Abstract 13677: Unrequested Imaging Findings From Routine Chest CT Identify Subjects at High-risk of Future Cardiovascular Events
Background: Through the increased use of chest Computed Tomography (CT)scans radiologists are routinely confronted with unrequested findings reflecting (preclinical) manifestations of atherosclerosis. We investigated the contribution of unrequested findings on chest CT in detecting subjects at high-risk for cardiovascular disease (CVD) by derivating and validating a CT imaging based prediction rule.
Methods: The cohort comprised 10 410 patients, who underwent diagnostic chest CT for non-cardiovascular indications. During a mean follow-up of 3.7 years (max. 7.0 years), 1148 CVD events (cases) were identified. Using a case-cohort approach, CT scans from the cases and from a ≈10% random sample of the baseline cohort (n=1366) were visually graded for several cardiovascular findings. Multivariable Cox proportional hazards analysis with backward elimination technique was used in the derivation cohort to derivate the best-fitting parsimonious prediction model.
Results: The final model included age, gender, CT indication, left anterior descending coronary artery calcifications, mitral valve calcifications, descending aorta calcifications and heart size (figure). The model demonstrated to have a good discriminative ability with a c-statistic of 0.71 (95%CI 0.68-0.74) and a good overall calibration as assessed in a separate validation cohort (n=1653, mean follow-up 3.7 years).
Conclusion: Using data obtained in a large cohort of patients undergoing CT scanning for a non-cardiovascular indication we were able to show that the use of unrequested information available in the scans may help to adequately detect subjects at a high-risk of future cardiovascular events. The resulting prediction rule may be utilized to assist clinicians to refer these patients for timely preventive cardiovascular risk management.
- © 2013 by American Heart Association, Inc.