Atherogenic index of plasma and the risk of advanced subclinical coronary artery disease beyond traditional risk factors: An observational cohort study

Abstract Background Atherogenic lipoprotein profile of plasma is an important risk factor for atherosclerosis. The atherogenic index of plasma (AIP) has been suggested as a novel marker for atherosclerosis. Hypothesis AIP is a useful marker of advanced subclinical coronary artery disease (CAD) in subjects without overt renal dysfunction. Methods A total of 6928 subjects with estimated glomerular filtration rate > 60 mL/minutes/1.73 m2 evaluated by coronary computed tomography angiography (CCTA) for health check‐up were included. The relation of AIP to advanced CAD (heavy coronary calcification, defined as coronary artery calcium score [CACS] >100 or obstructive coronary plaque [OCP], defined as plaque with >50% stenosis) was evaluated. Results All participants were stratified into four groups based on AIP quartiles. The prevalence of CACS >100 (group I [lowest] 4.7% vs group II 7.0% vs group III 8.8% vs group IV 10.0%) and OCP (group I 3.7% vs group II 6.4% vs group III 8.8% vs group IV 10.9%) (all P < .001) increased with elevating AIP quartiles. Higher AIP (per 0.1 unit increase) was associated with an increased risk of CACS >100 (odds ratio [OR] 1.057, 95% confidence interval (CI) 1.010 to 1.106, P = .017; relative risk (RR) 1.048, 95% CI 1.009‐1.089, and P = .015) and OCP (OR 1.079, 95% CI 1.033‐1.127, P = .001; RR 1.069, 95% CI 1.031‐1.108, P < .001) after adjusting for age > 60 years, male sex, hypertension, diabetes mellitus, dyslipidaemia, obesity, and proteinuria. Conclusion AIP is independently associated with advanced subclinical CAD beyond traditional risk factors.


| INTRODUCTION
Atherogenic lipoprotein profile of plasma is an important risk factor for atherosclerosis. The atherogenic index of plasma (AIP) has been suggested as a marker of plasma atherogenicity based on its strong and positive association with cholesterol esterification rates, lipoprotein particle size, and remnant lipoproteinaemia. [1][2][3] In addition, previous data have shown that AIP is more closely related to cardiovascular (CV) risk than individual lipoprotein cholesterol fractions or other atherogenic indices. [4][5][6] However, data on the association of AIP with the advanced subclinical coronary artery disease (CAD) beyond traditional risk factors are limited in clinical practice.
Coronary artery calcium score (CACS) has been regarded as a useful marker of coronary atherosclerosis because it represents the degree of atheromatous plaque burden. 7,8 Recently, coronary computed tomography angiography (CCTA) has been used as an effective and noninvasive tool to assess the presence of plaques, severity of stenosis, and subtypes of plaque in coronary arteries with a robust ability to predict major adverse CV events. [9][10][11] Thus, this study evaluated the association of AIP with advanced subclinical CAD in subjects without overt renal dysfunction using CCTA.

| Study population and design
This observational, retrospective, single-center registry consisted of 8648 consecutive participants evaluated by CCTA for health check-up with 64-slice multidetector computed tomography between January 2004 and April 2009 at Severance CV hospital. 12 Of them, 1720 subjects were excluded because of (a) age < 30 years (n = 69), (b) overt renal dysfunction defined as glomerular filtration rate (GFR) estimated by modification of diet in renal disease (MDRD) method <60 mL/ minutes/1.73 m 2 (n = 958), and (c) insufficient medical records for traditional risk factor and AIP calculation or poor image quality (n = 693). As a result, 6928 participants were finally included in the present study.
The study protocol was approved by the ethics committee of severance CV Hospital, and informed consent for the procedure was obtained from each participant. Patients or the public were not involved in the design, conduct, reporting, or dissemination of our research.
Medical history of hypertension, diabetes mellitus, and dyslipidaemia was systematically acquired. Height, body weight, and blood pressure (BP) were measured during visits. Height and weight measurements were obtained while the subjects wore light clothing without shoes. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. After the participants rested for ≥5 minutes, the BP was measured at the right arm using an automatic manometer with an appropriate cuff size. Serum levels of total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), glucose, and creatinine were measured after a minimum of 12 hours fasting period. The AIP was calculated as the base 10 logarithm of the ratio of the triglycerides to HDL-C concentrations. 1 The kidney function was ascertained by estimated GFR calculated using the MDRD formula. 13 Hypertension was defined as systolic BP ≥140 mmHg and/or diastolic BP ≥90 mmHg or treatment with antihypertensive agents.
Diabetes mellitus was defined as treatment with hypoglycaemic agents or insulin, or fasting glucose ≥126 mg/dL. Dyslipidaemia was defined as serum levels of total cholesterol ≥240 mg/dL, LDL-C ≥ 130 mg/dL, HDL-C ≤ 40 mg/dL, triglycerides ≥150 mg/dL, and/or treatment with lipid lowering agents. Obesity was defined as BMI ≥25 kg/m 2 based on the cutoffs for Asian population. Proteinuria was defined based on the sex-specific urinary albumin/creatinine ratio. 14,15 Age ≥ 60 years, male sex, hypertension, diabetes mellitus, dyslipidaemia, obesity, and proteinuria were considered as traditional risk factors in the present study. All procedures were in accordance with the ethical standards of the institutional research committee (Giessen, AZ.: 127/16) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

