Prediction of coronary plaque progression using biomechanical factors and vascular characteristics based on computed tomography angiography

Figures & data

Figure 1. An example of fluid-structural interaction analyzed at baseline. The three-dimensional lumenal surface of the right coronary artery is created using coronary CTA. Inlet velocity and outlet pressure are entered into the fluid-structure interaction model to calculate wall shear stress and von Mises stress values.

Table 1. Baseline demographics and clinical characteristics.

Characteristic	All patients (n = 12)
Age, y	57 (51–65)
Male, n (%)	8 (67)
Body mass index, kg/m^2	25.5 (24.8–28.3)
Previous unstable angina, n (%)	9 (75)
Previous myocardial infarction, n (%)	3 (25)
Previous PCI, n (%)	4 (33)
Hypertension, n (%)	11 (92)
Diabetes mellitus, n (%)	5 (42)
Triglycerides, mmol/L	1.38 (1.23–1.69)
Total cholesterol, mmol/L	4.24 (3.29–5.09)
High-density lipoprotein, mmol/L	1.07 (0.99–1.33)
Low-density lipoprotein, mmol/L	2.48 (1.83–3.54)

Continuous-variable data are presented as median (interquartile range).

Figure 2. Effect of baseline WSS and VMS on vascular outcomes in coronary artery segments at follow-up. (A) WSS and change in minimal lumen area. (B) VMS and change in minimal lumen area. (C) WSS and change in plaque burden. (D) VMS and change in plaque burden. (E) WSS and change in plaque volume. (F) VMS and change in plaque volume. The p values refer to the univariate analysis and are adjusted for the clustering of segments within patients and for multiple comparisons. Error bars represent SEM. MLA = minimal luminal area; PB = plaque burden; PV = plaque volume; SEM = standard error of the mean; VMS = von Mises stress; WSS = wall shear stress.

Table 2. Independent baseline predictors of the change in anatomic outcomes in 3-mm segments (n = 365).

Outcome	Baseline predictor	β (95% CI)	p
Change in minimal lumen area^A (mm^2)	Wall shear stress	0.15 (−0.12 to 0.43)	0.275*
	Von Mises stress	0.29 (0.07 to 0.51)	0.009
	Minimal lumen area	−0.31 (−0.43 to −0.19)	<0.001
	Plaque volume	0.08 (0.04 to 0.11)	<0.001
Change in plaque burden^B (%)	Wall shear stress	−1.45 (–2.86 to −0.04)	0.044
	Von Mises stress	−0.40 (−1.46 to 0.65)	0.451*
	Minimal lumen area	−1.75 (–3.46 to 0.04)	0.045
	Plaque burden	−0.63 (−0.89 to −0.38)	<0.001
Change in plaque volume^C (mm^3)	Wall shear stress	0.45 (−0.36 to 1.25)	0.281*
	Von Mises stress	0.80 (0.15 to 1.44)	0.016
	Plaque burden	−0.11 (−0.18 to −0.04)	0.001
	Plaque volume	−0.13 (−0.24 to −0.02)	0.019

Variables entered in the multivariate model: (A) Change in minimal lumen area: wall shear stress, von Mises stress, minimal lumen area, plaque volume; (B) Change in plaque burden: wall shear stress, von Mises stress, minimal lumen area, plaque burden; (C) Change in plaque volume: wall shear stress, von Mises stress, plaque burden, plaque volume.

* The primary variables of interest (WSS and VMS) are not removed, even if not significant, from the backward stepwise process of multivariate analysis. CI = confidence interval.

Figure 3. (A) Frequency of serial remodeling patterns in low-, moderate-, and high-WSS segments. (B) Frequency of serial remodeling patterns in low-, moderate-, and high-VMS segments. The p values refer to the ordered logistic regression analysis and are adjusted for the clustering of segments within patients. The primary variable of interest (wall shear stress) is not removed, even if not significant, from the backward stepwise process of logistic regression analysis. CI = confidence interval.

Table 3. Independent baseline predictors of arterial remodeling in 3-mm segments (n = 365).

Baseline independent predictor	β (95% CI)	p
Wall shear stress	0.36 (−0.54 to 1.26)	0.435
Von Mises stress	0.61 (0.13 to 1.09)	0.012
Plaque burden	0.05 (0.01 to 0.09)	0.026