Ferritin levels, inflammatory biomarkers, and mortality in peripheral arterial disease: A substudy of the Iron (Fe) and Atherosclerosis Study (FeAST) Trial

Abstract

This study delineated correlations between ferritin, inflammatory biomarkers, and mortality in a cohort of 100 cancer-free patients with peripheral arterial disease (PAD) participating in the Veterans Affairs (VA) Cooperative Study #410, the Iron (Fe) and Atherosclerosis Study (FeAST). FeAST, a prospective, randomized, single-blind clinical trial, tested the hypothesis that reduction of iron stores using phlebotomy would influence clinical outcomes in 1227 PAD patients randomized to iron reduction or control groups. The effects of statin administration were also examined in the Sierra Nevada Health Care (SNHC) cohort by measuring serum ferritin levels at entry and during the 6-year study period. No difference was documented between treatment groups in all-cause mortality and secondary outcomes of death plus nonfatal myocardial infarction and stroke. Iron reduction in the main study caused a significant age-related improvement in cardiovascular disease outcomes, new cancer diagnoses, and cancer-specific death. Tumor necrosis factor (TNF)-alpha, TNF-alpha receptors 1 and 2, interleukin (IL)-2, IL-6, IL-10, and high-sensitivity C reactive protein (hs-CRP) were measured at entry and at 6-month intervals for 6 years. Average levels of ferritin and lipids at entry and at the end of the study were compared. The clinical course and ferritin levels of 23 participants who died during the study were reviewed. At entry, mean age of entry was 67 +/- 9 years for the SNHCS cohort, comparable to FeAST and clinical and laboratory parameters were equivalent in substudy participants randomized to iron reduction (n = 51) or control (n = 49). At baseline, 53 participants on statins had slightly lower mean entry-level ferritin values (114.06 ng/mL; 95% confidence interval [CI] 93.43-134.69) vs the 47 off statins (127.62 ng/mL; 95% CI, 103.21-152.02). Longitudinal analysis of follow-up data, after adjusting for the phlebotomy treatment effect, showed that statin use was associated with significantly lower ferritin levels (-29.78 ng/mL; Cohen effect size, -0.47 [t(df, 134) = 2.33, P = .02]). Mean follow-up average ferritin levels were higher in 23 participants who died (132.5 ng/mL; 95% CI, 79.36-185.66) vs 77 survivors (83.6 ng/mL; 95% CI, 70.34-96.90; Wilcoxon P = .05). Mean follow-up IL-6 levels were higher in dead participants (21.68 ng/mL; 95% CI, 13.71-29.66) vs survivors (12.61 ng/mL; 95% CI, 10.72-14.50; Wilcoxon P = .018). Ferritin levels correlated (Pearson) with average IL-6 levels (r = 0.1845; P = .002) and hsCRP levels (r = .1175; P = .04) during the study. These data demonstrate statistical correlations between levels of ferritin, inflammatory biomarkers, and mortality in this subset of patients with PAD.