Site vs. core laboratory variability in computed tomographic angiography-derived SYNTAX scores in the SYNTAX III trial

Abstract

To investigate the variability between site and core laboratory (CL) calculation of the anatomical SYNTAX score (SS) based on coronary computed tomography angiography (CTA) alone and functional SS based on coronary CTA and fractional flow reserve derived from computed tomography (FFRCT) in the SYNTAX III trial. The SYNTAX III trial was a multicentre, international study that included 223 patients with three-vessel disease with or without left main involvement. Functional SS was computed by subtracting non-flow limiting stenoses (FFRCT > 0.80) from anatomical SS. SS was combined with clinical information to generate the SYNTAX score II (SS II) that provides treatment recommendations. The mean anatomical SS based on coronary CTA alone was 33.4 ± 12.7 by sites and 37.1 ± 13.4 by CL (P < 0.001). The mean functional SS based on coronary CTA and FFRCT was 30.5 ± 13.0 by sites and 33.3 ± 13.6 by CL (P < 0.001). The intraclass correlation coefficient was 0.49 [95% confidence interval (CI) 0.37-0.59) in anatomical SS and 0.62 (95% CI 0.52-0.70) in functional SS. The Cohen's κ comparing treatment recommendation between sites and CL was 0.68 (95% CI 0.58-0.78) based on anatomical SS and 0.71 (95% CI 0.60-0.82) based on functional SS. The mean anatomical SS derived from coronary CTA alone and functional SS based on coronary CTA and FFRCT were higher when assessed by the CL than by the sites themselves. However, substantial agreement in treatment recommendation by SS II between sites and CL was demonstrated. NCT02385279.