Associations of atherogenic index of plasma and other nontraditional lipid parameters with the risk of coronary artery aneurysm

Abstract

Background

Aberrant lipid metabolism is associated with the development of coronary artery aneurysms (CAA). Research on the association between traditional and nontraditional lipid parameters, as well as CAA, in patients undergoing coronary angiography is limited. Therefore, this study aimed to investigate and compare the effects of lipid parameters on CAA development.

Methods

160 patients with CAA and 160 patients without CAA were included in this study. Traditional lipid parameters were calculated and compared between the two groups. Multivariate logistic analysis and restricted cubic spline (RCS) were used to explore the association between lipid indicators and CAA incidence. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of lipid parameters.

Results

Serum triglyceride (TG), atherogenic index of plasma (AIP), lipoprotein combine index (LCI), AC, Castelli’s index-I (CRI-I), and CRI-II were all significantly higher in patients with CAA, while high-density lipoprotein cholesterol (HDL-C) and remnant cholesterol (RC) were lower in CAA group compared with control group (p < 0.05). Multivariate logistic regression analysis showed that TG, AIP, LCI, AC, CRI-I, and CRI-II were all independent risk factors for CAA, whereas total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), non-HDL-C, and RC/HDL-C were not associated with CAA after adjusting for covariates. Among these traditional and nontraditional lipid parameters, AIP exhibited the highest area under the curve (AUC) for predicting CAA (AUC = 0.712, 95% confidence interval (CI): 0.633–0.791, p < 0.001), suggesting that nontraditional lipid parameters have outstanding predictive power for CAA.

Conclusions

Patients with CAA had higher levels of traditional lipid parameters. Among these lipid indicators, the AIP had the highest predictive potential for CAA.

Introduction

A coronary artery aneurysm (CAA) is defined as an atypical dilation with a dilated segment more than 1.5 times of the normal adjacent segment [1]. Epidemiological studies showed 0.3%-5.0% of patients undergoing coronary angiography are diagnosed with CAA [2]. Accumulating evidence indicates that CAA is closely associated with a poor prognosis [3]. In a prospective study including 1698 patients with myocardial infarction, CAA increased 2.25 the risk of major adverse cardiac events (MACE) during a 4-year follow-up [4]. Therefore, it is important to identify the risk factors for CAA and improve prognosis.

Although CAA is recognized as a different form of atherosclerosis, the underlying mechanisms remain largely unknown [5]. Various factors are involved in CAA pathogenesis by inducing the activation of matrix-degrading enzymes and subsequent excessive expansive remodeling. Abnormal lipid metabolism plays a vital role in this process. Oxidized LDL-C is engulfed by macrophages to form foam cells, whereas matrix metalloproteinases degrade the extracellular matrix layers, ultimately contributing to CAA [6]. Previous clinical studies have found an association between dysregulated lipid levels and increased incidence of CAA [7]. Patients with familial hypercholesterolemia (FH) were more likely to develop CAA [8, 9]. CAA occurs more frequently in patients with reduced high-density lipoprotein cholesterol (HDL-C) levels. Notably, reduction in serum low-density lipoprotein cholesterol (LDL-C) levels by repeated plasma exchange in a patient with FH led to angiographic improvement in CAA [10]. Boles et al. were the first to employ lipid profiling and demonstrated lower sphingomyelin concentration in CAA blood samples [11]. Nontraditional lipid parameters are derived from the calculations of traditional lipid indicators and have recently received increasing attention. Nontraditional lipid parameters can better reflect the balance between atherogenic and antiatherogenic lipoproteins and interactions between lipid components [12]. Previous studies have reported that total cholesterol (TC)/HDL-C (Castelli’s index-I, CRI-I) and LDL-C/HDL-C (CRI-II) are superior predictors of coronary artery disease (CAD) compared with conventional lipid parameters [13]. The atherogenic index of plasma (AIP) was identified as an independent risk factor for CAD risk in 4644 postmenopausal women cohort [14]. Additionally, elevated remnant cholesterol was linked to increased risk of both developing thoracic aortic aneurysm and CAA [15, 16]. However, the relationship between unconventional lipid parameters and CAA has not been investigated.

Based on the role of dysregulated lipid metabolism in the pathogenesis of CAA, the present study included 320 patients to investigate the association between various traditional lipid parameters, including lipoprotein combine index (LCI), AIP, non-high-density lipoprotein cholesterol (non-HDL-C), AC, Castelli’s index-I (CRI-I), CRI-II, remnant cholesterol (RC), RC/HDL-C, and CAA. Logistic regression analysis, restricted cubic spline (RCS), and receiver operating characteristic (ROC) curves were used to analyze the relationships between these nontraditional lipid parameters and CAA.

