Association Between Low-Density Lipoprotein-Cholesterol Level and Risk of Intracranial Atherosclerotic Stenosis

Results From the APAC Study

Abstract

Background: 

It is unknown whether plasma low-density lipoprotein-cholesterol level (LDL-C) can predict the occurrence of intracranial atherosclerotic stenosis (ICAS), especially asymptomatic (A)ICAS. This study investigated whether LDL-C level is independently associated with AICAS.

Methods: 

The Asymptomatic Polyvascular Abnormalities Community study is an ongoing community-based, prospective, long-term follow-up observational study with 3387 participants. AICAS was diagnosed by transcranial Doppler ultrasonography. The participants were divided into 3 groups based on LDL-C level. Cox regression was used to evaluate the association between LDL-C level and incidence of AICAS.

Results: 

During 2 years of follow-up, 9.98% of participants were diagnosed with AICAS. The incidence of AICAS (person-years with 95% CI) was 4.99% (4.48%-5.50%). AICAS incidence did not increase with increasing LDL-C level. Compared with the <2.6 mmol/l subgroup, the incidence of asymptomatic ICAS was not significantly higher in the 2.6 to 3.4 and >3.4 mmol/l subgroups after adjusting for confounding factors (hazard ratio=0.95, 95% CI: 0.86-1.03 and hazard ratio=0.96, 95% CI: 0.84-1.10, respectively).

Conclusions: 

LDL-C is not an independent predictor of AICAS incidence in the Chinese population.

BACKGROUND

Intracranial atherosclerotic stenosis (ICAS) is a major cause of ischemic stroke and transient ischemic attack (TIA).^1,2 ICAS is more common among Asians than in Western populations.³ Hospital-based study samples showed that the prevalence of ICAS in Asian countries is 9% to 65%, Brazil is 39%, Europe and the United States is 10% to 16%. The prevalence of ICAS depends on the characteristics of the study population and the diagnostic method used.⁴ In China, 30% to 40% patients with ischemic stroke and over 50% with TIA have ICAS;⁵ identifying risk factors of asymptomatic (A)ICAS and early implementation of interventions can reduce the incidence of these events.

The newly proposed conceptual framework for addressing residual atherosclerotic cardiovascular diseases risk in the era of precision medicine propose 5 broad residual risk targets: lipoproteins, inflammation, metabolism, platelets, and coagulation.⁶ Among them, low-density lipoprotein cholesterol (LDL-C) is the main etiological risk factor of atherosclerotic cardiovascular disease, including ischemic stroke caused by ICAS.⁷ Therefore, lowering LDL level has become an important risk factor for preventing recurrence and improving prognosis of patients with ischemic stroke.

We previously found in a cross-sectional study that LDL-C level was not associated with the occurrence of ICAS.⁸ However, the study design did not allow examination of the contribution of LDL-C to the development of ICAS. The current study investigated whether serum LDL-C is associated with risk of AICAS in the Chinese population.

METHODS

Study Population

The study enrolled participants from the community-based, prospective, long-term follow-up observational Asymptomatic Polyvascular Abnormalities Community (APAC) study investigating the epidemiology of asymptomatic polyvascular abnormalities in Chinese adults since June 2010 and is part of Kailuan study, which has been previously reported.^8–10 A total of 5440 participants (3264 men; mean age: 55.2 y) were eligible and were enrolled in the APAC study. Participants who underwent routine medical examination every 2 years were followed up through face-to-face interviews with hospital physicians and nurses. In the baseline survey and biennial routine medical examination, participants underwent extensive clinical examination, laboratory testing, and transcranial Doppler (TCD) examination. In the current study, we used 2012 follow-up data, and the participants were excluded for the following reasons: were missing TCD during follow-up (N=1233); had ICAS at baseline (N=698); had taken lipid-lowering agents (N=75); experienced TIA or stroke during the follow-up (N=32); and had incomplete LDL-C data (N=15). Ultimately, 3387 participants were included (Fig. 1). Written, informed consent was obtained from all participants. The study was performed in accordance with the tenets of the 1964 Helsinki Declaration and was approved by the Ethics Committees of the Kailuan General Hospital and Beijing Tiantan Hospital.

FIGURE 1: Flowchart of participant selection. APAC indicates Asymptomatic Polyvascular Abnormalities Community; ICAS, intracranial atherosclerotic stenosis; TCD, transcranial Doppler.

Measurement of Indicators

Details on the measurement of physiological indicators have been previously described.⁸ Hypertension was defined as systolic blood pressure ≥140 mm Hg, diastolic pressure ≥90 mm Hg, personal history of hypertension, or currently taking antihypertensive medication prescribed by a physician. Diabetes mellitus was diagnosed if the participant was undergoing treatment with insulin or oral hypoglycemic agents, fasting blood glucose level was ≥126 mg/dl, or if there was a personal history of diabetes mellitus. Smoking status was classified as nonsmoking or smoking according to self-reported information. Body mass index was calculated as body weight (kg) divided by the squared height (m²).

