P0422RED BLOOD CELL DISTRIBUTION WIDTH AND OUTCOME IN SUBJECTS WITH BIOPSY-PROVEN GLOMERULOPATHIES WITHOUT ANEMIA

Background and Aims Red blood cell distribution width (RDW) is a marker of anisocytosis and is mainly modified in anemia. However, elevated RDW has recently been reported to predict cardiovascular risk, was correlated with disease activity in inflammatory conditions such as systemic lupus erythematosus, rheumatoid arthritis and, not the least, was associated with increased risk of end stage kidney disease and proteinuria in chronic kidney disease (CKD) subjects. The current study aims to assess if RDW is an independent predictor for renal replacement therapy (RRT) and mortality in subjects with glomerulopathies (GP). Method This retrospective, single-center study, included 467 subjects with hemoglobin >12 g/dL at presentation, who were histologically diagnosed with primary and secondary glomerulopathies between 1st Jan. 2008 and 31st Dec. 2017 and who were followed for a mean of 52.8 (95%CI 50.7-55) months, until 31st May 2018. Subjects with inadequate biopsy sample were excluded. Those who deceased until the end of the follow-up were not included in the kidney survival analysis. Predictors of mortality were assessed by logistic regression. Kaplan-Meier method was used to evaluate kidney survival. Patients were stratified in two groups according to the median RDW value: low RDW group (RDW≤15%; n=342) and high RDW group (RDW>15%; n=125). RRT initiation and all-cause mortality were the primary endpoints of the study. Results Subjects with high RDW were older [53 (95%CI 48-56) vs. 45 (95%CI 43-47) years; p=0.03], had more frequent cerebrovascular disease (9.7% vs. 2.7%; p=0.003), congestive heart failure (8.1% vs. 1.8%; p=0.002), connective tissue disease (11.3% vs. 3%; p=0.001) and non-hematologic neoplasia (4.8% vs. 0.9%, p=0.01). Furthermore, they had more severe inflammation as suggested by higher erythrocyte sedimentation rate [54.5 (95%CI 43-60) vs. 35 (95%CI 30-40) mm/hour; p=0.002], fibrinogen [630 (95%CI 562-701) vs. 540 (95%CI 514-585 mg/dL; p=0.02], and C reactive protein [3 (95%CI 2-4) vs. 2 (95%CI 2-3); p=0.03], had lower hemoglobin [13.7 (95%CI 13.4-14) vs. 14.1 (95%CI 13.9-14.4) g/dL); p=0.008], and higher proteinuria [4.2 (95%CI 3.1-5.2) vs. 2.4 (95%CI 2.1-2.8) g/g creatinine; p=0.002]. However, the kidney function between the two groups was similar (56.5 vs. 55 ml/min MDRD; p=0.1). Secondary GP were more frequent encountered in high RDW group (42.4% vs. 18.1%; p<0.001), especially amyloidosis (19.2% vs. 4.7%; p<0.001) and lupus nephritis (9.6% vs. 3.5%; p=0.01). During the follow-up period, in high RDW group 26.4% of the subjects died, compared to 11.4% from the other group (p<0.001). In a multivariate logistic regression model, RDW was an independent predictor for mortality [OR 1.5 (95%CI 1.2-15); p=0.007]. Other independent predictors were older age [OR 1.06 (95%CI 1.03-1.08); p<0.001], presence of cerebrovascular disease [OR 3.31 (95%C I1.01-10.8); p=0.04], lower serum albumin [OR 0.48 (95%CI 0.43-0.70), p<0.001] and higher serum creatinine [OR 1.29 (95%CI 1.06-1.57); p<0.001]. There was no difference regarding the frequency of RRT initiation between groups (18% in high RDW group vs. 11% in low RDW group; p=0.07) and the time to kidney death was similar [79.7 (95%CI 72.8-86.6) vs. 87.8 (95%CI 84-91.6) months; log rank p=0.1]. Conclusion In non-anemic subjects with biopsy-proven glomerular disease, a higher RDW seems associated with a higher risk of mortality, but not with kidney survival. In addition, RDW above 15% might point out to a secondary GP, like amyloidosis and lupus nephritis.