185 Early extubation following cardiac surgery: Pre-operative scorecard using the Manitoba model

Abstract

BackgroundExtubation protocols following cardiac surgery vary depending on institutional practices. Early extubation has gained popularity as it has been shown to reduce ICU and hospital length of stay (LOS). In spite of its increasing practice, there remains no predictive pre-operative tool to help identify potential candidates for early extubation. The aim of this study was to develop a pre-operative bedside scorecard to aid in identifying potential candidates for early (defined as intra-operative) extubation following cardiac surgery.MethodsFollowing a retrospective cohort analysis of consecutive patients undergoing cardiac surgery at a single tertiary center from January 2007 to January 2008 (derivation cohort), independent predictors of failure to extubate in the operating room were identified by regression analysis. Independent predictors with the highest coefficients and clinical relevance were used to develop the predictive scorecard. Verification of the scorecard was performed using bootstrapping sampling analysis on the original derivation cohort, and utilizing the predictive scorecard on a validation cohort of patients undergoing cardiac surgery from February 2008 to September 2008.Results1,083 patients underwent cardiac surgery in the derivation cohort, of which 604 patients (55.8%) were successfully extubated intra-operatively. Following regression analysis, nine risk factors for failed intra-operative extubation were used from this cohort to create the predictive scorecard (Table 1). Verification of the scorecard via bootstrapping produced a C-statistic of 0.76 (95% C.I. 0.72-0.79). Verification with the validation cohort (n = 792) produced an observed-to-expected ratio of 0.97 (Figure 1).Tabled 1View Large Image Figure ViewerDownload Hi-res image Download (PPT)ConclusionsUsing the Manitoba model for early extubation, we present a validated assessment tool predicting successful intra-operative extubation following cardiac surgery. BackgroundExtubation protocols following cardiac surgery vary depending on institutional practices. Early extubation has gained popularity as it has been shown to reduce ICU and hospital length of stay (LOS). In spite of its increasing practice, there remains no predictive pre-operative tool to help identify potential candidates for early extubation. The aim of this study was to develop a pre-operative bedside scorecard to aid in identifying potential candidates for early (defined as intra-operative) extubation following cardiac surgery. Extubation protocols following cardiac surgery vary depending on institutional practices. Early extubation has gained popularity as it has been shown to reduce ICU and hospital length of stay (LOS). In spite of its increasing practice, there remains no predictive pre-operative tool to help identify potential candidates for early extubation. The aim of this study was to develop a pre-operative bedside scorecard to aid in identifying potential candidates for early (defined as intra-operative) extubation following cardiac surgery. MethodsFollowing a retrospective cohort analysis of consecutive patients undergoing cardiac surgery at a single tertiary center from January 2007 to January 2008 (derivation cohort), independent predictors of failure to extubate in the operating room were identified by regression analysis. Independent predictors with the highest coefficients and clinical relevance were used to develop the predictive scorecard. Verification of the scorecard was performed using bootstrapping sampling analysis on the original derivation cohort, and utilizing the predictive scorecard on a validation cohort of patients undergoing cardiac surgery from February 2008 to September 2008. Following a retrospective cohort analysis of consecutive patients undergoing cardiac surgery at a single tertiary center from January 2007 to January 2008 (derivation cohort), independent predictors of failure to extubate in the operating room were identified by regression analysis. Independent predictors with the highest coefficients and clinical relevance were used to develop the predictive scorecard. Verification of the scorecard was performed using bootstrapping sampling analysis on the original derivation cohort, and utilizing the predictive scorecard on a validation cohort of patients undergoing cardiac surgery from February 2008 to September 2008. Results1,083 patients underwent cardiac surgery in the derivation cohort, of which 604 patients (55.8%) were successfully extubated intra-operatively. Following regression analysis, nine risk factors for failed intra-operative extubation were used from this cohort to create the predictive scorecard (Table 1). Verification of the scorecard via bootstrapping produced a C-statistic of 0.76 (95% C.I. 0.72-0.79). Verification with the validation cohort (n = 792) produced an observed-to-expected ratio of 0.97 (Figure 1).Tabled 1 1,083 patients underwent cardiac surgery in the derivation cohort, of which 604 patients (55.8%) were successfully extubated intra-operatively. Following regression analysis, nine risk factors for failed intra-operative extubation were used from this cohort to create the predictive scorecard (Table 1). Verification of the scorecard via bootstrapping produced a C-statistic of 0.76 (95% C.I. 0.72-0.79). Verification with the validation cohort (n = 792) produced an observed-to-expected ratio of 0.97 (Figure 1). ConclusionsUsing the Manitoba model for early extubation, we present a validated assessment tool predicting successful intra-operative extubation following cardiac surgery. Using the Manitoba model for early extubation, we present a validated assessment tool predicting successful intra-operative extubation following cardiac surgery.