Background and Purpose Neuropsychiatric findings were examined in 91 patients with acute focal subcortical lesions to determine whether cognitive outcome would differ depending on whether the head of the caudate or other subcortical structures were injured.
Methods Patients were evaluated using the Mini-Mental State Examination (MMSE), Hamilton Rating Scale for Depression, and modified Present State Examination. Patients were reexamined at short-term (3 to 6 months) or long-term (1 to 2 years) follow-up.
Results There were no significant intergroup differences in the MMSE scores at the initial evaluation or at short-term follow-up. At long-term follow-up, however, patients with either right or left caudate lesions had significantly lower MMSE scores than patients with other subcortical lesions.
Conclusions A significant number of patients with caudate infarction deteriorate in their intellectual function between 1 and 2 years after stroke. This phenomenon could be mediated through disruption of cortical projections to the caudate.
Prior work has demonstrated that neuronal circuits connect prefrontal cortex to the caudate nuclei in a topographically organized manner and in turn connect the caudate to frontal cortical areas via other basal ganglia nuclei and thalamus.1 2 Specific behavioral and learning disturbances have also been shown to follow lesions of the prefrontal cortex. Presumably mediated through these cortical-caudate-cortical loops, the caudate may influence cognitive functions such as consolidation of memories. In single-unit recordings from monkey caudates, caudate cells appeared to play a role in the integration of visual information and memories.3
Patients with Huntington’s disease have subcortical degeneration, primarily in the caudate nucleus. These patients suffer not only from motor dysfunction4 but also from progressive dementia, personality change, and mood disorders.5 In positron emission tomography and SPECT studies, patients with Huntington’s disease have been found to have reduced cerebral blood flow and metabolism in the caudate and lentiform nuclei. Furthermore, a linear relationship has been found between blood flow or metabolism in the caudate nucleus and patients’ overall functional ability or cognitive capacity.6 7
Because neuropathological changes in Huntington’s disease involve structures other than the caudate nucleus, such as the putamen, thalamus, and cortex, these cognitive and behavioral changes may not be due to caudate dysfunction. Focal brain lesions, especially strokes, therefore may be more useful than Huntington’s disease for investigating the relationship between the caudate and cognitive or mood disorders. Studies of patients with caudate strokes have suggested that unilateral or bilateral damage may result in significant memory impairment, mood change, or a frontal behavioral syndrome.8 9 10 11 12 There are few studies, however, of patients with unilateral caudate lesions that have examined long-term outcome of memory and mood disorders. We therefore investigated the 1- to 2-year outcomes of neuropsychological function and mood disorder among patients with unilateral caudate strokes and compared them with outcomes in patients having other subcortical lesions. We hypothesized that caudate lesions would lead to long-term mood and cognitive disorders with greater frequency than other subcortical lesions.
Subjects and Methods
Patients were selected from consecutive admissions for acute treatment or rehabilitative treatment after cerebral thromboembolic infarction or intracerebral hemorrhage. Detailed neurological and psychiatric examinations were performed on all patients, excluding only those who had severe comprehension deficits or markedly decreased level of consciousness. For the present study, patients were included only if they had a single lesion demonstrated by CT scan that demonstrated an ischemic or hemorrhagic infarction restricted to basal ganglia, thalamus, or cerebral white matter; follow-up data on outcome was obtained at either short- or long-term follow-up. Patients with cortical lesion or brain stem lesions were excluded. A total of 91 patients met the selection criteria. Lesion volume (expressed as a percentage of total brain volume) was calculated from the ratio of the largest cross-sectional area of the lesion on any CT scan slice to the cross-sectional area of the whole brain on the slice passing through the body of the lateral ventricle.13
Patients were classified into four groups. A total of 9 patients had left caudate lesions (left caudate lesion group), 12 had right caudate lesions (right caudate lesion group), 27 had left subcortical lesions (left subcortical lesion group), and 43 had right subcortical lesions (right subcortical lesion group). Patients with caudate lesions may have had some extension outside of the caudate head, but more than half of the lesion had to be within the caudate head. Patients with other subcortical lesions had no involvement of the caudate head.
Patients were evaluated using neurological and psychiatric examinations in-hospital and at 3 to 6 months (short-term follow-up) or 12 to 24 months (long-term follow-up) after stroke.
