Paradoxical Long-Term Impact Between Serum Apolipoprotein E and High-Density Lipoprotein Cholesterol in Patients Undergoing Percutaneous Coronary Intervention

Abstract

Aim: Apolipoprotein E (ApoE) strongly affects arteriosclerosis but has atheroprotective effects in combination with high-density lipoprotein cholesterol (HDL-C). The impact of the quantitative relationship between serum ApoE and HDL-C levels in patients with coronary artery disease (CAD) remains unclear.

Methods: A total of 3632 consecutive patients who underwent their first intervention between 2000 and 2016 were included. They were categorized into normal and abnormal HDL-C groups based on the normal HDL-C value, and each group was subdivided into high and low ApoE subgroups based on the group-specific median ApoE value. We evaluated the incidence of major adverse cardiac and cerebrovascular events (MACCE), including cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, and all-cause death

Results: During a 6.4-year follow-up, 419 patients developed MACCE and 570 patients died. The interaction term between ApoE levels and HDL-C status in MACCE and all-cause death proved to be statistically significant. Kaplan–Meier analysis revealed that the cumulative incidence of MACCE was significantly higher for elevated pre-procedural ApoE levels than for reduced preprocedural ApoE levels in the normal HDL-C group. Conversely, the cumulative incidence of MACCE was significantly higher for reduced pre-procedural ApoE levels than for elevated pre-procedural ApoE levels in the abnormal HDL-C group. After adjustment for important covariates, multivariable Cox hazard analysis revealed that the serum ApoE level was a strongly independent predictor of MACCE; this was inversely related in both groups.

Conclusions: Serum ApoE levels may have a paradoxical impact on the future cardiovascular risk depending on the HDL-C status in patients with CAD.

Introduction

High-density lipoprotein cholesterol (HDL-C) is inversely associated with an increased risk of atherosclerotic cardiovascular disease¹⁾. HDL particles can counteract atherogenesis through mediation of cholesterol efflux via macrophages, maintenance of the endothelial function, protection against the oxidation of low-density lipoproteins, anti-inflammation, and immunomodulation^2-7). Above all, an elevated cholesterol efflux capacity is inversely associated with the development of coronary artery disease (CAD) and the occurrence of cardiovascular events^{8, 9)}. The main constituent proteins of HDL are apolipoprotein A-I (ApoA-I) and apolipoprotein A-II. However, large HDL comprises apolipoprotein E (ApoE)-rich particles and is thus called ApoE-containing HDL. These particles strongly contribute to the cholesterol efflux capacity of HDL^{10, 11)}. A previous study has reported that an increase in the ApoE-containing HDL-C to HDL-C ratio contributes to a reduction in the risk of coronary heart disease¹²⁾. However, ApoE usually affects the development of cardiovascular events not only qualitatively (at genotype level) but also quantitatively (at serum level) and has strongly positive correlations with triglyceride-rich lipoproteins (TRLs)¹³⁾. Nevertheless, high TRL levels are paradoxically associated with enhanced macrophage-mediated cholesterol efflux in individuals with low HDL-C levels, but not in those with normal HDL-C levels¹⁴⁾. We assumed that if contradictory interactions were verified between serum ApoE and HDL-C levels, the quantitative relationship between them may paradoxically affect the incidence of cardiovascular events in clinical practice. Thus, we aimed to investigate the long-term prognostic impact of serum ApoE levels according to the HDL-C status in patients with CAD who underwent percutaneous coronary intervention (PCI).

Methods

Study Population

This observational and retrospective cohort study was performed at our institution. We screened 4542 consecutive patients with CAD who underwent their first PCI for coronary artery lesions between January 2000 and December 2016. The exclusion criteria were as follows: (1) patients ＜20 years or ＞85 years of age, (2) patients with familial hypercholesterolemia, (3) patients with severely elevated triglyceride (TG) levels, (4) patients receiving hemodialysis, and (5) patients with missing data on ApoE.

CAD was categorized as either the acute coronary syndrome (ACS) or the chronic coronary syndrome (CCS). ACS included unstable angina and both ST-segment and non-ST-segment elevation myocardial infarction (MI). It was diagnosed on the basis of electrocardiographic abnormalities, positivity for myocardial deviant enzymes or cardiac troponin T, echocardiographic findings suggestive of an impaired left ventricular systolic function, and signs and symptoms of cardiac ischemia. Patients with ACS were hospitalized urgently, and an emergency PCI was performed. Conversely, CCS was diagnosed on the basis of the following clinical scenarios: classical history of anginal symptoms, new onset of heart failure or left ventricular dysfunction suggested CAD, and detection of CAD at screening even if asymptomatic. Patients with CCS were initially treated by lifestyle management (smoking cessation, healthy diet, physical activity, and healthy weight) and pharmacological management. Pharmacological management included treatment of ischemia through anti-ischemic drugs, such as nitrates, β-blockers, calcium channel blockers, or nicorandil, and prevention of ischemic events through antiplatelet drugs, anticoagulant drugs, statins, other lipid-lowering drugs, and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers.

This study was approved by the ethics committee of our institution, and all screened patients provided written informed consent. The investigation conformed to the principles outlined in the Declaration of Helsinki¹⁵⁾.

Data Collection and Definitions

Data on patient characteristics were collected from an institutional database. Blood samples were collected before the intervention in the morning after an overnight fast, and all blood tests were performed in the same laboratory. For lipid profiling, total cholesterol, HDL-C, low-density lipoprotein cholesterol (LDL-C), and TG levels were assayed at our institution using LABOSPECT 008α (Hitachi, Ltd., Tokyo, Japan). In addition, lipoprotein(a) [Lp(a)] levels were assayed using a latex-enhanced immunoturbidimetric assay, while ApoA-I, apolipoprotein B (ApoB), and ApoE were simultaneously measured through immunoassay on the same device. Estimated remnant cholesterol was calculated using the formula of non-HDL-C minus LDL-C, instead of direct measurement¹⁶⁾.

