360 - WINE CONSUMPTION, MEDITERRANEAN DIET, AND CARDIOVASCULAR RISK IN TWO SPANISH COHORTS
Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y Nutrición (CIBERObn); Departamento de Medicina Preventiva y Salud Pública, Universidad de Navarra; Instituto de Investigación Sanitaria de Navarra (IDISNA); Universitat de València; Unidad de Nutrición Humana, Universitat Rovira i Virgili; Institut d’Investigacions Biomèdiques August Pi Sunyer (IDIBAPS); Institut Hospital del Mar de Investigacions Mèdiques (IMIM); Universidad de Navarra, Instituto de Nutrición y Salud (INS); Servicio de Medicina Interna, Hospital Clinic de Barcelona.
Background/Objectives: The benefits of the Mediterranean diet (MedDiet) are well established. However, one component, wine, remains controversial. We assessed the association between MedDiet (with or without wine consumption) and long-term all-cause mortality in two Spanish cohorts.
Methods: The PREDIMED trial included 7,447 high-risk participants. Adherence to MedDiet was measured using a validated 14-item questionnaire, including one item on wine (cut-off: 7 glasses/week). All-cause mortality was tracked for 17 years. A younger Spanish cohort (the SUN project), including 23,133 participants followed-up for 22 years was also evaluated.
Results: In PREDIMED, compared to poor compliers with MedDiet (excluding wine), good compliers (excluding wine), had a multivariable-adjusted hazard ratio (HR) for all-cause mortality of 0.77 (CI: 0.68-0.87), which was 0.67 (CI: 0.57-0.78) for MedDiet compliers (including wine). In exploratory dose-response analyses, reduced risk for death was not present in PREDIMED participants who drank 3 or more glasses of wine/day. In the SUN cohort the HR was 0.94 (CI: 0.71-1.26) for MedDiet compliers (excluding wine); and 0.54 (CI: 0.28-1.04) for MedDiet compliers (including wine). However, multiplicative interactiosn between adherence to the MedDiet and the wine point did not reach statistical significance and several sensitivity analyses were non-significant. When pooling both cohorts, wine consumption within the MedDiet was associated with lower all-cause mortality (p = 0.01).
Conclusions/Recommendations: In PREDIMED, moderate wine consumption, as part of the MedDiet, appeared to be associated with lower all-cause mortality. However, some non-significant associations in sensitivity analyses and non-significant interactions advise caution in the interpretation of these findings.
Funding: PID2020-114022RB-I00, CIBEROBN, National Institutes of Health (NIH), Agencia Estatal de Investigación/Fondo Europeo de Desarrollo Regional, Generalitat de Catalunya, European Research Council (ERC).
