Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Nov 1.
Published in final edited form as: J Clin Lipidol. 2017 Oct 3;11(6):1458–1470.e4. doi: 10.1016/j.jacl.2017.09.002

Association of Low Density Lipoprotein Pattern with Mortality after Myocardial Infarction: Insights from the TRIUMPH Study

Yashashwi Pokharel 1,2, Yuanyuan Tang 2, Bhaskar Bhardwaj 1,2, Krishna K Patel 1,2, Mohammed Qintar 1,2, James H O’Keefe Jr 1,2, Krishnaji R Kulkarni 3, Peter H Jones 4, Seth S Martin 5, Salim S Virani 4,6, John A Spertus 1,2
PMCID: PMC5698121  NIHMSID: NIHMS910549  PMID: 29050980

Abstract

Background

Studies of incident coronary heart disease risk within low-density lipoprotein (LDL) subclass (small, dense vs. large, buoyant) have shown mixed results. No prospective cohort study has examined the association of small, dense or large, buoyant LDL with mortality after myocardial infarction (MI).

Objective

To examine association of LDL pattern after MI and death

Methods

In 2476 patients hospitalized for MI, LDL pattern [A (large, buoyant), A/B (mixed) and B (small, dense)] was established by ultracentrifugation using Vertical Auto Profile. Using time-to-event analysis, we examined the association with 5-year mortality within LDL patterns, after adjusting for important patient and treatment characteristics. We additionally adjusted for LDL cholesterol (LDL-C) and triglyceride levels and used directly measured LDL-C and non-HDL-C as exposures.

Results

Patterns A, A/B and B were present in 39%, 28% and 33% of patients, respectively, with incident rates (per 1000 patient-years) of 50, 34 and 24 for all-cause and 24, 19 and 10 for CV mortality. The HRs (95% CI) with LDL patterns A/B and B compared to pattern A were 0.77 (0.61, 0.99) and 0.67 (0.51, 0.88) for all-cause, 0.94 (0.67, 1.33) and 0.69 (0.46, 1.03) for cardiovascular, and 0.64 (0.45, 0.91) and 0.65 (0.45, 0.93) for non-cardiovascular mortality, respectively. Results were similar when further adjusted for LDL-C and triglycerides, or with LDL-C and non-HDL-C as exposures.

Conclusion

Compared with LDL pattern A, pattern B was significantly associated with reduced all-cause and non-CV mortality with a trend for lower CV mortality after MI, independent of LDL-C and triglycerides.

Keywords: Low density lipoprotein pattern, size, density, paradox, myocardial infarction, mortality

Introduction

Increased low-density lipoprotein cholesterol (LDL-C) level has long been recognized as an important causal risk factor for atherosclerotic cardiovascular (CV) disease.1,2 Despite marked improvement in CV prevention,3 of which statin therapy has been a cornerstone,4 many patients continue to have residual CV risk. In several large outcome trials of statin therapy, about a quarter to a third of patients continue to experience major coronary events,5 highlighting the importance of understanding the factors determining this residual risk.

Measurement of LDL-C provides the total cholesterol content of LDL particles, but heterogeneity exists with respect to the proportion of types of LDL particles present in each patient.6 LDL particle pattern based on size and density are of particular importance. Small, dense LDL particles (pattern B), may be more atherogenic than large, buoyant LDL particles (pattern A),7 and thus could partly explain the residual CV risk. Several studies have shown that small, dense LDL particles are associated with higher risk for incident CV events,8–13 while results of other studies are mixed.14–16

On the other hand, there are limited data on LDL particle sub-fractions and risk of CV events and death in patients with prior CV disease. In a small study of patients with acute ischemic stroke, small, dense LDL particles were associated with higher odds for in-hospital mortality.13 Conversely, in patients referred for coronary angiography, only large LDL was associated with higher mortality when compared with intermediate sized LDL.17 Similarly, most studies have shown greater mortality risk associated with higher serum cholesterol levels in patients with coronary heart disease (CHD),18–22 although a few studies have shown an inverse association of LDL-C with mortality after myocardial infarction (MI).23–25 We hypothesized that among patients recovering from MI, presence of LDL pattern B (small, dense LDL particles) would be associated with an increased risk of death as compared with LDL pattern A (large, buoyant LDL particles). We further hypothesized that such relationship would be present even in patients with relatively controlled LDL-C levels (i.e., <100 mg/dL).26

