Non-high-density lipoprotein cholesterol (non-HDLc) is an independent predictor of cardiovascular disease risk, with elevated levels signifying an increased risk beyond low-density lipoprotein. Previous data have shown inconsistent association of lipid subfractions with progression of coronary artery calcium (CAC), a surrogate marker of incident cardiovascular disease. We sought to evaluate the association between non-HDLc and development (incident) and progression of CAC in a cohort of multiethnic asymptomatic subjects. The cohort (n = 5,705) was derived from the limited access data set of the Multi-Ethnic Study of Atherosclerosis obtained from the National Heart Lung and Blood Institute. Multivariable regression analysis was performed to derive the association between non-HDLc and incident CAC (n = 2,927) and non-HDLc and progression of CAC (n = 2,778). In the population without CAC at baseline, non-HDLc, especially >190 mg/dl, was independently associated with incident CAC (relative risk 1.40, 95% confidence interval 1.09 to 1.79, p = 0.008) after adjustments with age, gender, race, systolic blood pressure, antihypertension medication use, smoking, diabetes, lipid-lowering therapy use, follow-up duration, and waist-hip ratio. Similarly, among those with CAC at baseline, non-HDLc levels >190 mg/dl were associated with significant CAC progression in the overall population (β 16.4, 95% confidence interval −5.63 to 27.2, p = 0.003) after adjustments. In conclusion, non-HDLc levels, especially >190 mg/dl, are consistently associated with increased risk of CAC progression. Our results suggest that among lipid fractions, non-HDLc may be best suited for the prediction of future CAC progression.
Coronary artery calcium (CAC) is a known surrogate of subclinical atherosclerosis and a strong, independent predictor of cardiovascular morbidity and mortality. The precise location and quantification of CAC allow serial analysis of progression of CAC and the potential to capture the effects of continued exposure to risk factors over a period of time. Furthermore, CAC progression over time also enables us to predict the risk of incident coronary heart disease events and, thereby, assessment of the effectiveness of medical therapy aimed at reducing cardiovascular events. Traditional lipid parameters (low-density lipoprotein cholesterol [LDLc], high-density lipoprotein cholesterol [HDLc], triglycerides [TGs]) have been studied for their association with progression of CAC with inconsistent results. No data exist evaluating the association between non-HDLc and CAC progression. Therefore, we sought to evaluate the association between non-HDLc and the incidence and progression of CAC among a multiethnic population asymptomatic adults at baseline.
Methods
The Multi-Ethnic Study of Atherosclerosis (MESA) is a population-based study (n = 6,814) of varied ethnicities, aged 45-84 years, with no cardiovascular disease (CVD). Details of the study have been reported previously. We used the National Heart Lung and Blood Institute’s (NHLBI) limited access data set for our study. The cohort with data on non-HDLc was included. Furthermore, participants without data on CAC score (n = 1032) at baseline or on follow-up and those with missing information on the study variables (n = 77) were excluded for respective analyses (please refer to the Appendix ), leaving a total of 5,705 subjects for the final analyses.
Information on demographics, smoking, medical conditions, and family history was collected by questionnaire at the initial examination. Hypertension was defined as systolic blood pressure (BP) ≥140 mm Hg or diastolic BP ≥90 mm Hg or a history of medical treatment for hypertension. Diabetes was defined as fasting blood glucose ≥126 mg/dl or a history of insulin or an oral hypoglycemic agent use. Hyperlipidemia was defined as total cholesterol levels ≥240 mg/dl or use of lipid-lowering therapy.
Fasting blood samples were obtained to measure plasma lipids including TGs, total cholesterol, and HDLc using Center for Disease Control standardized methods. LDLc was estimated using Friedewald equation in patients with TG <400 mg/dl. Non-HDLc was calculated using the equation:
CAC was measured using either electron-beam computed tomography or multidetector computed tomography at 3 field centers. The methodology for acquisition and interpretation of the scans has been documented previously. Each participant was scanned twice consecutively, and these scans were read independently at a centralized reading center. The results from the 2 scans were averaged to provide a more accurate point estimate of the amount of calcium present. The amount of calcium was quantified using the Agatston scoring method. Calcium scores were adjusted using a standard calcium phantom that was scanned along with the participant. Measurements were made of all participants at baseline (2000 to 2002). Follow-up examinations were performed in 50% of participants at examination 2 and the remaining 50% at examination 3, with an average of 22 and 40 months between examinations, respectively. Incident CAC was defined as detectable CAC at the follow-up examination (either examination 2 or 3) in a participant free of detectable CAC at examination 1 (n = 2,927), whereas CAC progression was defined as a change in CAC volume score in participants who had detectable CAC at examination 1 (n = 2,778).
