Ultra-processed Food Intake and Long-Term Risk of Cardiovascular Disease Mortality among Patients with Myocardial Infarction: A Prospective Analysis of the Alpha Omega Cohort
TL;DR
In patients with prior myocardial infarction, a high UPF diet was lower in whole, cardioprotective foods and associated with a higher risk of stroke mortality than a high ultra-processed diet.
Abstract
Background The impact of a diet high in ultra-processed foods (UPFs) on mortality risk in patients with cardiovascular disease (CVD) remains unclear. Objectives We examined the diet quality of a high UPF diet and its association with risk of total CVD mortality, coronary heart disease (CHD) mortality, and stroke mortality in patients with prior myocardial infarction (MI). Methods We analyzed data from 4365 patients with a history of MI in the Alpha Omega Cohort (79% males). Dietary intake at baseline was collected using a 203-item food-frequency questionnaire. Diet quality was assessed using the Dutch Healthy Diet Index adapted for patients with CVD (DHD-CVD, range 0–160). UPFs were identified and categorized according to the Nova classification system. Q1 and Q4 of %g/d of UPF represented a low and a high ultra-processed diet, respectively. Hazard ratios (HRs) for CVD, CHD, and stroke mortality were obtained from multivariable Cox regression models. Results Mean UPF contribution to the diet was 18%g/d (Q1 compared with Q4 = 9%g/d compared with 29%g/d). Compared with Q1, patients in Q4 had higher intakes of total energy and lower intakes of fruit and vegetables. DHD-CVD scores decreased slightly across quartiles (Q1 compared with Q4 = 95 compared with 82). Over a median follow-up of 14 y, 1012 patients died from CVD (659 from CHD, 208 from stroke). Compared with Q1, a high ultra-processed diet was not consistently or significantly associated with CVD and CHD mortality. However, it was associated with a higher risk of stroke mortality (HRper10% point increase: 1.21; 95% confidence interval: 1.02, 1.42). This association was stronger in older patients, females, and those with a higher BMI, lower physical activity levels, and lower socioeconomic status and robust to sensitivity analysis. Conclusions In patients withprior MI, a high UPF diet was lower in whole, cardioprotective foods. Our study highlights the adverse association of a high UPF diet with stroke mortality risk among patients who have previously experienced an MI.