LAUSR

Automated plaque quantification derived from coronary CT angiogragphy datasets provides exact and reliable assessment of coronary atherosclerosis burden. To investigate the potential for category based reclassification of patients based upon quantified coronary plaque volume in patients with 10 years of follow-up. Coronary PV was quantified with dedicated software in 1577 patients with suspected coronary artery disease. Cardiac death and acute coronary syndrome were defined as endpoint. Patients were initially classified as low, intermediate or high risk based upon the Morise score. Quantified PV was used to reclassify patients as shown in Figure 1 Panel A. The applied cutoffs (PV=0, PV0–110.5 mm3 and PV>110.5mm3) were established by previous work of our group. Categorical net reclassification improvement was used to compare the initial and updated patient stratification. Patients were followed for 10.4 years. The combined endpoint occurred in 59 patients, of whom 36 suffered from cardiac death, 18 had non-fatal myocardial infarction and 5 presented with unstable angina requiring recascularisation. The Morise score classified the majority of patients as intermediate risk patients (71%) and smaller proportions as low risk (21.9%) or high risk (7.1%). Quantified PV based reclassification resulted in reclassification of 800 (51%) patients. Of those, the majority was classified into a lower risk category (n=502). Calculation of the categorical NRI proved a significantly superior risk stratification when compared to the initial risk groups (0.48 with 95% CI 0.13 and 0.68, p<0.001). The reclassification matrix is shown in Figure 1 Panel B. After reclassification, the estimated 10-year event rates for low, intermediate and high risk patients were 0.6% (95% CI 0 and 1.3%), 4.8% (95% CI 2.4 and 7.2%) and 11.3% (95% CI 6.6 and 13.9%) respectively. Quantified coronary PV permits an effective and useful approach to reclassify patients with suspected coronary artery disease into superior risk categories. Type of funding source: None