Advertisement

High-density lipoprotein function is associated with atherosclerotic burden and cardiovascular outcomes in type 2 diabetes

Open AccessPublished:July 06, 2018DOI:https://doi.org/10.1016/j.atherosclerosis.2018.07.005

      Abstract

      Background and aims

      Measures of HDL function are emerging tools for assessing cardiovascular disease (CVD) event risk. HDL-apoA-I exchange (HAE) reflects HDL capacity for reverse cholesterol transport.

      Methods

      HAE was measured in 93 participants with type 2 diabetes (T2D) and at least one additional CVD risk factor in the Asker and Bærum Cardiovascular Diabetes study. At baseline and after seven years, the atherosclerotic burden was assessed by invasive coronary angiography. Major CVD events were registered throughout the study.

      Results

      Linear regression analysis demonstrated a significant inverse association between HAE and atherosclerotic burden. Cox proportional hazard regression analysis showed a significant association between HAE and a composite of major CVD events when controlling for waist-hip ratio, HR = 0.89, 95% CI = 0.80–1.00 and p=0.040.

      Conclusions

      Despite the relatively small size of the study population and the limited number of CVD events, these findings suggest that HAE provides valuable information in determining CVD risk.

      Keywords

      1. Introduction

      There is a well established inverse relationship between high-density lipoprotein cholesterol (HDL-C) levels and coronary heart disease events and mortality [
      • Miller N.E.
      • Thelle D.S.
      • Forde O.H.
      • et al.
      The Tromsø heart-study. High-density lipoprotein and coronary heart-disease: a prospective case-control study.
      ,
      • Gordon T.
      • Castelli W.P.
      • Hjortland M.C.
      • et al.
      High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study.
      ,
      • Gordon D.J.
      • Probstfield J.L.
      • Garrison R.J.
      • et al.
      High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies.
      ]. Nevertheless, almost 40% of men with coronary heart disease have HDL-C levels in the normal range [
      • van der Steeg W.A.
      • Holme I.
      • Boekholdt S.M.
      • et al.
      High-density lipoprotein cholesterol, high-density lipoprotein particle size, and apolipoprotein A-I: significance for cardiovascular risk: the IDEAL and EPIC-Norfolk studies.
      ]. Furthermore, merely raising plasma HDL-C does not necessarily reduce risk for cardiovascular events, as evidenced by recent drug trials of niacin and cholesteryl ester transfer protein (CETP) inhibitors that significantly elevated HDL-C levels but failed to reduce the risk of CVD events [
      • Barter P.J.
      • Caulfield M.
      • Eriksson M.
      • et al.
      Effects of torcetrapib in patients at high risk for coronary events.
      ,
      • HPS2-THRIVE C.G.
      • Landray M.J.
      • Haynes R.
      • et al.
      Effects of extended-release niacin with laropiprant in high-risk patients.
      ,
      • HPS3/TIMI55–REVEAL C.G.
      • Bowman L.
      • Hopewell J.C.
      • et al.
      Effects of anacetrapib in patients with atherosclerotic vascular disease.
      ]. This suggests that circulating HDL-C levels do not directly reflect the atheroprotective properties of HDL. As a result, assays of HDL function have been proposed as more relevant and reliable measurements than HDL-C levels in determining CVD risk [
      • Oda M.N.
      High-density lipoprotein cholesterol: origins and the path ahead.
      ].
      HDL has both anti-inflammatory and anti-oxidative properties, but its primary means of atheroprotection has been attributed to its ability to mediate reverse cholesterol transport (RCT) [
      • Fisher E.A.
      • Feig J.E.
      • Hewing B.
      • et al.
      High-density lipoprotein function, dysfunction, and reverse cholesterol transport.
      ]. Cholesterol efflux capacity (CEC) is the most widely used assay for quantifying HDL's role in RCT. CEC has been found to be inversely associated with CVD events and improves CVD risk prediction models when included [
      • Khera A.V.
      • Cuchel M.
      • de la Llera-Moya M.
      • et al.
      Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis.
      ,
      • Rohatgi A.
      • Khera A.
      • Berry J.D.
      • et al.