Methods

Study population

We used data from the Translational Research Investigating Underlying disparities in acute Myocardial infarction Patients’ Health Status (TRIUMPH)27 registry. TRIUMPH investigators obtained institutional review board approval at each participating institution and informed consent from study participants for the collection of data from patients’ index hospitalization and follow-up. The methodology of the TRIUMPH study has been previously described.27 In brief, TRIUMPH was a prospective observational, 24-center MI registry of 4,340 patients hospitalized for acute MI from diverse geographical regions throughout the United States between April 11, 2005 and December 31, 2008. Inclusion criteria were age ≥18 years, elevated cardiac enzymes (creatinine kinase-MB or Troponin-I) within 24 hours of hospital admission and a diagnosis of an AMI supported by prolonged ischemic symptoms or electrocardiographic ST changes in ≥2 contiguous leads. MI occurring in the setting of elective coronary revascularization was excluded. Patients transferred to the enrolling hospital from another facility after more than 24 hours, patients who were incarcerated, refused to participate, were not able to provide informed consent, or did not speak English or Spanish, were excluded. The baseline for the current analysis was the index hospitalization for MI during which baseline VAP testing was performed. A voluntary component of the TRIUMPH study was to participate in a blood work analysis, with collection during the index hospitalization (median [25th, 75th percentile] of 2 [1, 3] days after admission and 1 [0, 2] days prior to discharge) and optionally at 30-days and 6-months later through home visits by trained paramedics. Our analytic cohort consisted of 2,476 patients with LDL pattern available at baseline and 1,286 patients with LDL pattern available at 30 days (Figure 1). Since participation in blood work analysis (either at baseline, 30-days or 6-months) was voluntary, not all of the participants at 30 days had an LDL pattern assessment at baseline.

Figure 1.

Figure 1

Open in a new tab

Cohort development

LDL = low-density lipoprotein

Baseline for this study was the index hospitalization for myocardial infarction

Lipid assessment

Blood samples were processed, serum separated, refrigerated and sent by overnight mail in freezer packs to the core laboratory (Clinical Reference Laboratory, Lenexa, KS). Serum specimens were aliquoted, stored frozen at −70 degree centigrade and then sent to Atherotech (Birmingham, AL) by overnight mail on dry ice.

Comprehensive lipoprotein profile was measured by Atherotech’s Vertical Auto Profile (VAP),28,29 which rapidly separates lipoproteins according to their density by ultracentrifugation using a vertical rotor and single density gradient spin. Cholesterol content is quantified using an enzymatic reaction and spectrophotometric absorbance. There were four distinct LDL subclasses: LDL 1, LDL 2, LDL 3 and LDL 4. The density increased and size decreased from LDL 1 through LDL 4, where in LDL 1 was the largest and the most buoyant LDL subclass and LDL 4 was the smallest and most dense LDL subclass. Eventually, three broad groups of LDL patterns were identified: pattern A consisted of a preponderance of large, buoyant LDL subclass, while pattern B consisted mainly of small, dense LDL subclass, and others with a mix of both were assigned pattern A/B.29