Blood samples were processed and analyzed using an autoanalyzer (Hitachi 747; Hitachi, Tokyo, Japan) at the central laboratory of Kailuan General Hospital. Serum total cholesterol (TC) level was measured using the endpoint test method. High-density lipoprotein cholesterol (HDL-C) and LDL-C were measured with the direct test method. Triglyceride level was measured with the glycerol phosphate oxidase method.

Two experienced neurologists performed TCD using a portable device (EME Companion; Nicolet, Madison, WI, USA). The diagnostic criteria for ICAS were a peak systolic flow velocity >140 cm/s for the middle cerebral artery, >120 cm/s for the anterior cerebral artery, >100 cm/s for the posterior cerebral and vertebra-basilar arteries, and >120 cm/s for the siphon internal carotid artery. Age, disturbance in echo frequency, turbulence, and whether the abnormal velocity was segmental were also taken into consideration for ICAS diagnosis.¹¹

Statistical Analysis

All statistical analyses were performed using SAS software (SAS Institute, Cary, NC, USA). Participants were grouped as follows according to serum LDL-C level: group 1, <2.6 mmol/l; group 2, 2.6 to 3.4 mmol/l; and group 3, ≥3.4 mmol/l.¹² Analysis of variance was used to analyze continuous variables and the χ² test was used to compare categorical variables among the 3 groups. Age-adjusted and sex-adjusted or multivariable-adjusted hazard ratio (HR) and 95% CI were calculated using Cox regression models. Age, sex, hypertension, diabetes, current smoking status, and TC level were adjusted for in the multivariable-adjusted model. In addition, some baseline characteristics including age, sex, hypertension, diabetes, and smoking status were evaluated to assess whether the interaction between these variables and the relationship between LDL-C level and ICAS incidence was significant. The threshold for statistical significance was a P value <0.05.

RESULTS

Incidence of ICAS

A total of 3387 participants (1468 women and 1919 men) were included in the study. The median (interquartile range) age of the participants was 51.65 (45.33 to 59.83) years. During the follow-up, 338 patients developed ICAS based on TCD results. The incidence (person-years with 95% CI) of AICAS was 4.99% (4.48%-5.50%) for all participants, 3.83% (3.23%-4.42%) for men, and 6.50% (5.64%-7.36%) for women.

Baseline Characteristics

The baseline characteristics of the study population are presented in Table 1. Participants with AICAS were older and had higher rates of hypertension and diabetes than those without ICAS. There are more females and smokers among participants with AICAS. The median (interquartile range) LDL-C level was 2.66 (2.33-3.10) mmol/l among participants with ICAS and 2.57 (2.14-3.01) mmol/l among those without ICAS (P=0.042).

TABLE 1 - Baseline Characteristics of Study Participants

Characteristic	Total (N=3387)	AICAS (N=338)	Without AICAS (N=3049)	P
Female	1468 (43.34)	191 (56.51)	1277 (41.88)	<0.001
Age, years	51.65 (45.33-59.83)	52.42 (45.93-58.65)	50.87 (45.04-58.65)	0.006
Smoking	1229 (36.29)	99 (29.29)	1130 (37.06)	0.005
Hypertension	732 (21.61)	91 (26.92)	641 (21.02)	0.012
Diabetes	219 (6.47)	34 (10.06)	185 (6.07)	0.005
BMI, kg/m2	24.69 (22.64-26.96)	24.49 (22.26-26.85)	24.69 (22.67-26.81)	0.609
Glucose, mmol/l	5.20 (4.82-5.72)	5.10 (4.75-5.70)	5.17 (4.81-5.70)	0.291
LDL-C, mmol/l	2.59 (2.15-3.04)	2.66 (2.23-3.10)	2.57 (2.14-3.01)	0.042
TC, mmol/l	4.90 (4.34-5.59)	5.11 (4.46-5.78)	4.90 (4.34-5.55)	0.001
HDL-C, mmol/l	1.57 (1.31-1.90)	1.62 (1.31-1.95)	1.57 (1.31-1.90)	0.334
TG, mmol/l	1.29 (0.92-1.91)	1.30 (0.94-1.86)	1.28 (0.91-1.90)	0.865

Values are shown as n (%) or median (interquartile range).

AICAS indicates asymptomatic intracranial arterial stenosis; BMI, body mass index; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglycerides.

We examined the association between baseline lipid level and incidence of ICAS (Table 2). Compared with the <2.6 mmol/l subgroup, the incidence of asymptomatic ICAS was not significantly higher in the 2.6 to 3.4 and >3.4 mmol/l subgroups after adjusting for confounding factors (HR=0.95, 95% CI: 0.86-1.03 and HR=0.96, 95% CI: 0.84-1.10, respectively). The presence or absence of age, sex, and other potential indicators did not affect the relationship between LDL-C level and incidence of AICAS (Table 3).

