Association of coronary computed tomography angiography-derived plaque characteristics and brain white matter hyperintensity: an observational study

Abstract

Background

White matter hyperintensity (WMH) is a common finding in brain magnetic resonance imaging (MRI), but its relationship with coronary plaque characteristics remains unclear.

Purpose

To evaluate the association between coronary computed tomography angiography (CCTA)-derived plaque characteristics and WMH in brain MRI.

Material and Methods

CCTA and brain MRI of 392 consecutive patients were retrospectively collected. Degree of total WMH was evaluated based on Fazekas scale (0–6) and classified into mild (0–2) and moderate-to-severe (3–6) groups. Besides demographic and clinical data, morphological and quantitative parameters (e.g. aortic ulcer, Agatston Score, diameter stenosis, plaque volume, plaque length) were also evaluated based on CCTA. Chi-square (or Fisher's exact) test and Student's t-test (or Wilcoxon's signed-rank test) were used for comparing variables between two groups when appropriate. Multivariate logistic regression analyses were applied to identify the independent variables associated with moderate-to-severe WMH.

Results

Patients with moderate-to-severe WMH showed older age, larger plaque burden, plaque length, volume of noncalcified and calcified plaque, higher total cholesterol, triglyceride, low-density lipoprotein, Agatston Score, and higher proportion of male, hypertension, diabetes, aortic ulcer, and obstructive coronary artery disease (CAD) (all P < 0.05). Age (odds ratio [OR]=1.061, 95% confidence interval [CI]=1.013–1.111; P = 0.012), plaque burden (OR=15.259, 95% CI=2.466–94.412; P = 0.003), obstructive CAD (OR=8.020, 95% CI=3.628–17.727; P <0.001), and Agatston Score (OR=1.004, 95% CI=1.002–1.006; P <0.001) were found to be independently associated with moderate-to-severe WMH.

Conclusion

Older age, larger plaque burden, higher Agatston score, and obstructive CAD were prone to moderate-to-severe WMH.

Keywords

Coronary artery disease coronary computed tomography angiography white matter hyperintensity coronary atherosclerosis coronary plaque burden cerebral small vessel disease

Introduction

White matter hyperintensity (WMH) appears as lesions of increased signal intensity on T2-weighted image (T2WI) or fluid-attenuated inversion recovery (FLAIR) images of the brain (1). It has been indicated to be associated with cognitive impairment, Alzheimer's disease, stroke, and dementia (2 –4). Although the exact mechanism of WMH remains unclear, previous studies have indicated that some atherosclerosis related risk factors (e.g. hypertension, diabetes, high cholesterol, smoking, and aging) might be associated with the occurrence of WMH (5 –9). Meanwhile, some other studies have reported that the use of statins might be effective in limiting the progression of WMH (10,11). Therefore, atherosclerosis might be an important potential mechanism of WMH.

The coronary artery is another crucial target organ of atherosclerosis. Atherosclerosis can lead to the stenosis or obstruction of the coronary artery and subsequent myocardial ischemia, which has been proven to be a leading cause of mortality and disability (12). Coronary computed tomography angiography (CCTA) is a non-invasive imaging technique that is widely used for evaluating and following up patients with coronary artery disease (CAD) (13). Because of the concept of “pan-vascular medicine” and the potentially common mechanism of atherosclerosis (14), several previous studies have explored the association of CCTA-derived imaging features and the characteristics of brain WHM.

The association between CAD and WMH has garnered increasing attention in medical research. Emerging evidence has demonstrated a significant positive correlation between coronary artery calcium (CAC) score and WMH burden in the eastern population (15,16), with CAC quantification serving as a prognostic indicator for WMH progression (17). Furthermore, a western study reported the association between WMH and coronary plaque volume (calcified, noncalcified, low attenuated) (18). However, previous studies have predominantly emphasized the relationship between CAC score and WMH burden, with insufficient exploration of non-calcified lesions, plaque burden, and other characteristics of coronary plaques. Therefore, if we could add to the exploration of the relationship between the more comprehensive coronary plaque characteristics and WMH, it could provide deeper insights into their shared pathophysiological mechanisms. This enhanced understanding would hold promise for improving future co-management strategies for patients with cardiovascular and cerebrovascular disease.

Therefore, the aim of the present study was to evaluate the association between the overall CCTA-derived plaque characteristics and WMH in the brain MRI based on a study general population from an eastern country.

Material and Methods

Study population

Between January 2021 and December 2021, 1895 participants who underwent both brain MRI and CCTA in our hospital were enrolled. These patients typically underwent these two examinations prompted by symptoms suggestive of cardiovascular or cerebrovascular disease, including dizziness, headache, chest tightness, and chest pain. The research protocol was approved by the Ethics Committee of our hospital (reference no. 2023-SR-885).

Clinical information

Based on the electronic medical record system, we retrospectively collected the following clinical information: sex, age, body mass index (BMI), smoking history (defined as an average smoking of more than one cigarette per day and lasting for more than 6 months) (19), drinking history (defined as an average alcohol consumption of ≥100 g/week, with a duration of more than 1 year) (20), history of medication (anti-hypertensive medication, anti-diabetic medication, aspirin, and statins) use, concomitant hypertension (previous use of anti-hypertensive medication, or systolic blood pressure >140 mmHg or diastolic blood pressure >90 mmHg) (21), concomitant diabetes (previous use of anti-diabetic medication or fasting blood glucose >126 mg/dL, or HbA1c of ≥6.5%) (22). The level of total cholesterol, triglyceride, low-density lipoprotein (LDL), and creatinine were also recorded.

