The association between in-hospital hemoglobin changes, cardiovascular events, and mortality in acute decompensated heart failure: Results from the ESCAPE trial

Introduction The effect of acute changes of hemoglobin during index heart failure admission on long-term outcomes remains unknown. Methods We examined 433 patients enrolled in the ESCAPE trial. Results Of the 433 patients, 324 (75%) had baseline and discharge hemoglobin available for analysis. Of those, 64 (20%) had at least 1 g/dL drop of hemoglobin by time of discharge. Compared to patients without hemoglobin changes (g/dL), patients with hemoglobin drop were older (59 vs. 55, p = 0.011), had lower systolic BP (mm Hg) (99 vs. 106, p = 0.017), lower sodium (mg/dL) (136 vs. 137 (mg/dL), p = 0.025), higher BUN (mg/dL) (37 vs. 26, p < 0.001), higher creatinine (mg/dL) (1.6 vs. 1.3, p < 0.001) and higher hospital length of stay (10 days vs. 6 days, p = < 0.001). Higher hemoglobin drop was observed in the pulmonary artery catheter (PACs) (vs. clinical care) randomized arm of the trial (2 g/dL: 10% versus 3%, p = 0.010; 3 g/dL: 5% versus 0%, p = 0.005). After adjustments, a drop of hemoglobin with at least 1 g/dL was associated with increased mortality risk (Adjusted HR 2.38, p = 0.003) and higher hemoglobin concentrations by the time of discharge was associated with lower mortality rate (Adjusted HR 0.79, p = 0.003). PACs insertion was not associated with adverse clinical outcomes by quartiles of % change of hemoglobin. However, PACs use was an independent predictor of hemoglobin drop during heart failure admission (Adjusted OR: Hb Drop 1 g/dL: 1.88, p = 0.043; Hb Drop 2 g/dL: 3.6 p = 0.025). Conclusion In-hospital decrease in hemoglobin is independently associated with increased long-term mortality and hospital length of stay in ADHF. The ideal hemoglobin levels in ADHF patients should be investigated and the insertion of PACs to direct therapy should be weighed against bleeding risks.