Patients with a blood pressure ＞140/90 mmHg or those receiving antihypertensive drugs were regarded as being hypertensive¹⁷⁾. Patients were considered to have dyslipidemia if they had a TG level ≥ 150 mg/dL, LDL-C level ≥ 140 mg/dL, or HDL-C level ＜40 mg/dL or if they were receiving lipid-lowering therapy¹⁸⁾. Patients were considered to have diabetes mellitus if their hemoglobin A1c level was ≥ 6.5% or if they received oral hypoglycemic agents or insulin injections¹⁹⁾. Chronic kidney disease (CKD) was defined as an estimated glomerular filtration rate (eGFR) of ＜60 mL/min/1.73 m², as calculated by the Modification of the Diet in Renal Disease equation, which was modified with a Japanese coefficient using the baseline serum creatinine²⁰⁾. Patients were classified as anemic based on their hemoglobin levels using the World Health Organization definition (＜12.0 g/dL and ＜13.0 g/dL in women and men, respectively)²¹⁾. A positive family history for cardiovascular disease was defined as the presence of any first-degree relative with premature cardiovascular disease (age ＜55 years and ＜65 years for men and women, respectively)²²⁾.

HDL-C levels of ≥ 40 mg/dL are considered as normal; therefore, we categorized patients with HDL-C levels ≥ 40 and ＜40 mg/dL into the normal HDL-C group and the abnormal HDL-C group, respectively¹⁷⁾.

Study Endpoint

The study endpoint comprised major adverse cardiac and cerebrovascular events (MACCE) and all-cause death. MACCE was defined as a composite of cardiovascular death, non-fatal MI, and non-fatal stroke. Cardiovascular death was defined as death resulting from acute MI, sudden cardiac death, heart failure, stroke, cardiovascular procedures, cardiovascular hemorrhage, and other cardiovascular causes. MI included ST-segment elevation MI (ST elevation, abnormal biomarkers) and non-ST elevation MI (no ST elevation, abnormal biomarkers)^{23, 24)}.

Clinical follow-up data were collected from the patients’ medical records or by contacting the patients or their families if they had not been followed up at our institution after the intervention. Information on the circumstances and date of death was obtained from the families of patients who died at home, and details of events associated with the cause of death were supplied by staff of other hospitals or clinics to which the patient had been admitted. Blinded investigators collected all data.

Statistical Analysis

Categorical data are presented as numbers and percentages and were compared using the chi-square test. Continuous variables are expressed as mean±standard deviation or as median and interquartile range and were compared using a one-way analysis of variance or the Kruskal–Wallis test. The Kolmogorov–Smirnov test was used to examine whether scores were likely to follow a certain distribution in all patients. In case of p＜0.05, the variable was not considered to be distributed normally. A univariate analysis of the relationship between the ApoE and HDL-C levels was performed using Pearson’s correlation analysis. Furthermore, to determine whether the association between the ApoE levels and the clinical outcomes was dependent on the HDL-C status, the interaction between the ApoE levels (presented as a continuous variable) and the HDL-C status (dichotomized into normal and abnormal HDL-C) was analyzed by a regression model. Kaplan–Meier analyses were performed to determine the cumulative incidences of MACCE and all-cause death. In these analyses, comparisons were performed between two groups categorized on the basis of the normal HDL-C value of 40 mg/dL, and these groups were further subdivided into two additional groups on the basis of the median group-specific ApoE value and then compared. The intergroup differences were assessed using a log-rank test. To assess whether the serum ApoE level was a predictor of clinical outcomes, we performed a multivariable Cox regression analysis after adjustment for important covariates. During the regression analysis, multi-collinearity of lipid profiles was evaluated using pairwise correlation coefficients between ApoE and variance inflation factor, due to considering confounding factors. The lipid profiles were added as covariates for regression analysis after it was proved that they were not a multi-collinearity with ApoE.

All probabilities were expressed as two-tailed values, with statistical significance inferred at p＜0.05. All confidence intervals (CIs) were computed at the 95% level. All data were analyzed using JMP version 14.2 for Macintosh (SAS Institute, Cary, NC, USA).

Results

Baseline Characteristics of Study Population

We studied a total of 3632 consecutive patients with CAD who underwent their first intervention. We excluded 101 patients ＜20 years or ＞85 years of age, 53 with familial hypercholesterolemia, 17 with severely elevated TG levels (TG ≥ 500 mg/dL), 251 undergoing hemodialysis, and 765 with missing data on ApoE. Based on the normal range of HDL-C ≥ 40 mg/dL, 2220 patients (61.1%) were allocated to the normal HDL-C group, and 1412 (38.9%) were allocated to the abnormal HDL-C group. Additionally, each of these two groups was further divided into two groups based on the median pre-procedural serum ApoE value. Thus, as shown in Fig.1, the following four subgroups were obtained: the normal HDL-C/high ApoE group comprised patients with HDL-C ≥ 40 mg/dL and ApoE ≥ 4.0 mg/dL (n=1069, 29.4%), the normal HDL-C/low ApoE group with HDL-C ≥ 40 mg/dL and ApoE ＜4.0 mg/dL (n=1151, 31.7%), the abnormal HDL-C/high ApoE group with HDL-C ＜40 mg/dL and ApoE ≥ 3.8 mg/dL (n=698, 19.2%), and the abnormal HDL-C/low ApoE group with HDL-C ＜40 mg/dL and ApoE ＜3.8 mg/dL (n =714, 19.7%).

Fig.1. Study flowchart

Among 4542 patients, we excluded patients under 20 or over 85 years old, patients with familial hypercholesterolemia, patients with severely elevated triglyceride levels, patients undergoing hemodialysis, and patients with lacking data of ApoE at the time of first intervention. A total of 3632 patients were categorized into the normal and abnormal HDL-C groups according to the normal HDL-C value of ≥ 40 mg/dL, and each group was further subdivided into high and low ApoE subgroups based on the group-specific median ApoE value.

Normal HDL-C/high ApoE group, patients with HDL-C ≥ 40 mg/dL and ApoE ≥ 4.0 mg/dL (n=1069, 29.4%).

Normal HDL-C/low ApoE group, patients with HDL-C ≥ 40 mg/dL and ApoE ＜4.0 mg/dL (n=1151, 31.7%).

Abnormal HDL-C/high ApoE group, patients with HDL-C ＜40 mg/dL and ApoE ≥ 3.8 mg/dL (n=698, 19.2%).

Abnormal HDL-C/low ApoE group, patients with HDL-C ＜40 mg/dL and ApoE ＜3.8 mg/dL (n=714, 19.7%).

