- Top of page
Aim: To investigate the associations of Clock-Drawing Test (CDT) score with neuropsychological test scores and regional cerebral blood flow.
Method: Twenty-five patients (normal aging, n = 2; mild cognitive impairment, n = 7; Alzheimer's disease [AD], n = 16) participated in this study. Their average age was 74.8 years.
Results: CDT score correlated well with the neuropsychological test scores of Mini-Mental State Examination, Clinical Dementia Rating, immediate visual memory, delayed visual memory, and IQ evaluated by Koh's block design. CDT score also had a statistically significant correlation with the regional blood flow in the left hippocampal region as evaluated on 3-D stereotaxic region-of-interest template analysisapplied to single-photon emission computed tomography images. Using a cut-off point of 8/9 in the CDT, the high-CDT group had significantly higher delayed visual memory and IQ scores than the low-CDT group. Moreover, the high-CDT group had significantly higher regional blood flows in the left parietal, left angular and bilateral hippocampal regions than the low-CDT group.
Conclusion: CDT score correlates well with regional cerebral blood flow that is decreased in the early stage of AD.
RECENTLY, THE NUMBERS of dementia patients and elderly people in Japan have increased.1 The early diagnosis of Alzheimer's disease (AD) and the development of various therapeutic approaches to delay the progression of this disease are greatly needed to reduce the cost of care insurance. The Clock-Drawing Test (CDT) is a widely used cognitive screening tool and there have been several important reports on its psychometric properties and clinical applications.2–7 The CDT is a simple and easy test to administer even to AD patients and is also useful for screening AD patients3,5 and for monitoring cognitive decline in AD.6 But CDT cannot be used to distinguish early stage AD patients from normal elderly people.8 Nevertheless, it is easy to administer without causing undue stress to recipients, and it is a reliable screening test on some occasions. A former study found that by using a cut-off point of 8/9 in the CDT, normal aging and AD patients can be well differentiated,9 but the AD group had both high- and low-CDT subjects, as reported previously.9,10 The reason why such CDT performance difference occurs even in the same AD patient group is still unknown.
In contrast, the cerebral blood flow of early stage AD patients was characterized by reduction in the posterior cingulate and parieto-occipital regions.11,12 To date there have been only a few reports on the relationship between CDT score and regional cerebral blood flow.10,13 The previous reports demonstrated that CDT performance impairment correlates with the regional cerebral blood flow reduction of left posterolateral temporal lobe.10,13 Regardless of CDT performance requiring spatial construction, semantic memory and performance ability,10 there are no reports showing the association between impairment of CDT performance and reduction of parietal or frontal lobe blood flow. Moreover, there are no reports comparing the regional cerebral blood flow between high- and low-CDT-score groups divided using a cut-off point, to the best of our knowledge.
In the present study we examined the association of CDT score with scores in other neuropsychological tests and regional cerebral blood flow of early stage AD, mild cognitive impairment (MCI), and normal-aging patients. Moreover, we examined the psychometric characteristics and regional cerebral blood flow of patients in high- and low-CDT groups.
- Top of page
As shown in Table 2, CDT score had a moderate positive correlation with MMSE, immediate and delayed visual memory and IQ scores. CDT score also had a moderate negative correlation with total CDR score and the general grading score of CDR, which increase as dementia stage advances. From the MMSE results, figure imitation had a strong positive correlation (r = 0.79) with the CDT score. Time orientation, attention and calculation, delayed recall memory and language all showed moderate correlation with the CDT score. From the CDR results, memory, orientation, judgment and social relationship showed negative correlation with the CDT score. But home and hobbies, and personal care items of CDR were not statistically significantly associated with CDT.
Table 2. Correlation for CDT score and other neuropsychological tests
|Neuropsychological test||Spearman's correlation coefficient|
| Time orientation||0.39*|
| Place orientation||0.23|
| Immediate memory||0.00|
| Attention and calculation||0.42*|
| Figure imitation||0.79**|
|Total CDR score||−0.56**|
| Social relationship||−0.49*|
| Home and hobbies||−0.34|
| Personal care||−0.25|
|General grading score of CDR||−0.66**|
|Immediate visual memory||0.61**|
|Delayed visual memory||0.67**|
The correlation coefficients between the regional cerebral blood flow and CDT score are shown in Table 3. The CDT score had a moderately strong positive correlation only with the regional blood flow in the left hippocampus (r = 0.43). The CDT score and regional cerebral blood flows in the left parietal, pericallosal and bilateral angular regions had weak positive correlation.
