583407
research-article2015
CNU0010.1177/1474515115583407European Journal of Cardiovascular NursingLopes et al.
EUROPEAN SOCIETY OF CARDIOLOGY ®
Original Article
Predictive factors for bleeding-related re-exploration after cardiac surgery: A prospective cohort study
European Journal of Cardiovascular Nursing 1–8 © The European Society of Cardiology 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1474515115583407 cnu.sagepub.com
Camila T Lopes1, Evelise H Fadini Reis Brunori1, Vinicius Batista Santos1, Sue A Moorhead2, Juliana de Lima Lopes1 and Alba L Bottura Leite de Barros1
Abstract Background: Bleeding-related re-exploration is a life-threatening complication after cardiac surgery. Nurses must be aware of important risk factors for this complication so that their assessment, monitoring and evaluation activities can be prioritized, focused and anticipated. Aims: To identify the predictive factors for bleeding-related re-exploration after cardiac surgery and to describe the sources of postoperative bleeding. Methods: This is a prospective cohort study at a tertiary cardiac school-hospital in São Paulo/SP, Brazil. Adult patients (n=323) submitted to surgical correction of acquired cardiac diseases were included. Potential risk factors for bleedingrelated re-exploration within the 24 hours following admission to the intensive care unit were investigated in the patients’ charts. A univariate analysis and a multiple analysis through logistic regression were conducted to identify the outcome predictors. The area under the receiver-operating characteristic curve was calculated as a measure of accuracy considering the cut-off points with the highest sensitivity and specificity. Results: The univariate factors significantly associated with bleeding-related re-exploration were a lower preoperative platelet count, a lower number of bypasses in coronary artery bypass surgery and postoperatively, a lower body temperature, infusion of lower intravenous volume, a higher positive end-expiratory pressure during mechanical ventilation and transfusion of blood products. The independent predictors of bleeding-related re-exploration included postoperative red blood cell transfusion, and transfusion of fresh frozen plasma, platelet or cryoprecipitate units. These predictors had a sensitivity of 87.5%, a specificity of 99.28% and an accuracy of 97.93%. Conclusions: Blood product transfusion postoperatively is an independent predictor of bleeding-related re-exploration. Surgical errors prevailed as sources of bleeding. Keywords Cardiac surgery, cardiac surgical procedures, postoperative hemorrhage, reoperation, risk factors Date received: 29 January 2015; revised: 12 March 2015; accepted: 31 March 2015
Introduction During the 2-h recovery period after cardiac surgery, 21% of patients present with bleeding complications that require interventions.1 Excessive bleeding can occur in 6.4% to 52.9% of patients submitted to coronary artery bypass grafting (CABG).2–4 Excessive bleeding after cardiac surgery is a determinant of emergency re-exploration, which is associated with 5.9% mortality within the first 24 h.5 Bleeding-related re-exploration (BLR) is associated with higher incidences of hospital mortality and morbidity and prolonged length of stay.6–10 The cost of hospitalization
increases to €6290 after isolated first-time CABG due to re-exploration.11 1Paulista
Nursing School, Federal University of Sao Paulo (EPEUNIFESP), Brazil 2University of Iowa, Iowa City, USA Corresponding author: Camila Takao Lopes, Paulista Nursing School, Federal University of São Paulo (EPE-UNIFESP), 754 Napoleão de Barros St, 04024-002, São Paulo-SP, Brazil. Email: [email protected]
2 To the best of our knowledge, a few studies have investigated pre-, intra- and postoperative predictors of BLR after cardiac surgery through multiple/multivariate analysis or propensity-matched groups to control for possible confounders since 2009. The frequency of this complication is highly variable, ranging from 1.3% to 14.6%.7,9,12–17 In addition to the highly variable frequency of BLR, the independent predictors also differ across institutions. The independent predictors of BLR associated with patient characteristics include male gender;8,17 older age;8,17 a lower body mass index (BMI) or body surface area;7,12,13,17 higher preoperative serum creatinine, preoperative dialysis or decreased glomerular filtration rate;7,13,14,18 a low left ventricle ejection fraction (LVEF), advanced New York Heart Association classes,15 cardiogenic shock or congestive heart failure;17 peripheral vascular disease;8,17 lower preoperative hemoglobin levels (Hb);15 being non-White, being nondiabetic, submission to previous cardiovascular interventions, cerebrovascular disease, immunosuppressive therapy, and chronic lung disease – moderate to severe.18 Preoperative exposure to anticoagulants has also been found as an independent predictor of the outcome: aspirin;14,16 clopidogrel or aspirin within five and two days, respectively, before surgery;16 clopidogrel use prior to CABG14,17 and thienopyridine and glycoprotein IIb/IIIa receptor antagonists before CABG.17 The independent predictors of BLR associated with the procedure include urgent/emergency surgery;12,17 an individual surgeon;13 increased cardiopulmonary bypass (CPB) time;7 three-vessel coronary artery disease and percutaneous coronary intervention 97 patients presented excessive bleeding, which was also investigated as an outcome for a parallel study (data not shown; in press).
