Computed Tomographic Coronary Angiography-Derived Plaque Characteristics Predict Major Adverse Cardiovascular Events: A Systematic Review and Meta-Analysis.

BACKGROUND: Computed tomographic coronary angiography is a noninvasive imaging modality that permits identification and characterization of coronary plaques. Despite consensus statements supporting routine reporting of computed tomographic coronary angiography plaque characteristics, there remains uncertainty whether these data convey prognostic information. We performed a systematic review and meta-analysis assessing the strength of association between computed tomographic coronary angiography-derived plaque characterization and major adverse cardiovascular events (MACE). METHODS AND RESULTS: Electronic databases were searched for studies reporting computed tomographic coronary angiography plaque characterization and MACE. Data were gathered on plaque morphology (noncalcified, partially calcified, and calcified) and high-risk plaque (HRP) features, including low-attenuation plaque, napkin-ring sign, spotty calcification, and positive remodeling. Of 5496 citations, 13 studies met inclusion criteria. Five hundred fifty-two (3.9%) MACE occurred in 13977 patients with mean follow-up ranging between 1.3 and 8.2 years. In terms of plaque morphology, the strongest association was observed for noncalcified plaque (hazard ratio [HR], 1.45; 95% confidence interval [CI], 1.24-1.70; P<0.001), with weaker associations found for partially calcified (HR, 1.37; 95% CI, 1.18-1.60; P<0.001) and calcified plaques (HR, 1.23; 95% CI, 1.16-1.30; P<0.001). All HRP features were strongly associated with MACE, including napkin-ring sign (HR, 5.06; 95% CI, 3.23-7.94; P<0.001), low-attenuation plaque (HR, 2.95; 95% CI, 2.03-4.29; P<0.001), positive remodeling (HR, 2.58; 95% CI, 1.84-3.61; P<0.001), and spotty calcification (HR, 2.25; 95% CI, 1.26-4.04; P=0.006). The presence of >=2 HRP features had highest risk of MACE (HR, 9.17; 95% CI, 4.10-20.50; P<0.001). CONCLUSION(S): These data demonstrate that HRP is most likely an independent predictor of MACE, which supports the inclusion of HRP re