LDL-C/HDL-C and carotid plaques in patients with coronary heart disease: A Chinese cohort study

Background: Evidence of the association the low-/high-density lipoprotein cholesterol ratio (LDL-C/HDL-C) with the carotid plaques remains limited. The purpose of this study is to examine the association between LDL-C/HDL-C and carotid plaques of coronary heart disease (CHD), and to study what extent a healthy lifestyle reduces the risk of carotid plaques. Methods: In this large-scale and multi-center retrospective study, a total of 9,426 participants to explore the association between LDL-C/HDL-C and carotid plaques. No smoking and no drinking were considered healthy lifestyle. Generalised estimating equation models and conditional logistic regressions were used in statistical analyses. Results: In all the 9,426 participants, there were 6,989 (74.15%) patients having identied carotid plaques. High levels of LDL-C/HDL-C had a higher risk of carotid plaques than other lipid variables (OR:1.63; 95%CI:1.43-1.86). In stratied analyses by LDL-C/HDL-C triplet, participants in the LDL-C/HDL-C ((cid:0) 3mmol/L) group had a higher risk of carotid plaques compared to other two groups. Compared with the unfavourable lifestyle, intermediate lifestyle or favourable lifestyle was associated with a signicant 30% or 67% decrease in carotid plaques risk among patients with the LDL-C/HDL-C((cid:0)3mmol/L) respectively. There were signicantly additive and multiplicative interactions between lifestyle and LDL-C/HDL-C on carotid plaques. Conclusion: Our ndings provide evidence that a high level of LDL-C/HDL-C can increase the risk of carotid plaques in patients with CHD. And adhering to a healthy lifestyle has additive benecial effects on reducing the risk of carotid plaques. test. systolic total cholesterol; triglycerides; lipoprotein cholesterol; low-density lipoprotein


Introduction
Cardiovascular diseases (CVD) is a class of diseases Whose morbidity and mortality are highest in the world, and CVD mortality is expected to remain the world's highest until 2030, with coronary heart disease (CHD) second only to stroke in CVD [1] . Coronary atherosclerosis (AS), as the basis of CHD, usually occurs at the same time as carotid atherosclerosis [2] .
Studies have shown that peripheral vascular atherosclerosis severely affects the most signi cantly prediction-related for deadliness and death of CVD. It has been reported that when the conventional lipid parameters triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and total cholesterol (TC) remain apparently normal, other lipids such as lipid ratios, including TC/HDL-C, LDL-C/HDL-C, TG/HDL-C, and non-HDL-C/HDL-C are the diagnosticate alternatives that have been shown to predict the risk of a cardiovascular event [3][4][5] . Studies have shown LDL-C that plays a key role in the pathogenesis of AS. However, clinical studies suggest that despite the treatment of reduced LDL-C, signi cant CVD events continue to occur, may be residual risks associated with lipid abnormalities, particularly dyslipidemia causing atherosclerosis, containing elevated TG and reduced Page 4/19 HDL-C [6,7] . Therefore, new goals need to be provided to complement the measures to prevent CVD. Recent studies have shown that LDL-C/HDL-C is a better indicator of AS than individual LDL-C/HDL-C [8,9] . Nevertheless, there is limited data on the association of LDL-C/HDL-C with carotid plaques in CHD.
The healthy lifestyle factors can prevent CVDs. Drinking and smoking, two very common and concurrent risk factors, are associated with a signi cant proportion of mortality of CVD [10] . Studies have shown that the need for a variety of health factors can effectively reduce the risk of complications of CVD [11] . To what extent to which a jointly healthy lifestyle can reduce the risk of CHD associated with carotid plaques is unknown.
Therefore, this study aims to explore the relationship between the clinical indicators of LDL-C/HDL-C and carotid plaques in CHD, and investigate to what extent a healthy lifestyle could mitigate the risk of CHD related to carotid plaques.

