Giuseppe Scrascia1, Pietro Guida2, Sergio Maria Caparrotti3, Giuseppe Capone4, Marco Contini5, Mauro Cassese6, Vitantonio Fanelli5, Gianluca Martinelli6, Valerio Mazzei3, Salvatore Zaccaria7, Domenico Paparella8. 1. Division of Cardiac Surgery, Department of Emergency and Organ Transplant (D.E.T.O.), University of Bari "Aldo Moro," Bari, Italy; Division of Cardiac Surgery, Vito Fazzi Hospital, Lecce, Italy. 2. Health Regional Agency of Puglia (AReS Puglia), Puglia, Italy. 3. Division of Cardiac Surgery, Villa Bianca Hospital, Bari, Italy. 4. Division of Cardiac Surgery, Department of Emergency and Organ Transplant (D.E.T.O.), University of Bari "Aldo Moro," Bari, Italy. 5. Division of Cardiac Surgery, Villa Verde Hospital, Taranto, Italy. 6. Division of Cardiac Surgery, Santa Maria Hospital, Bari, Italy. 7. Division of Cardiac Surgery, Vito Fazzi Hospital, Lecce, Italy. 8. Division of Cardiac Surgery, Department of Emergency and Organ Transplant (D.E.T.O.), University of Bari "Aldo Moro," Bari, Italy. Electronic address: domenico.paparella@uniba.it.
Abstract
BACKGROUND: Anemia is a risk factor for adverse events after cardiac operations. We evaluated the incremental value of preoperative anemia over the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II to predict hospital death after cardiac operations. METHODS: Data for 4,594 consecutive adults (1,548 women [33.7%]), aged 67 ± 11 years, who underwent cardiac operations from January 2011 to July 2013 were extracted from the Regional Cardiac Surgery Registry of Puglia. The last preoperative hemoglobin value was used, according to World Health Organization criteria, to classify anemia as mild (hemoglobin 11.0 to 12.9 g/dL in men and 11.0 to 11.9 g/dL in women) in 1,021 patients (22.2%) and as moderate to severe (hemoglobin <11.0 g/dL) in 593 patients (12.9%). The EuroSCORE II was used to evaluate predicted hospital death after operations. Logistic regression analysis for in-hospital death was performed including EuroSCORE II risk factors and anemia, with model discrimination quantified by C statistic and risk classification by the use of net reclassification improvement (NRI). RESULTS: Overall expected and observed mortality rates were 4.4% and 5.9%. Anemia was significantly associated with a mortality rate of 3.4% in patients without anemia, 7.7% in mild anemia, and 15.7% in moderate to severe anemia (p < 0.001) and also at multivariate analysis correcting for EuroSCORE II (p < 0.001). When anemia was analyzed with EuroSCORE II, the model improved in discrimination (C statistic = 0.852 vs 0.860; p = 0.007) and reclassification (category free-NRI, 0.592; p < 0.001), preserving the calibration with good concordance between predicted probabilities and outcome. CONCLUSIONS: Preoperative anemia has strong association with operative death in cardiac surgical patients. Anemia provides significant incremental value over the EuroSCORE II and should be considered for assessment of cardiac surgical risk.
BACKGROUND:Anemia is a risk factor for adverse events after cardiac operations. We evaluated the incremental value of preoperative anemia over the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II to predict hospital death after cardiac operations. METHODS: Data for 4,594 consecutive adults (1,548 women [33.7%]), aged 67 ± 11 years, who underwent cardiac operations from January 2011 to July 2013 were extracted from the Regional Cardiac Surgery Registry of Puglia. The last preoperative hemoglobin value was used, according to World Health Organization criteria, to classify anemia as mild (hemoglobin 11.0 to 12.9 g/dL in men and 11.0 to 11.9 g/dL in women) in 1,021 patients (22.2%) and as moderate to severe (hemoglobin <11.0 g/dL) in 593 patients (12.9%). The EuroSCORE II was used to evaluate predicted hospital death after operations. Logistic regression analysis for in-hospital death was performed including EuroSCORE II risk factors and anemia, with model discrimination quantified by C statistic and risk classification by the use of net reclassification improvement (NRI). RESULTS: Overall expected and observed mortality rates were 4.4% and 5.9%. Anemia was significantly associated with a mortality rate of 3.4% in patients without anemia, 7.7% in mild anemia, and 15.7% in moderate to severe anemia (p < 0.001) and also at multivariate analysis correcting for EuroSCORE II (p < 0.001). When anemia was analyzed with EuroSCORE II, the model improved in discrimination (C statistic = 0.852 vs 0.860; p = 0.007) and reclassification (category free-NRI, 0.592; p < 0.001), preserving the calibration with good concordance between predicted probabilities and outcome. CONCLUSIONS: Preoperative anemia has strong association with operative death in cardiac surgical patients. Anemia provides significant incremental value over the EuroSCORE II and should be considered for assessment of cardiac surgical risk.
Authors: Sophie Provenchère; Arnaud Chevalier; Walid Ghodbane; Claire Bouleti; Philippe Montravers; Dan Longrois; Bernard Iung Journal: PLoS One Date: 2017-11-16 Impact factor: 3.240