Log in to MyACC
Preoperative NT-proBNP and Cardiovascular Events After Noncardiac Surgery
Study Questions:
Previous studies have suggested that the use of preoperative natriuretic peptide measurements can improve identification of surgical patients who are at risk for adverse cardiovascular outcomes. Can preoperative N-terminal pro–B-type natriuretic peptide (NT-proBNP) measurements, together with the Revised Cardiac Risk Index (RCRI), improve prediction of adverse postoperative cardiovascular outcomes compared to use of the RCRI alone in a very large surgical population?
Methods:
Across 16 centers in nine countries, 18,920 patients ≥45 years of age undergoing major noncardiac surgery were enrolled in the VISION trial. From this cohort, investigators recruited a subpopulation of subjects for a nested prospective study in which preoperative NT-proBNP levels were measured. During the study period, both investigators and clinicians were blinded to the preoperative NT-proBNP results. In all subjects, troponin T or high-sensitivity cardiac troponin T were measured at postoperative 6 and 12 hours, and days 1-3. The primary outcome was a composite of myocardial injury (MINS) and vascular mortality. MINS included both clinical myocardial infarction and myocardial injury based on troponin elevation. Optimum NT-proBNP thresholds were determined iteratively to establish ranges that maximized performance of the modeled relationship. The predictive value of adding NT-proBNP to a multivariate RCRI model was determined using net absolute reclassification improvement.
Results:
Among the subpopulation of 10,402 patients participating in this nested study, the primary outcome occurred in 12.25%, with 1,237 cases of MINS (11.9%) and 54 cases of vascular mortality (0.5%) within 30 days of surgery. The addition of NT-proBNP thresholds to the RCRI model improved its predictive capability, with the c-statistic increasing from 0.65 to 0.73, and overall 26% improved net risk reclassification.
In a Cox proportional hazard model, with NT-proBNP <100 pg/ml as a reference group and 5.2% incidence of primary composite outcome, those in the NT-proBNP ranges of 100-199, 200-1,499, and ≥1500 pg/ml had 12.3%, 20.8%, and 37.5% incidence of primary outcome, with respective adjusted hazard ratios of 2.27, 3.63, and 5.82. These NT-proBNP thresholds were also independently predictive of secondary outcomes, including 30-day myocardial infarction and all-cause mortality.
Conclusions:
In a large multinational cohort of patients undergoing noncardiac surgery, preoperative NT-proBNP levels within specific ranges improved the predictive value of RCRI in determining 30-day postoperative risk of myocardial injury and vascular mortality.
Perspective:
Although previous studies have demonstrated significant associations between natriuretic peptide elevation and adverse postoperative cardiovascular outcomes, this study had the advantage of a very large population size that allowed the investigators to define thresholds of NT-proBNP elevation that improved risk prediction significantly. Although these findings are highly relevant, they are likewise context-dependent. Depending on the representation of factors including patient comorbidity, surgical urgency, and procedural complexity, risk stratification thresholds for natriuretic peptides will likely require ongoing recalibration to define optimum ranges to fit any specific population of interest. A more interesting question will be how and whether improved risk prediction models like these can facilitate better clinical decision making, communication, and outcomes for our surgical patients.
Clinical Topics: Anticoagulation Management, Prevention
Keywords: General Surgery, Myocardial Infarction, Natriuretic Peptide, Brain, Natriuretic Peptides, Peptide Fragments, Preoperative Care, Risk Assessment, Risk Factors, Secondary Prevention, Troponin, Troponin T, Vascular Diseases
< Back to Listings