Imaging scan

Brain MRI was performed using a 3.0-T unit (Skyra; Siemens Healthineers, Erlangen, Germany) equipped with a 20-channel head coil. MRI protocols of the brain included axial and sagittal T1-weighted image (T1WI), axial T2WI, axial FLAIR, and axial diffusion weighted imaging (DWI). Detailed imaging parameters for FLAIR and DWI were as follows: (i) FLAIR: repetition time (TR) = 8000 ms, echo time (TE) = 97 ms, slice thickness = 5 mm, inversion time = 2300 ms, matrix = 256 × 256, field of view (FOV) = 230 × 230 mm²; and (ii) DWI: TR = 4800 ms, TE = 100 ms, slice thickness = 5 mm, FOV = 230 × 230 mm², b values = 0 and 1000 s/mm².

CCTA was performed using a first-generation or third-generation dual-source CT (DSCT) unit (Definition and Force; Siemens Medical Systems, Erlangen, Germany). Images were acquired with the scanning range from the tracheal carina to 1 cm below the diaphragm. An individualized injection of 50–80 mL of iopromide (Ultravist 370 mg iodine/mL; Bayer Schering Pharma, Berlin, Germany) was used at 4.0–6.0 mL/s according to the patient’s weight, heart rate, and cardiac function. The typical scan parameters were as follows: 32 × 0.6 mm (Definition) or 96 × 0.6 mm (Force) detector collimation, pitch adapted to the heart rate (0.2–0.5), 330 ms (Definition) or 250 ms (Force) gantry rotation time, and tube voltage of 80–120 kVp. Images were reconstructed at the best diastolic and systolic phases with a section thickness of 0.75 mm and increment of 0.5 mm.

Imaging analysis

Degree of WMH was evaluated in the periventricular (PVWMH) and deep location (DWMH), based on the Fazekas scale (23). Degree of DWMH was rated as follows: 0 = absent, 1 = punctate foci, 2 = beginning confluence, and 3 = large confluent areas. Degree of PVWMH was graded as follows: 0 = absent, 1 = “caps” or pencil-thin lining, 2 = smooth “halo” and 3 = irregular periventricular signal extending into the deep white matter. The score of total WMH was obtained by adding the score of DWMH and that of PVWMH. Degree of WMH was evaluated by two radiologists (with 11 and 3 years of experiences in neuroradiology, respectively) who were blinded to clinical information, CCTA images, and study design. If disagreements occurred, consensus was achieved by discussion with another senior radiologist (with 27 years of experience in neuroradiology). According to the total WMH score, our study cohort was classified into mild (0–2) and moderate-to-severe WMH group (3–6) (24). In addition, according to the DWMH and PVWMH scores, the study groups were divided into mild (0–1) and moderate-to-severe groups (2–3), respectively.

All CCTA datasets were transferred to a post-processing workstation (syngo.via; Siemens Healthineers) for subsequent analysis. Quantitative plaque analyses were performed by two radiologists (with 3 and 5 years of experience in cardiovascular imaging, respectively) who were blinded to clinical information and brain MRI scans using semi-automated plaque analysis software (Coronary Plaque Analysis v5.0.2, syngo.via Frontier; Siemens Healthineers) with manual correction. Axial, cross-sectional, and auto-generated curved planar reformation (CPR) images were used for the evaluation. Vascular volume corresponding to the length of the plaque were summed to calculate plaque burden. The readers manually defined the proximal and distal ends of the plaque, as well as the site of maximal stenosis of the coronary artery, on the CPR images. The software then semi-automatically delineated the inner and outer contours of the plaque and vessel lumen. The radiologists also assessed the presence of an aortic ulcer—defined as an outpouching of contrast material into or adjacent to the plaque measuring at least 2 mm in the aortic arch—within the scanning range (25) and recorded the Agatston Score.

The conventional plaque parameters included diameter stenosis, lesion length, calcified plaque volume, non-calcified plaque volume, and plaque burden (the ratio of plaque volume to vascular volume). Each coronary segment was assessed for maximal diameter stenosis according to the 18-segment model of the coronary tree, and the obstructive CAD was defined as diameter stenosis ≥50% in one or more coronary arteries (26,27). Plaque volumes across all coronary segments were incorporated, with plaque burden computed per site. The results exhibited total plaque volume, plus length and burden of peak stenosis plaque.