Lipoprotein (a) [Lp(a)] cholesterol and intermediate density lipoprotein cholesterol cannot be separately measured by the commonly used standard lipid panel, but are included in the calculated LDL cholesterol using Friedewald formula.30 However, VAP can separate the cholesterol in Lp(a) and intermediate density lipoprotein from LDL by ultracentrifugation and can thus measure all directly.29 In order to provide LDL cholesterol that is equivalent to the standard lipid panel, the VAP method adds Lp(a) cholesterol and intermediate density lipoprotein cholesterol to LDL cholesterol to give the total LDL cholesterol (referred here as LDL-C). The LDL cholesterol without Lp(a) cholesterol and intermediate density lipoprotein cholesterol is referred as “real” LDL cholesterol for clarity. The VAP method measures cholesterol concentration of lipoprotein (a) particles,29 unlike most other methods that measure apolipoprotein (a) or Lp (a) particle concentration. The VAP method also reports non-high density lipoprotein cholesterol (non-HDL-C), which is total cholesterol minus high-density lipoprotein cholesterol (HDL-C).

Study outcomes

Our primary outcomes of interest were time to all-cause mortality as well as CV mortality over 5 years follow up after the index hospitalization for MI. Both all-cause and CV mortality data was obtained directly from the Center for Disease Control and Prevention’s National Death Index. CV mortality was defined using International Classification of Diseases, Tenth Revision (ICD-10, codes 00–78).31

Other study variables

Details about study variables used in the TRIUMPH registry have been described previously.27 Briefly, patient characteristics were abstracted from medical records, patient interviews, additional specific blood work and ICD codes. Baseline characteristics were assessed during the index MI admission. Both lipids and high-sensitivity C-reactive protein were assessed at baseline and at 30 days after the MI.

We examined several baseline variables. These included demographic, lifestyle, vital signs, patient characteristics at index hospitalization, medical and surgical history, laboratory data, treatment characteristics and disease severity. We assessed disease severity using the Global Registry of Acute Coronary Events (GRACE) risk score.32 The GRACE risk score provides a risk score for 6-month mortality for patients with acute coronary syndrome and higher score confers greater risk.32

We examined the use of lipid-lowering therapy (LLT) [statin or non-statin therapy (cholesterol binding resins, ezetimibe, fibrates, niacin or omega-3-acid ethyl esters [Lovaza, GlaxoSmithKline]) at both hospital admission and discharge. Intensity of statin therapy was defined as recommended by the 2013 American College of Cardiology/American Heart Association Cholesterol Management Guideline.4 Other reported medications were abstracted at hospital discharge.

Statistical methods

We compared baseline patient characteristics by LDL pattern (A, A/B and B). We also compared the characteristics of the study population included in the study (N=2476) with those who did not participate in blood work, including VAP testing, and were consequently excluded (N=1864) to assess for selection bias. We used one-way analysis of variance and Wilcoxon Rank Sum tests for continuous variables and chi-square or Fisher’s exact tests for categorical variables.

In our primary analysis, we examined the hazards for all-cause and CV mortality in patients with LDL pattern B or A/B compared with pattern A, using Cox proportional hazard models. We used hierarchical models and treated hospital site as a random effect. This allowed us to control for the presence of potential differences in care for MI patients seen at different hospitals. We examined the relationship in an unadjusted model, followed by a multivariable-adjusted model. We selected the adjustment variables a priori, based on clinical knowledge and existing literature. We adjusted for GRACE risk score (age, history of congestive heart failure, history of prior MI, initial heart rate, initial systolic blood pressure, ST segment depression on electrocardiogram at admission, initial creatinine, elevated troponin and in-hospital percutaneous coronary intervention), sex, race, body mass index, history of hypertension, history of diabetes, history of alcohol abuse, current smoking, blood glucose, high-sensitivity C-reactive protein, metabolic syndrome, insurance status, presentation (ST segment elevation myocardial infarction vs. non-ST segment elevation myocardial infarction), in-hospital coronary artery bypass grafting and leisure time physical activity. Since LLT, including statin therapy, can alter lipoproteins, we examined interaction by use of statin or any lipid lowering therapy at hospital admission. In additional analysis, we also examined association of LDL pattern for non-CV mortality.