      HDL cholesterol efflux capacity and incident cardiovascular events.
      ].
      HDL carries 1–4 molecules of apolipoprotein A-I (apoA-I), its primary protein component [
      • Wald J.H.
      • Krul E.S.
      Jonas, A, Structure of apolipoprotein A-I in three homogeneous, reconstituted high density lipoprotein particles.
      ]. Lipid-poor apoA-I is the preferred substrate for the ATP-binding cassette transporter A1 (ABCA1) and is essential for de novo HDL biogenesis in the intimal layer of the arterial wall [
      • Mulya A.
      • Lee J.Y.
      • Gebre A.K.
      • et al.
      Minimal lipidation of pre-beta HDL by ABCA1 results in reduced ability to interact with ABCA1.
      ,
      • Okuhira K.
      • Tsujita M.
      • Yamauchi Y.
      • et al.
      Potential involvement of dissociated apoA-I in the ABCA1-dependent cellular lipid release by HDL.
      ,
      • Pownall H.J.
      • Hosken B.D.
      • Gillard B.K.
      • et al.
      Speciation of human plasma high- density lipoprotein (HDL): HDL stability and apolipoprotein A-I partitioning.
      ]. Because apoA-I is synthesized primarily in the liver, lipid-poor apoA-I arrives in the intima associated with HDL and must exchange off the HDL particle. The exchange/dissociation of apoA-I off HDL is a rate-limiting step of RCT [
      • Okuhira K.
      • Tsujita M.
      • Yamauchi Y.
      • et al.
      Potential involvement of dissociated apoA-I in the ABCA1-dependent cellular lipid release by HDL.
      ]. ApoA-I undergoes significant conformational change as it exchanges between HDL-bound and lipid-free states, and this conformational change can be reliably quantified by electron paramagnetic resonance (EPR) [
      • Handa D.
      • Kimura H.
      • Oka T.
      • et al.
      Kinetic and thermodynamic analyses of spontaneous exchange between high-density lipoprotein-bound and lipid-free apolipoprotein A-I.
      ,
      • Borja M.S.
      • Zhao L.
      • Hammerson B.
      • et al.
      HDL-apoA-I exchange: rapid detection and association with atherosclerosis.
      ]. Lipid-free apoA-I readily exchanges with apoA-I on HDL, and measuring the conformational change of exogenous spin-labeled apoA-I added to blood plasma is the basis of the HDL-apoA-I exchange (HAE) assay. HAE correlates strongly with CEC (r = 0.69, p < 0.001) in normolipidemic subjects [
      • Borja M.S.
      • Ng K.F.
      • Irwin A.
      • et al.
      HDL-apolipoprotein A-I exchange is independently associated with cholesterol efflux capacity.
      ].
      Cardiovascular disease is the leading cause of death in patients with type 2 diabetes (T2D), and the presence of both CVD and diabetes decreases lifespan in these patients by several years compared with the general population [
      • Emerging R.F.C.
      • Di Angelantonio E.
      • Kaptoge S.
      • et al.
      Association of cardiometabolic multimorbidity with mortality.
      ]. Even when LDL-C is well managed, there remains a significant residual risk of CVD, particularly in T2D [
      • Reiner Z.
      Managing the residual cardiovascular disease risk associated with HDL-cholesterol and triglycerides in statin-treated patients: a clinical update.
      ]. The randomized-controlled Asker and Bærum Cardiovascular Diabetes (ABCD) study showed that 2 years of structured multi-intervention significantly reduced the estimated 10-year absolute risk for coronary heart disease [
      • Johansen O.E.
      • Gullestad L.
      • Blaasaas K.G.
      • et al.
      Effects of structured hospital- based care compared with standard care for Type 2 diabetes-The Asker and Baerum Cardiovascular Diabetes Study, a randomized trial.
      ], driven by between-group differences in LDL-C, triglycerides, HbA1c, and systolic blood pressure, but with no long-term benefit on CVD events or death [
      • Ofstad A.P.
      • Ulimoen G.R.
      • Orvik E.
      • et al.
      Long-term follow-up of a hospital- based, multi-intervention programme in type 2 diabetes mellitus: impact on cardiovascular events and death.
      ]. In the present analysis we measure HDL function at the 7-year follow-up of these patients and examine associations with study outcomes.
      We hypothesize that loss of HDL function is a major contributor to residual risk of CVD in T2D patients, exhibited by an inverse association of HAE with atherosclerotic burden and CVD events.