ApoE, apolipoprotein E; HDL-C, high-density lipoprotein cholesterol

Table 1 summarizes the baseline clinical characteristics. The mean age was 66±10 years; 83% of patients were men,; and 26% of patients had ACS. The prevalence of hypertension, dyslipidemia, diabetes mellitus, current smoker, and family history was 89%, 89%, 41%, 25%, and 29%, respectively. Although the prevalence of patients with CKD was 25%, the mean eGFR of 72.7 mL/min/1.73 m² was within normal limits.

Table 1. Patients’ baseline clinical characteristics

	Overall (n=3632)	Normal HDL-C			Abnormal HDL-C
		High ApoE group (n=1069)	Low ApoE group (n=1151)	p value	High ApoE group (n=698)	Low ApoE group (n=714)	p value
Clinical characteristics
Age, years	66±10	66±10	67±10	0.043	64±10	66±10	0.010
Male, n (%)	3002 (83)	794 (74)	961 (83)	＜0.001	604 (87)	643 (83)	0.039
BMI, kg/m2	24.3±3.4	24.2±3.4	23.7±3.3	＜0.001	25.4±3.5	24.6±3.2	＜0.001
SBP, mmHg	136±23	136±23	137±24	0.268	135±22	134±23	0.424
DBP, mmHg	74±14	74±15	75±14	0.385	74±14	73±13	0.424
TC level, mg/dL	180±38	201±36	173±32	＜0.001	183±36	155±31	＜0.001
TG level, mg/dL	132±68	139±71	104±48	＜0.001	182±79	119±50	＜0.001
HDL-C level, mg/dL	45±13	54±14	50±9	＜0.001	33±5	34±4	0.118
Non-HDL-C level, mg/dL	135±36	147±36	123±31	＜0.001	150±35	121±31	＜0.001
LDL-C level, mg/dL	108±33	119±34	102±29	＜0.001	114±35	97±29	＜0.001
Lp(a) level, mg/dL	19 (10, 32)	19 (10, 34)	19 (9, 32)	0.457	18 (10, 31)	19 (10, 33)	0.036
ApoA-I level, mg/dL	122±25	139±23	128±20	＜0.001	107±16	102±14	＜0.001
ApoB level, mg/dL	90±23	98±23	80±18	＜0.001	103±22	82±19	＜0.001
ApoE level, mg/dL	4.1±1.2	5.0±1.0	3.3±0.5	＜0.001	4.9±1.2	3.1±0.5	＜0.001
LDL-C / ApoB ratio	1.18 (1.06, 1.32)	1.22 (1.08, 1.35)	1.23 (1.12, 1.37)	＜0.001	1.10 (0.96, 1.22)	1.15 (1.04, 1.28)	＜0.001
HDL-C / ApoA-I ratio	0.36 (0.32, 0.39)	0.37 (0.34, 0.42)	0.38 (0.35, 0.42)	0.009	0.31 (0.29, 0.34)	0.33 (0.31, 0.36)	＜0.001
Estimated RC level, mg/dL	27±15	28±15	21±11	＜0.001	36±17	24±11	＜0.001
Hemoglobin, g/dL	13.5±1.7	13.5±1.8	13.3±1.7	0.011	13.7±1.6	13.3±1.8	＜0.001
Estimated GFR, mL/min/1.73m2	72.7±20.1	73.8±20.7	73.8±19.7	0.991	71.8±19.5	70.1±20.2	0.099
FPG, mg/dL	117±45	116±42	117±47	0.579	117±43	117±47	0.818
HbA1c level, %	6.3±1.2	6.3±1.2	6.2±1.1	0.552	6.5±1.3	6.4±1.2	0.253
Hs-CRP level, g/dL	0.11 (0.05, 0.34)	0.10 (0.05, 0.30)	0.10 (0.03, 0.26)	＜0.001	0.16 (0.06, 0.42)	0.13 (0.06, 0.47)	0.569
EF, %	61.6±12.1	62.2±12.1	62.1±11.4	0.975	61.6±12.1	60.1±13.4	0.039
ACS, n (%)	929 (26)	278 (26)	350 (30)	0.021	128 (18)	173 (24)	0.007
Comorbidity
Hypertension, n (%)	3246 (89)	947 (89)	1037 (90)	0.250	541 (78)	533 (75)	0.208
Dyslipidemia, n (%)	3224 (89)	906 (85)	906 (79)	＜0.001	698 (100)	714 (100)
Diabetes mellitus, n (%)	1505 (41)	398 (37)	451 (39)	0.344	329 (47)	327 (46)	0.615
CKD, n (%)	895 (25)	248 (23)	265 (23)	0.922	179 (26)	203 (28)	0.239
Current smoker, n (%)	909 (25)	241 (23)	251 (22)	0.667	207 (30)	210 (29)	0.892
Family history, n (%)	1054 (29)	287 (27)	365 (32)	0.012	186 (27)	216 (30)	0.142
Medication
Statin, n (%)	2512 (69)	710 (66)	842 (73)	＜0.001	444 (64)	516 (72)	＜0.001
Fibrate, n (%)	125 (3)	46 (4)	26 (2)	0.006	32 (5)	21 (3)	0.103
Ezetimibe, n (%)	79 (2)	26 (2)	24 (2)	0.582	14 (2)	15 (2)	0.900
EPA, n (%)	88 (2)	24 (2)	20 (2)	0.392	28 (4)	16 (2)	0.054
Probucol, n (%)	32 (1)	9 (1)	5 (1)	0.223	15 (2)	3 (1)	0.003
Aspirin, n (%)	3423 (95)	1002 (95)	1072 (95)	0.791	658 (96)	691 (98)	0.041
β-blocker, n (%)	1757 (49)	506 (48)	512 (45)	0.262	377 (55)	362 (52)	0.216
CCBs, n (%)	1432 (40)	431 (41)	438 (39)	0.380	281 (41)	282 (40)	0.764
ACE-i/ARB, n (%)	1825 (51)	507 (48)	583 (52)	0.072	355 (52)	380 (54)	0.359
OHA, n (%)	1211 (33)	313 (29)	364 (32)	0.230	269 (39)	265 (37)	0.581
Insulin, n (%)	223 (6)	64 (6)	59 (5)	0.376	35 (5)	65 (9)	0.003

Normal HDL-C / high ApoE group, patients with HDL-C ≥ 40 mg/dL and ApoE ≥ 4.0 mg/dL (n= 1069).