Participants
This large-scale and multi-center retrospective study included 107,301 patients with CHD who were hospitalized in the First A liated Hospital of Tianjin University of Traditional Chinese Medicine, the Second A liated Hospital of Tianjin University of Traditional Chinese Medicine, Tianjin Nankai Hospital, Tianjin Chest Hospital, and Tianjin Medical University General Hospital from January 1, 2014, to September 30, 2020. Information was obtained from the medical records in the Hospital Information System (HIS). Participants were excluded as followed: (1) age less than 35 or greater than 75 years; (2) oncological, infectious, or serious liver or renal disease; (3) lack of lipid data or carotid ultrasound measurements. Finally, a total of 9,426 participants were enrolled in the study. A ow chart of patient selection is shown in Fig. 1.

Lipid and carotid ultrasound assessment
According to the HIS, the levels of TC, TG, HDL-C, and LDL-C were measured using standard procedures.
The carotid ultrasound results were evaluated by trained and certi ed technicians. The carotid intimamedia thickness (CIMT) was de ned as the mean of the IMT of the right and left common carotid arteries. The number and echoproperties of carotid plaques were recorded by a professional physician based on the analysis of carotid color Doppler ultrasound results. The number of carotid plaques were classi ed as single (n = 1) and multiple (n ≥ 2). Echoproperties of carotid plaques were varied from hypoechoic, isoechoic, hyperechoic, and mixed.

Covariates
HIS was used to collect information about the demographic characteristics, and medical history of participants. Assessment of lifestyle factors on smoking and drinking from the personal history. We de ned smoking status and drinking as no and yes. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) was measured by trained physicians using an electronic device. In the study, we considered no smoking and no drinking healthy lifestyle factors. Participants were divided into three lifestyles based on smoking and drinking: Unfavourable was de ned as participants who both smoking and drinking, intermediate was de ned as those who smoking only or drinking only, and favourable was de ned as those who no smoking and no drinking.

Statistical analyses
Using the Kolmogorov-Smirnov tested the normality of the continuous variable distribution. Descriptive data was represented in quartile (interquartile) for continuous variables and number (proportion) for classi cation variables. In conditional logistic regressions, the odds ratio (OR) and 95% con dence interval (CI) were estimated for the association between LDL-C/HDL-C and the risk of carotid plaques formation, the number of carotid plaques, and carotid plaques echogenicity. Age, sex, DBP, SBP, Hb1Ac and other lipid were considered as potential confounders in the logistics analysis. Missing values on smoking (n = 19), drinking (n = 19), SBP (n = 30), DBP (n = 30), Hb1Ac (n = 765) and carotid plaques echogenicity (n = 104) were imputed by multiple interpolation method.
The joint impact of LDL-C/HDL-C and lifestyle on the risk of carotid plaques was assessed by creating virtual variables based on the combined exposure of both factors. The existence of additional interactions was examined by estimating the associated excess risk (RERI), attributable ratio (AP), and synergistic index (SI). Furthermore, we studied the multiplicative interaction by merging two variables in the same model and their cross-product terms. All statistical analyses were used by SAS 9.4 (SAS Institute, Cary, NC, USA) and SPSS 24.0 (IBM Corp, New York, NY, USA).

Baseline characteristics
All of the 9,426 participants, the average age of participants was 59.00 ± 5 years old, Males constituted 48.96% overall, and patients with carotid plaque accounted for 74.15%. The baseline characteristics of the participants were divided into three groups of LDL-C/HDL-C < 2mmol/L (T1), 2.15mmol/L ≤ LDL-C/HDL-C ≤ 3mmol/L (T2) and LDL-C/HDL-C > 3mmol/L (T3) according to the tridigits of LDL-C/HDL-C. LDL-C/HDL-C (T3) was more likely to be male and had higher SBP, DBP, HbA1c, TG, TC, LDL-C, TC/HDL-C, TG/HDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, and a higher percentage of patients with smoking, drinking, and carotid plaques compared with participants in LDL-C/HDL-C (T1). The odds of carotid plaques, the number of carotid plaques, and carotid plaques echogenicity also differed among the different LDL-C/HDL-C groups. (P-values 0.001). (Table 1)

