Although Alzheimer's disease (AD) is a neurodegenerative disease characterized by impairment in memory, language, visuospatial function, and executive function, various accompanying neuropsychiatric symptoms are also the important aspects of this disease. Among them, depression is one of the most common psychiatric complications of AD, with 30% to 50% prevalence. In AD, depression often has different clinical manifestations when compared with early-life depression. While classical symptoms of early-life depression such as sad mood, guilt feelings, and self-depreciation are less prominent, nonspecific inner feelings, such as lack of pleasure or not feeling well, may be the primary manifestations of depression in patients with AD. What causes these differences is not yet clear. Cognition is especially primarily and persistently affected during disease progression in AD; hence, the atypical symptomatology of AD may be attributed to this impairment.

Even in cognitively normal younger or older adults, depression may be associated with cognitive impairments.12 Although the results of studies are somewhat mixed, slowed mental processing and deficits in attention and executive function seem to be more common in depressed individuals than in non-depressed individuals in the absence of dementia. Individuals with late-onset depression have more significant cognitive impairment.3 In AD, the relationship between cognitive impairment and depression remains controversial. Some previous studies have reported the negative impact of depression on general cognition,4 measures of dementia severity, working memory, processing speed,5 attention, motor functioning, visuospatial perception and construction.6 Other investigators have found no cognitive differences between AD patients with and without depressive symptoms.7 Due to this lack of a consistent relationship or a weak relationship between cognitive abnormalities and depression, it is uncertain whether depression is secondary to cognitive impairment or epiphenomena of AD.

These inconsistent results may be due to diverse study designs, and among these confounding factors, the inclusion of patients receiving psychoactive medications in the study may be critical. Depression is not only affected by antidepressants, but also by other psychoactive medications. Due to the longstanding Korean tradition of caring for dementia patients by family members, there are a considerable number of patients with mild to moderate AD who visit dementia clinics with never been on medication status. Therefore, we can build databases for drug-naïve probable AD patients, which will thus overcome this limitation.

However, we do not know much about whether depression in AD patients is associated with impaired cognitive performance beyond the effect of dementia itself. That is, if dementia severity was held constant, would AD patients with depression perform more poorly than AD patients without depression? In this study, we used the information about the stage of dementia to group people with a similar degree of dementia severity, and examined their performance on several specific neuropsychological tests. Therefore, the aim of this study was to determine whether diagnosed depression or reports of depressive features in AD patients were associated with additional cognitive impairment, particularly on timed tasks or tasks requiring effortful attention.

Cognitive impairment not only changes how we perceive ourselves and the world around us, it also changes the way we feel. Previous studies in patients with early-onset depression have focused on cognitive processes and the content of depressive cognition in an attempt to understand depression. Negative views of the self, the world, and the future, as well as recurrent and uncontrollable negative thoughts are debilitating symptoms of depression. Moreover, biases in cognitive processes such as attention and memory may not only be correlates of depressive episodes; they may also play a critical role in increasing an individual's vulnerability for depression.

Three important mechanisms have been postulated as potential links between biased attentional/memory processes and emotion dysregulation in early-onset depression. However, the true pattern of cognitive impairment in patients with early-onset depression remains a matter of controversy because most studies are modest or small in terms of their sample size, and hence cognitive deficits tend to differ between and within studies, in both their nature and severity.12

The cognitive implications of depression in patients with AD have been less studied compared to those of early-onset depression. Previous studies provide evidence for both a negative relation and no significant relation of depression with cognition. These inconsistencies in results of previous studies were mainly attributable to the study subjects (medicated vs. nonmedicated, or homogeneous AD group vs. heterogeneous dementia group). Fortunately, we could overcome these limitations due to the aforementioned reasons.

