The ability of computed tomography angiography to predict 5-year mortality in the SYNTAX III REVOLUTION trial

Abstract

Background
The feasibility of using coronary computed tomography angiography (CCTA) for long-term prediction of vital prognosis post-revascularization remains unknown.

Objectives
To compare the prognostic value of the SYNTAX score II 2020 (SS-2020) derived from invasive coronary angiography (ICA) or CCTA in patients with three-vessel disease and/or left main coronary artery disease undergoing percutaneous or surgical revascularization.

Methods
In the SYNTAX III REVOLUTION trial, the probability of death at five years was retrospectively assessed by calculating the SS-2020 using ICA and CCTA. High- and low-risk patients for mortality were categorized according to the median percentages of predicted mortality based on both modalities. The discriminative abilities of the SS-2020 were assessed using Harrell's C statistic.

Results
The vital status at five years of the 215 patients revascularized percutaneously (64 patients, 29.8%) or surgically (151 patients, 70.2%) was established through national registries. In patients undergoing revascularization, the SS-2020 was possibly helpful in discriminating vital prognosis at 5 years, with similar results seen with ICA and CCTA (C-index with ICA ​= ​0.75, intercept ​= ​−0.19, slope ​= ​0.92 and C-index with CCTA ​= ​0.75, intercept ​= ​−0.22, slope ​= ​0.99). In high- and low-risk patients, Kaplan-Meier estimates showed significant, and almost identical relative differences in observed mortality, irrespective of imaging modality (ICA: 93.8% vs 78.7%, log-lank P ​< ​0.001; CCTA: 93.7% vs 78.5%, log-lank P ​< ​0.001).

Conclusions
The predictive ability of the SS-2020 for five-year all-cause mortality derived from ICA and CCTA was comparable, and could helpfully discriminate vital prognosis in high- and low-risk patients.