Methods and materials

Study population

This was a retrospective cohort study that included CAA patients and patients without CAA who underwent coronary angiography for suspected coronary artery disease at the Department of Cardiology, Zhongda Hospital affiliated with Southeast University, from January 2021 to December 2024. The exclusion criteria were as follows: (1) age less than 18 or over 80 years; (2) people who had taken lipid-modifying drugs; (3) lack of triglyceride (TG), TC, HDL-C, LDL-C, or other critical clinical data; (4) allergy to contrast media, severe hepatic and renal insufficiency, infectious diseases, and tumors; and (5) pregnant and lactating women. The study was approved by the Ethics Committee of Zhongda Hospital affiliated with Southeast University (No: 2023ZDSYLL264-P01). Written informed consent was obtained from all the patients.

Definitions

Coronary artery aneurysm (CAA)

Abnormal coronary segment dilation of more than 1.5-fold the diameter of the adjacent normal segments is diagnosed as CAA [3].

Nontraditional lipid parameter

Traditional lipid parameters, including HDL-C, LDL-C, TC, and TG levels, were measured using an automatic biochemical analyzer. Nontraditional lipid parameters were calculated using the following formula [17–22]:

$$\text{AIP}=\text{lg}\left(\text{TG}/\text{HDL - C}\right)$$

$$\text{Non - HDL - C}=\text{TC - HDL - C}$$

$$\text{AC}=\text{non - HDL - C}/\text{HDL - C}$$

$$\text{CRI - I}=\text{TC}/\text{HDL - C}$$

$$\text{CRI - II}=\text{LDL}/\text{HDL - C}$$

$$\text{LCI}=\text{TC}\times \text{TG}\times \text{LDL - C}/\text{HDL - C}$$

$$\text{RC}=\text{TC - HDL - C - LDL - C}$$

Statistical analysis

Normally and non-normally distributed continuous variables were presented as mean ± standard deviation and mean (interquartile range). Categorical variables are presented as frequencies and percentages. Student’s t-test, Mann–Whitney U test, and chi-square test were used to compare data between the CAA and control groups. Logistic regression models were established to investigate the association between traditional and nontraditional lipid parameters and CAA: Model 1 (unadjusted), Model 2 (adjusted for age and sex), and Model 3 (adjusted for age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, and serum creatinine). The nontraditional lipid parameters were then divided into three tertiles, and odds ratios (OR) and 95% confidence intervals (CI) were calculated to evaluate the relationship between tertiles and CAA, with the lowest tertile as a reference. Restricted cubic splines (RCS) were used to explore the association between lipid indicators and CAA incidence. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of lipid parameters. All statistical analyses were performed using R version 4.2.2 and IBM SPSS Statistics 26.0. A two-tailed p < 0.05 was considered statistically significant.

Results

Baseline characteristics of subject characteristics

In the present study, 320 participants were included in the present study, with an average age of 64.74 ± 11.93 years, and 234 (73.13%) were male. Table 1 shows the baseline characteristics of the population groups based on CAA. No significant differences were found in age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, or serum creatinine (Scr). In terms of traditional lipid parameters, subjects with CAA have higher TG and lower HDL-C levels. Compared with those without CAA, participants with CAA exhibited higher LCI, AIP, AC, CR-I, and CR-II, as well as lower RC levels, while no significant differences were found in non-HDL-C and RC/HDL-C levels between the two groups. Among the 160 CAA participants, 64 lesions were located in the right coronary artery (RCA), 54 in the left anterior descending coronary artery (LAD), 26 in the left circumflex artery (LCX), and 8 in the left main coronary artery (LM). 2 patients had CAA at both the LM and LAD, 2 had CAA at both the LAD and LCX, and 4 had CAAs at both the LAD and RCA.

Table 1.

The comparison of demographic and coronary angiography data in patients with and without coronary artery aneurysm