Statistical analyses

Normality of distribution was evaluated using the Shapiro–Wilk test. If normally distributed, continuous variables were expressed as the mean ± standard deviation (SD). Otherwise, they were reported as median (interquartile range [IQR]). Categorical variables were expressed as percentages and frequencies. The ANOVA test was used to compare the differences in clinical characteristics and CCTA-derived parameters in each group (total WMH = 0–6 group; DWMH and PVWMH = 0–3 group). Post-hoc adjustments (Bonferroni test) were applied to identify where differences, if they existed, lay between WMH groups. Correlations between the clinical characteristics, CCTA-derived parameters and the score of WMH were assessed using Pearson correlation analyses for normally distributed data; Spearman correlation analyses were used for non-normally distributed data. After dichotomizing the study cohort into mild and moderate-to-severe WMH groups, comparisons of categorical variables between two groups were performed using chi-square or Fisher's exact tests as appropriate. Student's t-test or Wilcoxon's signed-rank test was used for the comparisons of continuous variables between two groups, as appropriate. Variables with a P value <0.10 in the foregoing analyses were entered into the multivariate logistic regression analysis to identify the independent variables associated with moderate-to-severe WMH. Odds ratio (OR) and 95% confidence interval (CI) were calculated. Logistic stepwise regressions were used to integrate the independent variables for establishing three models: model 1 (clinical characteristics), model 2 (CCTA-derived parameters), and model 3 (combination of clinical characteristics and CCTA-derived parameters). Receiver-operating characteristic (ROC) curves analyses and Delong tests were used to evaluate and compare the performances of each model in predicting the occurrence of moderate-to-severe WMH. Similar statistical analyses were also performed in the subgroup analysis of PVWMH and DWMH. Statistical analysis was performed using MedCalc version 20.1 (MedCalc Software Ltd., Ostend, Belgium) and SPSS version 26.0 (IBM Corp., Armonk, NY, USA) software. A two-sided P value <0.05 was considered to be statistically significant.

Results

Patient characteristics

A total of 392 participants were included in our study. A detailed flow diagram is shown in Fig. 1. We excluded 842 patients according to the following exclusion criteria: (i) missing clinical data (n = 675); (ii) history of atrial fibrillation, valve disease, hypertrophic cardiomyopathy, or dilated cardiomyopathy (n = 52); (iii) history of coronary artery bypass grafting (CABG) or stent implantation (n = 47); (iv) poor image quality (n = 41); and (v) concomitant brain tumor (n = 27).

Fig. 1.

Patient selection flow chart. CABG, coronary artery bypass grafting; CCTA, coronary computed tomography angiography; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy.

The characteristics of the enrolled 392 individuals (male = 53.8%; mean age = 65.6 ± 10.4 years) are summarized in Table 1. Hypertension, diabetes, smoker, drinker, Aspirin user, and statins user were found in 229 (58.4%), 115 (29.3%), 93 (23.7%), 62 (15.8%), 37 (9.4%), and 53 (13.5%) patients, respectively. The median triglyceride and creatinine were 1.27 mmol/L (IQR = 0.91–1.82 mmol/L) and 66.35 µmol/L (IQR = 57.73–76.18 µmol/L), respectively. Mean total cholesterol and LDL cholesterol were 4.32 ± 1.00 mmol/L and 2.70 ± 0.74 mmol/L, respectively.

Table 1.

Participant characteristics (n = 392).

Characteristics	Overall (n = 392)
Demographic variables
Male	211 (53.8)
Age (years)	65.6 ± 10.4
BMI (kg/m²)	24.26 ± 3.28
Clinical variables
Hypertension	229 (58.4)
Diabetes	115 (29.3)
Smoking history	93 (23.7)
Drinking history	62 (15.8)
Aspirin users	37 (9.4)
Statins users	53 (13.5)
Laboratory variables
Total cholesterol (mmol/L)	4.32 ± 1.00
Triglyceride (mmol/L)	1.27 (0.91–1.82)
LDL (mmol/L)	2.70 ± 0.74
Creatinine (µmol/L)	66.35 (57.73–76.18)
Coronary artery parameters
Aortic ulcer	18 (4.6)
Agatston Score	1.5 (0.00–118.75)
Obstructive CAD	98 (25.0)
Plaque length (mm)	4.90 (0–13.95)
Plaque burden	0.37 ± 0.33
Non-calcified plaque volume (mm³)	20.15 (0.00–81.54)
Calcified plaque volume (mm³)	7.07 (0.00–52.59)

Values are given as n (%), mean ± SD, or median (IQR).

BMI, body mass index; CAD, coronary artery disease; IQR interquartile range; LDL, low-density lipoprotein.

This study included 18 patients with aortic ulcers and 98 patients with obstructive CAD. The number of cases with the maximal stenosis plaque per coronary segment distributed as follows: RCA1 (n = 11), RCA2 (n = 7), RCA3 (n = 5), LM (n = 2), LAD1 (n = 40), LAD2 (n = 15), LAD3 (n = 3), DA1 (n = 3), DA2 (n = 1), LCX1 (n = 5), LCX2 (n = 4), OM1 (n = 1), and RI (n = 1). Median volumes of non-calcified and calcified plaque were 20.15 mm³ (IQR = 0.00–81.54 mm³) and 7.07 mm³ (IQR = 0.00–52.59 mm³), respectively. Mean plaque burden was 0.37 ± 0.33.

Variables associated with total WMH

As the degree of WMH increased from 0 to 6, significant differences were observed among the groups in sex, age, hypertension, diabetes, smoking history, and total cholesterol levels (all P <0.05). After dichotomizing the cohort into mild (0–2) and moderate-to-severe WMH (3–6) groups, the moderate-to-severe WMH group showed larger volumes of non-calcified and calcified plaque, greater plaque burden, longer plaque length, higher Agatston Score, and higher proportion of obstructive CAD (all P <0.05). Representative cases of a patient with severe WMH and that with mild WMH are shown in Fig. 2.