Associations between univariate and the risk of carotid plaques
In comparison with patients without carotid plaque, male, age, SBP, HbA1c LDL-C, non-HDL-C-HDL-C, smoking, drinking, and elevation were observed associated with carotid plaques. All lipid variables were risk factors for carotid plaques formation. HDL-C might be a protective factor associated with the number and echo properties of carotid plaques. LDL-C/HDL-C remained the highest risk associated with the number and echo properties of carotid plaques. (Fig. 2)

Association between LDL-C/HDL-C and carotid plaques
Three logistic regression models were constructed to assess the impact of LDL-C/HDL-C on carotid plaque. (Fig. 3 In the unadjusted and further adjustment model, logistic regression suggested that the carotid plaques risk of LDL-C/HDL-C (T3) was 1.67, 1.75 and 1.18 fold that of the LDL-C/HDL-C (T1). In the further analysis, convert continuous LDL-C/HDL-C to classi cation variables (tridigits), and the P for trend of the LDL-C/HDL-C with carotid plaques in the unadjusted or adjusted model was consistent with the results when the LDL-C/HDL-C served as a continuous variable. (P 0.001 or P 0.05) Furthermore, We only adjusted sex, age, SBP, DBP and HbA1c to avoid excessive adjusted, multivariate logistic regression analysis showed that TC, LDL-C, non-HDL-C, TC/HDL-C, LDL-C/HDL-C and non-HDL-C/HDL-C except TG and TG/HDL-C, HDL-C were the risk factors for individual carotid plaques. And all lipid variables except HDL-C were risk factors for multiple carotid plaques. (ESM Table 1) In further analysis, All other lipid variables except TC and TG/HDL-Cwere risk factors for plaques of hypoechoic, Isoechoic. All lipid variables were risk factors for plaques of hyperechoic and Mixture. HDL-C might be a protective factor associated the number and echo properties of carotid plaques. LDL-C/HDL-C remained the highest risk associated with the number and echo properties of carotid plaques. (ESM Table 2) Compared with no carotid plaque: * P < 0.05, ** P < 0.01.

Association between lifestyle-related factors and carotid plaques
Association between lifestyle factors and carotid plaques after adjusted models, being a no smoking and no drinking were interrelated with a reduced risk of carotid plaques.