      2. Patients and methods

      2.1 Study population

      The design, intervention and results from the ABCD-study (clinicaltrials.gov NCT00133718) have been described previously [
      • Johansen O.E.
      • Gullestad L.
      • Blaasaas K.G.
      • et al.
      Effects of structured hospital- based care compared with standard care for Type 2 diabetes-The Asker and Baerum Cardiovascular Diabetes Study, a randomized trial.
      ]. Briefly, 120 patients with T2D and ≥1 additional CVD risk factor were randomized to two years of either standard care by their general practitioner or structured multi-intervention care at a hospital outpatient clinic comprising an initial 6-month lifestyle program followed by intensified treatment to reach guideline targets for glycemic control, blood pressure and lipid profile [
      • Johansen O.E.
      • Gullestad L.
      • Blaasaas K.G.
      • et al.
      Effects of structured hospital- based care compared with standard care for Type 2 diabetes-The Asker and Baerum Cardiovascular Diabetes Study, a randomized trial.
      ]. Clinical, laboratory and cardiac assessments, including invasive coronary angiography were performed at baseline. Seven years after baseline, 93 patients were re-examined, 85 of which had completed the intervention. Major cardiovascular events were registered throughout the duration of the study and until May 29, 2012.
      This study was conducted in accordance with the Declaration of Helsinki and was approved by the Norwegian Regional Committee for Medical and Health Research Ethics. All participants provided their written informed consent.

      2.2 Clinical and laboratory methods

      Venous blood samples were drawn in the morning after an overnight fast and routine laboratory analyses performed by the local laboratory. All intra- and inter-assay coefficients of variations were <10%. Urinary albumin and creatinine concentrations were determined in timed overnight urine samples.

      2.3 HAE assay

      The HAE assay was performed on 93 samples from the 7-year follow-up. Freshly thawed plasma was apoB depleted by precipitation with polyethylene glycol 6000. The apoB-depleted plasma was combined with 3 mg/ml nitroxide spin-labeled apoA-I probe as previously described [
      • Borja M.S.
      • Zhao L.
      • Hammerson B.
      • et al.
      HDL-apoA-I exchange: rapid detection and association with atherosclerosis.
      ]. Samples were incubated for 15 min at 37 °C and scanned at 37 °C. The peak amplitude of the nitroxide signal from HAE probe in the sample (3462–3470 Gauss) was compared to the peak amplitude of a proprietary internal standard (3507–3515 Gauss). The internal standard is contained within the eScan spectrometer (Bruker BioSpin GmbH, Karlsruhe, Germany) cavity and does not contact the sample. The %HAE activity was calculated by comparison with a standard curve determined for each production lot of apoA-I probe. The final result is a measure of the relative efficiency of HDL-apoA-I exchange. All samples were read in duplicate and averaged. The intra-assay coefficient of variation was 4%.

      2.4 Invasive coronary angiography

      At baseline and at the 7-year follow-up, invasive coronary angiography was performed according to standard procedures [
      • Johansen O.E.
      • Birkeland K.I.
      • Orvik E.
      • et al.
      Inflammation and coronary angiography in asymptomatic type 2 diabetic subjects.
      ]. The angiograms were analyzed visually by an experienced radiologist with the support of a quantitative coronary analysis program (Sectra Cardiology Package 1.0, Sectra Imtec, Sweden). They were divided into 16 segments and graded from 0 to 4. Grade 0: <25% stenosis, 1: 25–50% stenosis, 2: 50–75% stenosis, 3: >75% stenosis and 4: Occlusion. The atherosclerotic burden was described by the Severity score, the average grade of the diseased segments graded ≥1, and the Extent score, the number of coronary segments graded ≥1 adjusted to 16 coronary segments [
      • Ledru F.
      • Ducimetière P.
      • Battaglia S.
      • et al.
      New diagnostic criteria for diabetes and coronary artery disease: insights from an angiographic study.
      ].

      2.5 Outcome

      The CVD outcome was characterized as time to CVD death, non-CVD death, stroke, acute myocardial infarction, hospitalization for unstable angina or congestive heart failure, amputation or revascularization procedures including percutaneous transluminal angioplasty. Revascularization procedures occurring in relation to the study angiography were excluded. All events underwent adjudication by two of the authors (K.I.B., L.G.) blinded to treatment allocation.