Outcome variable The outcome investigated was bleeding-related return to the operating room as a dichotomous variable, namely reexploration, required within the 24 h following admission to the postoperative ICU. Assessment for the occurrence of this complication was performed 24 h after admission to the ICU. The occurrence of the outcome was evaluated in the immediate postoperative period because around 77% of cardiac surgery patients requiring re-exploration return to the operating room in this period.5 Also, because we adopted a prospective design, we did not have available time and human resources to extend the study to more than 24 h postoperative. The decision for re-exploration was an individual prerogative of the surgeon. However, it was generally considered if the patient had excessive chest/mediastinal bleeding or evidence of cardiac tamponade on transthoracic echocardiogram accompanied by hemodynamic instability non-responsive to intravenous fluids requiring increased inotropic and vasoactive support.
Independent variables Methods Study design and settings This is a prospective cohort study. Data collection was performed at a tertiary school-hospital, which is a reference in cardiac care in São Paulo, SP, Brazil, with over 2.5 million procedures a year.
The selection of potential risk factors for BLR resulted from the findings of studies investigating predictive factors for excessive bleeding after cardiac surgeries through multiple/multivariate analysis.18,20–34 Patient-related factors (Table 1), surgical factors (Table 2) and factors (Table 3) were investigated in order to depict a real-world unselected cohort.
3
Lopes et al. Table 1. Patient-related factors associated with bleeding-related re-exploration in adult patients after cardiac surgery. Variable
Re-exploration (n=18)
No re-exploration (n=305)
p-value
Age, years Weight, kg Height, m Body mass index, kg/m2 Preoperative platelet count, × 103/mm3 Preoperative platelet count
research-article2015
CNU0010.1177/1474515115583407European Journal of Cardiovascular NursingLopes et al.
EUROPEAN SOCIETY OF CARDIOLOGY ®
Original Article
Predictive factors for bleeding-related re-exploration after cardiac surgery: A prospective cohort study
European Journal of Cardiovascular Nursing 1–8 © The European Society of Cardiology 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1474515115583407 cnu.sagepub.com
Camila T Lopes1, Evelise H Fadini Reis Brunori1, Vinicius Batista Santos1, Sue A Moorhead2, Juliana de Lima Lopes1 and Alba L Bottura Leite de Barros1
Abstract Background: Bleeding-related re-exploration is a life-threatening complication after cardiac surgery. Nurses must be aware of important risk factors for this complication so that their assessment, monitoring and evaluation activities can be prioritized, focused and anticipated. Aims: To identify the predictive factors for bleeding-related re-exploration after cardiac surgery and to describe the sources of postoperative bleeding. Methods: This is a prospective cohort study at a tertiary cardiac school-hospital in São Paulo/SP, Brazil. Adult patients (n=323) submitted to surgical correction of acquired cardiac diseases were included. Potential risk factors for bleedingrelated re-exploration within the 24 hours following admission to the intensive care unit were investigated in the patients’ charts. A univariate analysis and a multiple analysis through logistic regression were conducted to identify the outcome predictors. The area under the receiver-operating characteristic curve was calculated as a measure of accuracy considering the cut-off points with the highest sensitivity and specificity. Results: The univariate factors significantly associated with bleeding-related re-exploration were a lower preoperative platelet count, a lower number of bypasses in coronary artery bypass surgery and postoperatively, a lower body temperature, infusion of lower intravenous volume, a higher positive end-expiratory pressure during mechanical ventilation and transfusion of blood products. The independent predictors of bleeding-related re-exploration included postoperative red blood cell transfusion, and transfusion of fresh frozen plasma, platelet or cryoprecipitate units. These predictors had a sensitivity of 87.5%, a specificity of 99.28% and an accuracy of 97.93%. Conclusions: Blood product transfusion postoperatively is an independent predictor of bleeding-related re-exploration. Surgical errors prevailed as sources of bleeding. Keywords Cardiac surgery, cardiac surgical procedures, postoperative hemorrhage, reoperation, risk factors Date received: 29 January 2015; revised: 12 March 2015; accepted: 31 March 2015