As a secondary analysis, we repeated the analysis for patients with guideline recommended LDL-C levels (i.e., LDL-C <100 mg/dL),26 as it has been shown that even among this group, patients with higher levels of small, dense LDL-C have increased risk for incident coronary event.33,34

We performed several sensitivity analyses. We examined associations of cholesterol content in small, dense to large, buoyant LDL particle ratio, cholesterol concentration of each LDL subclass (i.e., LDL 1 to LDL 4), non HDL-C, LDL-C, patients not on LLT at hospital admission and LDL pattern at 30 days, with CV and all-cause mortality. The ratio of small, dense to large, buoyant LDL particles’ associated cholesterol ([LDL 3 + LDL 4]/[LDL 1 + LDL 2]) was transformed into a logarithm scale, as this ratio is not normally distributed.35 A higher (more positive) ratio in the logarithm scale means presence of denser LDL and vice versa. We examined the relationship between each 1 standard deviation increase in the cholesterol concentration of each LDL subclass (i.e., LDL 1 to LDL 4) with CV and all-cause mortality. We examined the associations of 10 mg/dL higher non-HDL-C, which is a close proxy for total atherogenic lipoprotein particles,6,36 as well as 10 mg/dL higher LDL-C. Because LDL is a known negative acute phase reactant and so can decrease at the time of acute MI,37 we assessed the association of LDL pattern at 30 days after MI. For analysis using 30-day lipids, we used high-sensitivity C-reactive protein measured at 30 days, but all other variables were based from baseline assessment.

Linearity was confirmed for continuous variables and proportionality hazard assumption was verified. The CV death was analyzed using a competing risk approach (competing risk being non-CV death), which provides hazards for CV death among patients who haven’t yet experienced either CV or non-CV death.38 We considered a 2-sided p<0.05 as statistically significant and did not adjust for multiple comparisons. All statistical analyses were performed with SAS, version 9.4 (SAS Institute, Inc., Cary, NC).

Results

Study population and baseline characteristics

After excluding patients who did not volunteer for blood testing, we included 2476 patients in our primary analysis (Figure 1). Compared with patients who participated in optional blood work analysis at baseline, those who did not participate were older, less likely to be current smoker, more likely to have metabolic syndrome, medical insurance, dyslipidemia and higher GRACE score (Supplemental Table 1). However, for the most part these two groups had similar characteristics at baseline. Importantly, there were no significant differences in the rate of all-cause or CV death between the two groups.

In the study cohort, 967 (39.0%) had LDL pattern A, 700 (28.3%) had LDL pattern A/B and 809 (32.7%) had LDL pattern B. Compared with patients with LDL pattern A, patients with pattern B were younger, more likely to be men, white, obese, history of dyslipidemia, to have presented with an ST-segment elevation MI, lower GRACE score, more likely to be treated with percutaneous coronary intervention, antiplatelet agent, angiotensin converting enzyme inhibitor or angiotensin receptor blocker and lipid lowering therapy at hospital discharge (Tables 1–2). There were no significant differences in metabolic syndrome, diabetes, initial blood glucose and use of statin therapy at both hospital admission and discharge. Patients with LDL pattern B had higher “real” LDL cholesterol, LDL-C, triglyceride, non-HDL-C levels and lower Lp (a) and HDL-C levels than patients with LDL pattern A (Table 2).

Table 1.