Normal HDL-C / low ApoE group, patients with HDL-C ≥ 40 mg/dL and ApoE ＜4.0 mg/dL (n= 1151).

Abnormal HDL-C / high ApoE group, patients with HDL-C ＜40 mg/dL and ApoE ≥ 3.8 mg/dL (n= 698).

Abnormal HDL-C / low ApoE group, patients with HDL-C ＜40 mg/dL and ApoE ＜3.8 mg/dL (n= 714).

ACE-i, angiotensin-converting enzyme inhibitor; ACS, acute coronary syndrome; ApoA-I, apolipoprotein A-I; ApoB, apolipoprotein B; ApoE, apolipoprotein E; ARB, angiotensin receptor blocker; BMI, body mass index; CCBs, calcium channel blockers; CKD, chronic kidney disease; DBP, diastolic blood pressure; EF, ejection fraction; FPG, fasting plasma glucose; GFR, glomerular filtration rate; HbA1c, Hemoglobin A1c; HDL-C, high-density lipoprotein cholesterol; Hs-CRP, high-sensitivity C-reactive protein; LDL-C, low-density lipoprotein cholesterol; Lp(a), lipoprotein (a); OHA, oral hypoglycemic agent; RC, remnant cholesterol; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride.

The high ApoE subgroups had higher body mass index, total cholesterol levels, TG levels, LDL-C levels, estimated remnant cholesterol levels, and proportion of other lipid-lowering drugs (fibrate, ezetimibe, eicosapentaenoic acid, and probucol), as well as lower age, prevalence of ACS, proportion of statin use, LDL-C/ApoB ratio, and HDL-C/ApoA-I ratio than the low ApoE subgroups in both normal and abnormal HDL-C groups. The correlation between HDL-C and ApoE was significant although it was relatively weak (r=0.11, p＜0.001).

Clinical Outcomes

The median follow-up duration was 6.4 years (interquartile range: 2.6–11.1 years), and the prognostic data were fully documented during the entire follow-up period. During the follow-up, 419 patients (11.5%) developed MACCE, including cardiovascular deaths (n=199), non-fatal MI (n=106), and non-fatal stroke (n=147), and 570 patients (15.7%) died of any cause. To identify if the associations between ApoE levels and the incidences of MACCE and all-cause death were modified by the HDL-C status, we analyzed using a regression model by including ApoE levels×HDL-C status (HDL-C ＜40 or ≥ 40 mg/dL) as the interaction term. As a result, this interaction effect was statistically significant for the incidence of MACCE and all-cause death (p＜0.001 for both).

Kaplan–Meier analysis revealed that the high ApoE subgroup had a significantly higher cumulative incidence of MACCE than the low ApoE subgroup in the normal HDL-C group (17.1% vs. 13.1%, log-rank test, p=0.044); however, the Kaplan–Meier curves for all-cause death were not significantly different between both high and low ApoE subgroups in the normal HDL-C group (19.9% vs. 20.7%, log-rank test, p=0.669) (Fig.2). Conversely, as observed in Fig.3, the low ApoE subgroup had a significantly higher cumulative incidence of MACCE and all-cause death than the high ApoE subgroup in the abnormal HDL-C group (22.5% vs. 15.2%, log-rank test, p=0.022, and 25.2% vs. 18.3%, log-rank test, p=0.017).

Fig.2. Kaplan–Meier curves for MACCE and all-cause death in the normal HDL-C group

a. The cumulative incidence of MACCE

b. The cumulative incidence of all-cause death

ApoE, apolipoprotein E; HDL-C, high-density lipoprotein cholesterol; MACCE, major adverse cardiac and cerebrovascular events

Fig.3. Kaplan–Meier curves for MACCE and all-cause death in the abnormal HDL-C group

a. The cumulative incidence of MACCE

b. The cumulative incidence of all-cause death

ApoE, apolipoprotein E; HDL-C, high-density lipoprotein cholesterol; MACCE, major adverse cardiac and cerebrovascular events

Table 2 shows the results of the multivariable Cox analysis for MACCE after adjusting for the important risk factors, namely, the lipid-related covariates (statin use, other lipid-lowering drugs use, and levels of TG, LDL-C, non-HDL-C, estimated remnant cholesterol, ApoB, Lp(a), LDL-C/ApoB ratio, and HDL-C/ApoA-I ratio) and other traditional coronary risk factors (age, sex, body mass index, hypertension, diabetes mellitus, family history, current smoking status, and CKD). A 1 mg/dL increase in the serum ApoE level was a significantly strong and independent predictor of MACCE in the normal HDL-C group (hazard ratio [HR]: 1.22, 95% CI: 1.06–1.39, p=0.005) and the abnormal HDL-C group (HR: 0.81, 95% CI: 0.66–0.99, p=0.049). Table 3 shows the results of the multivariable Cox analysis for all-cause death after adjusting for the important risk factors. A 1 mg/dL increase in the serum ApoE level was a significantly strong and independent predictor of all-cause death in the normal HDL-C group (HR: 1.25, 95% CI: 1.11–1.41, p＜0.001), but not in the abnormal HDL-C group (p=0.150).

Table 2. Multivariable Cox hazards analysis for MACCE

Covariate	Normal HDL-C group		Abnormal HDL-C group
	HR (95% CI)	p value	HR (95% CI)	p value
ApoE as a continuous variable (HR per 1-mg/dL increase)
Crude	1.18 (1.05-1.31)	0.004	0.83 (0.72-0.94)	0.004
Model 1	1.21 (1.08-1.34)	0.001	0.84 (0.72-0.96)	0.009
Model 2	1.21 (1.06-1.38)	0.007	0.81 (0.65-0.99)	0.043
Model 3	1.22 (1.06-1.39)	0.005	0.81 (0.66-0.99)	0.049

Model 1: adjusted for age, sex, and BMI.

Model 2: adjusted for age, sex, BMI, statin use, other lipid lowering drugs use, TG, LDL-C, non-HDL-C, estimated RC, ApoB, Lp(a), LDL-C/ ApoB ratio, and HDL-C/ApoA-I ratio.