Table 3. Correlation for CDT score and regional cerebral blood flow
|Brain region|| ||Spearman's correlation coefficient|
The histogram of CDT score for each diagnostic group is shown in Fig. 1. Both the MCI and AD group had low- and high-CDT patients. To clarify the factors affecting the CDT performance, we divided patients into high- and low-CDT groups using a cut-off point of 8/9 in the CDT. The results of the neuropsychological tests of the high- and low-CDT groups are also shown in Table 1. The high-CDT group had a higher average score in all the tests than the low-CDT group, and statistically significant differences were recognized in the delayed visual memory and IQ scores between the two groups.
Figure 1. Clock-Drawing Test (CDT) scores for (□) normal, () mild cognitive impairment (MCI) and (▪) Alzheimer's disease (AD) subjects. Both MCI and AD contain h high- and low-CDT patients, according to the 8/9 cut-off point.
Download figure to PowerPoint
The regional cerebral blood flows of all the 12 brain regions, compared between high- and low-CDT groups, are shown in Table 4. The blood flows in the left parietal and angular regions and the bilateral hippocampi and pericallosal regions of the low-CDT group were significantly reduced compared with those of the high-CDT group.
Table 4. rCBF and CDT score (means ± SD)
|Brain region||High-CDT group (n = 11)||Low-CDT group (n = 14)|
| || || || ||(mL/100 g per min)|
|Callosomarginal||38.4 ± 4.3||37.1 ± 5.4||37.6 ± 4.8||36.8 ± 5.4|
|Precentral||40.2 ± 6.0||39.0 ± 5.4||38.4 ± 4.3||37.1 ± 5.4|
|Central||42.3 ± 6.1||42.0 ± 5.7||38.6 ± 4.8||37.6 ± 5.3|
|Parietal||40.5 ± 6.6*||41.5 ± 6.3||35.0 ± 3.5*||36.5 ± 4.4|
|Angular||42.1 ± 7.0*||44.3 ± 7.3||36.7 ± 2.9*||39.3 ± 5.2|
|Temporal||37.4 ± 5.7||38.4 ± 5.5||34.6 ± 3.7||36.1 ± 4.7|
|Posterior||46.0 ± 6.4||46.9 ± 6.4||41.5 ± 4.8||42.6 ± 5.6|
|Pericallosal||42.4 ± 6.3*||42.8 ± 6.1*||37.8 ± 5.1*||37.2 ± 4.5*|
|Lenticular nucleus||46.1 ± 6.0||43.8 ± 5.4||45.1 ± 6.8||42.2 ± 6.9|
|Thalamus||43.4 ± 8.3||43.6 ± 7.6||39.5 ± 4.9||38.2 ± 6.5|
|Hippocampus||35.1 ± 5.0*||34.1 ± 5.3*||30.5 ± 5.0*||30.4 ± 5.2*|
|Cerebellum||53.3 ± 8.0||53.9 ± 7.6||48.1 ± 5.4||49.4 ± 4.9|
- Top of page
First, we examined all subjects together to investigate whether the CDT score correlates with the other neuropsychological test scores or regional cerebral blood flows. As shown in Table 2, CDT score had a moderate positive correlation with the scores in the other neuropsychological tests. In the present study the minimum MMSE score was 16 points. This means that the dementia is mild. The present study showed that CDT score was associated with the following MMSE categories: orientation, recall, attention and calculation, sentence writing and figure imitation. Moreover, CDT score had positive correlations with the immediate and delayed visual memory scores. The CDT score also had a correlation with the severity of dementia as evaluated using CDR and MMSE. These results are in agreement with those of a previous study13 and suggest that CDT reflects memory and other cognitive functions. In the present study we limited the subjects to only AD, MCI and normal aging subjects to exclude vascular factors and other regional cognitive impairments. The results suggest that mild but substantial cognitive impairment was detected in the early stage of AD as previously reported.24–27 The present result showing that CDT score correlates with immediate and delayed visual memory scores suggests that CDT score may be affected by memory and construction ability. The present study also showed that CDT score has a positive moderate correlation with left hippocampal blood flow. Previous studies demonstrated that reduced hippocampal regional blood flow correlated with memory impairment.26,28,29 The present results also indicate the relationship between CDT score and memory impairment in AD subjects. Previous reports demonstrated mild dementia (CDR = 1) was associated with CDT impairment, but very mild dementia (CDR = 0.5) was not associated with CDT impairment using the originally reported cut-off point.8 Therefore, CDT impairment may occur after memory impairment occurs. AD pathology begins with the hippocampal region and spreads to the temporal lobe and parietal lobe.24 Therefore the present result that CDT score was associated with memory impairment and the reduction of hippocampal blood flow was compatible with the AD time course.
