Results of this investigation indicate that higher levels of cerebrovascular risk are associated with poorer neuropsychological functioning among adults with MDD. We found that greater levels of CVRFs, including diabetes, hypertension, and smoking, were associated with poorer performance on measures of executive functioning and working memory. Higher levels of HDL cholesterol, in contrast, were actually associated with better executive functioning. Three noninvasive measures of presymptomatic vascular dysfunction—IMT, FMD, and GTN-D—were also associated with decrements in neuropsychological performance. Increasing levels of IMT, an index of subclinical atherosclerosis, were associated with poorer executive functioning but were not associated with working memory after controlling for CVRFs. Poorer endothelial function, indexed by lower FMD, was associated with poorer executive functioning and working memory after controlling for both CVRFs and IMT. Similarly, poorer overall vascular response, indexed by lower GTN-D, was associated with poorer executive functioning after controlling for CVRFs and IMT, although GTN-D was not significantly related to working memory.
Previous population studies have shown that greater levels of CVRFs may be associated with a poorer cognitive functioning (44
). In a sample of approximately 11,000 individuals from the Atherosclerotic Risk in Communities study, Knopman and colleagues found that the presence of diabetes and hypertension at baseline testing were associated with more rapid cognitive decline (44
). Similarly, Elias and colleagues (3
) found that both hypertension and diabetes were associated with poorer neuropsychological performance on a variety of measures, most notably memory. Furthermore, Elias and colleagues (5
) found that CVRFs, measured using the FSRP, were associated with poorer performance across a range of neuropsychological tests, including measures of visual-spatial and executive functioning. More recently, Sheline and colleagues (28
) found that higher levels of CVRFs were associated with poorer processing speed, memory, and executive functioning in a study of individuals with late-onset depression.
Neuropsychological deficits (14
) and CVRFs (9
) are common among individuals with MDD and are associated with cognitive impairment in the general population (45
). Individuals with MDD have been shown to exhibit neuropsychological deficits over a broad range of cognitive domains (47
), particularly on tasks associated with frontal lobe activity, such as executive functioning (9
). Deficits in executive functioning have also been demonstrated among individuals with vascular depression, a late-onset subtype of MDD that is particularly resistant to conventional therapeutic techniques (9
) and is associated with ischemic damage to the prefrontal-subcortical loops that control executive functions (50
The existing literature on the relationship between serum cholesterol and neuropsychological performance has yielded mixed results. Elias and colleagues found that higher levels of total cholesterol were associated with better neuropsychological performance in the Framingham Heart Study (52
). Interestingly, Muldoon and colleagues also reported a positive relationship between total cholesterol and fluid intelligence levels, whereas higher cholesterol was associated with poorer scores on measures of crystallized intelligence (4
). Generally, previous studies among older individuals have noted a protective effect of elevated HDL level (53
), although the relationship between LDL and cognitive functioning is less clear (52
). Discrepant findings may result, in part, from the dual role of cholesterol as a risk factor for atherosclerosis as well as an agent for delivering nutrients necessary for cognitive functioning (55
). Interestingly, several previous studies have reported lower levels of serum cholesterol among individuals with depression (56
), although not all studies have supported this finding (62
Our observation that higher IMT is associated with poorer neuropsychological functioning is consistent with previous studies demonstrating an association between greater IMT and lower performance on tests associated with executive functioning (22
). Auperin and colleagues (23
) found that higher levels of IMT measured from the common carotid artery were associated with lower scores on the DSST and COWAT among men, whereas no consistent relationship was noted in women. Mathiesen and colleagues (22
) also found that common carotid artery stenosis was associated with poorer scores on a range of neuropsychological tests including the DST, Verbal Paired Associates immediate recall, and Trail Making Tests, although results were strongest for TMT-B. Notably, these associations were not significantly attenuated after controlling for white matter hyperintensities measured by magnetic resonance imaging. In contrast, Knopman and colleagues (44
) found that, although cognitive decline was associated with greater CVRFs, carotid IMT was not associated with more rapid decline in neuropsychological performance. Several previous pharmacological interventions have also reported an association between improved vascular functioning and increases in overall cognitive performance (63
). Furthermore, Muller and colleagues (65
) recently reported that higher pulse-wave velocity was associated with poorer performance on cognitive tests assessing processing speed and executive functioning in a sample of middle-aged and older men.
