White matter disease (WMD) of the brain is associated with incident stroke. Similarly, subclinical calcified coronary artery plaque has been associated with incident coronary artery disease (CAD) events. Although atherogenesis in both vascular beds may share some common mechanisms, the extent to which subclinical CAD is associated with WMD across age ranges in subjects with a family history of early-onset CAD remains unknown. We screened 405 apparently healthy participants in the Genetic Study of Atherosclerotic Risk for CAD risk factors and for the presence of noncalcified and calcified coronary plaque using dual-source multidetector cardiac computed tomographic angiography. The presence and volumes of WMD were assessed by 3-Tesla brain magnetic resonance imaging. Participants were 60% women, 36% African-American, mean age 51.6 ± 10.6 years. The overall prevalence of coronary plaque was 43.0%. Subjects with coronary plaque had significantly greater WMD volumes (median 1,222 mm 3 , interquartile range 448 to 3,871) compared with those without coronary plaque (median 551 mm 3 , interquartile range 105 to 1,523, p <0.001). In multivariate regression analysis, adjusting for age, gender, race, traditional risk factors, total brain volume, and intrafamilial correlations, the presence of coronary plaque was independently associated with WMD volume (p = 0.05). This study shows a significant association between WMD and noncalcified and calcified coronary plaque in healthy subjects, independent of age and risk factors. In conclusion, these findings support the premise of possible shared causal pathways in 2 vascular beds in families at increased risk for early-onset vascular disease.
Apparently healthy subjects with a family history of early-onset coronary artery disease (CAD) are at marked increased risk of developing clinically manifest CAD, independent of traditional risk factors. We recently demonstrated a high prevalence of early silent CAD in younger-age apparently healthy siblings of persons with early-onset CAD and a high 10-year incidence of clinically manifest CAD events. We have also reported a high prevalence of cerebral white matter disease (WMD) on magnetic resonance imaging (MRI) in young healthy siblings of early-onset CAD probands that was comparable to that of older subjects participating in the Atherosclerosis Risk in Communities study, suggesting an early subclinical atherosclerotic disease of the brains in those with a strong family history of CAD. Thus, the present study was designed to determine the association between coronary plaque on computed tomographic angiography (CTA) and WMD in young apparently healthy asymptomatic subjects with a strong family history of early-onset CAD.
Methods
Participants (n = 405) were recruited from the ongoing Genetic Study of Atherosclerosis Risk (GeneSTAR), a prospective study begun in 1982 to characterize genetic and biologic factors associated with incident cardiovascular and cerebrovascular disease in families with early-onset CAD. Briefly, hospitalized probands with acute myocardial infarction, unstable angina with coronary revascularization, or acute angina with flow-limiting stenosis of >50% diameter in at least 1 coronary artery at age <60 years were identified, and their healthy siblings <60 years of age were recruited and screened for risk factors and occult CAD using nuclear perfusion imaging. Commencing in 2002, adult offspring of both the probands and the siblings were also enrolled and underwent CTA at the same time as the original siblings. For the present study, the initially healthy siblings and offspring were included if they were 30 to 75 years of age and had no known history of CAD, stroke, or documented transient ischemic attacks. At the time of return for the CTA measurements, siblings and offspring were excluded if they had any serious chronic illness such as systemic autoimmune disease, chronic kidney disease, or neurologic diseases (dementia, Parkinson’s disease, multiple sclerosis, or life-threatening co-morbidity such as acquired immunodeficiency syndrome, cancer). Subjects were excluded if they reported a history of allergy to iodinated contrast material or implanted metal precluding MRI testing. The study was approved by the Johns Hopkins Medicine Institutional Review Board, and all participants gave informed consent.
Subjects underwent a comprehensive screening with all testing performed on the same day. Medical history and current medication use were assessed, and a physical examination was performed by a study physician. Anthropometric measures included height in inches and weight in kilograms; body mass index was calculated as weight in kilograms divided by height in meters squared. Current cigarette smoking was assessed using a standardized questionnaire and/or by expired carbon monoxide levels of ≥8 ppm on 2 measurements. Blood pressure was measured according to the American Heart Association guidelines 3 times over an 8-hour screening visit. Hypertension was defined as an average blood pressure ≥140 mm Hg systolic or ≥90 mm Hg diastolic, and/or use of an antihypertensive drug. Blood was taken for measurement of lipid and glucose levels after subjects had fasted overnight for 8 to 12 hours. Total cholesterol, high-density lipoprotein cholesterol, and triglyceride levels were measured using the United States Centers for Disease Control standardized methods. Low-density lipoprotein cholesterol was estimated using the Friedewald formula for those with triglyceride levels <400 mg/dl. Direct measurement of low-density lipoprotein cholesterol using ultracentrifugation was used for those with triglyceride levels ≥400 mg/dl (n = 5). Glucose concentration was measured using the glucose oxidase method ; type 2 diabetes was defined as a physician-diagnosed history, a fasting glucose level ≥126 mg/dl, and/or use of hypoglycemic medications.
