Longitudinal association between iron stores biomarkers and major adverse cardiovascular events in chronic kidney disease

Background and Aims Iron deficiency is a frequent condition in moderate to severe chronic kidney disease (CKD), often left under-screened, and infrequently supplemented. Iron deficiency is a major contributor to anemia, which is known to negatively impact outcomes and quality of life in CKD. However, evidence of an anemia-independent effect of iron deficiency on cardiovascular morbidity and all-cause mortality has emerged. Current challenges of studying the association between iron deficiency and the risk of events in CKD are the definition of critical thresholds for these biomarkers, the potential interaction between them, and their evolution over time. Previous observational cohort studies have only focused on baseline values of transferrin saturation and ferritin, the two traditional biomarkers of iron deficiency. Our aim was to study the longitudinal association between iron stores biomarkers and cardiovascular risk in CKD by evaluating the total effect of these biomarkers on major adverse cardiovascular events (MACE). Method We used data from the CKD-REIN cohort, which included 3033 patients with moderate to severe CKD from 40 nationally representative nephrology clinics in France between 2013 and 2016. All patients with at least one measurement of TSAT and ferritin were included in the present study. Our primary endpoint was MACE, defined as the first event among cardiovascular death, myocardial infarction, stroke or hospitalization for acute heart failure. Repeated measures of TSAT and ferritin were included as time-dependent variables in cause-specific Cox proportional hazard models for risk of MACE, with initiation of kidney replacement therapy (KRT) and death as competitive events. Models were adjusted for time-varying estimated glomerular filtration rate (eGFR), white cells and baseline age, sex, iron supplementation, and erythropoietin stimulating agents. Analyses were not adjusted for hemoglobin level because the aim of the study was to investigate the total effect of iron deficiency on MACE. Non-linear effects of TSAT and ferritin and their interaction were modeled using natural splines functions. Results A total of 2670 patients were included in the analysis: 65% were men, median age was 68 years (interquartile range IQR 60-76). Median baseline eGFR was 34 ml/min/1.73 m2 (IQR 25-44), TSAT was 24% (IQR 18-30) and ferritin was 127 ng/ml (IQR 72-230). Over a median follow-up period of 4.4 years, participants had a median of 5 (IQR 2-8) and 4 (IQR 2-6) measurements of ferritin and TSAT, respectively; 396 MACE before KRT were recorded. The negative nonlinear association of iron deficiency with MACE was essentially captured by TSAT, and the interaction terms between TSAT and ferritin were not statistically significant (p = 0.94). Compared with patients with TSAT of 20% and ferritin of 100 ng/ml, those with TSAT values < 20% had a higher adjusted hazard of MACE independently of ferritin values (See Figure 1); for example, for TSAT = 10% and ferritin = 100 ng/ml, the hazard ratio (HR) was 1.96, 95% CI (1.60-2.39). Conclusion In CKD patients under nephrology care, current value of TSAT below 20% was associated with higher risk of MACE regardless of ferritin current value. Before investigating strategies of treatment in clinical trials as PIVOTAL in dialysis patients, further studies clarifying critical thresholds of TSAT and ferritin in non-dialysis CKD patients, in presence or absence of anemia and among specific CKD subpopulations are highly desirable.