Model 3: adjusted for age, sex, BMI, statin use, other lipid lowering drugs use, TG, LDL-C, non-HDL-C, estimated RC, ApoB, Lp(a), LDL-C/ ApoB ratio, HDL-C/ApoA-I ratio, HT, DM, family history, current smoker, and CKD.

ApoB, apolipoprotein B; ApoE, apolipoprotein E; BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease; DM, diabetes mellitus; HDL-C, high-density lipoprotein cholesterol; HT, hypertension; HR, hazard ratio; LDL-C, low-density lipoprotein cholesterol; Lp(a), lipoprotein (a); MACCE, major adverse cerebral and cardiovascular events; RC, remnant cholesterol; TG, triglyceride.

Table 3. Multivariable Cox hazards analysis for all-cause death

Covariate	Normal HDL-C group		Abnormal HDL-C group
	HR (95% CI)	p value	HR (95% CI)	p value
ApoE as a continuous variable (HR per 1-mg/dL increase)
Crude	1.08 (0.98-1.19)	0.117	0.80 (0.70-0.91)	＜0.001
Model 1	1.18 (1.07-1.30)	0.002	0.85 (0.74-0.96)	0.011
Model 2	1.22 (1.07-1.37)	0.002	0.85 (0.69-1.02)	0.083
Model 3	1.25 (1.11-1.41)	＜0.001	0.87 (0.71-1.05)	0.150

Model 1: adjusted for age, sex, and BMI.

Model 2: adjusted for age, sex, BMI, statin use, other lipid lowering drugs use, TG, LDL-C, non-HDL-C, estimated RC, ApoB, Lp(a), LDL-C/ ApoB ratio, and HDL-C/ApoA-I ratio.

Model 3: adjusted for age, sex, BMI, statin use, other lipid lowering drugs use, TG, LDL-C, non-HDL-C, estimated RC, ApoB, Lp(a), LDL-C/ ApoB ratio, HDL-C/ApoA-I ratio, HT, DM, family history, current smoker, and CKD.

ApoB, apolipoprotein B; ApoE, apolipoprotein E; BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease; DM, diabetes mellitus; HDL-C, high-density lipoprotein cholesterol; HT, hypertension; HR, hazard ratio; LDL-C, low-density lipoprotein cholesterol; Lp(a), lipoprotein (a); RC, remnant cholesterol; TG, triglyceride.

Discussion

The following are the major findings of this study. First, the interactions between the ApoE levels and the HDL-C status for MACCE and all-cause death were statistically significant (p＜0.001 for both). Second, an elevated pre-procedural ApoE level was strongly associated with an increased incidence of MACCE in patients with a normal HDL-C level (≥ 40 mg/dL); conversely, a reduced pre-procedural ApoE level was strongly associated with an increased incidence of MACCE in patients with an abnormal HDL-C level (＜40 mg/dL). Third, of all the four groups, the abnormal HDL-C/low ApoE group had the worst cardiovascular prognosis. Finally, multivariable Cox regression analysis after adjustment for important covariates revealed that the ApoE levels, as continuous variables, had a similar paradoxical tendency for each HDL-C status.

In this study, we investigated the prognostic effects of serum ApoE levels after dichotomization on the basis of a normal HDL-C value (≥ 40 mg/dL). It is extremely important to verify the relationship between the HDL-C levels and the incidence of MACCE because it is well known that cardiovascular events increase in individuals with HDL-C ＜35 mg/dL²⁵⁾. The area under the receiver operating characteristic (ROC) curve was computed to test the predictive discrimination of MACCE. For an HDL-C cutoff value of 37 mg/dL, the area under the ROC curve was 0.53 (Supplemental Fig.1); this had the highest discriminating sensitivity (0.36) and specificity (0.70). After dichotomization based on this cutoff value, Kaplan–Meier analysis revealed that the low HDL-C group (HDL-C ≤ 37 mg/dL) had a significantly higher cumulative incidence of MACCE than the high HDL-C group (HDL-C ＞37 mg/dL) (20.5% vs. 14.7%, log-rank test, p=0.004) (Supplemental Fig.2). Furthermore, the corresponding thresholds of ApoE were 5.0 and 3.5 mg/dL in the high and low HDL-C groups, respectively.

Supplemental Fig.1. ROC curve analysis of HDL-C levels for predicting incidence of MACCE

AUC, area under the curve; HDL-C, high-density lipoprotein cholesterol; MACCE, major adverse cardiac and cerebrovascular events; ROC, receiver operating characteristics

Supplemental Fig.2. Kaplan–Meier curves for MACCE according to the cutoff value of HDL-C

HDL-C, high-density lipoprotein cholesterol; MACCE, major adverse cardiac and cerebrovascular events

ApoE, a major apolipoprotein, is a part of lipoprotein particles of several classes, including chylomicron remnants, very low-density lipoproteins, and HDL. It is a major ligand for low-density lipoprotein receptors and plays a role in cholesterol metabolism²⁶⁾. ApoE has three isoforms, namely, ApoE2, ApoE3, and ApoE4. Among these, ApoE2 and ApoE4 increase the risk for cardiovascular diseases. While ApoE2 is associated with increased plasma TG levels, ApoE4 also reportedly increases the plasma LDL-C levels and the risk of atherosclerosis²⁷⁾. Additionally, it has been reported that elevated ApoE levels are involved in the development of cardiovascular events^{13, 28)}. Conversely, ApoE-containing HDL exhibits several protective properties against arteriosclerosis, including promotion of cholesterol efflux, inhibition of platelet aggregation, stimulation of endothelial heparin sulfate synthesis, and maintenance of arterial elasticity^{11, 29, 30)}. Thus, the association between ApoE and HDL-C for cardiovascular events would be influenced by two major factors: the development of arteriosclerosis caused by atherogenic lipoproteins and the atheroprotective effects through cholesterol efflux capacity. Some studies have shown that the HDL-induced recycling of TRLs with ApoE2 and ApoE3 results in cholesterol efflux and formation of ApoE-containing HDL; however, ApoE4 does not produce an effective net cholesterol efflux due to enhanced surface biding and reuptake of cholesterol-rich particles^{31, 32)}. Actually, TG and estimated remnant cholesterol levels were significantly higher, but LDL-C levels were significantly lower, in the abnormal HDL-C group than in the normal HDL-C group (Supplemental Table 1). Therefore, it is possible that ApoE2 was strongly involved in the abnormal HDL-C group, while ApoE4 was strongly involved in the normal HDL-C group.

