Gassan Moady1, Zaza Iakobishvili2, Roy Beigel3, Nir Shlomo4, Shlomi Matetzky5, Doron Zahger6, Shaul Atar7. 1. Department of Cardiology, Galilee Medical Center, Nahariya, Azreili Faculty of Medicine, Bar-Ilan University, Safed, Israel. 2. The Heart Institute, Rabin Medical Center, Petah-Tikvah, Sackler School of Medicine, Tel-Aviv University, Tel Aviv, Israel. 3. The Leviev Heart Center, Sheba Medical Center, Tel Hashomer, Sackler School of Medicine, Tel-Aviv University, Tel Aviv, Israel. 4. Neufeld Cardiac Research Institute, Sheba Medical Center, Tel Hashomer, Israel. 5. The Leviev Heart Center, Sheba Medical Center, Tel Hashomer, Sackler School of Medicine, Tel-Aviv University, Tel Aviv, Israel; Neufeld Cardiac Research Institute, Sheba Medical Center, Tel Hashomer, Israel. 6. Department of Cardiology, Soroka Medical Center, Faculty of Health Sciences, Ben Gurion University of the Negev, Beer-Sheva, Israel. 7. Department of Cardiology, Galilee Medical Center, Nahariya, Azreili Faculty of Medicine, Bar-Ilan University, Safed, Israel. Electronic address: shaul.atar@gmail.com.
Abstract
BACKGROUND: The GRACE risk score is currently recommended as the major score for risk prediction on admission in patients with acute coronary syndrome (ACS). Anemia in patients with ACS adversely affects their clinical outcomes, yet hemoglobin level on admission is not included as a parameter in the GRACE score. We hypothesized that hemoglobin level on admission would improve the predictive value of the GRACE score. METHODS: We retrospectively studied one-year mortality in consecutive ACS patients included in the ACSIS (acute coronary syndrome Israeli Survey) registry between the years 2008 and 2013. Patients were classified into groups according to the GRACE score - ≥140 or <140, and according to the hemoglobin level: severe anemia - <8g/dl; mild anemia - 8-12g/dl; no anemia - >12g/dl. We analyzed the incremental predictive value of admission hemoglobin levels over the GRACE score. RESULTS: We studied 11,505 patients. The GRACE score predicted 1-year mortality with an area under the curve (ROC) of 0.68 (95% CI 0.66-0.7). When hemoglobin level on admission was incorporated into the model, the ROC increased to 0.73 (95% CI 0.71-0.75, p<0.001). The incremental value of hemoglobin levels on admission was significant only in the low (<140) GRACE score group. CONCLUSIONS: In patients a with low GRACE score (<140), anemia on admission has additional predictive value for one-year mortality. In contrast, in patients with a high GRACE risk score, hemoglobin level on admission did not improve prediction accuracy.
BACKGROUND: The GRACE risk score is currently recommended as the major score for risk prediction on admission in patients with acute coronary syndrome (ACS). Anemia in patients with ACS adversely affects their clinical outcomes, yet hemoglobin level on admission is not included as a parameter in the GRACE score. We hypothesized that hemoglobin level on admission would improve the predictive value of the GRACE score. METHODS: We retrospectively studied one-year mortality in consecutive ACS patients included in the ACSIS (acute coronary syndrome Israeli Survey) registry between the years 2008 and 2013. Patients were classified into groups according to the GRACE score - ≥140 or <140, and according to the hemoglobin level: severe anemia - <8g/dl; mild anemia - 8-12g/dl; no anemia - >12g/dl. We analyzed the incremental predictive value of admission hemoglobin levels over the GRACE score. RESULTS: We studied 11,505 patients. The GRACE score predicted 1-year mortality with an area under the curve (ROC) of 0.68 (95% CI 0.66-0.7). When hemoglobin level on admission was incorporated into the model, the ROC increased to 0.73 (95% CI 0.71-0.75, p<0.001). The incremental value of hemoglobin levels on admission was significant only in the low (<140) GRACE score group. CONCLUSIONS: In patients a with low GRACE score (<140), anemia on admission has additional predictive value for one-year mortality. In contrast, in patients with a high GRACE risk score, hemoglobin level on admission did not improve prediction accuracy.