Neurological and Psychiatric Examination
The neurological examinations and diagnoses for all patients were performed using the criteria established by the National Institute of Neurological and Communicative Disorders and Stroke’s Stroke Data Bank.14 All neurological evaluations and CT scan readings were carried out blinded to the findings on the psychopathologic examination. After giving informed consent, patients were administered a series of standardized test instruments that quantitatively measured cognitive function and mood impairment. We have shown15 16 17 that these instruments give reliable and valid measurements in this stroke population. The instruments included the MMSE,18 an 11-task examination that yields a total maximum score of 30 points and has been found to be reliable and valid in assessing a limited range of cognitive functions in stroke patients19 ; scores may range from 0 to 30, and a score of 23 or below has been found to be indicative of significant cognitive impairment. The HDS,20 a 17-item interviewer-rated scale, measures psychological and physiological symptoms of depression. The modified PSE,17 a semistructured psychiatric interview, was used to assess depression and anxiety and was scored by the examiner. Using the symptoms elicited by the PSE, a diagnosis of major or minor depression was made using DSM-IV21 symptom criteria. The PSE was modified to specifically identify the presence or absence of all DSM-IV symptoms of major depression.
Statistical analysis was done using mean and standard deviations, Student’s t tests, ANOVA, and Pearson correlation coefficients for the parametric data and intergroup comparisons. Nonparametric data such as frequencies of symptoms were compared using χ2 tests.
Demographic data of the study patients are shown in Table 1⇓. The groups were not significantly different in their age, socioeconomic status,23 marital status, education, personal or family history of psychiatric disorder, time since stroke or frequency of hemorrhagic lesions, and lesion volume. There were slightly fewer women in the left caudate lesion group and slightly more African Americans in the left subcortical lesion group; however, these differences were not statistically significant (Table 1⇓).
Neurological and Neuropsychological Findings
On neurological examination (Table 2⇓), each group showed similar frequencies of motor, sensory, and visual field deficits, as well as dysarthria. Although 7 patients had language disturbances, none had severe aphasia. There was a significantly higher frequency of neglect (assessed by failure to identify one stimulus only during double simultaneous tactile auditory or visual stimulation) in the right caudate lesion patients than in the other three groups (χ2=10.7, df=3, P=.013). There were no significant differences in the frequencies of depression.
Cognitive Impairment and Depression Scores
Mean±SD in-hospital MMSE scores were 25.7±4.6 for the 6 patients with left caudate lesions who were seen at short-term follow-up, 26.1±3.2 for the 9 right caudate lesion patients, 22.6±4.4 for 14 left subcortical lesion patients, and 25.4±4.9 for the 28 right subcortical lesion patients. At 3- to 6-month follow-up, the scores for these same patients were 25.0±3.8, 25.1±3.7, 23.3±3.8, and 26.2±5.2, respectively. The mean±SD time from stroke to follow-up was 3.8±1.3 months for the subcortical lesion patients and 3.8±1.4 months for the caudate lesion patients. There were no significant intergroup differences in MMSE scores at the initial evaluation or short-term follow-up (Fig 1⇓). Because we have reported in previous publications that cognitive impairment is associated with major depression,19 24 we examined the depression scores of these patients. Mean±SD in-hospital HDS scores were 9.5±7.0 for the left caudate lesion group (n=6), 7.0±6.1 for the right caudate lesion group (n=9), 6.7±5.6 for the left subcortical lesion group (n=14), and 9.6±6.1 for the right subcortical lesion group (n=28). At 3- to 6-month follow-up, the scores were 4.0±2.4, 7.6±4.8, 8.4±5.9, and 8.7±5.8, respectively. A two-way repeated measures ANOVA (factors of caudate versus other subcortical lesion, right versus left, and initial versus follow-up) demonstrated no significant main effect and no interaction (Fig 2⇓).