Supplemental Table 1. Patients’ baseline clinical characteristics according to HDL-C status

	Overall (n= 3632)	Normal HDL-C group (n= 2220)	Abnormal HDL-C group (n= 1412)	p value
Clinical characteristics
Age, years	66±10	67±10	65±10	＜0.001
Male, n (%)	3002 (83)	1755 (79)	1247 (88)	＜0.001
BMI, kg/m2	24.3±3.4	23.9±3.4	25.0±3.4	＜0.001
Lipid profiles
TC level, mg/dL	180±38	186±37	169±37	＜0.001
TG level, mg/dL	132±68	121±62	150±73	＜0.001
HDL-C level, mg/dL	45±13	52±12	34±4	＜0.001
Non-HDL-C level, mg/dL	135±36	134±36	135±36	0.394
LDL-C level, mg/dL	108±33	110±32	105±33	＜0.001
Lp(a) level, mg/dL	19 (10, 32)	19 (10, 33)	18 (10, 32)	0.914
ApoA-I level, mg/dL	122±25	133±23	104±15	＜0.001
ApoB level, mg/dL	90±23	89±22	93±23	＜0.001
ApoE level, mg/dL	4.1±1.2	4.1±1.1	4.0±1.3	0.019
LDL-C / ApoB ratio	1.18 (1.06, 1.32)	1.23 (1.10, 1.36)	1.13 (1.01, 1.24)	＜0.001
HDL-C / ApoA-I ratio	0.36 (0.32, 0.39)	0.38 (0.35, 0.42)	0.32 (0.30, 0.35)	＜0.001
Estimated RC level, mg/dL	27±15	24±13	30±16	＜0.001

Normal HDL-C group, patients with HDL-C ≥ 40 mg/dL (n= 2220).

Abnormal HDL-C group, patients with HDL-C ＜40 mg/dL (n= 1412).

ApoA-I, apolipoprotein A-I; ApoB, apolipoprotein B; ApoE, apolipoprotein E; BMI, body mass index; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; Lp(a), lipoprotein (a); RC, remnant cholesterol; TC, total cholesterol; TG, triglyceride.

Furthermore, although an elevated TRL level is usually known as a cardiovascular risk factor, it has been reported that elevated serum TG levels may efficiently promote cholesterol efflux from macrophages at lower HDL-C levels^{33, 34)}. Additionally, a basic report revealed differences in the cholesterol efflux capacity based on an HDL-C cutoff value of 40 mg/dL¹⁴⁾; the correlation between the TG level and the cholesterol efflux capacity was significantly positive in patients with an HDL-C level of ＜40 mg/dL, but not in patients with an HDL-C level of ≥ 40 mg/dL. Although this causal relationship remains unclear, the cholesterol efflux capacity of HDL may be attributed to factors related to the metabolism of increased TRLs (e.g., elevated ApoE levels) in individuals with abnormal HDL-C levels and is not always synergized with HDL-C concentration. Meanwhile, high ApoE levels in patients with normal HDL-C levels may contribute to the transformation of HDL dysfunction and promote the development of atherosclerosis and cardiovascular disease³⁵⁾. As mentioned above, serum ApoE predominantly promotes HDL-mediated cholesterol efflux during anti-atherogenesis in patients with abnormal HDL-C levels, while it predominantly promotes arteriosclerosis as reflected by TRL and LDL-C levels in those with normal HDL-C levels. Thus, this study identified a paradoxical impact of ApoE levels on the cardiovascular risk depending on the HDL-C status.

Limitations

This study had several limitations that should be considered. First, because this study had a single-center, retrospective, and observational design, unknown confounding factors may have affected the outcomes regardless of the analytical adjustments. In addition, the relatively small number of included patients may have limited the statistical power of the study. Second, this study included only Japanese patients. Thus, it is unclear whether this association between the serum ApoE and HDL-C levels can be applied to the patients with CAD who underwent PCI in other countries. Third, we performed this study as a clinical investigation and faithfully analyzed our findings in the backdrop of previous basic research. However, this study did not include any molecular investigations (e.g., ApoE-containing HDL). Fourth, 765 patients were excluded from this study due to missing data on ApoE; among these, 488 patients cleared the exclusion criteria. The HDL-C levels, LDL-C levels, proportion of current smoker, and prevalence of ACS were higher, while the ejection fraction was lower, in this subgroup than in the 3632 patients analyzed in the main study (p＜0.05 for all). The reason for this result may be considered that patients with ACS required an urgent treatment; therefore, detailed measurements of lipoproteins and apolipoproteins were often not possible.

Conclusion

In this study, serum ApoE levels were observed to have a paradoxical impact on the future long-term cardiovascular risk depending on the HDL-C status in patients with CAD who underwent PCI. Therefore, the relationship between the serum ApoE and HDL-C levels may be useful in the prediction of patient prognosis. Furthermore, to the best of our knowledge, this study is the first to present the clinical association between the serum ApoE and HDL-C levels as a risk factor for adverse cardiovascular events; thus, these findings warrant further investigation.

Author Contributions

Analysis: T.F. Concept and design: T.F., T.D. Interpretation of the data: T.F., T.D., R.N., M.T., N.T., Y.C., H.E., S.D., H.N., I.O., H.I., S.O., K.M., H.D., and T.M. Drafting of the manuscript: T.F. Clinical data acquisition: T.F., R.N., M.T., N.T., Y.C., H.E., and S.D. Writing-editing: T.D. Writing-review and editing: T.M. Supervision: T.D., and T.M.

Financial Support

This research received no grant from any funding agency in the public, commercial, or not-for-profit sectors.

Acknowledgments

The authors are grateful to the staff of the Department of Cardiovascular Medicine at Juntendo University. The authors also appreciate the secretarial assistance of Ms. Yumi Nozawa.

Conflict of Interest

The authors declare that there are no conflicts of interest.

Data Availability

All data generated or analyzed during this study are included in this published article. Due to the nature of this research, supporting data cannot be used outside at this time. If necessary, we will discuss the reconsideration of the data availability at ethics committee in our institution.

