Direct comparison of the accuracy of preoperative high-sensitivity cardiac troponin T to predict mortality, acute heart failure and perioperative myocardial infarction/injury after non-cardiac surgery

Background

Death, acute heart failure (AHF) and perioperative myocardial infarction/injury (PMI) are the most relevant cardiovascular complications following non-cardiac surgery. Unfortunately, the incidence of these complications are higher than expected. Currently available tools to predict these complications have only modest accuracy.

Purpose

To determine the accuracy of preoperative high-sensitivity cardiac troponin T (hs-cTnT) concentrations for prediction of mortality, AHF and PMI after non-cardiac surgery.

Methods

We prospectively included 4,709 patients at high cardiovascular risk undergoing non-cardiac surgery. Hs-cTnT concentrations were measured before surgery and, daily after surgery, for two days. PMI was defined as an absolute increase of 14ng/L (the 99th percentile of the assay used) from hs-cTnT baseline values. The primary endpoint was the diagnostic accuracy of preoperative hs-cTnT concentration to predict death, AHF and PMI within 30 days, as quantified by the area under the receiving-operating curve (AUC). Multivariate logistic regression analysis was performed to test the association between preoperative hs-cTnT and each endpoint.

Results

All-cause mortality occurred in 133 (3%), AHF in 84 (2%) and PMI in 742 (16%) patients. Preoperative hs-cTnT concentrations had good accuracy for prediction of death, AHF and PMI (AUC = 0.75 [95% CI, 0.71–0.79], 0.72 [95% CI, 0.67–0.77] and 0.73 [95% CI, 0.71–0.75], respectively). After adjusting for confounders, hs-cTnT remained an independent predictor for death with an adjusted odds ratio (aOR) of 2.1 (95% CI, 1.7–2.7, P<0.001) and for PMI (aOR 2.2, 95% CI, 1.9–2.4, P<0.001), but not for AHF (aOR 1.0, 95% CI, 0.7–1.4, P=0.99). An hs-cTnT concentration below 5ng/L was found in 526 (11%) patients, and this cut-off yielded a negative predictive value of 99.6% for the occurrence of death, 99.2% for AHF and 95.6% for PMI.

Conclusion

The preoperative hs-cTnT concentration has a good accuracy to predict mortality, AHF and PMI after non-cardiac surgery, but is an independent predictor only for mortality and PMI. A cut-off value of 5ng/L identifies a subgroup of patients at low risk for these complications.

Funding Acknowledgement

Type of funding source: Foundation. Main funding source(s): Swiss National Foundation, Swiss Herat Foundation