Diet quality and long-term cardiovascular mortality after myocardial infarction in the Alpha Omega Cohort

Abstract

Abstract Introduction A healthy diet is important for the prevention and treatment of cardiovascular disease (CVD). The Dutch Health Council recently defined dietary guidelines for patients with atherosclerotic CVD (ASCVD), which we investigated in relation to long-term mortality after myocardial infarction (MI). Purpose We developed the Dutch Healthy Diet index for CVD patients (DHD-CVD index) and examined its relation with mortality in post-MI patients. Methods We included 4,365 patients of the Dutch Alpha Omega Cohort (60-80 y, 80% male) with a MI <10 y prior to study enrolment. Dietary intake was assessed at baseline (2002-2006), using a validated 203-item food frequency questionnaire. To create the DHD-CVD index, the existing 2015 Dutch Healthy Diet index was adapted using the 2023 Dutch dietary guidelines for ASCVD patients (scale from 0 to maximal adherence [0-150]). Patients were followed for cause-specific mortality through December 2018. Hazard Ratios (HRs) across quartiles of the DHD-CVD-index (ref = Q1, i.e. low overall diet quality) and per 1-SD increment in score were estimated using Cox proportional hazard models. Associations were adjusted for age, sex, total energy intake, smoking, education and physical activity. Results were additionally stratified for sex, obesity and diabetes. Results Of the cohort, 17% smoked, 21% had diabetes and >90% used statins or other CVD drugs. During a follow-up of 48,660 person-years, 2,048 deaths occurred of which 911 were due to CVD. The mean DHD-CVD score was 78.9 (SD 13) points. High diet quality (Q4) was associated with a 29% lower risk of CVD mortality compared to low diet quality (HR: 0.71, 95% CI: 0.56, 0.84), with an HR of 0.88 (95% CI: 0.82, 0.94) per 1-SD. For all-cause mortality, we observed a 14% lower risk for high vs. low diet quality (HR: 0.86, 95% CI: 0.76, 0.98), with an HR of 0.94 (95% CI: 0.90, 0.98) per 1-SD. HRs did not essentially differ between male of female patients, or between obese and non-obese patients. HRs for CVD mortality were stronger in non-diabetic patients (HR: 0.86, 95% CI: 0.79, 0.93; per 1-SD), and no longer statistically significant in patients with diabetes (HR: 0.97, 95% CI: 0.84, 1.11; per 1-SD). Conclusion A high diet quality, measured with the DHD-CVD index, was associated with a lower risk of CVD and all-cause mortality after MI in Dutch patients. Adherence to dietary guidelines for CVD patients should be considered on top of medication use.