All participants underwent intracranial MRI and coronary CTA imaging. MRI was performed using a Philips 3.0-T scanner (Andover, Massachusetts). The series included the following imaging sequences (1) Axial T1-weighted magnetization prepared rapid gradient echo: repetition time 10 ms, time to echo 6 ms, inversion time voxel size 0.75 × 0.75 × 1.0 mm 3 , contiguous slices, with field of view imaging 240 mm, matrix 256 × 256 × 160 mm and (2) Axial turbo spin echo fluid attenuation inversion recovery: repetition time 11,000 ms, inversion time 2,800 ms, time to echo 68 ms, voxel size 0.47 × 0.47 × 3.0 mm 3 , contiguous slices, field of view imaging 240 mm, matrix 256 × 256 mm. An expert neuroradiologist evaluated all images for clinical pathology (DY). Volumetric analysis of white matter hyperintensities was performed using MIPAV software (Center for Information Technology, National Institutes of Health, Rockville, Maryland) as previously described. We implemented the topology-preserving anatomical segmentation algorithm (Lesion-TOADS software, Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, Maryland) as a module in the MAPS to simultaneously segment major brain structures and delineate white matter lesions. Segmented brain volumes and the WMD volume were quantified automatically using a multichannel classifier according to a support vector machine approach. The total volume of WMD was the primary dependent variable.
All participants underwent coronary CTA using a dual-source multidetector scanner (Definition Flash, Siemens Medical Solutions, Forchheim, Germany) to detect coronary artery plaque. A noncontrast scan was first performed to determine the coronary artery calcium (CAC) score. Coronary CTA was then performed with prospective electrocardiographic gating, 128 × 0.6 mm detector collimation, 280 ms gantry rotation, 850 mA tube current, and 120 kV tube voltage. Subsequently, 0.75-mm-thick axial slices were reconstructed at 0.5-mm intervals with B26 kernel using a half-scan reconstruction algorithm with resulting temporal resolution of 75 ms. All CTA scans were evaluated with the reader blinded to the participants’ risk factor, clinical, and MRI profiles. Using the noncontrast-enhanced images, the extent of CAC, defined as pixels of >130 HU, using the Agatston method was measured on a workstation (Leonardo Multimodality Workstation, Syngo, Siemens Medical Solutions, Malvern, Pennsylvania). The contrast-enhanced multidetector computed tomographic scans were then evaluated for plaque by examining the axial slices, curved multiplanar reformations, and thin-slab maximum intensity projections. Plaques were classified as exclusively noncalcified, primarily calcified, or of mixed composition.
All variables were examined using standard descriptive statistics, t tests were used for group comparisons for normally distributed variables and Wilcoxon rank sum tests for non-normally distributed variables; the chi-square statistic was used for testing categorical variables. Previous studies have examined associations of CAC with WMD only in older subjects using the age cut point of ≥55 years. We thus examined WMD volumes by strata of gender and age dichotomized at <55 or ≥55 years to garner new information about the younger age group and also by CAC Agatston score groups (0, 1 to 99, 100 to 299, and ≥300). White matter lesion volume was logarithmically transformed to achieve normality for multivariate linear regression analysis. One half the minimum detectable lesion volume (i.e., 0.5 × 13 = 6.5) was added to zero readings to allow for logarithmic transformation (39 of 405 subjects with zero detectable lesion volume). The multivariate model using generalized estimating equations to correct for nonindependence of families was performed to determine the association of CAC with WMD volume, adjusting for total brain volume and traditional risk factors, including age, gender, race, hypertension, diabetes, current smoking, and low-density lipoprotein cholesterol. Sensitivity analyses were performed using alternate transformations of WMD volume: (1) a random number from 1 to 12 was added to the zero values to allow for logarithmic transformation and (2) Tobit analysis was used when zero WMD volume was considered censored, with 1,000 bootstrapped iterations with family resampling to correct for intrafamilial correlations.
