Abstract 14819: The Red Cell Distribution Width Predicts Mortality among Patients Free from Systemic Inflammation
Background: The causal mechanism is unknown for why the red cell distribution width (RDW) predicts mortality and morbidity outcomes. One explanation is an inflammatory process: statistically significant, weak magnitude correlations (r<0.30) have been found between RDW and markers of inflammation. This study evaluated the association of RDW with mortality in patients with normal levels of inflammation, as indicated by normal high-sensitivity C-reactive protein (hsCRP) and white blood cell count (WBC).
Methods: Intermountain Heart Collaborative Study patients undergoing coronary angiography (N=650) from 1994-2000 were evaluated if they never smoked, were free of acute myocardial infarction (MI), and had baseline hsCRP≤3 mg/L, WBC>4 K/μL, and WBC≤10.6 K/μL. RDW and WBC were tested clinically at index hospitalization via the complete blood count; hsCRP testing used stored research samples. Subjects were followed until December, 2013, and all-cause mortality was determined from hospital records, Utah death certificates, and US Social Security data.
Results: Age averaged 66.7±11.6 years, 31.1% were female, and 62.9% died during a mean follow-up of 15.2±1.6 years (range:12.6-19.6 years). Continuous RDW predicted mortality after adjustment for demographics, risk factors, comorbidities, and baseline treatments (hazard ratio [HR]=1.15 per +1%, 95% CI=1.06, 1.26; p=0.002). In quartiles, this was significant for Q4 vs. Q1 (survival 23.6% vs. 43.4%, HR=1.56, CI=1.17, 2.08; p=0.003), but not Q2 (HR=1.18, p=0.29) or Q3 (HR=1.08, p=0.60) vs.Q1. In subjects with hsCRP<1 mg/L (n=259), RDW (adj. HR=1.13 per +1%, CI=0.99, 1.29, p=0.08) and RDW quartiles [Q2: adj. HR=1.70 (p=0.036), Q3: HR=1.88 (p=0.020), Q4: HR=1.97 (p=0.013) vs. Q1] predicted mortality. Among subjects free from heart failure (LVEF≥40%), diabetes, prior or current coronary disease, prior MI, stroke, renal failure, COPD, and depression (n=66), RDW had adjusted HR=1.70 per +1% (CI=1.06, 2.75; p=0.029).
Conclusions: RDW predicted mortality in patients with normal hsCRP and WBC. This suggests that RDW marks the risk of a breadth of health concerns, likely including but not limited to inflammation. Further investigation is required to explain the causes of RDW elevation.
Author Disclosures: B.D. Horne: None. J.B. Muhlestein: None. S.T. Bennett: None. J.L. Anderson: None.
- © 2014 by American Heart Association, Inc.