Longitudinal Association of Carotid Plaque Presence and Intima-Media Thickness With Depressive Symptoms in the ElderlySignificance
The Three-City Study
Objective—To investigate prospectively whether subclinical vascular disease is associated with future depressive symptoms in the elderly.
Approach and Results—A multicenter cohort of community-dwelling individuals aged 65 to 85 years was examined for carotid plaque presence and common carotid artery intima-media thickness at baseline and followed up after 2, 4, 7, and 10 years. At baseline and follow-up examinations, depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D). High level of depressive symptoms was defined as a CES-D score >16 in men and >22 in women. Among 4125 participants (58% women) at baseline, men more frequently showed carotid plaque presence and had higher mean common carotid artery intima-media thickness than women. After adjustment for major cardiovascular risk factors, carotid plaque presence was associated with a higher CES-D score at the 10-year follow-up in men (+1.46; 95% confidence interval, 0.71–2.20; P<0.001), but not in women. When restricting analyses to individuals without cardiovascular disease at baseline, carotid plaque presence increased the likelihood of high level of depressive symptoms at follow-up examinations in men (odds ratio, 1.47; 95% confidence interval, 1.06–2.05; P=0.022), but not in women. One SD increase in log-transformed common carotid artery intima-media thickness was associated with a higher CES-D score at the 10-year follow-up in women (+0.55; 95% confidence interval, 0.16–0.95; P=0.006) and men (+0.40; 95% confidence interval, 0.02–0.78; P=0.037). Common carotid artery intima-media thickness did not increase the likelihood of high level of depressive symptoms at follow-up in both sexes.
Conclusions—Subclinical vascular disease is associated with the progression of depressive symptoms in elderly men and women and the occurrence of high level of depressive symptoms in elderly men.
Clinically relevant depressive symptoms are common in the elderly population and have been associated with the onset and prognosis of cardiovascular disease (CVD).1,2 Unraveling the mechanisms contributing to depression development in late life may identify targets for intervention and thus carry substantial implications for CVD prevention. Various age-related processes possibly contribute to the development of depression among the elderly.3 In particular, vascular disease has been postulated to increase the risk of late life depression.4 This so-called vascular depression hypothesis is supported by neuroimaging studies describing positive associations of cerebral white matter lesions with future depressive symptoms.5,6 Although the origin of these cerebral hyperintensities in late life depression is thought to be vascular, it may also be heterogenous.7 Therefore, additional and more specific evidence on early markers of vascular disease with regard to depression development is needed. Carotid plaque and carotid intima-media thickness represent 2 distinct biologically and genetically determined markers of subclinical atherosclerosis and arteriosclerosis,8,9 which can be measured inexpensively and noninvasively by ultrasound imaging. Carotid plaque reflects the cumulative exposure to atherosclerotic risk factors, whereas carotid intima-media thickness is more the result of aging and accumulated exposure to blood pressure. To date, few studies have investigated the relationship between subclinical vascular disease and depression in older adults: 1 cross-sectional study showed an association for generalized extracoronary atherosclerosis,10 but this finding was not confirmed by the only prospective study.11 Because depressive states change over time in the elderly,12 repeated measurements of depressive symptoms may allow to better examine their relationship with subclinical vascular disease. We, therefore, investigated the association of carotid plaque presence (CPP) and common carotid artery intima-media thickness (CCA-IMT) with future depressive symptoms in elderly individuals from the Three-City Study. The objective of our study was to examine whether carotid plaque and intima-media thickness are prospectively related to depressive symptoms evaluated ≤5 times during a period of 10 years.
Materials and Methods
Materials and Methods are available in the online-only Data Supplement.
The study population included 4125 (93.0%) of eligible participants at baseline with depressive symptoms measured at least once at follow-up examinations. Mean age was 73.41 (SD, 4.87) years at baseline and 57.8% were women (Table 1).
Subclinical Vascular Disease at Baseline
At baseline, men more frequently showed CPP (54.0% versus 43.3%; P<0.001) and had higher mean CCA-IMT (0.73 versus 0.70 mm; P<0.001) than women (Table 1). A history of CVD at baseline was present in 14.6% of men and 4.3% of women (P<0.001).
Depressive Symptoms at Baseline and Follow-Up
Women had a higher median Epidemiologic Studies Depression Scale (CES-D) score than men at baseline (8 versus 5; P<0.001). Mean CES-D scores at baseline and follow-up and the frequency of high level of depressive symptoms (HLDS) and antidepressant use at follow-up examinations are shown in Table I in the online-only Data Supplement.
In light of higher CES-D scores and lower levels of CPP and CCA-IMT in women than men at baseline, as well as different cutoffs for HLDS in men and women (CES-D >16 and >22, respectively), longitudinal associations were quantified by sex.