      2.6 Statistical analysis

      Clinical and metabolic data are presented as either proportions, means with their standard deviations (SD) or medians with 25th and 75th percentiles. Student's t-test was applied to compare means between groups with normally distributed data, and Mann-Whitney U test to compare medians in case of non-normal distributions. Linear regression analysis was performed to identify associations between HAE as the exposure variable and Severity score and Extent score as outcome variables. Potential confounding variables were required to be associated significantly with both the exposure and outcome variable with p < 0.05. Qualified confounders were included in a multivariate regression analysis with a backward elimination procedure to determine a final model. All major CVD end points were combined in a composite variable. The association between HAE and the end point variable was determined by Cox proportional hazard regression analysis. Potential confounders were identified as described for linear regression analysis. The proportional hazard assumptions were checked by Schoenfeld tests and found to be satisfied. The primary outcome was the composite endpoint. Follow-up data were censored with closing date May 29, 2012.
      Statistical significance was assumed for p < 0.05. IBM SPSS Statistics for Macintosh, version 24 (Armonk, NY: IBM Corp) was applied for all statistical analyses.

      3. Results

      At inclusion, there were 25.6% and 19.0% females in the structured care and standard care groups respectively. The clinical and metabolic characteristics of the participants were similar at baseline (Table 1).
      Table 1Clinical and metabolic characteristics.
      Baseline2-year follow-up7-year follow-up
      Structured care N = 43Standard care N = 42p-valueStructured care N = 43Standard care N = 42p-valueStructured care N = 43Standard care N = 42p-value
      Age (years)58.5 (7.1)57.3 (10.0)0.52660.6 (7.0)59.1 (10.0)0.42965.7 (7.1)64.4 (9.9)0.491
      Diabetes duration (years)a4.0 (2.0, 9.0)3.5 (1.0, 12.0)0.8496.1 (4.6, 11.2)5.5 (3.7, 14.5)0.85411.4 (9.7, 16.3)10.6 (8.8, 19.6)0.930
      BMI (kg/m2)30.6 (4.8)29.0 (5.0)0.14030.8 (5.2)29.4 (4.9)0.19630.9 (4.9)28.4 (4.3)0.015
      Waist-hip ratio0.99 (0.08)0.97 (0.07)0.1931.00 (0.08)0.99 (0.08)0.1961.00 (0.07)0.97 (0.07)0.047
      Systolic blood pressure (mmHg)141 (18)141 (17)0.995136 (21)140 (21)0.316138 (16)138 (17)0.908
      Diastolic blood pressure (mmHg)83 (10)82 (9)0.79478 (9)80 (8)0.45379 (10)82 (10)0.341
      HbA1c (%)7.5 (1.6)7.8 (1.7)0.4676.8 (0.8)8.0 (1.6)<0.0017.1 (1.0)7.5 (1.1)0.056
      Total cholesterol (mmol/l)4.9 (1.0)4.9 (0.9)0.8314.2 (0.7)4.6 (0.9)0.0214.3 (1.1)4.2 (0.9)0.623
      HDL cholesterol (mmol/l)
      Median (25th and 75th percentile).
      for 7-year follow-up
      1.30 (0.39)1.31 (0.42)0.9341.36 (0.37)1.32 (0.43)0.6801.30 (1.0, 1.7)1.20 (1.0, 1.4)0.534
      LDL cholesterol (mmol/l)2.9 (0.9)2.8 (0.8)0.8572.2 (0.6)2.6 (0.9)0.0322.4 (0.9)2.2 (0.8)0.487
      Triglycerides (mmol/l)*1.6 (1.0, 2.5)1.3 (1.1, 2.3)0.4991.3 (0.9, 1.7)1.7 (1.0,