Results from our study indicate that subclinical levels of atherosclerosis, as measured by IMT, FMD, and GTN-D, may be associated with increased neuropsychological deficits. The finding that FMD and GTN-D were significantly associated with executive functioning after controlling for IMT is not surprising, given that IMT is a structural manifestation of systemic vascular disease, whereas endothelial and smooth muscle dysfunction precede the development of IMT (66
) and are potentially more sensitive indexes of early atherosclerotic disease (67
). The attenuation of the IMT and executive functioning relationship after controlling for FMD is therefore understandable, given the shared variance between these measures and the higher relative sensitivity of FMD. Furthermore, because impaired vascular function, assessed by FMD and GTN-D, may serve as early markers of a generalized vascular burden that precedes the development of manifest atherosclerotic disease, our findings may have important implications for understanding the relationship of subclinical vascular disease and neuropsychological performance.
The finding that CVRFs are associated with reduced performance on tasks of executive functioning may be particularly important among individuals with MDD because poorer executive performance has been associated with an increased risk of relapse among individuals with vascular depression as well as MDD (68
). Previous studies have reported that executive functioning is the only domain of cognitive functioning shown to differentiate depressed individuals in remission from healthy controls (69
) and has been associated with nonresponsiveness to pharmacologic treatment in patients with MDD.
Deficits in executive functioning may be the result of damage to the frontal-subcortical circuits of the brain (14
), a series of discrete pathways supporting executive cognitive functions (70
). This region is particularly vulnerable to ischemic injury due to its relatively long, small-diameter, penetrating branches deriving from the anterior and middle cerebral arteries (14
). Damage in this region could result in multiple deficits in executive functions because multiple areas of the frontal cortex are perfused by the same “watershed” arteries, thereby conferring a greater risk of microvascular injury during a state of hypoperfusion. Because ischemic damage to this area has been associated with the development of vascular depression (21
), this may have important implications for individuals with MDD. Furthermore, dysregulation between the prefrontal cortex and the limbic system has been implicated in the pathogenesis of MDD (71
) such that damage to this area might be associated with decreased regulatory functioning.
This study has several limitations. First, because we used a cross-sectional design, we are unable to determine if there is a causal relation between CVRFs and neuropsychological deficits. Second, we used a questionnaire to quantify CVRFs and relied on patient self-reports to document the presence of coronary heart disease (CHD) and diabetes. It is possible that other individuals with occult CHD or diabetes might have gone unidentified in our analysis. Third, the possibility exists that the association between MDD and impaired neuropsychological performance is due to a common pathway; for example, both may be a result of hypothalamic-pituitary-adrenal axis dysregulation, which has been associated with both MDD and neuropsychological deficits (72
). Finally, due to the limited range of depressive symptoms resulting from our inclusion criteria, it is unclear how depression severity may affect the relationship between neuropsychological performance, CVRFs, and vascular health among individuals with subclinical levels of depression.
In summary, higher levels of CVRFs were associated with deficits in neuropsychological functioning among individuals with MDD. Moreover, impaired vascular function predicted deficits in neurocognitive performance over and above the contribution of higher CVRFs. Future studies should investigate prospectively the relationship between CVRF, vascular functioning, and neuropsychological outcomes among individuals with MDD to determine the natural history of age-related declines in neurocognitive functioning, particularly in the context of impaired vascular function and CVRFs. Furthermore, the extent to which neurocognitive function may be improved post MDD treatment needs further study.