RESUMO
BACKGROUND: Patients undergoing non-cardiac surgery are at risk for peri-operative major cardiac events (PMCEs). The most common risk assessment tool is Revised Cardiac Risk Index (RCRI). N-terminal pro-brain natriuretic peptide (NT-proBNP) measured peri-operatively has prognostic information but the implication is uncertain. This study aimed to determine the accuracy of combining NT-proBNP and RCRI in predicting the PMCE in major non-cardiac surgery. METHODS: We performed a prospective cohort study to include non-cardiac surgical patients with moderate or high risk. PMCE included myocardial infarction, pulmonary edema, severe cardiac arrhythmias, and cardiac death occurring within 30 days post-operatively. Logistic regression models and the receiver operating characteristic (ROC) curves were used to determine the discriminative ability of NT-proBNP alone or incorporation with RCRI or its components in predicting PMCE. RESULTS: A total of 366 patients was included in the study with 48 PMCEs. When predicting PMCE, the area under the ROC curve (AUC) (95%-CI) of NT-proBNP alone and NT-proBNP incorporated with RCRI were 0.875 (0.819-0.932) and 0.882 (0.827-0.937), respectively. When incorporating NT-proBNP with the RCRI's components, the best four chosen models had the AUCs between 0.879 and 0.891. All these AUCs were not significantly different with the AUC of NT-proBNP alone. CONCLUSIONS: Higher preoperative NT-proBNP level leads to the increased risk of PMCE in patients undergoing non-cardiac surgery. Compared to NT-proBNP alone, the combination of NT-proBNP with the RCRI and other factors does not improve the accuracy in predicting PMCE. Future large studies are required to build a more accurate risk score.