Fig. 2.

White matter hyperintensity in two patients with varying degrees of CAD. The condition of CAD and FLAIR of two patients with varying degrees of CAD. (A–C) Patient 1 had mild CAD, with a Fazekas scale score of 0 (DWMH = 0, PVWMH = 0); (D–F) patient 2 had severe CAD, with a Fazekas scale score of 6 (DWMH = 3, PVWMH = 3). CAD, coronary artery disease; FLAIR, fluid-attenuated inversion recovery.

Correlation analyses

The degree of WMH showed significant correlations with age (r = 0.380; P <0.001), sex (r = 0.177; P <0.001), hypertension (r = 0.318; P <0.001), and diabetes (r = 0.215; P <0.001) (Fig. 3). All CCTA-derived parameters showed significant correlations with the degree of total WMH (all P <0.05).

Fig. 3.

Correlation analysis for different parameters and WMH. Blue = positive correlation; red = negative correlation; the darker the color, the greater the correlation coefficient between this parameter and WMH (deep and periventricular WMH). *P <0.05, **P <0.01, ***P <0.001. WMH, white matter hyperintensity.

Univariable and multivariable analyses

Significant differences were found in age, sex, hypertension, and diabetes between patients with mild WMH and those with moderate-to-severe WMH (all P <0.05) (Table 2). Compared with the patients with mild WMH, those with moderate-to-severe WMH showed larger median volume of non-calcified plaque (155.86 mm³ [IQR = 77.23–242.83] vs. 0.00 mm³ [IQR = 0.00–36.67]; P <0.001), larger median volume of calcified plaque (127.72 mm³ [IQR = 52.40–274.37] vs. 0.00 mm³ [IQR = 0.00–17.11]; P <0.001), higher mean plaque burden (0.68 ± 0.15 vs. 0.28 ± 0.31; P <0.001), longer median plaque length (25.20 mm [IQR = 13.60–40.10] vs. 0.00 mm [IQR = 0.00–8.00]; P <0.001), higher median Agatston Score (369.50 [IQR = 132.50–775.40] vs. 0.00 [IQR = 0.00–25.10]; P <0.001), and higher proportion of patients with aortic ulcers (12.40% vs. 2.30%; P <0.001) and CAD (80.90% vs. 8.60%; P <0.001).

Table 2.

Distribution of baseline characteristics in relation to severity of WMH.

	Univariate			Multivariate
	Mild WMH (n = 303)	Moderate-to-severe WMH (n = 89)	P value	OR (95%CI)	P value
Demographic variables
Male	152 (50.2)	59 (66.3)	0 . 007
Age (years)	64.1 ± 10.5	70.6 ± 8.2	<0.001	1.061 (1.013–1.111)	0 . 012
BMI (kg/m²)	24.22 ± 3.26	24.39 ± 3.38	0.672
Clinical variables
Hypertension	159 (52.5)	70 (78.7)	<0.001
Diabetes	71 (23.4)	44 (49.4)	<0.001
Smoking history	65 (21.5)	28 (31.5)	0.051
Drinking history	45 (14.9)	17 (19.1)	0.334
Aspirin users	25 (8.3)	12 (13.5)	0.138
Statins users	39 (12.9)	14 (15.7)	0.488
Laboratory variables
Total cholesterol (mmol/L)	4.41 ± 0.95	4.02 ± 1.10	0 . 004
Triglyceride (mmol/L)	1.30 (0.95–1.87)	1.14 (0.85–1.56)	0 . 015
LDL (mmol/L)	2.75 ± 0.68	2.53 ± 0.88	0 . 014
Creatinine (µmol/L)	66.60 (57.90–75.50)	64.90 (57.15–79.05)	0.891
Coronary artery parameters
Aortic ulcer	7 (2.3)	11 (12.4)	0 . 001
Agatston Score	0.00 (0.00–25.10)	369.50 (132.50–775.40)	<0.001	1.004 (1.002–1.006)	<0.001
Obstructive CAD	26 (8.6)	72 (80.9)	<0.001	8.020 (3.628–17.727)	<0.001
Plaque length (mm)	0.00 (0.00–8.00)	25.20 (13.60–40.10)	<0.001
Plaque burden	0.28 ± 0.31	0.68 ± 0.15	<0.001	15.259 (2.466–94.412)	0 . 003
Non-calcified plaque volume (mm³)	0.00 (0.00–36.67)	155.86 (77.23–242.83)	<0.001
Calcified plaque volume (mm³)	0.00 (0–17.11)	127.72 (52.40–274.37)	<0.001

Values are given as n (%), mean ± SD, or median (IQR).

Adjusting covariates for risk factor adjusted model: male (n, %), hypertension (n, %), diabetes (n, %), total cholesterol (mmol/L), triglyceride (mmol/L) and LDL (mmol/L).

BMI, body mass index; CAD, coronary artery disease; IQR, interquartile range; LDL, low-density lipoprotein; WMH, white matter hyperintensity.

