An increased mortality risk is associated with abnormal iron status in diabetic and non-diabetic Veterans with predialysis chronic kidney disease.

Iron parameters have not been well characterized in pre-dialysis patients with chronic kidney disease (CKD), and it remains unclear if abnormal iron balance is associated with increased mortality. Therefore, we performed a historical cohort study using data from the Veterans Affairs Corporate Data Warehouse to evaluate the relationship between iron status and mortality. We identified a pre-dialysis CKD cohort with at least one set of iron indices between 2006-2015. The cohort was divided into four iron groups based on the joint quartiles of serum transferrin saturation (percent) and ferritin concentration (ng/ml): reference (16-28%, 55-205 ng/ml), low iron (0.4-16%, 0.4-55 ng/ml), high iron (28-99.6%, 205-4941 ng/ml), and function iron deficiency (0.8-16%, 109-2783 ng/ml). We compared mortality risk between the iron groups using matching weights based on multinomial propensity score models and Poisson rate-based regression. We also evaluated if the association between iron groups and mortality differs between the diabetic and non-diabetic subgroups. Of the 80,067 eligible veterans, 32,489 were successfully matched. During the mean follow-up period of 4.0 years, adjusted relative rate (95% confidence interval) for all-cause mortality in three abnormal iron groups were increased compared to the reference: functional iron deficiency [1.21 (1.17, 1.25)], low iron [1.10 (1.07, 1.14)], and high iron [1.09 (1.06, 1.13)]. The mortality risk was similar between diabetic and non-diabetic subgroups for each iron group. Thus, an abnormal iron balance, particularly functional iron deficiency, is associated with increased mortality in CKD.