Year: 2010
Type: letter
Abstract: We read the recent paper by Al-Najjar et al.1 on the prognostic value of red cell distribution width (RDW) in heart failure (HF) with great interest. The authors have shown in a population of 1087 ambulatory patients with HF due to left ventricular systolic dysfunction that RDW is a potent prognostic marker of all-cause mortality and has similar prognostic power to that of NT-proBNP. Two important open questions were raised at the end of the manuscript: whether the effect of RDW is independent of erythropoietin (EPO) levels, and what are the mechanisms of elevated RDW values in HF. The authors suggested that this new information may lead to novel approaches in treatment. We have also published similar results showing the prognostic value of RDW in chronic HF2 that appear to be in parallel with the study of Al-Najjar et al. Besides showing the prognostic power of RDW for all-cause mortality in our cohort, we were able to show the same for rehospitalization due to worsening HF symptoms. Furthermore, according to our analysis, RDW has NT-proBNP-independent predictive power for clinical events in chronic HF. In addition, we also provided observational data about the laboratory correlates of RDW in chronic HF and described novel data for delineation of the underlying mechanisms behind this observation. According to our data, RDW values are strongly related to signs of ineffective erythropoiesis, inflammation, impaired renal function, and under nutrition. Higher RDW values were associated with significantly lower serum iron and ferritin levels and decreased transferrin saturation and also with increased soluble transferrin receptor levels. The strongest correlate of RDW was soluble transferrin receptor concentration in the multiple linear regression model. Thus, based on these observations, RDW seems to be a prominent marker of anaemia of chronic diseases complicated by iron deficiency in patients with chronic HF. Furthermore, high serum EPO levels were associated with high RDW values in our cohort, indicating that the bone-marrow effects of EPO may also be compromised. In response to the questions raised by Al-Najjar et al., we have now performed an analysis to evaluate whether the prognostic effect of RDW is independent of EPO levels. The results of our Cox regression models (Table 3 in our paper2) after addition of EPO as a further covariate suggest that the prognostic power of RDW is independent of EPO levels (for all-cause mortality HR 1.586 (95% CI 1.288–1.951; χ2 13.165 and P < 0.0001; for all-cause mortality or HF hospitalization HR 1.424 (95% CI 1.153–1.757; χ2 8.65 and P = 0.003). Taken together, the results of these two recent independent studies1,2 in combination with the pioneering work of Felker et al.3 have now convincingly shown that future HF prognostic models should utilize RDW, a test available as part of the complete blood count. The effectiveness of this marker may lie, as supported by our observational data, on its relationship with ineffective red cell production, inflammation, impaired renal function, and under nutrition.
Institutions: Semmelweis University, Hungarian Academy of Sciences
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Keywords: red cell distribution width, heart failure, prognostic factors, erythropoietin, iron deficiency anemia, anemia of chronic disease, soluble transferrin receptor, NT-proBNP, all-cause mortality, hospital readmission, left ventricular systolic dysfunction, erythropoiesis, inflammation, renal function, malnutrition, transferrin
Open Access status: bronze
OpenAlex ID: W2000550498