In the multivariable analyses, sex, age, hypertension, and diabetes were found to be independent clinical variables associated with moderate-to-severe WMH and adopted into model 1 (0.916 × male + 0.083 × age + 0.859 × hypertension + 1.086 × diabetes + −8.354). Meanwhile, plaque burden, obstructive CAD, and Agatston Score were found to be independent CCTA-derived parameters associated with moderate-to-severe WMH and used to establish model 2 (2.247 × obstructive CAD + 2.718 × plaque burden + 0.004 × Agatston Score + −4.172). In addition, after integrating clinical and imaging parameters, we found that age, plaque burden, obstructive CAD, and Agatston Score were independently associated with moderate-to-severe WMH and adopted into model 3 (0.059 × age + 2.082 × obstructive CAD + 2.725 × plaque burden + 0.004 × Agatston Score + −8.764).

The ROC curves of three models in predicting the moderate-to-severe WMH are showed in Fig. 4. Model 3 showed the optimal performance (AUC = 0.953), followed by model 2 (AUC = 0.948). Both outperformed model 1 (AUC = 0.785; model 3 vs. model 1; P <0.001; model 2 vs. model 1; P <0.001) (Table 3).

Fig. 4.

ROC curves of three models. (a) Model 1: composed of clinical parameters. The AUC was 0.785 (95% CI = 0.741–0.825; blue line). (b) Model 2: composed of CCTA-derived parameters. The AUC was 0.948 (95% CI = 0.921–0.968; green line). (c) Model 3: composed of clinical and CCTA-derived parameters. The AUC was 0.953 (95% CI = 0.927–0.972; orange line). AUC, area under the ROC curve; CCTA, coronary computed tomography angiography; CI, confidence interval; ROC, receiver operating characteristic.

Table 3.

Comparison and evaluation of the performance of different models.

		AUC	95% CI	SEN (%)	SPE (%)	ACC (%)	Vs model 1	Vs model 2	Vs model 3
WMH*	Model 1	0.785	0.741–0.825	23.60	95.05	78.83		<0.001	<0.001
	Model 2	0.948	0.921–0.968	74.16	93.73	89.29	<0.001		0.183
	Model 3	0.953	0.927–0.972	76.40	95.38	91.07	<0.001	0.183
DWMH^†	Model 1	0.766	0.721–0.807	31.86	89.96	73.21		0.053	<0.001
	Model 2	0.822	0.781–0.859	58.41	89.24	80.36	0.053		0 . 003
	Model 3	0.854	0.816–0.888	60.18	88.89	80.61	<0.001	0 . 003
PVWMH^‡	Model 1	0.808	0.765–0.845	19.12	97.22	83.67		<0.001	<0.001
	Model 2	0.935	0.906–0.957	61.76	95.37	89.54	<0.001		0.062
	Model 3	0.945	0.918–0.965	67.65	94.75	90.05	<0.001	0.062

*WMH model 1: male (OR = 2.499), age (OR = 1.087), hypertension (OR = 2.360), diabetes (OR = 2.961); model 2: obstructive CAD (OR = 9.462), plaque burden (OR = 15.156), Agatston Score (OR = 1.004); model 3: age (OR = 1.061), obstructive CAD (OR = 8.020), plaque burden (OR = 15.259), Agatston Score (OR = 1.004).

†

DWMH model 1: age (OR = 1.083), hypertension (OR = 2.538), diabetes (OR = 1.816); model 2: obstructive CAD (OR = 3.450), plaque burden (OR = 9.103); model 3: age, (OR = 1.062) hypertension (OR = 2.134), obstructive CAD (OR = 3.734), plaque burden (OR = 12.171).

‡

PVWMH model 1: age (OR = 1.060), triglyceride (OR = 0.556), total cholesterol (OR = 0.482), hypertension (OR = 2.294), diabetes (OR = 3.242); model 2: obstructive CAD (OR = 7.211), plaque burden (OR = 10.103); model 3: triglyceride (OR = 0.485), obstructive CAD (OR = 6.583), plaque burden (OR = 10.384).

ACC, accuracy; AUC, area under the receiver operating characteristic curve; CI, confidence interval; DWMH, deep location white matter hyperintensity; PVWMH, periventricular white matter hyperintensity; SEN, sensitivity; SPE, specificity; WMH, white matter hyperintensity.

Analysis of WMH by location

A comparison of clinical and CCTA-derived features among each group of DWMH and PVWMH is presented in Supplementary Tables 1 and 2. Different from the results about the total WMH, significant differences were found in total cholesterol, triglyceride, and LDL among each DWMH and PVWMH group (P <0.050). Supplementary Figs. 1 and 2 show results of pairwise comparison between groups (DWMH and PVWMH = 0–3). The results of the correlation analysis were similar to those of WMH (Fig. 3).

Supplementary Tables 1 and 2 show the results of the univariable analysis. Notably, the percentage of smoking history was higher in the moderate-to-severe DWMH group (20.80% vs. 31.00%; P = 0.032) and the percentage of Aspirin users was higher in the moderate-to-severe PVWMH group (8.00% vs. 16.20%; P = 0.037).