      2.1)
      0.0601.2 (1.0, 1.8)1.2 (0.9, 1.9)0.944
      *Mean (standard deviation).
      Bold indicates significant p-values.
      a Median (25th and 75th percentile).
      Two years of intervention in the structured care group resulted in significant changes in CVD risk factors, with significantly lower HbA1c, total cholesterol and LDL cholesterol than in the control group, as reported previously [
      • Johansen O.E.
      • Gullestad L.
      • Blaasaas K.G.
      • et al.
      Effects of structured hospital- based care compared with standard care for Type 2 diabetes-The Asker and Baerum Cardiovascular Diabetes Study, a randomized trial.
      ]. These differences were no longer present at the 7-year follow-up, with the possible exception of HbA1c, which was near significantly lower in the structured care group (mean between group difference = −0.44, 95% CI = −0.90 - 0.01 and p = 0.056). BMI and waist-hip ratio were similar in both groups at baseline and after the intervention, but after seven years, both were significantly higher in the structured care group.
      At the 7-year follow-up, 85% of the participants were treated with statins and 91% with anti-diabetic medication, with no significant differences between the groups. The median number of anti-diabetic drugs was two in both groups. The median number of drugs to treat hypertension was two in the structured care group and one in the standard care group, p = 0.012. There was no significant difference between the two groups in median (25th, 75th percentile) albumin-creatinine ratio, 0.9 (0.4, 4.0) in the structured vs. 0.8 (0.5, 2.7) in the standard care group.
      The Severity score and the Extent score were assessed in 72 participants at baseline and 69 at the 7-year follow up. Their distributions are shown in Fig. 1.
      Fig. 1
      Fig. 1The distribution of Severity score and Extent score.
      The atherosclerotic burden in all participants assessed by the Severity score and the Extent score at baseline and after seven years of follow-up.
      Throughout the present study, from January 17. 2002 to May 29. 2012, twelve major CVD events occurred (Table 2). Seven in the structured care group, four in the standard care group and one in a participant that did not complete the intervention.
      Table 2Major cardiovascular events.
      CV death0
      Non-CV death2
      Stroke1
      Acute myocardial infarction4 (3)
      Unstable angina2 (1)
      Congestive heart failure0
      Revascularization3 (2)
      N (N before the 7-year follow-up).
      There was no significant difference between the groups (p = 0.354). The mean follow-up time was 8.7 (SD = 2.0) years.
      At the 7-year follow-up, %HAE was 46.8 (SD = 8.9) in the structured care group compared with 46.6 (SD = 7.4) in the standard care group (p = 0.914). A linear regression analysis showed a significant inverse association between HAE and Severity score (B = −0.028, p = 0.043 and R2 = 0.431), controlling for Severity score at baseline and age. There was no significant association between HAE and Extent score. A Cox regression model, adjusted for waist-hip ratio, showed a significant association between %HAE and CVD outcome (HR = 0.89, 95% CI = 0.80–1.00 and p = 0.040).
      Similarly, serum levels of HDL-C at the 7-year follow-up were significantly and inversely associated with Severity Score (B = −0.520, p = 0.039 and R2 = 0.433), controlling for Severity score at baseline and age. There was no significant association between HDL-C and Extent score. For comparison, serum levels of LDL- C at the 7-year follow-up were significantly associated with Extent score (B = 0.413, p = 0.014 and R2 = 0.548), controlling for Extent score at baseline. There was no significant association between LDL-C and Severity score. In Cox regression models, HDL-C was significantly associated with CVD outcome (HR = 0.48, 95% CI = 0.004–0.54 and p = 0.014) adjusted for HbA1c, total cholesterol and diabetes duration, while there was no association between LDL and CVD outcome.

      4. Discussion

      This study reports the observation that HDL function, as measured by HAE, is significantly inversely associated with atherosclerotic burden, as assessed by coronary angiography. HAE was also significantly inversely associated with CVD end-points. These relationships emerge despite the relatively low number of patients and few registered CVD events in this study.
      Inverse relationships between HDL function and both atherosclerotic burden and CVD end points have previously been demonstrated for CEC [
      • Khera A.V.
      • Cuchel M.
      • de la Llera-Moya M.
      • et al.
      Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis.
      ,
      • Rohatgi A.
      • Khera A.
      • Berry J.D.
      • et al.
      HDL cholesterol efflux capacity and incident cardiovascular events.
      ]. Confirmation in large clinical studies is certainly needed, but the present study indicates for the first time that HAE may be similarly discriminating. This is important for two reasons. First, measures of HDL function, both CEC and HAE, are emerging as valuable tools for the assessment of residual risk after LDL-C reduction. Second, the simplicity and precision of the HAE assay makes HDL function analysis more convenient and accessible, facilitating its clinical application.
      Clinical intervention in the ABCD study significantly modified the CVD risk factor profile at two years, but these differences were no longer present at seven years. Similarly, there was no difference in HAE levels between the groups. Despite a substantial reduction in LDL cholesterol after two years, the intervention did not reduce mortality [
      • Ofstad A.P.
      • Ulimoen G.R.
      • Orvik E.
      • et al.
      Long-term follow-up of a hospital- based, multi-intervention programme in type 2 diabetes mellitus: impact on cardiovascular events and death.
      ]. This suggests the presence of a residual risk in T2D potentially involving reduced HDL function.
      There are several limitations to this study, particularly the relatively low number of participants and clinical events. Serial HAE measurements from baseline and after the clinical intervention were unfortunately not available to construct prediction models. Previous work with HAE has shown an increased discriminatory ability of HAE-apoA-I ratio compared with HAE alone [
      • Borja M.S.
      • Ng K.F.
      • Irwin A.
      • et al.
      HDL-apolipoprotein A-I exchange is independently associated with cholesterol efflux capacity.
      ,
      • Heier M.
      • Borja M.S.
      • Brunborg C.
      • et al.
      Reduced HDL function in children and young adults with type 1 diabetes.
      ], but apoA-I was not measured in this study. Half of the registered major CVD events occurred before HAE measurement at the 7-year follow-up. This precludes any assessment of the predictive value of HAE. This timing might also attenuate the relationship between HAE and atherosclerotic burden, as most patients after a CVD event will be provided intensified treatment and potentially improved HDL function.
      In conclusion, this is the first study to demonstrate the inverse relationship between HAE, as a measure of HDL function, and atherosclerotic burden and major CVD events in T2D.