References

• 1)  Di Angelantonio E, Sarwar N, Perry P, Kaptoge S, Ray KK, Thompson A, Wood AM, Lewington S, Sattar N, Packard CJ, Collins R, Thompson SG, Danesh J. Major lipids, apolipoproteins, and risk of vascular disease. JAMA, 2009; 302: 1993-2000
• 2)  Rosenson RS, Brewer HB, Jr., Davidson WS, Fayad ZA, Fuster V, Goldstein J, Hellerstein M, Jiang XC, Phillips MC, Rader DJ, Remaley AT, Rothblat GH, Tall AR, Yvan-Charvet L. Cholesterol efflux and atheroprotection: advancing the concept of reverse cholesterol transport. Circulation, 2012; 125: 1905-1919
• 3)  Rosenson RS, Brewer HB, Jr., Ansell B, Barter P, Chapman MJ, Heinecke JW, Kontush A, Tall AR, Webb NR. Translation of high-density lipoprotein function into clinical practice: current prospects and future challenges. Circulation, 2013; 128: 1256-1267
• 4)  Kuhn FE, Mohler ER, Satler LF, Reagan K, Lu DY, Rackley CE. Effects of high-density lipoprotein on acetylcholine-induced coronary vasoreactivity. Am J Cardiol, 1991; 68: 1425-1430
• 5)  Kontush A, Chantepie S, Chapman MJ. Small, dense HDL particles exert potent protection of atherogenic LDL against oxidative stress. Arterioscler Thromb Vasc Biol, 2003; 23: 1881-1888
• 6)  Barter PJ, Nicholls S, Rye KA, Anantharamaiah GM, Navab M, Fogelman AM. Antiinflammatory properties of HDL. Circ Res, 2004; 95: 764-772
• 7)  Murphy AJ, Akhtari M, Tolani S, Pagler T, Bijl N, Kuo CL, Wang M, Sanson M, Abramowicz S, Welch C, Bochem AE, Kuivenhoven JA, Yvan-Charvet L, Tall AR. ApoE regulates hematopoietic stem cell proliferation, monocytosis, and monocyte accumulation in atherosclerotic lesions in mice. J Clin Invest, 2011; 121: 4138-4149
• 8)  Khera AV, Cuchel M, de la Llera-Moya M, Rodrigues A, Burke MF, Jafri K, French BC, Phillips JA, Mucksavage ML, Wilensky RL, Mohler ER, Rothblat GH, Rader DJ. Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis. N Engl J Med, 2011; 364: 127-113
• 9)  Rohatgi A, Khera A, Berry JD, Givens EG, Ayers CR, Wedin KE, Neeland IJ, Yuhanna IS, Rader DR, de Lemos JA, Shaul PW. HDL cholesterol efflux capacity and incident cardiovascular events. N Engl J Med, 2014; 371: 2383-2393
• 10)  Wilson HM, Patel JC, Russell D, Skinner ER. Alterations in the concentration of an apolipoprotein E-containing subfraction of plasma high density lipoprotein in coronary heart disease. Clin Chim Acta, 1993; 220: 175-187
• 11)  Horiuchi Y, Ohkawa R, Lai SJ, Yamazaki A, Ikoma H, Yano K, Kameda T, Tozuka M. Characterization of the cholesterol efflux of apolipoprotein E-containing high-density lipoprotein in THP-1 cells. Biol Chem, 2019; 400: 209-218
• 12)  Qi Y, Liu J, Wang W, Wang M, Zhao F, Sun J, Liu J, Zhao D. Apolipoprotein E-containing high-density lipoprotein (HDL) modifies the impact of cholesterol-overloaded HDL on incident coronary heart disease risk: A community-based cohort study. J Clin Lipidol, 2018; 12: 89-98.e2
• 13)  Rasmussen KL, Tybjærg-Hansen A, Nordestgaard BG, Frikke-Schmidt R. Plasma levels of apolipoprotein E, APOE genotype, and all-cause and cause-specific mortality in 105 949 individuals from a white general population cohort. Eur Heart J, 2019; 40: 2813-2824
• 14)  Weibel GL, Drazul-Schrader D, Shivers DK, Wade AN, Rothblat GH, Reilly MP, de la Llera-Moya M. Importance of evaluating cell cholesterol influx with efflux in determining the impact of human serum on cholesterol metabolism and atherosclerosis. Arterioscler Thromb Vasc Biol, 2014; 34: 17-25
• 15)  Rickham PP. HUMAN EXPERIMENTATION. CODE OF ETHICS OF THE WORLD MEDICAL ASSOCIATION. DECLARATION OF HELSINKI. Br Med J, 1964; 2: 177
• 16)  Varbo A, Benn M, Tybjærg-Hansen A, Jørgensen AB, Frikke-Schmidt R, Nordestgaard BG. Remnant cholesterol as a causal risk factor for ischemic heart disease. J Am Coll Cardiol, 2013; 61: 427-436
• 17)  Umemura S, Arima H, Arima S, Asayama K, Dohi Y, Hirooka Y, Horio T, Hoshide S, Ikeda S, Ishimitsu T, Ito M, Ito S, Iwashima Y, Kai H, Kamide K, Kanno Y, Kashihara N, Kawano Y, Kikuchi T, Kitamura K, Kitazono T, Kohara K, Kudo M, Kumagai H, Matsumura K, Matsuura H, Miura K, Mukoyama M, Nakamura S, Ohkubo T, Ohya Y, Okura T, Rakugi H, Saitoh S, Shibata H, Shimosawa T, Suzuki H, Takahashi S, Tamura K, Tomiyama H, Tsuchihashi T, Ueda S, Uehara Y, Urata H, Hirawa N. The Japanese Society of Hypertension Guidelines for the Management of Hypertension (JSH 2019). Hypertens Res, 2019; 42: 1235-1481
• 18)  Kinoshita M, Yokote K, Arai H, Iida M, Ishigaki Y, Ishibashi S, Umemoto S, Egusa G, Ohmura H, Okamura T, Kihara S, Koba S, Saito I, Shoji T, Daida H, Tsukamoto K, Deguchi J, Dohi S, Dobashi K, Hamaguchi H, Hara M, Hiro T, Biro S, Fujioka Y, Maruyama C, Miyamoto Y, Murakami Y, Yokode M, Yoshida H, Rakugi H, Wakatsuki A, Yamashita S. Japan Atherosclerosis Society (JAS) Guidelines for Prevention of Atherosclerotic Cardiovascular Diseases 2017. J Atheroscler Thromb, 2018; 25: 846-984
