The predictive value of coronary computed tomography angiography-derived fractional flow reserve for perioperative cardiac events in lung cancer surgery

Zhao Ma,Shuo Dong,Songlei Ou,Xuchen Ma,Linqi Liu,Ziyu An,Feng Xu,Dongfeng Zhang,Chenchen Tu,Xiantao Song,Hongjia Zhang

DOI: https://doi.org/10.1016/j.ejrad.2024.111688

2024-08-22

Abstract:Purpose: As a non-invasive coronary functional examination, coronary computed tomography angiography (CCTA)-derived fractional flow reserve (CT-FFR) showed predictive value in several non-cardiac surgeries. This study aimed to evaluate the predictive value of CT-FFR in lung cancer surgery. Method: We retrospectively collected 227 patients from January 2017 to June 2022 and used machine learning-based CT-FFR to evaluate the stable coronary artery disease (CAD) patients undergoing lung cancer surgery. The major adverse cardiac event (MACE) was defined as perioperative myocardial injury (PMI), myocardial infarction, heart failure, atrial and ventricular arrhythmia with hemodynamic disorder, cardiogenic shock and cardiac death. The multivariate logistic regression analysis was performed to identify risk factors for MACE and PMI. The discriminative capacity, goodness-of-fit, and reclassification improvement of prediction model were determined before and after the addition of CT-FFR≤0.8. Results: The incidence of MACE was 20.7 % and PMI was 15.9 %. CT-FFR significantly outperformed CCTA in terms of accuracy for predicting MACE (0.737 vs 0.524). In the multivariate regression analysis, CT-FFR≤0.8 was an independent risk factor for both MACE [OR=10.77 (4.637, 25.016), P<0.001] and PMI [OR=8.255 (3.372, 20.207), P<0.001]. Additionally, we found that the performance of prediction model for both MACE and PMI improved after the addition of CT-FFR. Conclusions: CT-FFR can be used to assess the risk of perioperative MACE and PMI in patients with stable CAD undergoing lung cancer surgery. It adds prognostic information in the cardiac evaluation of patients undergoing lung cancer surgery.