Vegetable and Fruit Intake Frequency and Mortality in Patients With and Without Chronic Kidney Disease: A Hospital-Based Cohort Study

Objective: Patients with advanced chronic kidney disease (CKD) are generally discouraged from consuming high amounts of vegetables and fruits given the potential risk of hyperkalemia. In the general population, however, lower vegetable and fruit intake is associated with higher mortality. This Japanese hospital-based prospective cohort study including non-CKD and CKD participants examined whether the frequency of vegetable and fruit intake is associated with mortality, and whether the presence of CKD modifies this association.

Methods: Participants were 2,006 patients who visited the outpatient department of a general hospital between June 2008 and December 2016 (55% men; mean age, 69 years). Among these participants, 902 (45%) and 131 (7%) were non-dialysis-dependent patients with CKD and hemodialysis patients, respectively. The frequency of vegetable and fruit intake was determined by a self-report questionnaire using an ordinal scale, "never or rarely," "sometimes," and "every day." Multivariable Cox proportional hazards analysis was conducted, adjusting for potential confounders.

Results: Vegetable and fruit intake frequency decreased with worsening CKD stage (P for trend < .001). Baseline serum potassium levels stratified by CKD stage were similar across the three vegetable and fruit intake frequency groups. During a median follow-up of 5.7 years, 561 participants died (47.1/1,000 person-years). Adjusted hazard ratios relative to the "every day" group were 1.25 (95% confidence interval, 1.04-1.52) and 1.60 (95% confidence interval, 1.23-2.08) for the "sometimes" and "never or rarely" groups, respectively, after adjusting for demographic factors, comorbidities, and CKD status. When stratified by CKD status, a similar, albeit non-significant, dose-dependent relationship was observed between vegetable and fruit intake frequency and all-cause mortality irrespective of CKD status, with no effect modification by CKD status (P_interaction = .69).

Conclusion: A lower frequency of vegetable and fruit intake is significantly associated with a higher risk of death regardless of CKD status.