	CAA (n = 160)	Non-CAA (n = 160)	p value
Age (year, mean ± SD)	64.89 ± 11.60	64.04 ± 11.87	0.694
Sex (n, %)			0.801
Male	116 (72.5%)	118 (73.8%)
Female	44 (27.5%)	42 (26.2%)
BMI (n, %)			0.400
BMI < 18	2 (1.3%)	2 (1.3%)
18 ≤ BMI < 25	76 (47.5%)	76 (47.5%)
25 ≤ BMI < 30	66 (41.3%)	74 (46.3%)
BMI ≥ 30	16 (10.0%)	8 (5.0%)
SBP (mmHg)	133.57 ± 17.89	133.84 ± 18.91	0.900
DBP (mmHg)	78.19 ± 12.05	77.38 ± 13.84	0.579
Smoking (n, %)	56 (35.0%)	24 (15.0%)	< 0.001
Hypertension (n, %)	118 (73.8%)	116 (72.5%)	0.801
Diabetes (n, %)	50 (31.3%)	34 (21.3%)	0.042
Scr (μmol/L)	80.14 ± 23.76	78.98 ± 17.40	0.623
TG (mmol/L)	1.80 ± 0.98	1.25 ± 0.64	< 0.001
TC (mmol/L)	4.12 ± 1.28	4.13 ± 0.88	0.943
HDL-C (mmol/L)	1.17 ± 0.55	1.30 ± 0.33	0.015
LDL-C (mmol/L)	2.47 ± 1.04	2.37 ± 0.76	0.326
LCI	14.43 (8.01–26.41)	8.17 (4.10–15.85)	< 0.001
AIP	0.19 (− 0.03–0.40)	− 0.05 (− 0.26–0.13)	< 0.001
Non-LDL-C	2.95 ± 1.42	2.83 ± 0.85	0.374
AC	2.88 ± 1.26	2.36 ± 0.94	< 0.001
CR-I	3.88 ± 1.26	3.36 ± 0.94	< 0.001
CR-II	2.34 ± 1.00	1.97 ± 0.83	< 0.001
RC	0.57 (0.26–1.19)	0.52 (0.21–0.93)	0.013
RC/HDL-C	0.50 (0.20–0.86)	0.44 (0.12–0.96)	0.565
CAA lesion
LM (n, %)	8 (5.0%)
LAD (n, %)	54 (35.1%)
LCX (n, %)	26 (14.3%)
RCA (n, %)	64 (40.3%)
LM + LAD	2 (1.3%)
LAD + LCX	2 (1.3%)
LAD + RCA	4 (2.6%)

Univariate analysis of CAA

Table 2 displays the results of the univariate analysis of CAA, illustrating the association between each variable and CAA risk. Smoking was positively associated with an increased risk of CAAs (OR =  3.051, 95% CI 1.418–6.568). A unit increase in TG level was associated with a 2.555-fold risk of developing CAA (95% CI 1.807–3.611). Positive correlations were also found between CAA and nontraditional lipid parameters (OR =  1.037, 95% CI 1.018–1.056, p < 0.001for LCI; OR =  2.558, 95% CI 1.877–3.486, p < 0.001 for AIP; OR =  1.534, 95% CI 1.245–1.889, p < 0.001 for AC; OR =  1.534, 95% CI 1.245–1.889, p <  0.001 for CR-I; OR =  1.550, 95% CI 1.209–1.989, p =  0.001 for CR-II).

Table 2.

Univariate analysis of CAA

Variable	OR	95%CI	p value
Age	1.006	0.988–1.025	0.523
Sex
Female	Reference
Male	1.066	0.650–1.747	0.801
BMI
BMI <  18	Reference
18  ≤ BMI  < 25	0.500	0.024–10.251	0.653
25  ≤ BMI < 30	0.487	0.135–1.756	0.271
BMI ≥ 30	0.459	0.127–1.665	0.236
SBP	0.999	0.982–1.106	0.929
DBP	1.005	0.981–1.029	0.694
Smoking	3.051	1.418–6.568	0.004
Hypertension	1.066	0.530–2.144	0.801
Diabetes	1.684	0.824–3.442	0.153
Scr	1.003	0.988–1.018	0.727
TG	2.555	1.807–3.611	<  0.001
TC	0.993	0.813–1.213	0.960
HDL-C	0.513	0.293–0.898	0.020
LDL-C	1.13	0.885–1.443	0.488
LCI	1.037	1.018–1.056	<  0.001
AIP	2.558	1.877–3.486	<  0.001
Non-LDL-C	1.090	0.833–1.427	0.833
AC	1.534	1.245–1.889	<  0.001
CR-I	1.534	1.245–1.889	<  0.001
CR-II	1.550	1.209–1.989	0.001
RC	0.679	0.497–0.926	0.015
RC/HDL-C	0.892	0.622–1.277	0.658