| CCTA protocol
During CCTA examination, subjects with an initial heart rate > 65 beats/minutes received a single oral dose of 50 mg metoprolol tar- Bracco, Milan, Italy) injected at 5 mL/s, followed by an injection of 30 mL diluted contrast, and then 30 mL saline at 5 mL/s using an injector (Envision CT; Medrad, Indianola, Pennsylvania). The images were evaluated by two experienced cardiac radiologists. CACS, which was measured using the scoring system described by Agatston et al, 16 was categorized into four groups based on the following scores: 0, 1 to 10, 11 to 100, and >100. Coronary plaques were defined as structures ≥1 mm 2 within or adjacent to the vessel lumen, which were clearly distinguishable from the lumen and the surrounding pericardial tissue. Plaques without calcium were classified as noncalcified, those with calcified tissue involving ≥50% of the plaque area (density > 130 HU) were classified as calcified, and those with <50% calcium were classified as mixed plaques. 17 Figure 2. The prevalence of coronary plaque sub-types according to AIP quartiles is presented in Supplementary Figure 1.

| Clinical variables and the risk of advanced CAD
The association of clinical variables with CACS >100 and OCP is presented in Table 2. Age, male sex, hypertension, diabetes mellitus, dyslipidaemia, obesity, and proteinuria were significantly associated with an increased risk of both CACS >100 and OCP. With regard to the association between AIP and the risk of advanced CAD, the risk of CACS >100 and OCP was significantly higher in group II, group III, and group IV compared with that in group I, respectively.

| Predictive value of AIP for advanced CAD
In ROC curve analysis, the optimal AIP cutoff for predicting CACS  Table 1).

| DISCUSSION
In the present study, a significant association was observed between AIP and the risk of advanced subclinical CAD in subjects beyond traditional risk factors. Recently, CCTA has been widely used for the comprehensive evaluation of coronary atherosclerosis, including lesion location, severity, and plaque characteristics in clinical practice. 18 The CONFIRM (Coronary CT angiography evaluation for clinical outcomes: An international multicenter registry) studies previously revealed that the severity of CAD as revealed by CCTA has a strong predictive value for adverse clinical outcomes. 19,20 We considered a CACS >100 an indicator of advanced coronary artery calcification based on the facts that (1) Nasir et al. 21 reported that the frequency of a CACS >100 in the Asian population is significantly lower than that in Western populations and (2) the overall proportion of CACS >100 was 7.6% in our participants. Additionally, we defined OCP as a plaque with ≥ 50% stenosis because 1) all participants received a CCTA examination for health check-up and (2) the prevalence of OCP was 7.4% in this study.
High levels of the triglycerides-to-HDL-C ratio have been associated with obesity and metabolic syndrome. 22 (VLDL-4) study showed that a higher ratio of triglycerides to HDL-C was associated with an increasingly atherogenic lipid phenotype, characterized by higher remnant lipoprotein particle cholesterol along with higher nonHDL-C and LDL-C density. 3 Millán et al 25  Several limitations to the study should be acknowledged. First, this was a retrospective, cross-sectional, and observational study.
Thus, selection bias might be present. Second, potential roles of other environmental risk factors and the diet or exercise in subclinical atherosclerosis were not evaluated. Finally, this study included only the Korean adults, which may limit the generalization of obtained results to other populations. Despite these limitations, this study is unique in that an independent association was identified between AIP and advanced subclinical coronary atherosclerosis beyond traditional risk factors in an Asian population. Furthermore, the present study investigated the association of AIP with coronary plaque subtypes.
In conclusion, elevated AIP levels are significantly associated with a higher risk of advanced subclinical CAD as revealed by CCTA in Korean adults with near-normal renal function even after adjusting for traditional risk factors. With consideration of traditional risk factors together, AIP may be an effective predictive marker for advanced coronary atherosclerosis in general population without overt renal dysfunction.

ACKNOWLEDGMENT
The Medical Information Center of Ulsan University Hospital supported this work.