Relation of MRI Aortic Wall Area and Plaque to Incident Cardiovascular Events: The Framingham Heart Study

Abstract

Background Arterial arteriosclerosis and atherosclerosis reflect vascular disease, the subclinical detection of which allows opportunity for cardiovascular disease (CVD) prevention. Larger cohort studies simultaneously quantifying anatomic thoracic and abdominal aortic pathologic abnormalities are lacking in the literature. Purpose To investigate the association of aortic wall area (AWA) and atherosclerotic plaque presence and burden as measured on MRI scans with incident CVD in a community sample. Materials and Methods In this prospective cohort study, participants in the Framingham Heart Study Offspring Cohort without prevalent CVD underwent 1.5-T MRI (between 2002-2005) of the descending thoracic and abdominal aorta with electrocardiogram-gated axial T2-weighted black-blood acquisitions. The wall thickness of the thoracic aorta was measured at the pulmonary bifurcation level and used to calculate the AWA as the difference between cross-sectional vessel area and lumen area. For primary or secondary analyses, multivariable Cox proportional hazards regression models were used to examine the association of aortic MRI measures with risk of first-incident CVD events or stroke and coronary heart disease, respectively. Results In 1513 study participants (mean age, 64 years ± 9 [SD]; 842 women [56%]), 223 CVD events occurred during follow-up (median, 13.1 years), of which 97 were major events (myocardial infarction, ischemic stroke, or CVD death). In multivariable analysis, thoracic AWA and prevalent thoracic plaque were associated with incident CVD (hazard ratio [HR], 1.20 per SD unit [95% CI: 1.05, 1.37] [P = .006] and HR, 1.63 [95% CI: 1.12, 2.35] [P = .01], respectively). AWA and prevalent thoracic plaque were associated with increased hazards: 1.32 (95% CI: 1.07, 1.62; P = .01) and 2.20 (95% CI: 1.28, 3.79; P = .005), for stroke and coronary heart disease, respectively. Conclusion In middle-aged community-dwelling adults, thoracic aortic wall area (AWA), plaque prevalence, and plaque volumes measured with MRI were independently associated with incident cardiovascular disease, with AWA associated in particular with stroke, and plaque associated with coronary heart disease. Clinical trial registration no. NCT00041418 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Peshock in this issue.

Graphical abstract

Figure 1:

Sample selection flowchart. CVD = cardiovascular disease, Exam = examination.

Figure 2:

Measurements of aortic arterio- and atherosclerosis at aortic MRI for cardiovascular disease (CVD) risk prediction. (A) Three-dimensional rendered MRI aortic angiogram depicts the aortic anatomy. The aortic wall area (AWA) was measured at the pulmonary artery bifurcation level (upper line), and the diaphragm (lower line) was used to separate the thoracic and abdominal aorta. (B–D) Axial fat-suppressed, black-blood, two-dimensional, T2-weighted MRI images of the thoracic aorta visualize aortic measurements. Aortic plaque was defined as characteristic luminal protrusions of 1 mm or greater in radial thickness (dotted line in B). Wall thickness (WT) (lines in C) and lumen area (circle in D) were used to calculate AWA. (E) Diagram shows AWA equation and cross-sectional depiction of lumen diameter (D) (dotted line) and cross-sectional vessel area (VA) (orange circle).

Figure 3:

Risk of elevated aortic wall measures and plaque for incident cardiovascular disease (CVD). Forest plot shows the hazard ratios (HRs) of CVD events for each unit increase in aortic wall and plaque measures after age/sex or multivariable adjustments (age, sex, hypertension, hypercholesterolemia, diabetes mellitus, and smoking status). Continuous variables were calculated per 1 SD unit. CVD occurred in 223 of 1513 Framingham Heart Study Offspring Cohort participants free of disease at baseline (incidence rate, 12.4 [95% CI: 10.8, 14.1]). Ln = natural-logarithmic–transformed.

Figure 4:

Kaplan-Meier plots for event-free survival. Kaplan-Meier curves show survival free of cardiovascular disease for participants with and without (A) aortic wall area (AWA) greater than the 80th percentile, (B) prevalence of thoracic plaque, and (C) two (AWA >80th percentile and thoracic plaque), one (AWA >80th percentile or thoracic plaque), or no composite variables. P values reflect log-rank tests. perc. = percentile.

Figure 5:

Reproducibility of aortic wall area (AWA) measurements. (A) Line graphs show intra- and interobserver correlation. Solid blue lines indicate the linear regression line, and dotted red lines show the 95% CI. (B) Bland-Altman plots show AWA measurements. Solid blue lines indicate the bias, and dotted red lines represent the limits of agreement (1.96 × SD).