of speed of information processing, just as the TMT-B and WAIS-II digit span can be viewed as measures of executive function. 52 Because the way in which the tests were categorized would have influenced the results, we chose to follow the same categorization used in our earlier study, 3 keeping in mind the element of arbitrariness in some of the categorization. It is unfortunate that we were not able to make use of extensive neuroimaging techniques that would have enabled us to understand the process of plasticity and recovery better. More light could have been thrown on the relationship with factors such as the size or exact location of the stroke or metabolic changes that may occur in the brain during recovery. A final critical remark concerns the control group. Although neuropsychologic studies hardly use control groups, it would have been methodologically more elegant if we had also reexamined the group of healthy subjects. That would have enabled us to determine the range of variation in test scores not related to stroke and recovery. CONCLUSIONS The results of our study are relevant for clinical practice because they can enhance the therapy. Indeed, there are a number of different approaches that can be used in the rehabilitation of cognitive dysfunction for example, restorative therapies, compensation or strategy training, and behavioral approaches. 53 Perhaps in the future it will be possible to use the increasing knowledge about recovery to select the most appropriate treatment regimen. Maybe a restorative rehabilitation approach would be more effective for aspects that show a considerable ability for recovery, whereas for other aspects, a compensatory approach might be more appropriate. References 1. Hom J, Reitan RM. Generalized cognitive function after stroke. J Clin Exp Neuropsychol 1990;12: Tatemichi TK, Desmond DW, Stern Y, Paik M, Sano M, Bagiella E. Cognitive impairment after stroke: frequency, patterns, and relationship to functional abilities. J Neurol Neurosurg Psychiatry 1994;57: Hochstenbach J, Mulder T, van Limbeek J, Donders R, Schoonderwaldt H. Cognitive decline following stroke: a comprehensive study of cognitive decline following stroke. J Clin Exp Neuropsychol 1998;20: Kase CS, Wolf PA, Kelly-Hayes M, Kannel WB, Beiser A, D Agostino RB. Intellectual decline after stroke: the Framingham study. Stroke 1998;29: Wade DT, Wood VA, Hewer RL. Recovery after stroke: the first 3 months. J Neurol Neurosurg Psychiatry 1985;48: Wade DT, Hewer RL, David RM, Enderby PM. Aphasia after stroke: natural history and associated deficits. J Neurol Neurosurg Psychiatry 1986;49: Ferrucci L, Bandinelli S, Guralnik JM, et al. Recovery of functional status after stroke. A postrehabilitation follow-up study. Stroke 1993;24: Samuelsson M, Soderfeldt B, Olsson GB. Functional outcome in patients with lacunar infarction. Stroke 1996;27: Hajek VE, Gagnon S, Ruderman JE. Cognitive and functional assessments of stroke patients: an analysis of their relation. Am J Phys Med Rehabil 1997;78: Sonoda S, Chino N, Domen K, Saitoh E. Changes in impairment and disability from the third to the sixth month after stroke and its relationship evaluated by an artificial neural network. Am J Phys Med Rehabil 1997;76: Roth EJ, Heinemann AW, Lovell LL, Harvey RL, McGuire JR, Diaz S. Impairment and disability: their relation during stroke rehabilitation. Am J Phys Med Rehabil 1998;79: Tatemichi TK, Desmond DW, Mayeux R, et al. Dementia after stroke: baseline frequency, risks, and clinical features in a hospitalized cohort. Neurology 1992;42: Tatemichi TK, Foulkes MA, Mohr JP, et al. Dementia in stroke survivors in the Stroke Data Bank cohort: prevalence, incidence, risk factors, and computed tomographic findings. Stroke 1990;21: Tatemichi TK, Paik M, Bagiella E, et al. Risk of dementia after stroke in a hospitalized cohort: results of a longitudinal study. Neurology 1994;44:

6 1504 RECOVERY OF COGNITION AFTER STROKE, Hochstenbach 15. Kotila M, Waltimo O, Niemi ML, Laaksonen R. Dementia after stroke. Eur J Neurol 1986;25: Pohjasvaara T, Erkinjuntti T, Ylikoski R, Hietanen M, Vataja R, Kaste M. Clinical determinants of poststroke dementia. Stroke 1998;29: Desmond DW, Moroney JT, Bagiella E, Sano M, Stern Y. Dementia as a predictor of adverse outcomes following stroke: an evaluation of diagnostic methods. Stroke 1998;29: Prencipe M, Ferretti C, Casini AR, Santine M, Giubilei F, Culasso F. Stroke, disability, and dementia: results of a population survey. Stroke 1998;28: Meerwaldt JD. Spatial disorientation in right-hemisphere infarction: a study of the speed of recovery. J Neurol Neurosurg Psychiatry 1983;46: David RM, Skilbeck CE. Raven