Modest U-Shaped Association between Dietary Acid Load and Risk of All-Cause and Cardiovascular Mortality in Adults

Abstract

Background: Conflicting evidence associates diet acidity with the incidence of chronic diseases such as hypertension, diabetes, kidney disease, and bone-mineral disorders. It is currently unknown whether dietary acidity is associated with death.Objective: We investigated the association of dietary acid load with the risk of all-cause and cardiovascular disease (CVD) mortality.Methods: We used data from 2 prospective cohorts, the Swedish Mammography Cohort and the Cohort of Swedish Men, which included 36,740 women and 44,957 men aged 45–84 y at the start of a 15-y follow-up period (1998–2012). Acid load was estimated from food-frequency questionnaires by use of the validated potential renal acid load (PRAL) algorithm. Deaths were ascertained via record linkage. Associations of PRAL with mortality were modeled by use of restricted cubic splines.Results: The median PRAL was 0.65 mEq/d (range: −109 to 81.5 mEq/d) in women and 12.3 mEq/d (−111 to 121 mEq/d) in men. During a mean of 13.5 ± 3.3 y of follow-up, there were 8576 and 13,332 deaths, of which 3203 and 5427 were attributed to cardiovascular causes in woman and men, respectively. In both sexes, a nonlinear U-shaped relation was observed, with higher mortality rates for both dietary acid and alkali excess. Compared with neutral PRAL (0 mEq/d), the HRs for all-cause mortality for the 10th and 90th percentiles of PRAL were 1.05 (95% CI: 1.01, 1.10) and 1.03 (95% CI: 0.98, 1.08), respectively, in women. The corresponding results for men were HRs 1.01 (95% CI: 1.00, 1.02) and 1.04 (95% CI: 1.00, 1.08) respectively. This relation was slightly stronger for CVD mortality.Conclusions: Excess diet alkalinity and acidity both showed weak associations with higher mortality in Swedish adults. An acid-base balanced diet was associated with the lowest mortality, but the magnitude of mortality reduction was modest. The Swedish Mammography Cohort was registered at as NCT01127698. The Cohort of Swedish Men was registered at clinicaltrials.gov as NCT01127698. The Cohort of Swedish Men was registered at clinicaltrials.gov as NCT01127711.