Prospective associations of major depressive disorder with atherosclerotic cardiovascular disease, cognitive impairment and mortality in the community

Background: Major depressive disorder (MDD) has been consistently found to be prospectively associated with atherosclerotic cardiovascular diseases (ASCVD) and cognitive decline.

Given the high lifetime-prevalence of MDD, its strong links with ASCVD and cognitive decline, two major disease burden, have major public health significance.

However, the underlying mechanisms of these links have not been adequately studied with repetitive measures and simultaneous assessment of multiple mechanisms.

Moreover, mechanisms are likely to vary across MDD subtypes, which have shown differential associations with cardiovascular risk factors (CVRF).

Aim: To conduct a fourth follow-up (FU4) evaluation of our population-based cohort approximately 4-years after the last assessment to gain a better understanding of the associations of MDD with ASCVD, cognitive decline and death related to ASCVD and all-cause mortality as well as the role of factors that could be involved in these associations.

Although rates of ASCVD or cognitive decline are still low, this follow up is likely to detect a steep increase of these outcomes.

The proposed data collected at FU4 will allow us to address the following questions: 1) Does MDD or its subtypes (atypical, melancholic, unspecified) increase vulnerability to ASCVD (ischemic heart disease, stroke and peripheral artery disease), ASCVD death or all-cause mortality? 2) Does MDD or its subtypes increase vulnerability to subsequent cognitive decline (changes of cognitive test scores, minor cognitive impairment)? 3) Are sleep and activity patterns, inflammation, HPA axis regulation, diet and cardio-metabolic factors involved in the association of MDD or its subtypes and ASCVD or cognitive decline either as mediators of a causal relationship or as shared etiological factors?

We hypothesize that both atypical and melancholic MDD are associated with ASCVD or cognitive decline, but that differential mechanisms may underlie these associations.

Methods: We propose to perform a comprehensive 4th follow-up (FU4) of the elderly participants (year of birth 1930-1955; n=3881) of the CoLaus|PsyCoLaus cohort, which was randomly selected between 2003 and 2006 from the residents of the City of Lausanne.

Participants underwent four thorough physical, biochemical and genetic (GWAS) evaluations of CVRF and ASCVD as well as comprehensive psychiatric assessments including a semi-structured diagnostic interview, cognitive tests and MRI. At the end of 2020, 9.5% of the cohort were deceased. FU3, which was slightly delayed by the covid pandemic will be completed by spring 2021.

At each evaluation we have elicited information on behavioral, physiological, and metabolic factors that may be involved in the association between MDD and ASCVD. In addition, genetic information is available on all participants (Affymetrix GWAS).

A particular focus was placed on actigraphy in combination with ecological momentary assessments, which provide an objective measure of activity and sleep patterns.

FU4 (expected n=2100 for the physical and n=1600 for the psychiatric evaluations) will include the measures of previous assessments.

For incident ASCVD, dementia or death, information will be elicited from medical records or from the Swiss National Death Registry.

Interoperability of data is assured by integration into the Maelström meta-data catalogue through a SPHN driver project.Main expected outcome: The proposal will extend current knowledge on the complex associations of the heterogeneous category of MDD with ASCVD and cognitive decline by 1) establishing the specific prospective associations of subtypes of MDD with ASCVD cognitive decline, and 2) characterizing the role of modifiable factors involved therein, which will contribute to the development of more targeted strategies of prevention of both ASCVD and cognitive decline.Potential relevance: Our study has a series of unique and novel features including 1) comprehensive psychiatric phenotyping of the full range of mental disorders and subtypes of MDD in a prospective population-based cohort, 2) repetitive assessment of depressive episodes and their timing that can elucidate their dose-response relationship with ASCVD and cognitive decline, 3) repeated and simultaneous assessments of factors potentially involved in the association of MDD with ASCVD and cognitive decline enhancing our ability to characterize their independent roles, 4) availability of complementary genetic and familial aggregation data allowing us to triangulate causality by combining different approaches, and 5) combined objective measures of sleep and physical activity by actigraphy with subjective tracking of life context and emotional states that can characterize dynamic processes to supplement traditional clinical measures.