Participants were compared across the categories of non-HDLc (<160 mg/dl, 160 to 189 mg/dl, and ≥190 mg/dl). Categorical variables were compared using chi-square test and are presented as percentages. Continuous variables were compared using analysis of variance test and are presented as mean ± SD. To evaluate the association between incident CAC and non-HDLc, we performed generalized linear regression with log-link and Poisson regression model with robust error variance, an approach that avoids convergence issues at times faced using a binomial model. Relative risk (RR) regression was used rather than logistic regression because the incidence of CAC was >10%, and therefore, the odds ratio can be an overestimation of the RR. To estimate the progression of CAC among those with detectable CAC in examination 1, a median regression analysis was performed that provides advantages of being more robust to the outliers and being semiparametric in nature. These 2 study outcomes were studied separately using following models: model 1-adjusted for age, gender, race, and follow-up duration. Model 2-adjusted for variables in model 1 and systolic BP (continuous), antihypertension medications, smoking, diabetes, lipid-lowering therapy, and waist-hip ratio. Formal statistical analysis was performed to evaluate the interaction between non-HDLc and gender and non-HDLc and ethnicity, which was found to be NS (p >0.05). All the statistical analysis was performed using STATA Version 10 (STATA Corp., College Station, Texas), and statistical significance was defined as p <0.05.
Results
The mean age of the participants was 62.0 years (±10.2 years) and consisted of nearly 48% men, 23% Hispanics, 38% Caucasians, 12% Chinese, and 27% African-Americans. Table 1 lists the baseline characteristics compared across non-HDLc categories. Participants with higher non-HDLc levels were younger in age with higher prevalence of metabolic syndrome, had higher waist-hip ratio, diastolic BP, and had a worse lipid profile. Table 2 lists the association of non-HDLc with CAC incidence and progression. As shown, non-HDLc exhibits a linear association with incident and progression of CAC. Non-HDLc levels, especially ≥190 mg/dl, were independently associated with incident CAC in the overall population (RR: 1.40; 95% confidence interval 1.09 to 1.79; p = 0.008). Similarly, among those with CAC at baseline, non-HDLc levels ≥190 mg/dl were significantly associated with CAC progression in the overall population (β: 16.4; 95% confidence interval 5.63 to 27.2; p = 0.003) after adjustments per model 2.
| Non-High-Density Lipoprotein Cholesterol Levels (mg/dl) | ||||
|---|---|---|---|---|
| <160 (n = 4,048) | 160-189 (n = 1,133) | >190 (n = 524) | p Value | |
| Age (yrs) | 62 ± 10 | 61 ± 10 | 61 ± 9 | 0.0006 ∗ |
| Men | 47.4% | 50.0% | 44.5% | 0.097 |
| Hispanic | 19.4% | 26.0% | 28.4% | |
| Caucasian | 39.5% | 39.6% | 40.7% | <0.001 ∗ |
| Chinese | 12.2% | 12.0% | 10.1% | |
| African-American | 29.0% | 22.3% | 20.8% | |
| Waist-to-hip ratio | 0.92 ± 0.10 | 0.94 ± 0.10 | 0.94 ± 0.10 | <0.001 ∗ |
| Systolic BP (mm Hg) | 126 ± 2 | 127 ± 19 | 128 ± 22 | 0.163 |
| Diastolic BP (mm Hg) | 71 ± 10 | 72 ± 10 | 73 ± 11 | <0.001 |
| Antihypertensive medication use | 38.9% | 30.8% | 28.2% | <0.001 ∗ |
| Smoker | 49.8% | 47.4% | 47.5% | 0.253 |
| Diabetes mellitus | 13.1% | 11.5% | 13.9% | 0.257 |
| Metabolic syndrome | 28.2% | 37.5% | 49.6% | <0.001 ∗ |
| Hyperlipidemia | 19.4% | 17.5% | 83.4% | <0.001 ∗ |
| Lipid lowering therapy | 19.0% | 10.3% | 8.8% | <0.001 ∗ |
| Total cholesterol (mg/dl) | 178 ± 24 | 219 ± 14 | 260 ± 29 | <0.001 ∗ |
| LDLc (mg/dl) | 103 ± 21 | 142 ± 15 | 174 ± 23 | <0.001 ∗ |
| Triglycerides (mg/dl) | 114 ± 62 | 155 ± 80 | 207 ± 157 | <0.001 ∗ |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