Discussion
In this large-scale and multi-center retrospective database of CHD, our ndings provide evidence that a high level and independent positive correlation between LDL-C/HDL-C and carotid plaques. After adjusting other covariates, the participants increased carotid plaques by 1.63 fold. The classi cation variables also showed that LDL-C/HDL-C (T3) had a higher risk of carotid plaques that of LDL-C/HDL-C (T1). Moreover, Patients with LDL-C/HDL-C (T3) who reported maintained a healthy lifestyle including no smoking and no drinking, and had a signi cantly lower risk of carotid plaques than patients with both smoking and drinking.
Studies show that coronary atherosclerosis and carotid stenosis are closely related [14] . Most acute cardiovascular events are attributed to vulnerable carotid atherosclerotic plaque rupture and secondary thrombosis [15][16] . Dyslipidemia is the main risk factor for atherosclerosis [17][18] . LDL-C is considered a major cardiovascular risk factor, with a denser reduction in LDL-C more associated with reduced total and cardiovascular mortality in patients at baseline [19] . Studies suggested a reverse relationship between the concentration of HDL-C in plasma and CHD. Studies showed that HDL-C neither has any protective function nor re ects the functionality of HDL-C. Therefore HDL-C protection is still controversial [20] . Studies proved that the LDL-C/HDL-C was closely related to the onset of CHD and the progression of atherosclerosis [21][22][23] . A prospective study showed that the LDL-C/HDL-C predicted the progress of CIMT better than HDL-C or LDL-C alone [24] . These results are consistent with our results, which further investigate the relationship between LDL-C/HDL-C and the risk of carotid plaques formation, the number of carotid plaques, and carotid plaques echogenicity. In our study, LDL-C/HDL-C > 3mmol/L is considered a risk factor for carotid plaques in CHD.
Atherosclerotic plaques are rich in lipids, and the lipid composition is thought to affect the stability of atherosclerotic plaques. Furthermore, the lipid content strongly determines plaque impnerability and the rate of stenosis [25] . Recent studies have found that neovascularization in plaques has become an important marker for evaluating plaque stability [26,27] . A relationship between neovascularization in plaques and different types of plaque echogenicity shown by conventional ultrasound has been reported, and neovascularization in isoechoic plaques and hypoechoic brous lipid plaques is more obvious than that in calci ed plaques [28] . Homogeneous hypoechoic plaques are mostly unstable plaques, rich in lipids, and prone to in ammatory reactions, which can promote the production of neovascularization in plaques. Neovascularization in hyperechoic plaques is rare [29] . In our study, an elevated LDL-C/HDL-C ratio was signi cantly correlation with the occurrence of hypoechoic and isoechoic plaques. A possible explanation may be that the characteristics of unstable plaques are associated with denser LDL subgroup typing. The greater the LDL density, the easier to reach the lower endothelium, where they are more likely to be oxidized. Furthermore, high density LDL particles have higher a nity for proteinosan, which extends its stay on the wall and inducthe development of atherosclerosis. In addition, the size of HDL-C particles generally decreases with the LDL-C/HDL-C, suggesting the blockade of HDL maturity, which may be responsible for the progression of atherosclerosis [30,31] .
Although studies have shown that moderate drinking in the literature is associated with reducing the risk of CHD. A case-control study showed that changes in alcohol consumption during life could distort the relationship between real alcohol and CHD [32] . A case-cohort study showed that drinking was inversely associated with non-fatal CHD risk but positively associated with the risk of different stroke subtypes. A multivariable Mendelian randomization study found smoking was a risk factor for CVD even after adjusting for drinking [33][34] . Accordance with our results that lifestyle interventions, being no smoking and no drinking may greatly reduce the risk of carotid plaques in CHD.

Strengths And Limitations
This study has several advantages and limitations. Firstly, The ultrasound may not be as accurate as high-resolution magnetic resonance imaging (MRI) or computed tomography (CT) in assessing the presence of a plaque. However, the safety, noninvasive nature of ultrasound should be acknowledged. Furthmore, BMI is an important confounding factor in CHD and carotid plaques. Because much BMI data were missing in this study, BMI was not included in the model. Finally, because this study is a multicenter, large-scale study, no causal relationship between LDL-C/HDL-C and carotid plaques, which requires further prospective research.

Conclusion
LDL-C/HDL-C is an independent risk factor for the occurrence of carotid plaques in CHD patients, which is a higher risk factor than other lipid variables. In clinical treatment, the impact of a high level of LDL-C/HDL-C should be considered. In addition to the basic lowering of blood lipid treatment, attention should be paid to improve patients' lifestyle, strengthen self-management, promote a healthy lifestyle, and reduce the risk of carotid plaque in CHD.
Declarations CY, SG and RY took responsibility for the study protocol and statistical analysis; ZLand QC analyzed the data together and drafting the article; YL, XC, SW, YH, MH, YL, XX conducted data collection; YX, LL and YZ revised the article critically. All authors revised the article for important intellectual content and approved the article. The nal version of the manuscript has been approved by all authors.

Financial support
This work was supported by the National Basic Research Program of China (973 project, grant numbers: 2014CB542902) Availability of data and materials The datasets used in the present study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate

Consent for publication
Not applicable.

Declaration of con ict of interest
The authors declare no competing interests.

Figure 1
Flow chart of patient recruitment.

Figure 2
Associations between univariate and the risk of carotid plaques.