Introduction During the 2-h recovery period after cardiac surgery, 21% of patients present with bleeding complications that require interventions.1 Excessive bleeding can occur in 6.4% to 52.9% of patients submitted to coronary artery bypass grafting (CABG).2–4 Excessive bleeding after cardiac surgery is a determinant of emergency re-exploration, which is associated with 5.9% mortality within the first 24 h.5 Bleeding-related re-exploration (BLR) is associated with higher incidences of hospital mortality and morbidity and prolonged length of stay.6–10 The cost of hospitalization
increases to €6290 after isolated first-time CABG due to re-exploration.11 1Paulista
Nursing School, Federal University of Sao Paulo (EPEUNIFESP), Brazil 2University of Iowa, Iowa City, USA Corresponding author: Camila Takao Lopes, Paulista Nursing School, Federal University of São Paulo (EPE-UNIFESP), 754 Napoleão de Barros St, 04024-002, São Paulo-SP, Brazil. Email: [email protected]
2 To the best of our knowledge, a few studies have investigated pre-, intra- and postoperative predictors of BLR after cardiac surgery through multiple/multivariate analysis or propensity-matched groups to control for possible confounders since 2009. The frequency of this complication is highly variable, ranging from 1.3% to 14.6%.7,9,12–17 In addition to the highly variable frequency of BLR, the independent predictors also differ across institutions. The independent predictors of BLR associated with patient characteristics include male gender;8,17 older age;8,17 a lower body mass index (BMI) or body surface area;7,12,13,17 higher preoperative serum creatinine, preoperative dialysis or decreased glomerular filtration rate;7,13,14,18 a low left ventricle ejection fraction (LVEF), advanced New York Heart Association classes,15 cardiogenic shock or congestive heart failure;17 peripheral vascular disease;8,17 lower preoperative hemoglobin levels (Hb);15 being non-White, being nondiabetic, submission to previous cardiovascular interventions, cerebrovascular disease, immunosuppressive therapy, and chronic lung disease – moderate to severe.18 Preoperative exposure to anticoagulants has also been found as an independent predictor of the outcome: aspirin;14,16 clopidogrel or aspirin within five and two days, respectively, before surgery;16 clopidogrel use prior to CABG14,17 and thienopyridine and glycoprotein IIb/IIIa receptor antagonists before CABG.17 The independent predictors of BLR associated with the procedure include urgent/emergency surgery;12,17 an individual surgeon;13 increased cardiopulmonary bypass (CPB) time;7 three-vessel coronary artery disease and percutaneous coronary intervention 97 patients presented excessive bleeding, which was also investigated as an outcome for a parallel study (data not shown; in press).
Outcome variable The outcome investigated was bleeding-related return to the operating room as a dichotomous variable, namely reexploration, required within the 24 h following admission to the postoperative ICU. Assessment for the occurrence of this complication was performed 24 h after admission to the ICU. The occurrence of the outcome was evaluated in the immediate postoperative period because around 77% of cardiac surgery patients requiring re-exploration return to the operating room in this period.5 Also, because we adopted a prospective design, we did not have available time and human resources to extend the study to more than 24 h postoperative. The decision for re-exploration was an individual prerogative of the surgeon. However, it was generally considered if the patient had excessive chest/mediastinal bleeding or evidence of cardiac tamponade on transthoracic echocardiogram accompanied by hemodynamic instability non-responsive to intravenous fluids requiring increased inotropic and vasoactive support.
Independent variables Methods Study design and settings This is a prospective cohort study. Data collection was performed at a tertiary school-hospital, which is a reference in cardiac care in São Paulo, SP, Brazil, with over 2.5 million procedures a year.
The selection of potential risk factors for BLR resulted from the findings of studies investigating predictive factors for excessive bleeding after cardiac surgeries through multiple/multivariate analysis.18,20–34 Patient-related factors (Table 1), surgical factors (Table 2) and factors (Table 3) were investigated in order to depict a real-world unselected cohort.
3
Lopes et al. Table 1. Patient-related factors associated with bleeding-related re-exploration in adult patients after cardiac surgery. Variable
Re-exploration (n=18)
No re-exploration (n=305)
p-value
Age, years Weight, kg Height, m Body mass index, kg/m2 Preoperative platelet count, × 103/mm3 Preoperative platelet count