Logistic regression analysis of associations between lipid parameters and CAA

Three logistic regression models were constructed to evaluate the association between the lipid parameters and CAA risk (Fig. 1). Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, and serum creatinine. Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, and serum creatinine. Among these traditional lipid parameters, TG was associated with a higher risk of CAA in the unadjusted and adjusted models (OR =  2.555, 95% CI 1.807–3.611 for model 1; OR =  2.884, 95% CI 2.009–4.138 for model 2; OR =  2.828, 95% CI 1.940–4.122 for model 3). HDL-C was found negatively associated with CAA, with OR =  0.439 (95% CI 0.233–0.824) after adjusting all confounding covariates. LCI (OR =  1.041, 95% CI 1.022–1.061, AIP (OR =  3.133, 95% CI 2.229–4.403, AC (OR =  1.612, 95% CI 1.300–1.999), , CR-I (OR =  1.612, 95% CI 1.300–1.999, and CR-II (OR =  1.617, 95% CI 1.253–2.087) exhibited associations with CAA after adjusting for age and sex. After further adjustment for age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, and Scr, the associations between LCI (OR =  1.041, 95% CI 1.019–1.063), AIP (OR =  3.071, 95% CI 2.144–4.398), AC (OR =  1.651, 95% CI 1.290–2.113), CR-I (OR =  1.651, 95% CI 1.290–2.113), CR-II (OR =  1.652, 95% CI 1.236–2.206), and CAA remained significant.

Fig. 1.

Forest Plot of traditional and nontraditional lipid parameters with CAA

The five independent risk nontraditional lipid parameters were divided into three tertiles as categorical variables to evaluate the association between these lipid parameters and CAA after adjusting the covariates including age, sex, BMI, SBP, DBP, smoking, hypertension, diabetes, and serum creatinine. As shown in Fig. 2, T2 and T3 was observed to have a 2.325-fold (95% CI 1.250–4.325, p =  0.008) and 3.955-fold risk (95% CI 2.068–7.565, p <  0.001) for CAA using T1 of LCI as a reference. Patients in the highest tertile of AIP, AC, CR-I and CR-II had significantly highest risk for CAA. Among these lipid parameters, AIP had highest predictive value for CAA (OR =  2.916, 95% CI 1.587–5.356, p =  0.001 for T2; OR =  6.105, 95% CI 2.636–10.429, p <  0.001 for T3) (Fig. 2).

Fig. 2.

Associations between tertiles of nontraditional lipid parameters and CAA

Restricted cubic spline (RCS) and ROC curve analysis

The RCS model revealed a J-shaped association between these lipid indices and CAA risk. In particular, the inflection point of the AIP was 0.3. When the AIP was ≥ 0.3, the risk of CAA increased remarkably (Fig. 3). ROC analysis was used to compare the predictive value of TG level and the five selected independent nontraditional lipid parameters for CAA (Fig. 4). All these indicators have superior predictive potential, and the cut-off value, corresponding sensitivity, specificity, and area under the curve (AUC) are listed in Table 3. Among them, AIP exhibited the highest predictive ability, with an AUC of 0.712 (95% CI 0.633–0.791), sensitivity of 58.8%, specificity of 75.0%, and an optimal cut-off value of 0.105.

Fig. 3.

Associations between traditional lipid parameters, nontraditional lipid parameters and CAA using a restricted cubic spline regression model

Fig. 4.

ROC curve analysis of the lipid parameters in predicting CAA

Table 3.

The AUC, cut-off value, sensitivity, and specificity of lipid parameters in predicting CAA

Variable	AUC	Cut-off value	95% CI	Sensitivity (%)	Specificity (%)	p value
TG	0.688	1.155	0.607–0.769	72.5	57.5	< 0.001
LCI	0.643	13.615	0.558–0.729	56.3	71.2	0.002
AIP	0.712	0.105	0.633–0.791	58.8	75.0	< 0.001
AC	0.642	2.531	0.557–0.728	60.0	66.2	0.002
CR-I	0.642	3.531	0.557–0.728	60.0	66.2	0.002
CR-II	0.618	2.134	0.531–0.705	60.0	62.5	0.001

Discussion

This study found that nontraditional lipid parameters play an important role in the development of CAA. Among these nontraditional lipid parameters, we demonstrated that LCI, AIP, AC, CR-I, and CR-II were independent risk factors for CAA. Patients in the T3 group had a 6.381-fold higher risk of CAA than those in the lowest AIP group. Additionally, the relationship between AIP and odds ratio appeared to be J-shaped. These findings revealed that AIP might be a promising novel biomarker for predicting CAA risk in patients who underwent coronary angiography.

Coronary angiography is considered the gold standard for the evaluation and diagnosis of CAA [23]. However, it is easy to miss because most CAA patients are asymptomatic. Some patients present with angina, while some patients with CAA may present with acute myocardial infarction as the first manifestation [24]. Unfortunately, little information is available from clinical studies on CAA. Therefore, the early identification and management of CAA is of great significance and remains a substantial challenge.