      Conflicts of interest

      APO and OEJ are presently employed by Boehringer Ingelheim Norway.
      MNO is a founder of and owns a significant stake in Seer Biologics, Inc. Potential benefit in no way influenced the thoroughness, stringency, interpretation and presentation of this manuscript's content.
      KIB has received grants to his institution for lectures and consulting from Astra Zeneca, Novo Nordisk, Sanofi, Lilly, Boehringer Ingelheim, Roche and Merck Sharp & Dohme.

      Financial support

      MH has received funds from Oslo Diabetes Research Centre and from the European Union Seventh Framework Programme ( FP7-PEOPLE-2013- COFUND ) under grant agreement no 609020 - Scientia Fellows . APO received a grant from the Bærum Hospital Internal Research foundation. MSB was supported by American Heart Association Western States Affiliate postdoctoral fellowship 14POST1833018 . Additionally, this work was supported by the California Tobacco Related Disease Research Program 21RT-0125 . The funding bodies played no direct roles in study design, data collection, analysis or interpretation or in writing the manuscript.

      Author contributions

      MH performed the HAE assay, conducted statistical analyses and wrote the manuscript. MSB and MNO analyzed data and provided technical support. MNO provided reagent and equipment resources. OEJ, KIB and LG designed the study. KE performed and analyzed the coronary angiographies. APO and OEJ collected the data. All authors reviewed and edited the manuscript.