Characteristics by LDL pattern at baseline

Characteristics Total LDL Pattern at Baseline P-Value
N = 2476 A
N = 967		 A/B
N = 700		 B
N = 809
Demographic
Age, years 58.3 ± 12.2 60.0 ± 12.9 58.7 ± 12.2 55.8 ± 10.8 < 0.001
Women 790 (31.9%) 387 (40.0%) 207 (29.6%) 196 (24.2%) < 0.001
Race (N=2472) < 0.001
    White 1672 (67.6%) 576 (59.7%) 476 (68.0%) 620 (76.8%)
    Black 632 (25.6%) 327 (33.9%) 173 (24.7%) 132 (16.4%)
    Other 168 (6.8%) 62 (6.4%) 51 (7.3%) 55 (6.8%)
Insured at baseline (N=2423) 1871 (77.2%) 715 (75.6%) 530 (77.7%) 626 (78.7%) 0.274
Lifestyle
History of Smoking 1483 (59.9%) 552 (57.1%) 429 (61.3%) 502 (62.1%) 0.070
Current Smoking (N=2460) 1027 (41.7%) 384 (39.9%) 288 (41.4%) 355 (44.2%) 0.186
History of Alcohol Abuse 260 (10.5%) 114 (11.8%) 66 (9.4%) 80 (9.9%) 0.236
Leisure time physical activity (N=2462) 0.006
    Mainly sedentary 1101 (44.7%) 463 (48.2%) 321 (46.1%) 317 (39.4%)
    Mild exercise 747 (30.3%) 283 (29.4%) 213 (30.6%) 251 (31.2%)
    Moderate exercise 517 (21.0%) 182 (18.9%) 138 (19.8%) 197 (24.5%)
    Strenuous exercise 97 (3.9%) 33 (3.4%) 25 (3.6%) 39 (4.9%)
Baseline Variables
Initial Heart Rate, per minute (N=2470) 82.8 ± 21.8 84.2 ± 21.9 82.3 ± 20.5 81.5 ± 22.5 0.028
Initial Systolic BP, mmHg (N=2462) 142.6 ± 29.8 142.1 ± 31.3 141.8 ± 28.0 143.9 ± 29.6 0.337
Initial Diastolic BP, mmHg (N=2462) 83.1 ± 19.0 81.9 ± 19.7 82.9 ± 18.1 84.8 ± 18.6 0.005
Metabolic Syndrome, (N=2448) 33 (3.5%) 31 (4.6%) 46 (5.9%) 110 (4.6%) 0.067
Body mass index, kg/m2 (N=2342) 29.8 ± 6.4 28.8 ± 6.4 29.9 ± 6.5 30.8 ± 6.3 < 0.001
MI presentation < 0.001
    STEMI 1054 (42.6%) 358 (37.0%) 312 (44.6%) 384 (47.5%)
    NSTEMI 1422 (57.4%) 609 (63.0%) 388 (55.4%) 425 (52.5%)
Medical and Surgical History
History of Hypertension 1637 (66.1%) 651 (67.3%) 460 (65.7%) 526 (65.0%) 0.573
History of Dyslipidemia 1176 (47.5%) 429 (44.4%) 328 (46.9%) 419 (51.8%) 0.007
History of Diabetes 787 (31.8%) 305 (31.5%) 230 (32.9%) 252 (31.1%) 0.760
History of Prior Myocardial Infarction 501 (20.2%) 190 (19.6%) 146 (20.9%) 165 (20.4%) 0.824
History of Chronic Heart Failure 216 (8.7%) 112 (11.6%) 60 (8.6%) 44 (5.4%) < 0.001
History of Prior PCI 474 (19.1%) 148 (15.3%) 154 (22.0%) 172 (21.3%) < 0.001
History of Prior CABG 280 (11.3%) 100 (10.3%) 94 (13.4%) 86 (10.6%) 0.110
Laboratory Variables
Initial Glucose, mg/dL (N=2459) 160.9 ± 91.3 158.2 ± 91.6 158.5 ± 81.2 166.2 ± 98.7 0.135
Hemoglobin A1c, % (N=2407) 6.6 ± 1.7 6.4 ± 1.6 6.6 ± 1.7 6.6 ± 1.8 0.053
High-sensitivity C-reactive Protein, mg/L (N=2422) 2.0 (0.7, 4.5) 2.0 (0.7, 4.7) 2.1 (0.9, 4.7) 1.9 (0.7, 4.2) < 0.001
Initial Creatinine, mg/dL (N=2470) 1.2 ± 1.1 1.4 ± 1.5 1.1 ± 0.7 1.1 ± 0.6 < 0.001
Ejection Fraction, % (N=2099) 48.6 ± 13.4 48.0 ± 14.3 47.9 ± 13.2 50.0 ± 12.3 0.005
Treatment
In-Hospital PCI 1592 (64.3%) 533 (55.1%) 466 (66.6%) 593 (73.3%) < 0.001
In-Hospital CABG 230 (9.3%) 109 (11.3%) 66 (9.4%) 55 (6.8%) 0.005
Beta Blocker at discharge (N=2465) 2217 (89.9%) 853 (88.9%) 624 (89.3%) 740 (91.8%) 0.094
Antiplatelet at discharge (N=2465) 2399 (97.3%) 913 (95.1%) 687 (98.3%) 799 (99.1%) < 0.001
ACE Inhibitor/ARB at discharge (N=2465) 1839 (74.6%) 688 (71.7%) 524 (75.0%) 627 (77.8%) 0.012
Disease Severity
GRACE 6 Month Mortality Risk Score 99.0 ± 29.6 105.5 ± 30.6 99.1 ± 29.2 91.3 ± 26.7 < 0.001
Open in a new tab

