P4401Depressive symptomatology, sex and 10-year cardiovascular disease; revealing the mediation ranking of lifestyle, sociodemographic and clinical factors in primary and secondary prevention spectrum

Abstract Background/Introduction Brain and heart interplay is highly discussed in healthy yet even more in cardiac population with inconclusive evidence regarding the sex-related interactions. Purpose The sex-specific effect of depressive symptomatology (DS) on 10-year first and recurrent cardiovascular disease (CVD) events was evaluated. Methods The samples of two cohorts were used; n=845 free of CVD males and females (2002–2012) and n=2,172 males and females with acute coronary syndrome (ACS) (2004–2014) with baseline psychological assessments (Zung Self-Rating Depression Scale and Centre of Epidemiological Studies-Depressive symptoms scale, respectively) were used. The percentage of excess mediated risk (PEMR) and the corresponding 95% Confidence Interval (95% CI) were used to evaluate the mediating effect of various factors on the examined association. Results ACS as well as free-of-CVD females scored significantly higher for DS. Males exceeded females against first (19.7% vs. 11.7%, p<0.001) and subsequent CVD events (38.8% vs. 32.9%, p=0.016) while in participants with DS a male-to-female first and recurrent CVD event rate ratio below -1- was noticed. Multivariate Cox regression analysis revealed that DS remained an independent aggravating factor for first (Hazard Ratio (HR) = 2.72, 95% Confidence Interval (95% CI) 1.50, 9.12, p=0.01) and recurrent (HR=1.31, 95% CI 1.01, 1.69, p<0.001) CVD events only in females. Mediation analysis in females revealed that 35% (23%, 44%) of excess first-CVD-event risk of DS was attributed to lifestyle, sociodemographic, clinical, anthropometric factors as well as lipid and inflammatory markers. The respective number for recurrent events was 46% (23%, 53%). In free-of-CVD females, factors mostly accounted for excess DS risk, close to the overall, were C-reactive protein [38% (31%, 51%)], waist-to-hip ratio [35% (31%, 42%)] and diabetes [32% (27%, 36%)]. Among lipid markers, the biggest mediating effect was observed for high density lipoprotein [28% (25%, 32%)] and triglycerides [26% (22%, 33%)]. As for non-clinical factors, financial status [23% (15%, 31%)] presented the biggest mediating effect followed by educational status, adherence to Mediterranean diet and sleep duration. In ACS females, diabetes [40% (27%, 51%)] and adherence to medication [40% (29%, 53%)] had the biggest mediating effect followed by hypertension [38% (27%, 48%)]. Patients' CVD history and discharge status, presented a very low mediating effect size (∼10%). Among the examined lifestyle factors, current smoking was revealed the strongest mediator, accounting for 33% (19%, 39%) of DS aggravating effect. Anthropometric parameters in terms of body mass index seemed to modestly mediate the examined association (∼29%). Conclusions The present work augments prior evidence that psychological stressors possess important drivers of CVD onset and progression mainly in females while it gives rise to research towards unidentified paths behind this claim. Acknowledgement/Funding The ATTICA study is supported by research grants from the Hellenic Cardiology Society [HCS2002] and the Hellenic Atherosclerosis Society [HAS2003].