Previous studies have reported that men are more vulnerable to CAA than are women [25]. Similarly, in the present study, 72.5% of the CAA patients were men. In accordance with previous reports, the RCA was the most affected artery (42.50%), followed by the LAD (38.75%) and LCX (17.50%) [26]. The incidence of CAA is low, but it is associated with a poor prognosis [27].

In terms of pathology, atherosclerosis is the one of the most common causes contributing to CAA [28, 29]. It has been acknowledged that the deposition of lipids results in a disordered arrangement of the intima and media, as well as the devastation of elastic components [30]. Lipid parameters are closely associated with the development of CAA. CAA are frequently found in patients with familial hypercholesterolemia. Dysregulated levels of HLD-C and LDL-C are also involved in the development of CAA [31]. Nontraditional lipid parameters have received increasing attention in recent years because they can better reflect the interactions between these lipid indicators. Although the implication of nontraditional lipid parameters has been established in cardiovascular diseases, their role in CAA remains largely undetermined. A Previous study by our team found that TG (OR =  3.736, 95% CI =  2.028–6.883, p < 0.001) and LDL-C/HDL-C (OR =  1.569, 95% CI =  1.229–2.419, p =  0.026) were independently associated with CAA risk. Furthermore, TG has the highest predictive value for CAA [32]. The present study comprehensively explored the relationship between traditional lipid parameters, nontraditional lipid parameters, and CAA. Notably, TG is the sole traditional lipid indicator independently associated with CAA, with a predictive potential similar to that of nontraditional parameters. Furthermore, it indicated that AIP contributed to the early identification of CAA among patients who underwent coronary angiography with a higher predictive value than TG.

AIP, calculated as the logarithm of the plasma TG/HDL ratio, was first proposed 20 years ago as an outstanding biomarker of atherosclerosis. Importantly, it was strongly correlated with the particle size of lipoprotein [33]. Significant correlation was found between AIP and cardiovascular risk factors [34, 35]. Of note, increasing large-sample research has shown that AIP is closely associated with cardiovascular risk and poor prognosis compared with traditional cardiovascular risk factors in different populations [36, 37]. Another prospective cohort study enrolling 3820 residents revealed that AIP was associated with the risk of adverse cardiovascular and cerebrovascular events [38]. The present study, for the first time, proposes that the AIP has an outstanding predictive ability for CAA. The possible underlying mechanisms of AIP leading to CAA are as follows: AIP is positively associated with the particle size and fractional esterification rate of lipoproteins, especially small dense LDL (sdLDL) [17]. sdLDL is more vulnerable to oxidative stress injury, is easily converted into oxidized LDL, triggers inflammatory responses, and contributes to atherosclerosis [17]. In addition, AIP reflects cholesterol transport, and excessive TG accumulation in the endothelium can lead to atherosclerosis [39].

The current study provides several notable contributions to the field First, we elucidated the association between nontraditional lipid parameters and CAA. Second, this study is the first to reveal a J-shaped relationship between the AIP and CAA risk. Third, this study identified the optimal cutoff value of AIP at 0.105 for CAA risk assessment. However, this study has several limitations need to be acknowledged. This was a small-sample, single-center study; therefore, the research findings of the study need to be confirmed in large-sample prospective studies. Secondly, our findings are more applicable to individuals who have not used lipid-modifying drugs to avoid the potential impact of these medications on AIP and other lipid-related parameters. Additionally, the underlying mechanisms of how AIP contributes to CAA remain unexplored, and some confounding factors were not considered in the present study.

Conclusion

TG levels and several nontraditional lipid parameters showed significant positive associations with CAA risk, among which AIP exhibited the highest correlation. AIP showed superior predictive potential for CAA compared with other parameters and is expected to serve as a simple and promising biomarker for identifying CAA in clinical practice.

Author contributions

QYH and YGL designed the study; QYH, QY, and WD collected and analyzed the data; and QYH wrote the manuscript. QY, YGL, and WD provided patient samples, and QYH and TCC provided funding and edited the manuscript. All the authors have read and approved the final manuscript.

Funding

This study was supported by the Natural Science Foundation of Jiangsu Province (No. BK20241682), National Natural Science Foundation of China (No.82170433), and the Research Talent Project of Zhongda Hospital affiliated with Southeast University (No. CZXM-GSP-RC13).

Data availability

The data used to support the findings of this study, as described in the research article, are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Zhongda Hospital, which is affiliated with Southeast University (No:2023ZDSYLL264-P01). Written informed consent was obtained from all the patients.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.