      References

        • Miller N.E.
        • Thelle D.S.
        • Forde O.H.
        • et al.
        The Tromsø heart-study. High-density lipoprotein and coronary heart-disease: a prospective case-control study.
        Lancet. 1977; 1: 965-968
        • Gordon T.
        • Castelli W.P.
        • Hjortland M.C.
        • et al.
        High density lipoprotein as a protective factor against coronary heart disease. The Framingham Study.
        Am. J. Med. 1977; 62: 707-714
        • Gordon D.J.
        • Probstfield J.L.
        • Garrison R.J.
        • et al.
        High-density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies.
        Circulation. 1989; 79: 8-15
        • van der Steeg W.A.
        • Holme I.
        • Boekholdt S.M.
        • et al.
        High-density lipoprotein cholesterol, high-density lipoprotein particle size, and apolipoprotein A-I: significance for cardiovascular risk: the IDEAL and EPIC-Norfolk studies.
        J. Am. Coll. Cardiol. 2008; 51: 634-642
        • Barter P.J.
        • Caulfield M.
        • Eriksson M.
        • et al.
        Effects of torcetrapib in patients at high risk for coronary events.
        N. Engl. J. Med. 2007; 357: 2109-2122
        • HPS2-THRIVE C.G.
        • Landray M.J.
        • Haynes R.
        • et al.
        Effects of extended-release niacin with laropiprant in high-risk patients.
        N. Engl. J. Med. 2014; 371: 203-212
        • HPS3/TIMI55–REVEAL C.G.
        • Bowman L.
        • Hopewell J.C.
        • et al.
        Effects of anacetrapib in patients with atherosclerotic vascular disease.
        N. Engl. J. Med. 2017; 377: 1217-1227
        • Oda M.N.
        High-density lipoprotein cholesterol: origins and the path ahead.
        Curr. Opin. Endocrinol. Diabetes Obes. 2015; 22: 133-141
        • Fisher E.A.
        • Feig J.E.
        • Hewing B.
        • et al.
        High-density lipoprotein function, dysfunction, and reverse cholesterol transport.
        Arterioscler. Thromb. Vasc. Biol. 2012; 32: 2813-2820
        • Khera A.V.
        • Cuchel M.
        • de la Llera-Moya M.
        • et al.
        Cholesterol efflux capacity, high-density lipoprotein function, and atherosclerosis.
        N. Engl. J. Med. 2011; 364: 127-135
        • Rohatgi A.
        • Khera A.
        • Berry J.D.
        • et al.
        HDL cholesterol efflux capacity and incident cardiovascular events.
        N. Engl. J. Med. 2014; 371: 2383-2393
        • Wald J.H.
        • Krul E.S.
        Jonas, A, Structure of apolipoprotein A-I in three homogeneous, reconstituted high density lipoprotein particles.
        J. Biol. Chem. 1990; 265: 20037-20043
        • Mulya A.
        • Lee J.Y.
        • Gebre A.K.
        • et al.
        Minimal lipidation of pre-beta HDL by ABCA1 results in reduced ability to interact with ABCA1.
        Arterioscler. Thromb. Vasc. Biol. 2007; 27: 1828-1836
        • Okuhira K.
        • Tsujita M.
        • Yamauchi Y.
        • et al.
        Potential involvement of dissociated apoA-I in the ABCA1-dependent cellular lipid release by HDL.
        J. Lipid Res. 2004; 45: 645-652
        • Pownall H.J.
        • Hosken B.D.
        • Gillard B.K.
        • et al.
        Speciation of human plasma high- density lipoprotein (HDL): HDL stability and apolipoprotein A-I partitioning.
        Biochemistry. 2007; 46: 7449-7459
        • Handa D.
        • Kimura H.
        • Oka T.
        • et al.
        Kinetic and thermodynamic analyses of spontaneous exchange between high-density lipoprotein-bound and lipid-free apolipoprotein A-I.
        Biochemistry. 2015; 54: 1123-1131
        • Borja M.S.
        • Zhao L.
        • Hammerson B.
        • et al.
        HDL-apoA-I exchange: rapid detection and association with atherosclerosis.
        PLoS One. 2013; 8: e71541
        • Borja M.S.
        • Ng K.F.
        • Irwin A.
        • et al.
        HDL-apolipoprotein A-I exchange is independently associated with cholesterol efflux capacity.
        J. Lipid Res. 2015; 56: 2002-2009
        • Emerging R.F.C.
        • Di Angelantonio E.
        • Kaptoge S.
        • et al.
        Association of cardiometabolic multimorbidity with mortality.
        J. Am. Med. Assoc. 2015; 314: 52-60
        • Reiner Z.
        Managing the residual cardiovascular disease risk associated with HDL-cholesterol and triglycerides in statin-treated patients: a clinical update.
        Nutr. Metabol. Cardiovasc. Dis. 2013; 23: 799-807
        • Johansen O.E.
        • Gullestad L.
        • Blaasaas K.G.
        • et al.
        Effects of structured hospital- based care compared with standard care for Type 2 diabetes-The Asker and Baerum Cardiovascular Diabetes Study, a randomized trial.
        Diabet. Med. 2007; 24: 1019-1027
        • Ofstad A.P.
        • Ulimoen G.R.
        • Orvik E.
        • et al.
        Long-term follow-up of a hospital- based, multi-intervention programme in type 2 diabetes mellitus: impact on cardiovascular events and death.
        J. Int. Med. Res. 2017; 45: 1535-1552
        • Johansen O.E.
        • Birkeland K.I.
        • Orvik E.
        • et al.
        Inflammation and coronary angiography in asymptomatic type 2 diabetic subjects.
        Scand. J. Clin. Lab. Invest. 2007; 67: 306-316
        • Ledru F.
        • Ducimetière P.
        • Battaglia S.
        • et al.
        New diagnostic criteria for diabetes and coronary artery disease: insights from an angiographic study.
        J. Am. Coll. Cardiol. 2001; 37: 1543-1550
        • Heier M.
        • Borja M.S.
        • Brunborg C.
        • et al.
        Reduced HDL function in children and young adults with type 1 diabetes.
        Cardiovasc. Diabetol. 2017; 16: 85