The changes in diagnosis were as follows: in the left caudate lesion group, 4 of the 6 patients had an in-hospital diagnosis of depression. The 3 major depressions had all resolved at follow-up, and only 1 patient with minor depression still had a minor depression diagnosis. In the right caudate lesion group, the 2 cases of in-hospital major depression had resolved, and no patients had a depression diagnosis at short-term follow-up. Among the 14 patients with other left subcortical lesions, there were 5 major depressions and 8 minor depressions in-hospital, while at short-term follow-up there was 1 major depression and 6 minor depressions. Among the 28 right subcortical lesion patients, there were 7 major depressions and 6 minor depressions in-hospital, while at short-term follow-up there were 1 major depression and 4 minor depressions. Thus, at short-time follow-up, depressions associated with subcortical lesions had virtually all resolved. Although there were 17 major depressions and 15 minor depressions in-hospital, by 4- to 6-month follow-up there were only 2 major depressions and 11 minor depressions. This is consistent with a previous study demonstrating the rapid remission of depression in patients with subcortical lesions.25
Mean in-hospital MMSE scores were 23.8±7.9 for 5 patients with left caudate lesions who were seen at long-term follow-up, 24.7±2.9 for 6 patients with right caudate lesions, 22.3±5.3 for 12 patients with left subcortical lesions, and 26.1±4.2 for 21 patients with right subcortical lesions. At 12- to 24-month follow-up, however, the scores for the left caudate group had gone from 23.8±7.9 to 19.4±5.4, and the scores in the right caudate group had gone from 24.7±2.9 to 23.5±4.1. The right subcortical group scores had gone from 31.3±5.3 to 25.7±3.2, and those of the left subcortical group had gone from 26.1±4.2 to 27.8±3.4. A two-way repeated measures ANOVA demonstrated a significant group-by-time interaction, and post hoc analysis showed that left subcortical lesion patients had significantly improved in their MMSE scores from in-hospital to long-term follow-up (t=3.3, df=11, P<.005). On the other hand, although there was not a significant group-by-time effect for caudate lesion patients, patients with left caudate lesions had significantly lower MMSE scores at long-term follow-up than patients with other subcortical lesions (ANOVA, F=7.63; P<.0005) (Fig 1⇑).
Mean in-hospital HDS scores were 7.8±6.5 for the left caudate lesion group, 4.8±3.2 for the right caudate lesion, 6.2±4.8 for the left subcortical lesion group, and 8.1±4.7 for the right subcortical lesion group. At 12- to 24-month follow-up, the scores were 7.8±7.5, 6.7±4.7, 10.4±7.3, and 7.3±8.4, respectively. A two-way repeated measures ANOVA demonstrated no effects of caudate versus subcortical, right versus left, or time (Fig 2⇑).
The changes in diagnosis were as follows: in the left caudate lesion group, 1 in-hospital minor depression changed to major depression, and 1 in-hospital major depression changed to minor depression at long-term follow-up. In the right caudate lesion group, the 2 cases of in-hospital major depression had resolved, and 1 patient with minor depression still had a minor depression diagnosis at long-term follow-up. In the left subcortical lesion group, there were 2 cases of major and 5 cases of minor depression, while at long-term follow-up there were 2 different patients who developed major depression and 3 with new minor depressions. In right subcortical lesion patients, there were 2 cases of major and 4 cases of minor depression in-hospital, while at long-term follow-up there were 2 new cases of major depression and 2 continuing cases of minor depressions. Thus, at long-term follow-up, the original in-hospital major and minor depressions had largely resolved, but other patients had developed depression. The total number of depressions remained stable, but the patients were different.
This study found that patients with caudate lesions deteriorated in their intellectual function but not their mood over the first 1 to 2 years after stroke. Patients with other subcortical lesions tended to improve in their intellectual function over the first 2 years after stroke. This phenomenon was not related to subsequent illness or infarction in the caudate lesion group. Only 1 patient among the caudate lesion patients had another infarction during follow-up. He had right parietal lobe infarction at about 9 months of follow-up. However, his intellectual function was not influenced by the stroke. His initial MMSE was 27, at 6 months it was 27, and at 2 years it was 28.
