Usefulness of red cell distribution width to predict mortality in patients undergoing endovascular repair of abdominal aortic aneurysms.

ABSTRACT

BACKGROUND: Red cell distribution width (RDW) reflecting impaired erythropoyesis, has been associated with poor prognosis and mortality in several conditions. The aim of this study was to determine the relationship between RDW and the 5-year survival after the endovascular repair of abdominal aortic aneurysms (EVAR) and its ability to improve the discriminative power of a survival predictive score. METHODS: Retrospective analysis of 284 patients undergoing EVAR at a single centre. The pattern of relationship between RDW and survival was assessed with penalized smoothing splines. Categorized RDW values were added to a predictive score based in standard preoperative variables, whose improvement in discriminative power was calculated on the basis of changes in the C-statistics and the continuous Net Reclassification Index (c-NRI). RESULTS: The survival rate at 5 years was 66.2% and was independently associated with hemoglobin (HR=0.85, P<0.004), statin intake (HR=0.54, P<0.004), heart failure (HR=2.53, P<0.018), atrial fibrillation (HR=2.53, P<0.000) and the non-revascularized coronary artery disease (HR=2.15, P<0.005). The relationship between RDW values and 5-year survival was linear. RDW-CV and RDW-SD were categorized to cut-off values of ≥15% (N.=83, 29.2%) and ≥50 fL (N.=82, 28.9%) that were independently associated with poorer 5-year survival rates (HR=2.03, CI 95%=1.29-3.19, P=0.002 and HR=1.89, CI 95%=1.21-2.95, P=0.005, respectively). The addition of the RDW CV or the RDW-SD to the baseline predictive score significantly improved the c-NRI (0.437, P<0.001 and 0.442, P<0.001, respectively). CONCLUSIONS: High preoperative RDW levels were linear and adversely related to 5-year survival after EVAR, improved the discriminative power of a predictive score based in standard preoperative variables and may help in decision-making at the time of surgical planning.