Different from the results of total WMH, the independent variables relating to DWMH and PVWMH do not include the Agatston Score. The independent variables associated with moderate-to-severe DWMH were age, hypertension, plaque burden, and obstructive CAD, while those associated with moderate-to-severe PVWMH were triglyceride, obstructive CAD, and plaque burden. The ROC curve is shown in Supplementary Figs. 3 and 4. Model 3 showed an AUC of 0.854 in predicting moderate-to-severe DWMH, which was higher than that of model 1 (AUC = 0.766; model 3 vs. model 1; P <0.001) and that of model 2 (AUC = 0.822; model 3 vs. model 2; P = 0.003). In PVWMH, both model 3 and model 2 showed a higher AUC than model 1 (model 3 vs. model 1, 0.945 vs. 0.808; P <0001; model 2 vs. model 1, 0.935 vs 0.808; P <0.001) in predicting moderate-to-severe PVWMH.

Discussion

There are three main findings in our study. First, we found that all CCTA-derived metrics showed significant correlations with the degree of total WMH. Patients with more severe CAD usually showed higher degrees of WMH. Second, among the CCTA-derived metrics, higher plaque burden, obstructive CAD, and higher Agatston Score were found to be independently associated with higher degrees of WMH. Third, after integrating with the clinical variables, CCTA-derived metrics showed a satisfactory performance in predicting the co-existence of higher degrees of WMH.

Our study found a close relationship between hypertension and WMH. A previous study showed that Fazekas scale scores for WMHs increased with increased blood pressure values, which supported our findings (6). Hypertension contributed to the remodel of arterial walls and disrupted the blood–brain barrier, which subsequently lead to chronic ischemia and hypoperfusion (28,29). Austin et al. reported that diabetes was associated with WMH, which was also in line with our findings (5). Large glucose variability could induce oxidative stress and correlated negatively with endothelial function (30,31); therefore, it was linked with WMH. In our study, WMH was also proved to be interdependently associated with increased age. Garnier-Crussard et al. demonstrated that WMH increased with age (32), which agreed with our research results. In addition, Skampardoni et al. found that the structural and functional degradation of blood vessels in the elderly population, combined with brain atrophy and WMH, leads to cognitive decline and accelerates the brain aging process (33). Based on the above researches, the degree of white matter lesions may be higher in the elderly population and those with hypertension and diabetes. We recommend strengthening follow-up monitoring and implementing early intervention measures for these patients, which may be of great significance in delaying the progression of WMH. In our study, we anomalously found that the mild WMH group showed greater TC, TG, and LDL compared to the moderate-to-severe WMH group. We deemed that this result might be associated with medication history. Patients with a higher burden of atherosclerosis or more severe WMH had a relatively higher proportion of statin or other lipid-lowering therapy history.

A previous study showed that coronary plaque volume was positively correlated with WMH; this association was particularly prominent in the elderly population (18). In addition, the volume of WMH showed an increasing trend with higher CAC score (15). Recent studies reported that ischemic heart disease was associated with brain aging and cognitive decline (34,35), which aligned with our results. Our results further supported the view that some common pathophysiological features, such as atherosclerosis, could exist in both cardiac and cerebral vascular diseases. In addition, our study first found that the proportion of aortic ulcer was higher in patients with moderate-to-severe WMH, which was seldom reported in previous studies. Aortic ulcers typically occur in the late stage of atherosclerosis, caused by plaque rupture due to progressive vascular inflammation, indicating severe damage to the vascular endothelium (36). Such patients may be older, with cardiovascular risk factors and severe atherosclerotic lesions (37), and similar manifestations could be also common in individuals with a heavy burden of white matter lesions.

Another strength of our study was that we evaluated the relationship between coronary plaque burden and WMH. The results supported that among the factors studied, plaque burden was a stronger independent risk factor for WMH than CAC score. Compared with the plaque volume, plaque burden assessed individual lesion risk by measuring the ratio of plaque volume to vessel volume at the site of the lesion, reflecting the impact of atherosclerosis on vascular remodeling more accurately (38). Positive remodeling served as one of the high-risk indicators of coronary plaque (39). As calcified or non-calcified coronary plaque progressed, it may exacerbate vascular stenosis, cause lumen obstruction, and subsequently induced myocardial ischemia. In the event that plaque fell off and obstructed small vessel branches, it could directly trigger myocardial infarction (39). In addition, a recent study showed that increase in plaque burden was associated with an increase in peri-coronary adipose tissue attenuation, supporting the hypothesis that inflammation could promote atherogenesis (40). Some studies demonstrated that severe plaque burden is an effective index to predict the occurrence of adverse cardiovascular events (41,42). Based on this, patients with a significant plaque burden were more prone to developing coronary inflammation and myocardial ischemia, while inflammation and small vessel pathology constituted key pathological mechanisms of WMH. Compared with previous studies conducted on western populations (18), our research revealed an independent correlation between plaque burden and WMH. On the basis of past research, this article specifically reported the association between plaque burden and WMH, rather than CAC score (15,16), providing a more precise assessment of coronary artery lesions. In summary, we supposed the degree of plaque burden might be helpful to identify those patients at increased risk of future WMH progression.

To date, numerous studies have reported the association between coronary artery calcification and cerebral small vessel diseases or WMH. Our study added to previous work by incorporating more precise quantitative parameters and exploring the independent risk factors most closely related to WMH in the eastern population. Based on our results, it could be inferred that for patients with severe CAD, clinical priority should be given to MRI of the brain, which not only helps to identify this subclinical symptom early, but also provides an important reference for personalized prevention and intervention. This study provided valuable enlightenment for the follow-up research direction and clinical practice; it was also expected to further optimize the management strategy of joint prevention and treatment of the two diseases in the future.