Before further discussion of these findings, methodological limitations of this study should be acknowledged. Patients were studied using the MMSE, a brief language-dominated cognitive examination. Therefore, subtle changes in neuropsychological functions among subcortical lesion patients may have been missed. Second, this study used a significant number of inner-city African American patients with limited education; therefore, these findings may not hold for all stroke patients. We did not conduct interrater reliability studies on our lesion classification, but the evaluations were made by an experienced neurologist (H.B.) who was blind to any of the clinical findings. The number of patients in each caudate group was small (ranging from 5 to 9 at each follow-up), and this reduced the power of our analysis. It is also possible that a recurrent stroke occurred that was not detected. Patients were given neurological and psychiatric examinations, and an interval history was taken at 3, 6, 12, and 24 months of follow-up. If a recurrent stroke was suspected, a CT scan was done. Although some silent stroke may have been missed, this risk should be equally shared across all groups and not be an explanation just for the caudate lesion patients. The fact that cognitive decline was not found at short-term follow-up but was found at long-term follow-up suggests that it may have been missed because of a type II error at short-term follow-up. We found the group difference at short-term follow-up (ie, 0.14) to be outside the 95% confidence range for our group difference at long-term follow-up. Furthermore, the intergroup difference at short-term follow-up was only 0.14, and a power analysis demonstrated that even dramatic increases in N would not have demonstrated a significant difference. This supports the conclusion that cognitive decline associated with caudate infarction occurred after the first 3 to 6 months after stroke. It is also possible that our exclusion criterion of no significant comprehension deficit may have excluded some patients with caudate lesions. In addition, some transient cognitive disturbance may have been missed during the initial evaluation, since patients were examined at a mean of 38 days after stroke. It is also possible that an analysis of subscores on the MMSE would have demonstrated long-term deterioration in specific aspects of cognitive function. A more detailed neuropsychological evaluation should be done to examine this possibility. In addition, other variables associated with the development of dementia include the existence of cerebral white matter leukoaraiosis,24 cerebral blood flow,25 quality of social supports available to the patient, and age.24 Moroney et al26 demonstrated that hypoxic-ischemic disorders, associated with medical conditions such as seizures, cardiac arrhythmias, or pneumonia, were independent risk factors for developing dementia after ischemic stroke. Although there was no significant intergroup difference in age, other factors were not considered in this study and may have contributed to the deterioration in the caudate lesion patients. It is not obvious, however, how these factors would affect the caudate group more than the other subcortical groups.
The caudate has been hypothesized to play a distinct role in memory. In monkeys, for example, stimulation of the caudate has been shown to impair memory.27 Mendez et al12 reported on 11 patients with unilateral or bilateral caudate infarction and 1 patient with a healed abscess who came to medical attention because of an acute behavioral disturbance. Among these patients, 1 patient with bilateral lesions was judged to have a “global dementia” with an MMSE score of 18. Moreover, among 7 patients with unilateral lesions, there were significant impairments in tasks requiring planning and sequencing compared with age-matched control subjects. They also had short attention spans and decreased free recall of episodic and semantic items. Caplan et al2 also reported on the clinical and CT findings in 18 patients with unilateral caudate infarction. Memory impairment was observed in 2 patients who had left-sided lesions, and they were also found to be “abulic and slow.” Richfield et al11 reported a patient with bilateral caudate damage examined at 8 months from onset or 1 year later who had cognitive impairment, especially in delayed recall. Tatemichi et al28 reported that a combination of psychomotor slowing or abulia and memory impairment was the most striking behavioral feature among stroke patients with dementia. In our study, 6 of 21 patients with caudate lesions showed abulia as determined by a report of lack of drive or motivation in response to a question on the PSE, and 3 of these 6 patients showed both memory impairment and MMSE scores below 20.
Neuropsychological and SPECT studies were conducted in 2 patients at 16 months and 3 years after caudate hemorrhage.10 One of the patients showed a slight verbal comprehension deficit and mild impairment of verbal memory, and SPECT study demonstrated decreased perfusion, particularly in the frontal cortex. Pozzilli et al10 suggested that the deficit in verbal comprehension and verbal memory may have been due to dysfunction of the corticocaudate connections. Censori et al29 also reported that aphasia was often associated with poststroke dementia. In our study, there was a relatively high percentage of aphasia in the left caudate lesion group compared with the other subcortical groups. Although this may have influenced verbal memory in 2 patients, it was not an explanation for the phenomenon of cognitive deterioration after caudate stroke.