Longitudinal Associations With CES-D Score
The Figure shows longitudinal associations of baseline CPP and CCA-IMT with CES-D score at follow-up examinations. Baseline CPP was associated with a significantly higher CES-D score at the 10-year follow-up in men (+1.46; 95% confidence interval [CI], 0.71–2.20; P<0.001), but not in women after adjustment for age, study center, antidepressant use, and major CVD risk factors. Furthermore, 1 SD increase in log-transformed baseline CCA-IMT was associated with a significantly higher CES-D score at the 10-year follow-up in women (+0.55; 95% CI, 0.16–0.95; P=0.006) and men (+0.40; 95% CI, 0.02–0.78; P=0.037). Results remained largely unchanged when further adjusting for cognitive performance at baseline and incident dementia, coronary heart disease and stroke during follow-up, and when restricting analyses to individuals without a history of CVD at baseline (Table II in the online-only Data Supplement).
Longitudinal Associations With HLDS
Table 2 presents longitudinal associations of baseline CPP and CCA-IMT with HLDS at follow-up examinations. Baseline CPP increased the likelihood of HLDS at follow-up in men (odds ratio, 1.36; 95% CI, 1.02–1.81; P=0.035), but not in women after adjustment for age, study center, and antidepressant use (model 1). The association among men was borderline significant when further adjusting for major CVD risk factors (odds ratio, 1.31; 95% CI, 0.99–1.75; P=0.062; model 2) and significant after exclusion of individuals with CVD history (odds ratio, 1.47; 95% CI, 1.06–2.05; P=0.022; model 5). In both sexes, no significant association was found between baseline CCA-IMT and HLDS at follow-up examinations.
When testing for effect modification by sex in pooled analyses, a borderline significant interaction with CPP (P=0.052) was observed in the association with HLDS.
In this study, we report on the association of CPP and CCA-IMT with future depressive symptoms in community-dwelling elderly who had no HLDS and who were free of current and past major depressive episodes at baseline examination. Baseline CCA-IMT was associated with a higher CES-D score at follow-up in women and men. Baseline CPP was associated with a higher CES-D score and an increased likelihood of HLDS at follow-up in men only. The association of CPP with HLDS in men was particularly significant and independent of major CVD risk factors among individuals without a history of CVD. Overall, subclinical vascular disease was associated with the progression of depressive symptoms in both sexes and the occurrence of HLDS in men.
Progression of Depressive Symptoms and Occurrence of HLDS
Our study results about the association of baseline CPP and CCA-IMT with the occurrence of HLDS complement recent findings of the only previous study that prospectively investigated this relationship. Within the Rotterdam Study, generalized extracoronary atherosclerosis including CPP and CCA-IMT did not increase the risk of incident HLDS.11 In contrast to this study, however, only 1 interim examination was performed to evaluate the presence of HLDS during a follow-up period of 6 years. Thus, we extend the results of this previous study using a much longer follow-up with ≤5 repeated assessments of depressive symptoms. Our results hint toward an effect of subclinical vascular disease on the occurrence of HLDS in men.
On the basis of our present data, some issues may explain the lack of a significant association of CPP with CES-D and HLDS among women. Although the majority of individuals under study were women, the prevalence of CPP was substantially lower in women than in men. This suggests that we may have lacked power in the analysis among women. A significant association of CPP with CES-D score was observed at the 7-year follow-up only. Whether this finding is spurious or reflects a time-dependent effect of CPP on the CES-D score remains to be investigated. That the association disappeared at the 10-year follow-up might be explained by the higher attrition rate at this examination.
We acknowledge that CPP does not capture the entire prognostic information provided by carotid plaque. Other quantitative markers such as plaque area and volume are more predictive of CVD and may have shown a significant association with depressive symptoms in women.13 However, this information was not available in this study. Finally, despite differences in exposure, a true sex difference in the association between CPP and depressive symptoms represents an alternative explanation for our findings.
Several pathophysiological pathways may underlie the observed association of CPP and CCA-IMT with depressive symptoms.
First, and in accordance with the vascular depression hypothesis, vascular disease because of atherosclerosis and arteriosclerosis may cause depression in late life. CCA-IMT and CPP predict future CVD events including stroke,14,15 which could be on the pathway between subclinical vascular disease and depressive symptoms; however, significant associations among individuals without a history of CVD at baseline independently from incident CHD and stroke indicate that the relationships unlikely reflect a psychological manifestation from overt CVD.11
Second, reverse causality may be an issue, in that depression preceded the development of CPP and the increase of CCA-IMT. Although we excluded participants with past and current depressive episodes, we cannot completely rule out misclassification related to the scales we used and, therefore, the possibility of such episodes before study inclusion.16 However, it is unlikely that such misclassification would be related to baseline CPP and CCA-IMT.
Third, residual confounding may account for the observed association, for instance by health conditions related to both CPP and CCA-IMT, as well as depressive symptoms. In this regard, cognitive impairment could contribute to the progression of depressive symptoms and prompt a lifestyle that favors increased CCA-IMT and plaque development. However, participants with prevalent dementia were excluded from the study and adjustment for global cognitive performance at baseline and incident dementia during follow-up did not attenuate the results.
Finally, data supporting a link between vascular disease and depression suggest that the relationship is bidirectional. This study, however, relates to the vascular depression hypothesis postulating that vascular disease increases the risk of late life depression. We thus examined the corresponding exposure–outcome relationship in community-dwelling elderly.
