P1473THE EFFECT OF SERUM C-REACTIVE PROTEIN AND FERRITIN LEVELS ON ALL-CAUSE AND CARDIOVASCULAR MORTALITY IN HAEMODIALYSIS PATIENTS

Abstract

Listen

Translate article

Abstract Background and Aims Ferritin, is commonly used as a surrogate marker of iron storage in haemodialysis (HD) patients. However, besides being an indicator of iron storage, serum ferritin levels are closely influenced by non-iron related factors, such as systemic inflammation. The aim of this study is to assess the combined effect of serum ferritin and C-reactive protein (CRP) levels on all-cause and cardiovascular (CV) mortality in a large national cohort of HD patients. Method This is a retrospective study which included all patients that were on standard HD therapy in 34 Romanian Fresenius dialysis units at January 1st 2014. The main outcomes were all-cause and CV mortality. Patients were censored at December 31st 2016. Results Our study included 1255 patients. The mean age of the population at baseline was 59 years. The median values for CRP and ferritin levels were 4.8 (IQR 1.9-11.0) mg/dL and 696.0 (IQR 448.5-996.8) ng/mL, respectively. Using the tertiles categories for CRP (<2.7, 2.7-8.1, and >8.1 mg/dL) and ferritin (<536.6, 536.6-876.1, and >876.1 ng/mL) our population was categorized into 9 groups of patients: Group 1 CRP<2.7 mg/dL, ferritin<536.6 ng/mL; Group 2 CRP<2.7 mg/dL, ferritin 536.6-876.1ng/mL; Group 3 CRP 2.7-8.1mg/dL, ferritin >876.1 ng/mL; Group 4 CRP 2.7-8.1 mg/dL, ferritin<536.6 ng/mL; Group 5 CRP 2.7-8.1 mg/dL, ferritin 536.6-876.1ng/mL; Group 6 2.7-8.1 mg/dL, ferritin >876.1 ng/mL; Group 7 CRP>8.1 mg/dL, ferritin<536.6 ng/mL; Group 8 CRP>8.1 mg/dL, ferritin 536.6-876.1ng/mL; Group 9 CRP>8.1 mg/dL, ferritin >876.1 ng/mL; During the follow-up (mean 1.75, median 1.94 years), 204 patients (16.25%) died with 103 (8.21%) from CV disease. The central point of our investigation was to assess the combined effect of CRP and ferritin levels on the two outcomes (Table 1). For the all-cause mortality, after adjustment for different demographic, clinical and biological risk factors, only patients with CRP levels in the highest tertile (CRP>8.1 mg/dL), irrespectively of ferritin levels (Groups 7, 8 and 9), remained significantly associated with this outcome. When considering the CV mortality, in the adjusted analysis, only patients from Groups 4 (2nd tertile CRP, 1st tertile ferritin) and 7 (3rd tertile CRP, 1st tertile ferritin) remained associated with a higher risk for the outcome as compared with patients from Group 1. Conclusion We show that increased levels of CRP are associated with an increased risk for all-cause mortality in the HD population, irrespective of ferritin levels. However, increasing CRP levels are associated with CV mortality only for the lowest tertile of ferritin levels. This work was supported by a grant of the Ministery of Research and Innovation, CNCS-UEFISCDI, project number PN-III-P1-1.1-PD-2016-0287, within PNCDI III and by grants of the “Grigore T. Popa” University of Medicine and Pharmacy, contract numbers 27495/2018 and 27505/2018.