On the basis of these prior studies, the present findings might be construed to indicate that caudate lesions lead to chronic deficits in frontal lobe function. This chronic frontal lobe dysfunction gradually leads to dysfunction in other connected cortical regions, and this is manifested by cognitive decline. Other explanations such as chronic dysfunction of the caudate might also be proposed as an explanation for these findings. This phenomenon, however, is worthy of further study. Although depression was not associated with this cognitive deterioration, antidepressant treatment has been shown to improve both cognitive function and depression after stroke.30 31 If antidepressants work in part by improving frontal lobe activity, this treatment could offer a potential method of preventing the cognitive deterioration associated with caudate lesions. This and other treatment options need to be explored in future research.
Selected Abbreviations and Acronyms
|DSM-IV||=||Diagnostic and Statistical Manual of Mental Disorders, 4th edition|
|HDS||=||Hamilton Rating Scale for Depression|
|MMSE||=||Mini-Mental State Examination|
|PSE||=||Present State Examination|
|SPECT||=||single-photon emission CT|
We wish to thank Teresa Kopel who helped with some of the patient evaluations and J. Todd Kosier for suggestions on the statistical analysis.
- Received September 19, 1996.
- Revision received March 5, 1997.
- Accepted March 5, 1997.
- Copyright © 1997 by American Heart Association
Hikosaka O, Sakamoto M. Cell activity in monkey caudate nucleus preceding saccadic eye movements. Exp Brain Res. 1986;8:454-461.
Marsden C. The mysterious motor function of the basal ganglia: the Robert Wartenberg lecture. Neurology. 1982;32:514-539.
Hasselbalch S, Oberg G, Sorensen S, Andersen A, Waldemar G, Schmidt J, Fenger K, Paulson O. Reduced regional cerebral blood flow in Huntington’s disease studied by SPECT. J Neurol Neurosurg Psychiatry. 1992;55:1018-1023.
Stein R, Kase C, Hier D, Caplan L, Mohr J, Hemmati M, Henderson K. Caudate hemorrhage. Neurology. 1984;34:1549-1554.
Pardal M, Micheli F, Asconape J, Paradiso G. Neurobehavioral symptoms in caudate hemorrhage: two cases. Neurology. 1985;35:1806-1807.
Mendez M, Adams N, Lewandowski K. Neurobehavioral changes associated with caudate lesions. Neurology. 1989;39:349-354.
Robinson R, Bolla-Wilson K, Kaplan E, Lipsey J, Price T. Depression influences intellectual impairment in stroke patients. Br J Psychiatry. 1986;148:541-547.
Kunitz S, Gross C, Heyman A, Kase C, Mohr J, Price T, Wolf P. The pilot Stroke Data Bank: definition, design, and data. Stroke. 1984;15:740-746.
Robinson R, Bolla-Wilson K, Kaplan E, Rao K, Lipsey J, Price T. Evidence for intellectual impairment related to depression in stroke patients. Br J Psychiatry. 1986;148:541-547.
Hamilton MA. A rating scale for depression. J Neurol Neurosurg Psychiatry. 1960;23:56-62.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Edition 4. Washington, DC: American Psychiatric Press Inc; 1994.
Hollingshead A, Redlich F. Social Class and Mental Illness: a Community Study. New York, NY: John Wiley & Sons Inc; 1958.
Liu C, Miller B, Cummings J, Mehringer C, Goldberg M, Howng S, Benson D. A quantitative MRI study of vascular dementia. Neurology. 1992;42:138-143.
Terayama Y, Meyer J, Kawamura J, Weathers S, Mortel K. Patterns of cerebral hypoperfusion compared among demented and nondemented patients with stroke. Stroke. 1992;23:686-692.
Moroney JT, Bagiella E, Desmond DW, Paik MC, Stern Y, Tatemichi TK. Risk factors for incident dementia after stroke. Stroke.. 1996;27:1283-1289.
Rosvold H, Delgado J. The effect on delayed-alternations test performance of stimulating or destroying electrical structures within the frontal lobes of monkey’s brain. J Comp Physiol Psychol. 1956;49:356-372.
Tatemichi T, Desmond D, Prohovnik I. Strategic infarcts in vascular dementia. Arzneimittelforschung/Drug Res. 1995;45:371-385.
Censori B, Manara O, Agostinis C, Camerlingo M, Casto L, Galavotti B, Partziguian T, Servalli M, Cesana B, Belloni G, Mamoli A. Dementia after first stroke. Stroke. 1996;27:1205-1210.