OBJECTIVE: Analyse risk factors and construct a predictive model for identification of patients at risk of early out-of-hospital mortality after coronary reoperations (RECABG). METHODS: 505 patients, discharged from hospital after a RECABG (1987-1998), were studied by univariate and multivariate analysis. A stepwise selective procedure (p<0.05) was used to identify a subset of variables with prognostic value for early out-of-hospital mortality. This subset was used to calculate a prognostic score 'S' and a predicted probability 'p' for early out-of-hospital mortality, p=1/1+ e-s. Sensitivity analysis was used for evaluation. RESULTS: The best predictive variables for early out-of-hospital mortality were diabetes (p=0.002), lung disease (p=0.05), emergency operation (p=0.0001) and a perioperative myocardial infarction (p=0.0001). Emergency operation (p=0.001) and antegrade/retrograde cardioplegia (p<0.0000) were independent predictors of a perioperative myocardial infarction. The prognostic accuracy (ROC area) was 86%. Patients were classified into low risk (5%), intermediate risk (15%), high risk (30%) and very high risk (≥40%). A predicted probability of ≥0.40 was used as cut-off point. The specificity of this test was 99%, sensitivity 33%, predictive value of a positive test 79%, and 95% for a negative test. CONCLUSION: The results show that patients discharged from hospital after RECABG can be stratified according to their early out-of-hospital risk. A perioperative myocardial infarction is the major independent risk factor and can be affected by use of retrograde cardioplegia.
OBJECTIVE: Analyse risk factors and construct a predictive model for identification of patients at risk of early out-of-hospital mortality after coronary reoperations (RECABG). METHODS: 505 patients, discharged from hospital after a RECABG (1987-1998), were studied by univariate and multivariate analysis. A stepwise selective procedure (p<0.05) was used to identify a subset of variables with prognostic value for early out-of-hospital mortality. This subset was used to calculate a prognostic score 'S' and a predicted probability 'p' for early out-of-hospital mortality, p=1/1+ e-s. Sensitivity analysis was used for evaluation. RESULTS: The best predictive variables for early out-of-hospital mortality were diabetes (p=0.002), lung disease (p=0.05), emergency operation (p=0.0001) and a perioperative myocardial infarction (p=0.0001). Emergency operation (p=0.001) and antegrade/retrograde cardioplegia (p<0.0000) were independent predictors of a perioperative myocardial infarction. The prognostic accuracy (ROC area) was 86%. Patients were classified into low risk (5%), intermediate risk (15%), high risk (30%) and very high risk (≥40%). A predicted probability of ≥0.40 was used as cut-off point. The specificity of this test was 99%, sensitivity 33%, predictive value of a positive test 79%, and 95% for a negative test. CONCLUSION: The results show that patients discharged from hospital after RECABG can be stratified according to their early out-of-hospital risk. A perioperative myocardial infarction is the major independent risk factor and can be affected by use of retrograde cardioplegia.
Authors: M A Borger; V Rao; R D Weisel; A A Floh; G Cohen; C M Feindel; H E Scully; L L Mickleborough; T M Yau Journal: J Thorac Cardiovasc Surg Date: 2001-01 Impact factor: 5.209