Comparison of High-Sensitivity Troponin T Versus N-Terminal Pro-B-Type Natriuretic Peptide for Prediction of Major Adverse Cardiovascular Events After Noncardiac Surgery.

Both preoperative hs-TnT and NT-proBNP were significantly associated with 30-day MACE and death after noncardiac surgery.

• Study pooled data from METREPAIR and MINSS studies involving 2251 patients with elevated cardiovascular risk undergoing elective noncardiac surgery.
• MACE occurred in 5.8% of patients (131/2251) and death in 1.6% (37/2251).
• Both biomarkers were significantly associated with outcomes in logistic regression models that included Revised Cardiac Risk Index (RCRI) and age.
• Patients had elevated cardiovascular risk and were undergoing elective noncardiac surgery.

Only hs-TnT statistically improved discrimination for predicting 30-day death when added to RCRI and age.

• AUC for RCRI+age+hs-TnT model for death = 0.680 versus AUC for RCRI+age = 0.674.
• Delta AUC: 0.006; P=0.047.
• NT-proBNP did not achieve a statistically significant improvement in discrimination for death.
• Neither biomarker produced a clinically large improvement in AUC for the death endpoint.

The contribution of both biomarkers to risk prediction over RCRI and age alone was limited for predicting MACE.

• Neither hs-TnT nor NT-proBNP significantly improved discrimination for the primary endpoint of 30-day MACE.
• Brier scores and calibration slopes showed good calibration for all models, including base models with RCRI and age only.
• Decision curve analysis was conducted at predefined thresholds for MACE and death.
• The incremental predictive value of both biomarkers beyond RCRI and age was described as limited.

Decision curve analysis showed no clinically meaningful advantage of hs-TnT over NT-proBNP or vice versa for predicting MACE or death.

• Benefit equivalents of RCRI+hs-TnT versus RCRI+NT-proBNP were -0.06 per 1000 and -0.17 per 1000 for MACE.
• For death at a 5% threshold, benefit equivalents were +0.06 per 1000 (hs-TnT) versus -0.08 per 1000 (NT-proBNP).
• For death at a 10% threshold, benefit equivalents were -0.35 per 1000 (hs-TnT) versus -0.06 per 1000 (NT-proBNP).
• These small and inconsistent differences across thresholds did not indicate a clear advantage for either biomarker.

No significant difference in predictive performance was found between hs-TnT and NT-proBNP, and a supporting literature review did not suggest differences between biomarkers.

• Direct comparison of hs-TnT versus NT-proBNP showed no statistically significant difference in overall predictive performance.
• A literature review was conducted as part of this secondary analysis and did not suggest differences between the two biomarkers.
• The study addressed a gap noted by conflicting guideline recommendations: the European Society of Cardiology recommends hs-TnT, while American cardiology and European anesthesiology societies support NT-proBNP.
• The analysis used area under the receiver operating characteristic curve, calibration slopes, Brier scores, and decision curve analysis to comprehensively assess prediction.

This was a secondary analysis pooling data from two prospective studies (METREPAIR and MINSS) of patients with elevated cardiovascular risk undergoing elective noncardiac surgery.

• Total pooled sample size was 2251 patients.
• Logistic regression models included RCRI and age as the base model, augmented by each biomarker.
• The primary endpoint was 30-day MACE; secondary endpoint was 30-day death.
• Prediction was assessed using AUC, calibration slopes, Brier scores, and decision curve analysis at predefined thresholds.