Periaortic Fat Attenuation on Nongated Noncontrast Chest CT Images to Assess Changes in Arterial Inflammation: Impact of Atorvastatin

BACKGROUND: Imaging markers of atherosclerotic inflammation are needed to enhance cardiovascular risk assessment and evaluate the impact of therapies. We sought to test the hypothesis that treatments impacting arterial inflammation can be evaluated using a simplified measure of periaortic fat attenuation (FA) assessed on noncontrast, nongated computed tomography (CT) of the descending thoracic aorta. METHODS: Measurements were performed on 18 F-fluorodeoxyglucose positron emission tomography/CT images from a double-blind, randomized trial conducted between 2008 and 2009 that assessed the impact of statin therapy on arterial inflammation. Periaortic adipose tissue quantification was performed on the chest CT images over a 10 cm portion of the descending aorta. FA was determined as the mean attenuation of the entire volume of delineated periaortic fat. Arterial inflammation (aorta) and leukopoietic activity (bone marrow and spleen) were assessed by measuring standardized uptake values on 18 F-fluorodeoxyglucose positron emission tomography images. Baseline relationships and changes from baseline to 12 weeks were assessed. All models evaluating FA were adjusted for baseline kilovoltage peak. RESULTS: Sixty subjects (79.9% male, mean age 60±8.9 years) with risk factors or established atherosclerosis (32 randomized to atorvastatin 10 mg, 28 randomized to atorvastatin 80 mg) were studied. On average, it took 88±17 seconds to assess FA per subject. At baseline, FA correlated with leukopoietic activity ( r =0.412; P =0.021 and r =0.442; P =0.013, for bone marrow and spleen, respectively). Furthermore, FA correlated with aortic inflammation assessed on 18 F-fluorodeoxyglucose positron emission tomography as quintiles ( r =0.274; P =0.043). Moreover, high dose (versus low dose) atorvastatin was associated with a significant reduction in FA after 12 weeks (standardized β=−0.603; P =0.010) after adjustment for baseline FA, kilovoltage peak, and prior statin use. CONCLUSIONS: Periaortic FA is a marker of atherosclerotic inflammation that can be easily measured on nongated, nonenhanced chest CT images and be used to provide insights into the impact of therapies on atherosclerotic inflammation.