TABLE 2 - Association Between Baseline LDL-C Level and Risk of AICAS

LDL-C, mmol/l	Cases (N)	Unadjusted	Model 1*	Model 2†
<2.6	2311	Reference	Reference	Reference
2.6-3.4	1705	0.95 (0.86-1.03)	0.93 (0.86-0.99)	0.95 (0.86-1.03)
≥3.4	581	0.93 (0.84-1.04)	0.90 (0.81-1.01)	0.96 (0.84-1.10)
LDL-C (continuous measure)	3387	0.96 (0.91-1.00)	0.93 (0.89-0.98)	0.95 (0.89-1.01)

^*Model 1: adjusted for age, sex.

^†Model 2: adjusted for age, sex, smoking, hypertension, diabetes, and total cholesterol.

AICAS indicates asymptomatic intracranial arterial stenosis; HR, hazard ratio; LDL-C, low-density lipoprotein cholesterol.

TABLE 3 - Association Between Baseline LDL-C Level and Risk of AICAS, Stratified by Selected Risk Factors

	LDL-C, mmol/l
Variable	<2.6	2.6-3.4	≥3.4	P for interaction
Sex
Male	Reference	0.88 (0.79-1.10)	0.83 (0.70-1.12)	0.061
Female	Reference	1.07 (0.95-1.22)	1.20 (0.97-1.47)
Age, years
≥60	Reference	0.97 (0.89-1.06)	0.94 (0.81-1.09)	0.118
<60	Reference	0.86 (0.73-1.02)	1.07 (0.81-1.42)
Hypertension
No	Reference	0.97 (0.89-1.06)	0.96 (0.82-1.11)	0.562
Yes	Reference	0.88 (0.74-1.03)	0.94 (0.69-1.26)
Diabetes
No	Reference	0.96 (0.88-1.04)	0.96 (0.83-1.10)	0.551
Yes	Reference	0.85 (0.63-1.15)	0.92 (0.55-1.55)
Smoking
No	Reference	1.00 (0.90-1.10)	1.04 (0.88-1.24)	0.129
Yes	Reference	0.90 (0.79-1.02)	0.86 (0.70-1.06)

Multivariable-adjusted hazard ratio adjusted for age, sex, smoking, hypertension, diabetes, and total cholesterol (Cox proportional hazard model).

AICAS indicates asymptomatic intracranial arterial stenosis; LDL-C, low-density lipoprotein cholesterol.

DISCUSSION

The results of this study show that LDL-C is not an independent predictor of AICAS incidence in the Chinese population. ICAS is far more prevalent among Asians and Africans and in individuals of Hispanic origin.¹³ Other studies conducted in the Chinese population have shown that 5.9% to 6.9% of subjects past their fifth decade of life had asymptomatic intracranial arterial disease by TCD.^11,14 The prevalence of AICAS by TCD was 12.6% in a cross-sectional study from Hong Kong;¹⁵ and 14.7% of subjects referred to a neurology clinic in Japan because of concern regarding risk of stroke had AICAS by magnetic resonance angiography.¹⁶

Hypertension, smoking, diabetes, and elevated LDL-C are potentially modifiable risk factors for ICAS;^14,17,18 nonmodifiable risk factors include race, age, certain angiotensin-converting enzyme polymorphisms, glutathione S-transferase omega-1 gene polymorphism, and increased plasma homocysteine level.¹³ The relationship between serum cholesterol and AICAS is controversial, with some studies reporting a positive correlation; previous studies in China showed that LDL-C,^15,19 non-HDL-C,⁸ and TC²⁰ were important risk factors for AICAS. However, a negative correlation has been reported by others, which is consistent with our results. A study conducted in Korea showed that LDL-C was not an independent Asymptomatic Korean Population ICAS risk factor,²¹ and a study conducted in Spain found no significant correlation between TC or LDL-C and AICAS.²²

Despite the rigorous study design, there were some limitations to our study. First, a poor TCD reading within the defined temporal window was defined as non-AICAS, which may have led to an underestimate of the incidence of ICAS. Second, 1233 participants were missing TCD during follow-up, which may have led to results migration. Third, although TCD is a noninvasive and convenient screening method for diagnosing ICAS, it is less accurate than digital subtraction angiography. Fourth, this cohort was a subgroup from the APAC study, so the findings might not be generalizable to other populations. Advanced studies are essential for further research. Nonetheless, this is the first large, prospective clinic trial to date that has investigated the relationship between LDL-C level and the occurrence of AICAS.

CONCLUSIONS

LDL-C is not an independent predictor of AICAS incidence in the Chinese population. However, more epidemiologic and experimental data are needed to confirm this finding.