Results
The study population consisted of 405 apparently healthy subjects identified from 245 families with early-onset CAD (one proband per family). Probands were 68.2% men, with a mean age of 46.7 ± 6.8 years for the first CAD event. Study subjects were siblings (n = 217) of the probands or adult offspring (n = 188) of the probands or siblings. The mean age of offspring was 43.6 ± 7.6 years, range 30 to 62 years, whereas siblings were 58.5 ± 7.4 years old, range 38 to 74 years. Sample characteristics are listed in Table 1 by the absence or presence of any coronary plaque and by WMD volume dichotomized at the median. The overall prevalence of calcified and/or noncalcified coronary plaque was 43.0%. Most participants had some degree of WMD with an overall prevalence of 90.4%. Older age and hypertension were significantly associated with both the presence of coronary plaque and WMD volume above the median. Additionally, triglycerides, high-density lipoprotein cholesterol, and diabetes were significantly associated with the presence of coronary plaque. Nothing beyond age and hypertension was associated with WMD volume above or below the median level. Subjects on statin therapy had a greater prevalence of coronary plaque compared with those not on statin therapy ( Table 1 ) and greater median WMD volumes (1,356, interquartile range [IQR] 422 to 3,696 vs 687, IQR 207 to 1,774, respectively; p = 0.0002.)
| Variable | Coronary Plaque Absent (n = 231) | Coronary Plaque Present (n = 174) | p | WMD Volume ≤799 mm 3 , n = 203 (Median) | WMD Volume >799 mm 3 , n = 202 (Median) | p |
|---|---|---|---|---|---|---|
| Age (yrs) | 47.3 ± 9.5 | 57.4 ± 8.9 | <0.001 | 47.8 ± 9.8 | 55.5 ± 9.8 | <0.001 |
| Men (%) | 30.7 | 53.4 | <0.001 | 41.9 | 39.1 | 0.57 |
| African-American (%) | 36.4 | 35.1 | 0.78 | 32.0 | 39.1 | 0.14 |
| Hypertension (%) | 30.3 | 58.6 | <0.001 | 34.0 | 51.0 | 0.001 |
| Diabetes (%) | 7.4 | 16.1 | 0.006 | 9.4 | 12.9 | 0.26 |
| Current smoking (%) | 16.5 | 20.1 | 0.34 | 18.7 | 17.3 | 0.72 |
| Statin therapy (%) | 34.3 | 65.7 | <0.001 | 35.4 | 64.7 | <0.001 |
| LDL cholesterol (mg/dl) | 114.1 ± 34.7 | 114.5 ± 40.1 | 0.91 | 117.3 ± 37.4 | 111.2 ± 36.5 | 0.10 |
| HDL cholesterol (mg/dl) | 60.3 ± 17.8 | 56.1 ± 17.3 | 0.02 | 57.8 ± 16.9 | 59.7 ± 18.6 | 0.28 |
| Triglycerides (mg/dl) ‡ | 92.0 (66.0–135.0) | 103.0 (73.0–151.0) | 0.002 | 93.0 (70.0–141.0) | 95.0 (66.0–138.3) | 0.81 |
| Body mass index (kg/m 2 ) | 29.7 ± 5.9 | 30.3 ± 5.4 | 0.34 | 30.2 ± 5.8 | 29.7 ± 5.5 | 0.42 |
| hs-CRP (mg/dl) ‡ | 2.8 (1.1–9.3) | 2.3 (1.1–7.9) | 0.73 | 2.8 (1.2–9.2) | 2.3 (1.1–7.8) | 0.36 |
∗ Coronary plaque defined by the presence of calcified or noncalcified plaque on CTA.
† Continuous variables are presented as mean ± 1 SD.
‡ Non-normally distributed continuous variables are presented as median (IQR).
Of the 174 subjects with coronary plaque on CTA, 11.3% had exclusively noncalcified plaque (no calcium), 31.2% had primarily calcified plaque, and 57.5% had both calcified and noncalcified plaque. In 63 subjects <55 years of age with coronary plaque, 14.4% had exclusively noncalcified plaque, compared with only 6.6% of those subjects ≥55 years of age. Subjects with coronary plaque had significantly greater volumes of WMD compared with those with no coronary plaque (median 1,222, IQR 105 to 1,523 vs median 551, IQR 448 to 3,871, respectively; p <0.001), as shown in Figure 1 . The gender-specific distributions of WMD volume by age group and the absence or presence of subclinical CAD are shown in Figure 2 . In subjects <55 years of age, WMD volume was significantly higher in men and women with coronary plaque compared with those without coronary plaque. There was a similar pattern in older men ≥55 years of age but not in older women. Overall the association of coronary plaque with greater WMD volume remained highly significant when adjusted for age, gender, and intrafamilial correlation (p = 0.004). This association remained significant after additional adjustment with statin use (p = 0.03). The distribution of WMD volume by calcified plaque extent defined by incremental categories of CAC (Agatston) is shown in Figure 3 . There was a strong association of incremental CAC score category with WMD volume (p <0.001 for trend).