Taken together, we observed robust longitudinal associations of CPP and CCA-IMT with future depressive symptoms suggesting that subclinical vascular disease may be on the pathophysiological pathway toward depression in late life.
A positive association of CPP and CCA-IMT with depressive symptoms may have major public health relevance, particularly in light of the aging population. However, before direct implications can be drawn from our study results, they need to be confirmed by independent studies that should include larger population samples, reassess CPP and CCA-IMT during follow-up, measure additional markers such as plaque area and volume, and investigate clinical depressive disorders. If confirmed, our findings may suggest that prevention of CPP and of its risk factors may forestall the progression of depressive symptoms and the occurrence of HLDS in late life. This may constitute complementary approaches for the prevention of late life depression and the reduction of CVD risk: similarly to patients at high risk of CVD, a systematic evaluation of depressive symptoms may need to be considered among elderly individuals with subclinical atherosclerosis and arteriosclerosis.17,18
Strengths and Limitations
This study has several strengths, including the length of follow-up with repeated assessments of study participants, the standardized measurement of CPP and CCA-IMT, the use of a validated instrument to evaluate depressive symptoms, and the adjustment for many potential confounding factors. Among study limitations, we note that measures of CPP and CCA-IMT at a single point in time disregard the dynamic nature of vascular remodeling. We, therefore, cannot exclude the possibility of regression dilution bias attenuating the findings toward the null result. Furthermore, we lacked qualitative (echolucency and calcification) and quantitative (surface area and volume) data with regard to carotid plaque. Finally, no data were available on clinical depressive disorders using a formal diagnostic evaluation. However, in the Rotterdam Study mentioned above, extracoronary atherosclerosis was not significantly related to incident clinical depressive disorders.
Among community-dwelling elderly, carotid plaque was associated with increments in CES-D score and the occurrence of HLDS in men during a period of 10 years, whereas carotid intima-media thickness was associated with increments in CES-D score in women and men. These results lend support to the hypothesis that vascular disease may contribute to the pathogenesis of depression in late life.
Sources of Funding
The Three-City Study is conducted under a partnership agreement between the Institut National de la Santé et de la Recherche Médicale (INSERM), the Victor Segalen-Bordeaux II University and Sanofi-Aventis. The Fondation pour la Recherche Médicale funded preparation and initiation of the study. The Three-City Study is also supported by the Caisse Nationale Maladie des Travailleurs Salariés, Direction Générale de la Santé, MGEN, Institut de la Longévité, Conseils Régionaux of Aquitaine and Bourgogne, Fondation de France, and Ministry of Research—INSERM Program Cohortes et collections de données biologiques.
Dr Helmer reports financial relationships with Ipsen and Schwab, outside the submitted work. The other authors report no conflicts.
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.114.305061/-/DC1.
- Nonstandard Abbreviations and Acronyms
- common carotid artery intima-media thickness
- Center for Epidemiologic Studies Depression Scale
- carotid plaque presence
- cardiovascular disease
- high level of depressive symptoms
- Received November 26, 2014.
- Accepted March 17, 2015.
- © 2015 American Heart Association, Inc.
- Nicholson A,
- Kuper H,
- Hemingway H
- Steffens DC,
- Krishnan KR,
- Crump C,
- Burke GL
- Baldwin RC,
- O’Brien J
- Spence JD
- Plichart M,
- Empana JP,
- Lambert JC,
- Amouyel P,
- Tiret L,
- Letenneur L,
- Berr C,
- Tzourio C,
- Ducimetière P
- Mathiesen EB,
- Johnsen SH,
- Wilsgaard T,
- Bønaa KH,
- Løchen ML,
- Njølstad I
- Lorenz MW,
- Markus HS,
- Bots ML,
- Rosvall M,
- Sitzer M
- Prabhakaran S,
- Rundek T,
- Ramas R,
- Elkind MS,
- Paik MC,
- Boden-Albala B,
- Sacco RL
- Thombs BD,
- de Jonge P,
- Coyne JC,
- Whooley MA,
- Frasure-Smith N,
- Mitchell AJ,
- Zuidersma M,
- Eze-Nliam C,
- Lima BB,
- Smith CG,
- Soderlund K,
- Ziegelstein RC
- Perk J,
- De Backer G,
- Gohlke H,
- et al
This is the first study to show that subclinical vascular disease is associated with future depressive symptoms in the elderly. During a period of 10 years, subclinical vascular disease was related to the progression of depressive symptoms in men and women and the occurrence of high level of depressive symptoms in men. As postulated in the vascular depression hypothesis, vascular disease may thus be on the pathophysiological pathway to depression in late life. Our findings raise the question whether intervening on vascular risk at the stage of subclinical vascular disease and its risk factors may prevent depression in late life. Given the high prevalence of late life depression and the increased cardiovascular risk associated with depression, this is of utmost significance. Randomized controlled studies may thus investigate whether such interventions will lower the incidence of late life depression.