• 19)  Nathan DM, Balkau B, Bonora E, Borch-Johnsen K, Buse JB, Colagiuri S, Davidson MB, DeFronzo R, Genuth S, Holman RR, Ji L, Kirkman S, Knowler WC, Schatz D, Shaw J, Sobngwi E, Steffes M, Vaccaro O, Wareham N, Zinman B, Kahn R. International Expert Committee report on the role of the A1C assay in the diagnosis of diabetes. Diabetes Care, 2009; 32: 1327-1334
• 20)  Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, Yamagata K, Tomino Y, Yokoyama H, Hishida A. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis, 2009; 53: 982-992
• 21)  World Health Organization. Nutritional anaemias. Report of a WHO scientific group. World Health Organ Tech Rep Ser, 1968; 405: 5-37
• 22)  Scheuner MT, Whitworth WC, McGruder H, Yoon PW, Khoury MJ. Familial risk assessment for early-onset coronary heart disease. Genet Med, 2006; 8: 525-531
• 23)  Garcia-Garcia HM, McFadden EP, Farb A, Mehran R, Stone GW, Spertus J, Onuma Y, Morel MA, van Es GA, Zuckerman B, Fearon WF, Taggart D, Kappetein AP, Krucoff MW, Vranckx P, Windecker S, Cutlip D, Serruys PW. Standardized End Point Definitions for Coronary Intervention Trials: The Academic Research Consortium-2 Consensus Document. Circulation, 2018; 137: 2635-2650
• 24)  Hicks KA, Mahaffey KW, Mehran R, Nissen SE, Wiviott SD, Dunn B, Solomon SD, Marler JR, Teerlink JR, Farb A, Morrow DA, Targum SL, Sila CA, Hai MTT, Jaff MR, Joffe HV, Cutlip DE, Desai AS, Lewis EF, Gibson CM, Landray MJ, Lincoff AM, White CJ, Brooks SS, Rosenfield K, Domanski MJ, Lansky AJ, McMurray JJV, Tcheng JE, Steinhubl SR, Burton P, Mauri L, O’Connor CM, Pfeffer MA, Hung HMJ, Stockbridge NL, Chaitman BR, Temple RJ. 2017 Cardiovascular and Stroke Endpoint Definitions for Clinical Trials. Circulation, 2018; 137: 961-972
• 25)  Assmann G, Schulte H, von Eckardstein A, Huang Y. High-density lipoprotein cholesterol as a predictor of coronary heart disease risk. The PROCAM experience and pathophysiological implications for reverse cholesterol transport. Atherosclerosis, 1996; 124 Suppl: S11-S20
• 26)  Mahley RW, Weisgraber KH, Huang Y. Apolipoprotein E: structure determines function, from atherosclerosis to Alzheimer’s disease to AIDS. J Lipid Res, 2009; 50 Suppl(Suppl): S183-S188
• 27)  Bennet AM, Di Angelantonio E, Ye Z, Wensley F, Dahlin A, Ahlbom A, Keavney B, Collins R, Wiman B, de Faire U, Danesh J. Association of apolipoprotein E genotypes with lipid levels and coronary risk. JAMA, 2007; 298: 1300-1311
• 28)  Fukase T, Dohi T, Chikata Y, Takahashi N, Endo H, Doi S, Nishiyama H, Kato Y, Okai I, Iwata H, Okazaki S, Isoda K, Miyauchi K, Daida H, Minamino T. Serum apolipoprotein E levels predict residual cardiovascular risk in patients with chronic coronary syndrome undergoing first percutaneous coronary intervention and on-statin treatment. Atherosclerosis, 2021; 333: 9-15
• 29)  Paka L, Kako Y, Obunike JC, Pillarisetti S. Apolipoprotein E containing high density lipoprotein stimulates endothelial production of heparan sulfate rich in biologically active heparin-like domains. A potential mechanism for the anti-atherogenic actions of vascular apolipoprotein e. J Biol Chem, 1999; 274: 4816-4823
• 30)  Kothapalli D, Liu SL, Bae YH, Monslow J, Xu T, Hawthorne EA, Byfield FJ, Castagnino P, Rao S, Rader DJ, Puré E, Phillips MC, Lund-Katz S, Janmey PA, Assoian RK. Cardiovascular protection by ApoE and ApoE-HDL linked to suppression of ECM gene expression and arterial stiffening. Cell Rep, 2012; 2: 1259-1271
• 31)  Cullen P, Cignarella A, Brennhausen B, Mohr S, Assmann G, von Eckardstein A. Phenotype-dependent differences in apolipoprotein E metabolism and in cholesterol homeostasis in human monocyte-derived macrophages. J Clin Invest, 1998; 101: 1670-1677
• 32)  Heeren J, Beisiegel U, Grewal T. Apolipoprotein E recycling: implications for dyslipidemia and atherosclerosis. Arterioscler Thromb Vasc Biol, 2006; 26: 442-448
• 33)  Miller M, Stone NJ, Ballantyne C, Bittner V, Criqui MH, Ginsberg HN, Goldberg AC, Howard WJ, Jacobson MS, Kris-Etherton PM, Lennie TA, Levi M, Mazzone T, Pennathur S. Triglycerides and cardiovascular disease: a scientific statement from the American Heart Association. Circulation, 2011; 123: 2292-2333
• 34)  Attia N, Ramaharo A, Paul JL, Cambillau M, Beaune P, Grynberg A, Simon A, Fournier N. Enhanced removal of cholesterol from macrophage foam cells to serum from type IV hypertriglyceridemic subjects. Atherosclerosis, 2008; 198: 49-56
• 35)  Corsetti JP, Gansevoort RT, Bakker SJ, Navis G, Sparks CE, Dullaart RP. Apolipoprotein E predicts incident cardiovascular disease risk in women but not in men with concurrently high levels of high-density lipoprotein cholesterol and C-reactive protein. Metabolism, 2012; 61: 996-1002