IQ and language recovery following stroke. J Clin Neuropsychol 1984;6: Perani D, Vallar G, Paulesu E, Alberoni M, Fazio F. Left and right hemisphere contribution to recovery from neglect after right hemisphere damage an [18F]FDG PET study of two cases. Neuropsychologia 1993;31: Vallar G, Perani D, Cappa SF, Messa C, Lenzi GL, Fazio F. Recovery from aphasia and neglect after subcortical stroke: neuropsychological and cerebral perfusion study. J Neurol Neurosurg Psychiatry 1988;51: Wade DT, Wood VA, Hewer RL. Recovery of cognitive function soon after stroke: a study of visual neglect, attention span and verbal recall. J Neurol Neurosurg Psychiatry 1988;51: Desmond DW, Moroney JT, Sano M, Stern Y. Recovery of function after stroke. Stroke 1996;27: Hochstenbach J, Anderson PG, van Limbeek J, Mulder T. Is there a relation between neuropsychological variables and quality of life after stroke? Arch Phys Med Rehabil 2001;82: Hochstenbach J, Mulder T. The role of neuropsychology in the relearning of motor skills following stroke. Int J Rehabil Res 1999;22: Duncan PW, Lai SM, Keighley J. Defining post-stroke recovery: implications for design and interpretation of drug trials. Neuropharmacology 2000;39: Hochstenbach J. Rehabilitation is more than functional recovery. Disabil Rehabil 2000;22: Jongbloed L. Prediction of function after stroke: a critical review. Stroke 1986;17: Kwakkel G, Wagenaar RC, Kollen BJ, Lankhorst GJ. Predicting disability in stroke: a critical review of the literature. Age Ageing 1996;25: Rey A. L examen clinique en psychologie. Paris: Presses Universitaires de France; Stinissen J, Willems PJ, Coetsier P, Hulsman WL. Handleiding bij de Nederlandstalige bewerking van de Wechsler Adult Intelligence Scale (WAIS). Amsterdam: Swets & Zeitlinger; Wechsler D. Wechsler Adult Intelligence Scale manual. New York: Psychological Corp; Wilson BA, Cockburn J, Baddeley AD. The Rivermead Behavioural Memory Test. Titchfield (UK): Thames Valley Test Co; Wilson BA, Cockburn J, Baddeley AD, Hiorns RW. The development and validation of a test battery for detecting and monitoring everyday memory problems. J Clin Exp Neuropsychol 1989; 11: Reitan RM. Validity of the Trail Making test as an indicator of organic brain damage. Percept Mot Skills 1958;8: Wilson BA, Cockburn J, Halligan P. Behavioural Inattention Test. Titchfield (UK): Thames Valley Test Co; Bobertag O. Uber Intelligenzprüfungen. Leipzig: Ambrosius Barth; Karpov BA, Luria AR, Yarbuss AL. Disturbances of the structure of active perception in lesions of the posterior and anterior regions of the brain. Neuropsychologia 1968;6: Friedman PJ. Clock drawing in acute stroke. Age Ageing 1991; 20: Money J. A standardized road map test of direction sense: manual. San Rafael (CA): Academic Therapy Publications; Deelman BG, Liebrand WB, Koning-Haanstra M, van den Burg W. [A test for aphasia for auditive language comprehension and verbal language use] [Dutch]. Lisse (The Netherlands): Swets & Zeitlinger; Kotila M, Waltimo O, Niemi ML, Laaksonen R, Lempinen M. The profile of recovery from stroke and factors influencing outcome. Stroke 1984;15: Enderby P, Wood VA, Wade DT, Hewer RL. Aphasia after stroke: a detailed study of recovery in the first 3 months. Int J Rehabil Med 1987;8: Ferro MJ, Crespo M. Young adult stroke: neuropsychological dysfunction and recovery. Stroke 1988;19: Weiller C, Isensee C, Rijntjes M, et al. Recovery from Wernicke s aphasia: a positron emission tomography study. Ann Neurol 1995; 37: van Zandvoort MJ, Kessels RP, Brouwer MW, Kappelle JL, de Haan EH. The feasibility and prognostic value of neuropsychological testing in the early stage after ischaemic stroke. In: van Zandvoort MJ, editor. Cognitive function following single lacunar infarct [dissertation]. Utrecht: Univ Utrecht; Skilbeck CE, Wade DT, Hewer RL, Wood VA. Recovery after stroke. J Neurol Neurosurg Psychiatry 1983;46: Wade DT, Hewer RL. Functional abilities after stroke: measurement, natural history, and prognosis. J Neurol Neurosurg Psychiatry 1987;50: Heller A, Wade DT, Wood VA, Sunderland A, Hewer RL. Arm function after stroke: measurement and recovery over the first three months. J Neurol Neurosurg Psychiatry 1987;50: Mulder T, Hochstenbach J. Adaptability and flexibility of the human motor system: implications for neurological rehabilitation. Neural Plast 2001;8: Spreen O, Strauss E. A compendium of neuropsychological testsadministration, norms, and commentary. New York: Oxford Univ Pr; p Sloan S, Ponsford J. Managing cognitive problems following TBI. In: Ponsford J, Sloan S, Snow P, editors. Traumatic brain injury: rehabilitation for everyday adaptive living. Hove (UK): Psychology Pr; p