In our study, AD patients with depression showed significantly lower scores on the digit forward, digit backward, calculation, and CWST tests compared to AD patients without depression. The digit span subtest of the Wechsler Memory Scale comprises two tests, digits forward and backward. Digits forward is the simpler and less effortful condition requiring the efficiency of attention rather than memory.16 It tends to be more susceptible to the left hemisphere impairment rather than either the right hemisphere impairment or diffuse brain damage.17 Whereas digits backward is the more complex and effortful condition requiring executive abilities,18 which is supported by the factor analytic study loaded digit backward onto a "frontal factor".19 In addition, in neuroimaging studies among healthy adults, doing digits backward has been associated with greater frontal activation than doing digits forward.20 Accordingly, digits backward is regarded as a better test for assessing executive aspects of working memory.21 However, it should be noted that the manipulation of information during working memory tasks such as digits backward also requires activation of posterior brain regions (e.g., the superior and inferior parietal cortex, the superior temporal cortex), suggesting a role for non-frontal brain regions as well.22 Impairment of performance on calculation tests was reported in patients with a left parietal lesion.23

The ability to restrain irrelevant information, suppress activation of an inappropriate response triggered by associated cues; thus focusing on relevant information is the one of the important roles of executive functioning.24 This cognitive inhibition is an essential part of goal-directed behavior, and thus a part of maintaining daily functioning. To measure this cognitive inhibition, Stroop test is widely used. The Stroop task presents incongruously colored ink (e.g., the word 'red' printed in blue ink). Participants are asked to name the ink color of each word, while attempting to ignore the meaning of the word. This attempt to suppress word meaning in order to name ink color has led to longer response latencies than those that result from color naming congruent stimuli (e.g., the word 'red' printed in red ink), a phenomenon that has been referred to as the Stroop effect. In the first neuropsychological study with the Stroop test in patients with focal brain lesions, impairment of performance on the Stroop test was associated with a left dorsolateral prefrontal lesion.25 Another study showed that patients with left dorsolateral lesions showed impaired performance on the Stroop test.26 Moreover, superior medial lesions, particularly involving the right supplementary motor area, were also reported to be associated with errors during the interference condition of the Stroop task.27

Considering the impairment of performance on the digit forward, backward, calculation and CWST tests, our study suggested that the frontal, left hemisphere and the left parietal dysfunctions were associated with depression. In studies of major depression without dementia, the medial prefrontal areas such as the anterior cingulate cortex, orbitofrontal cortex, and also the dorsolateral prefrontal regions, ventral striatum (caudate and putamen), middle temporal regions and limbic structures are the suggested candidates for affective regulation areas. For detailed anatomical localization, neuropsy-chological tests are of limited value, but our study suggested similar anatomic areas.2428

Correlation analysis between GDS15 and specific cognitive neuropsychological tests showed a significant correlation with these neurocognitive tests. These findings imply that these specific cognitive neuropsychological tests might be not only a trait marker, but also a state marker. A trait marker shows the properties of a certain symptom that plays the role of an antecedent and possibly plays a causative role in the pathophysiology of the symptom, whereas a state marker indicates the status of clinical manifestations in patients. Although it is not possible to elucidate the cognitive impairments that cause depression or epiphenomenon, certain pathophysiological anatomy may be involved in the depressive symptomatology of AD.

However, this study had several limitations. First, this study is a retrospective study and sample size is relatively small. Secondly, depression was defined with the GDS15, a questionnaire that reflects only the patient's subjective symptoms based on the questions asked by the examiner. AD patients frequently deny having depressive symptoms. Therefore, the assessment of depression based on patient self-reports could lead to biased outcomes. Finally, our study is a hospital-based study, and hence, it may not represent the real community.

In conclusion, our study shows that AD patients with depression have a disproportionate cognitive deficit, suggesting frontal (especially in the left dorsolateral), left hemisphere and left parietal dysfunction. Considering the neuropsychological differences between AD patients with depression and without depression, depressive symptoms may have specific anatomic substrates. In a future study, a higher number of drug-naïve patients and more specific tests that represent specific anatomical substrates may provide a better understanding of the biological pathophysiology of depression in patients with AD.