The present study has some limitations. First, the detailed volume of WMH was not calculated and compared in this study. Second, the microvascular function of cardiac and cerebral vascular beds was not analyzed and correlated in this study. Future studies with more detailed volumetric and micro-vessel–related functional parameters are needed to confirm the association between CAD and WMH.

In conclusion, our study indicated that there was potential association between CCTA-derived plaque characteristics and WMH in brain MRI. Integrating the clinical and CCTA-derived plaque characteristics could effectively predict the co-existence of moderate-to-severe WMH. Our study could serve as a potential cornerstone for promoting the screen of WMH in the patients with CAD, and co-management of CAD and WMH.

Supplemental Material

sj-docx-1-acr-10.1177_02841851261421597 - Supplemental material for Association of coronary computed tomography angiography-derived plaque characteristics and brain white matter hyperintensity: an observational study

Supplemental material, sj-docx-1-acr-10.1177_02841851261421597 for Association of coronary computed tomography angiography-derived plaque characteristics and brain white matter hyperintensity: an observational study by Jie Qin, Dandan Wu, Yue Chu, Yunfei Wang, Wen Qian, Xiaoquan Xu and Yi Xu in Acta Radiologica

Supplemental Material

sj-docx-2-acr-10.1177_02841851261421597 - Supplemental material for Association of coronary computed tomography angiography-derived plaque characteristics and brain white matter hyperintensity: an observational study

Supplemental material, sj-docx-2-acr-10.1177_02841851261421597 for Association of coronary computed tomography angiography-derived plaque characteristics and brain white matter hyperintensity: an observational study by Jie Qin, Dandan Wu, Yue Chu, Yunfei Wang, Wen Qian, Xiaoquan Xu and Yi Xu in Acta Radiologica

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received the following financial support for the research, authorship, and/or publication of this article: This work was supported by the grants from the Natural Scientific Foundation of China (Grant Nos. 82302163 for Yunfei Wang) and Young Scholars Fostering Fund with the First Affiliated Hospital of Nanjing Medical University (Grant Nos. PY2022036 for Yunfei Wang). This work was supported by Jiangsu Province Capability Improvement Project through Science, Technology and Education (Grant Nos. JSDW202243 for Xiaoquan Xu).

ORCID iDs

Xiaoquan Xu

Yi Xu

Supplementary material

Supplementary material for this article is available online.

References

Wardlaw

Smith

Biessels

, et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol 2013;12:822–838.

Weng

Liu

, et al. White matter hyperintensities: a possible link between sarcopenia and cognitive impairment in patients with mild to moderate Alzheimer's disease. Eur Geriatr Med 2023;14:1037–1047.

Vadinova

Sihvonen

Garden

, et al. Early subacute white matter hyperintensities and recovery of language after stroke. Neurorehabil Neural Repair 2023;37:218–227.

Oveisgharan

Poole

, et al. Association of white matter hyperintensities with pathology and progression of parkinsonism in aging. JAMA Neurol 2021;78:1494–1502.

Austin

Nasrallah

Erus

, et al. Association of brain volumes and white matter injury with race, ethnicity, and cardiovascular risk factors: the multi-ethnic study of atherosclerosis. J Am Heart Assoc 2022;11:e023159.

Zhao

, et al. Volume of white matter hyperintensities increases with blood pressure in patients with hypertension. J Int Med Res 2019;47:3681–3689.

de Havenon

Majersik

Tirschwell

, et al. Blood pressure, glycemic control, and white matter hyperintensity progression in type 2 diabetics. Neurology 2019;92:e1168–e1175.

Todate

Uwano

Yashiro

, et al. High prevalence of cerebral small vessel disease on 7T magnetic resonance imaging in familial hypercholesterolemia. J Atheroscler Thromb 2019;26:1045–1053.

Austin

Nasrallah

Erus

, et al. Association of brain volumes and white matter injury with race, ethnicity, and cardiovascular risk factors: the multi-ethnic study of atherosclerosis. J Am Heart Assoc 2022;11:e023159.

10.

Zhang

Cui

Zhao

, et al. Effects of sartans and low-dose statins on cerebral white matter hyperintensities and cognitive function in older patients with hypertension: a randomized, double-blind and placebo-controlled clinical trial. Hypertens Res 2019;42:717–729.

11.

Guo

Liu

, et al. Assessing the effectiveness of statin therapy for alleviating cerebral small vessel disease progression in people >/=75 years of age. BMC Geriatr 2020;20:92.

12.

Virani

Alonso

Benjamin

, et al. Heart disease and stroke statistics-2020 update: a report from the American Heart Association. Circulation 2020;141:e139–e596.

13.

Lee

Lin

Dey

, et al. Application of quantitative assessment of coronary atherosclerosis by coronary computed tomographic angiography. Korean J Radiol 2024;25:518–539.

14.

Zhou

Zhao

, et al. Panvascular medicine: an emerging discipline focusing on atherosclerotic diseases. Eur Heart J 2022;43:4528–4531.

15.

Jin

Qin

Zhao

, et al.

Is coronary artery calcium an independent risk factor for white matter hyperintensity?

BMC Neurol 2023;23:13.

16.

