Dataset Information

A novel machine learning-derived radiotranscriptomic signature of perivascular fat improves cardiac risk prediction using coronary CT angiography.

ABSTRACT:

Background

Coronary inflammation induces dynamic changes in the balance between water and lipid content in perivascular adipose tissue (PVAT), as captured by perivascular Fat Attenuation Index (FAI) in standard coronary CT angiography (CCTA). However, inflammation is not the only process involved in atherogenesis and we hypothesized that additional radiomic signatures of adverse fibrotic and microvascular PVAT remodelling, may further improve cardiac risk prediction.

Methods and results

We present a new artificial intelligence-powered method to predict cardiac risk by analysing the radiomic profile of coronary PVAT, developed and validated in patient cohorts acquired in three different studies. In Study 1, adipose tissue biopsies were obtained from 167 patients undergoing cardiac surgery, and the expression of genes representing inflammation, fibrosis and vascularity was linked with the radiomic features extracted from tissue CT images. Adipose tissue wavelet-transformed mean attenuation (captured by FAI) was the most sensitive radiomic feature in describing tissue inflammation (TNFA expression), while features of radiomic texture were related to adipose tissue fibrosis (COL1A1 expression) and vascularity (CD31 expression). In Study 2, we analysed 1391 coronary PVAT radiomic features in 101 patients who experienced major adverse cardiac events (MACE) within 5 years of having a CCTA and 101 matched controls, training and validating a machine learning (random forest) algorithm (fat radiomic profile, FRP) to discriminate cases from controls (C-statistic 0.77 [95%CI: 0.62-0.93] in the external validation set). The coronary FRP signature was then tested in 1575 consecutive eligible participants in the SCOT-HEART trial, where it significantly improved MACE prediction beyond traditional risk stratification that included risk factors, coronary calcium score, coronary stenosis, and high-risk plaque features on CCTA (Δ[C-statistic] = 0.126, P < 0.001). In Study 3, FRP was significantly higher in 44 patients presenting with acute myocardial infarction compared with 44 matched controls, but unlike FAI, remained unchanged 6 months after the index event, confirming that FRP detects persistent PVAT changes not captured by FAI.

Conclusion

The CCTA-based radiomic profiling of coronary artery PVAT detects perivascular structural remodelling associated with coronary artery disease, beyond inflammation. A new artificial intelligence (AI)-powered imaging biomarker (FRP) leads to a striking improvement of cardiac risk prediction over and above the current state-of-the-art.

SUBMITTER: Oikonomou EK

PROVIDER: S-EPMC6855141 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Purpose To evaluate the frequency and implications of perivascular fat stranding on coronary computed tomography (CT) angiograms obtained for suspected acute coronary syndrome (ACS). Materials and Methods This retrospective registry study was approved by the institutional review board. The authors reviewed the medical records and images of 1403 consecutive patients (796 men, 607 women; mean age, 52.8 years) who underwent coronary CT angiography at the emergency department from February 2012 to March 2016. Fat attenuation, length and number of circumferential quadrants of the affected segment, and attenuation values in the unaffected epicardial and subcutaneous fat were measured. "Cases" were defined as patients with perivascular fat stranding. Patients with significant stenosis but without fat stranding were considered control subjects. Baseline imaging characteristics, ACS frequency, and results of subsequent downstream testing were compared between cases and control subjects by using two-sample t, Mann-Whitney U, and Fisher tests. Results Perivascular fat stranding was seen in 11 subjects, nine with atherosclerotic lesions and two with spontaneous coronary artery dissections, with a mean fat stranding length of 19.2 mm and circumferential extent averaging 2.9 quadrants. The mean attenuation of perivascular fat stranding, normal epicardial fat, and normal subcutaneous fat was 17, -93.2, and -109.3 HU, respectively (P < .001). Significant differences (P < .05) between cases and control subjects included lower Agatston score, presence of wall motion abnormality, and initial elevation of serum troponin level. ACS frequency was 45.4% in cases and 3.8% in control subjects (P = .001). Conclusion Recognition of perivascular fat stranding may be a helpful additional predictor of culprit lesion and marker of risk for ACS in patients with significant stenosis or spontaneous coronary artery dissection. © RSNA, 2018 Online supplemental material is available for this article.

Project description:ObjectivesThis study sought to determine prognostic value of nonobstructive coronary artery disease (CAD) for atherosclerotic cardiovascular disease (ASCVD) events and to determine whether incorporation of this information into the pooled cohort equation reclassifies recommendations for statin therapy as defined by the 2013 guidelines for cholesterol management of the American College of Cardiology and American Heart Association (ACC/AHA).BackgroundDetection of nonobstructive CAD by coronary computed tomography angiography may improve risk stratification and permit individualized and more appropriate allocation of statin therapy.MethodsThis study determined the pooled hazard ratio of nonobstructive CAD for ASCVD events from published studies and incorporated this information into the ACC/AHA pooled cohort equation. The study calculated revised sex- and ethnicity-based 10-year ASCVD risk and determined boundaries corresponding to the original 7.5% risk for ASCVD events. It also assessed reclassification for statin eligibility by incorporating the results from meta-analysis to individual patients from a separate cohort.ResultsThis study included 2 studies (2,295 subjects; 66% male; prevalence of nonobstructive CAD, 47%; median follow-up, 49 months; 67 ASCVD events). The hazard ratio of nonobstructive CAD for ASCVD events was 3.2 (95% confidence interval: 1.5 to 6.7). Incorporation of this information into the pooled cohort equation resulted in reclassification toward statin eligibility in individuals with nonobstructive CAD, with an original ASCVD score of 3.0% and 5.9% or higher in African-American women and men and a score of 4.4% and 4.6% or higher in Caucasian women and men, respectively. The absence of nonobstructive CAD resulted in reclassification toward statin ineligibility if the original ASCVD score was as 10.0% and 17.9% or lower in African-American women and men and 13.7% and 14.3% or lower in Caucasian women and men, respectively. Reclassification is observed in 14% of patients.ConclusionsDetection of nonobstructive CAD by coronary computed tomography angiography improves risk stratification and permits individualized and more appropriate allocation of statin therapy across sex and ethnicity groups.