We also compared the high- and low-CDT patients to determine the differences in cognitive function between these two groups. As we expected, the high-CDT group had better cognitive functions than the low-CDT group (Table 1). But statistically significant differences were detected only for delayed visual memory and IQ scores measured using Koh's block design. These findings suggested that decreased performance in CDT was associated with the impairments of memory and cognitive functions in AD subjects. In other words, the low-CDT group could have more advanced AD than the high-CDT group because AD starts with memory impairment and progresses to the other impairments of cognitive functions such as construction ability. Therefore we should consider this possibility when we investigate the regional cerebral blood flow in the low-CDT group. Regional cerebral blood flow had some differences between the high- and low-CDT groups. The blood flow in the left parietal and angular regions and bilateral hippocampi of the low-CDT group was significantly lower than that of the high-CDT group. Previous studies showed that the cerebral blood flows in the parietal and angular regions correlate with figure design ability or spatial construction ability.30 Damage to the left angular region of the brain causes agraphia and impaired spatial recognition.31 Parietal lobe damage also causes impaired spatial construction ability.31 CDT is one of the psychological tests for evaluating the construction ability of a patient.31 Therefore, the association between CDT score and the reduction in cerebral blood flows in the left parietal and angular regions of the brain may be important for assessing cognitive function. Degenerative changes in the hippocampal region strongly correlate with memory impairment and AD stage.24,32,33 Medial temporal region and posterior cingulate hypoperfusions were observed in a non-demented person with memory impairment and in AD patients.11,34 AD converters also had parietal and anterior cingulate hypoperfusions.11,34 Taken together, the medial temporal and hippocampal regions are highly associated with memory function, and the degeneration of these regions occurs first in AD. Ueda et al. reported that CDT performance could be predicted from the left posterior temporal blood flow.13 Nagahama et al. reported that CDT performance is closely associated with regional blood flow in the left posterior region of the brain.10 The present results are partially similar to those of these studies. But these previous studies did not show any correlation between the CDT score and the hippocampal blood flow. The scoring system used by previous researchers provides an outline of the clock to be drawn; but the present scoring system provided no outline. Thus, the present patients needed to recall the appearance of a clock under the present scoring system. Nagahama et al. and Rouleau et al. reported that the CDT scores of AD patients were significantly improved on the copy conditions relative to the drawing-to-command condition and that semantic memory deficit may partly contribute to the impairment of CDT performance.10,16 Therefore, this memory recall system may affect the CDT performance, and the reduction of hippocampal blood flow in the low-CDT group may suggest that memory contributes to the CDT performance of AD patients. Further studies using an immediate system for detecting regional cerebral blood flow such as functional magnetic resonance imaging should be performed to clarify this point because of the paucity of CDT studies. The comparison study of the regional cerebral blood flow between high- and low-CDT groups under the condition of controlling the MMSE and IQ score may provide useful information for the mechanism of poor CDT performance.
In general medical practice, CDT is widely used and has a sensitivity of approximately 70–75% and a specificity of approximately 65–79% in AD patients.35,36 But in situations in which only a few affected patients are to be screened, CDT is not effective.8,35 The present study and previous studies demonstrated that CDT score correlates well with blood flow in the brain regions that are associated with memory and spatial recognition in AD patients.10,13 Moreover, temporoparietal blood flow reduction is more predictable than hippocampal blood flow reduction in AD patients.12,37,38 Therefore, if CDT is used together with other sensitive memory and cognitive evaluation tests, CDT can be easily administered and may serve as a simple tool for detecting memory and cognitive impairments in the early stage of AD. The present sample size was small, thus caution must be used in generalizing these findings to other patients with dementia. Further studies are necessary to confirm the association between CDT score and regional cerebral blood flow in AD patients. Moreover, the effectiveness of CDT for evaluating patients with other types of dementia should be determined.