Choi

Kim

Kwon

, et al. Association of cerebral white matter hyperintensities with coronary artery calcium in a healthy population: a cross-sectional study. Sci Rep 2022;12:21562.

17.

Jin

Hou

Qin

, et al. Predicting progression of white matter hyperintensity using coronary artery calcium score based on coronary CT angiography-feasibility and accuracy. Front Aging Neurosci 2023;15:1256228.

18.

Johansen

Gottesman

Kral

, et al. Association of coronary artery atherosclerosis with brain white matter hyperintensity. Stroke 2021;52:2594–2600.

19.

Yang

Chen

, et al. Smoking in China: findings of the 1996 National Prevalence Survey. JAMA 1999;282:1247–1253.

20.

Wood

Kaptoge

Butterworth

, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet 2018;391:1513–1523.

21.

Whelton

Carey

Aronow

, et al. 2017ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: executive summary: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. Hypertension 2018;71:1269–1324.

22.

Carson

Reynolds

Fonseca

, et al. Comparison of A1C and fasting glucose criteria to diagnose diabetes among U.S. adults. Diabetes Care 2010;33:95–97.

23.

Fazekas

Niederkorn

Schmidt

, et al. White matter signal abnormalities in normal individuals: correlation with carotid ultrasonography, cerebral blood flow measurements, and cerebrovascular risk factors. Stroke 1988;19:1285–1288.

24.

Zeng

Chen

Zhu

, et al. Severity of white matter hyperintensities: lesion patterns, cognition, and microstructural changes. J Cereb Blood Flow Metab 2020;40:2454–2463.

25.

Wang

Zheng

Bao

, et al. Predicting asymptomatic coronary artery disease in first-ever acute ischemic stroke patients: a cross-sectional study. Quant Imaging Med Surg 2024;14:2441–2454.

26.

Cury

Leipsic

Abbara

, , et al. CAD-RADS 2.0 - 2022 coronary artery disease-reporting and data system: an expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Cardiology (ACC), the American college of radiology (ACR), and the North America society of cardiovascular imaging (NASCI). J Cardiovasc Comput Tomogr 2022;16:536–557.

27.

Leipsic

Abbara

Achenbach

, et al. SCCT guidelines for the interpretation and reporting of coronary CT angiography: a report of the Society of Cardiovascular Computed Tomography guidelines committee. J Cardiovasc Comput Tomogr 2014;8:342–358.

28.

Wartolowska

Webb

AJS

. Midlife blood pressure is associated with the severity of white matter hyperintensities: analysis of the UK Biobank cohort study. Eur Heart J 2021;42:750–757.

29.

Brown

IAM

Diederich

Good

, et al. Vascular smooth muscle remodeling in conductive and resistance arteries in hypertension. Arterioscler Thromb Vasc Biol 2018;38:1969–1985.

30.

de Havenon

Majersik

Tirschwell

, et al. Blood pressure, glycemic control, and white matter hyperintensity progression in type 2 diabetics. Neurology 2019;92:e1168–e1175.

31.

Wang

Wei

, et al. Relationship between type 2 diabetes and white matter hyperintensity: a systematic review. Front Endocrinol (Lausanne) 2020;11:595962.

32.

Garnier-Crussard

Bougacha

Wirth

, et al. White matter hyperintensities across the adult lifespan: relation to age, Aβ load, and cognition. Alzheimers Res Ther 2020;12:127.

33.

Skampardoni

Nasrallah

Abdulkadir

, et al. Genetic and clinical correlates of AI-based brain aging patterns in cognitively unimpaired individuals. JAMA Psychiatry 2024;81:456–467.

34.

Rauseo

Salih

Raisi-Estabragh

, et al. Ischemic heart disease and vascular risk factors are associated with accelerated brain aging. JACC Cardiovasc Imaging 2023;16:905–915.

35.

Xie

Zheng

Yan

, et al. Cognitive decline before and after incident coronary events. J Am Coll Cardiol 2019;73:3041–3050.

36.

Sorber

Hicks

. Diagnosis and management of acute aortic syndromes: dissection, penetrating aortic ulcer, and intramural hematoma. Curr Cardiol Rep 2022;24:209–216.

37.

Vilacosta

Ferrera

San Román

. Acute aortic syndrome. Med Clin (Barc) 2024;162:22–28.

38.

Lee

Sung

Rizvi

, et al. Quantification of coronary atherosclerosis in the assessment of coronary artery disease. Circ Cardiovasc Imaging 2018;11:e007562.

39.

Tian

Chen

Xia

, et al. Pathways from insulin resistance to incident cardiovascular disease: a Bayesian network analysis. Cardiovasc Diabetol 2024;23:421.

40.

Tzolos

McElhinney

Williams

, et al. Repeatability of quantitative pericoronary adipose tissue attenuation and coronary plaque burden from coronary CT angiography. J Cardiovasc Comput Tomogr 2021;15:81–84.

41.

Dou

Shan

Wang

, et al. Integrating coronary plaque information from CCTA by ML predicts MACE in patients with suspected CAD. J Pers Med 2022;12:596.

42.

van Driest

Bijns

van der Geest

, et al. Utilizing (serial) coronary computed tomography angiography (CCTA) to predict plaque progression and major adverse cardiac events (MACE): results, merits and challenges. Eur Radiol 2022;32:3408–3422.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.00 MB

0.55 MB