Data based on total N of 2476, unless otherwise indicated.

All variables are based on baseline.

Continuous variables compared using one-way analysis of variance and categorical variables compared using chi-square or Fisher’s exact test.

High-sensitivity C-reactive protein and triglycerides are expressed as median (25th, 75th percentile)

ACE Inhibitor = angiotensin converting enzyme inhibitor, ARB = angiotensin receptor blocker, BP = blood pressure, CABG = coronary artery bypass grafting, GRACE = Global Registry of Acute Coronary Events, LDL = low-density lipoprotein, NSTEMI = non-ST segment elevation myocardial infarction, PCI = percutaneous coronary intervention, STEMI = ST segment elevation myocardial infarction

Table 2.

Lipids and lipid lowering medications by LDL pattern at baseline

Variables Total LDL Pattern at Baseline P-Value
N = 2476 A
N = 967		 A/B
N = 700		 B
N = 809
LDLR-C, mg/dL 79.2 ± 27.5 75.6 ± 27.8 79.6 ± 26.5 83.2 ± 27.5 < 0.001
LDL-C, mg/dL 95.5 ± 32.4 93.7 ± 33.1 94.5 ± 31.4 98.4 ± 32.1 0.006
LDL-C <100 mg/dL 1460 (59.0%) 595 (61.5%) 424 (60.6%) 441 (54.5%) 0.006
HDL-C, mg/dL 39.9 ± 10.6 42.4 ± 11.7 40.5 ± 10.4 36.5 ± 8.1 < 0.001
Triglycerides, mg/dL 131.0 (99.5, 178.0) 120 (93.0, 156.0) 123.5 (98.0, 158.5) 167.0 (119.0, 231.0) < 0.001
Non-HDL-C, mg/dL 116.3 ± 36.4 113.8 ± 36.5 114.2 ± 35.0 121.1 ± 37.1 < 0.001
Lp(a), mg/dL 6.4 ± 3.9 6.6 ± 3.8 6.4 ± 3.5 6.2 ± 4.2 0.035
LDL Ratio (logarithm scale) 0.0 (−0.4, 0.5) −0.6 (−0.9, −0.4) 0.1 (−0.1, 0.2) 0.7 (0.5, 0.9) < 0.001
Lipid Lowering Medication at admission (N=2465) 892 (36.0%) 307 (31.7%) 263 (37.6%) 322 (39.8%) 0.001
Lipid Lowering Medication at discharge (N=2465) 2197 (89.1%) 836 (87.1%) 620 (88.7%) 741 (91.9%) 0.004
Statin at admission 827 (33.4%) 294 (30.4%) 238 (34.0%) 295 (36.5%) 0.024
Statin at discharge (N=2465) 2156 (87.5%) 827 (86.1%) 607 (86.8%) 722 (89.6%) 0.079
Statin Intensity at discharge (N=2145) 0.375
    Low 81 (3.8%) 32 (3.9%) 21 (3.5%) 28 (3.9%)
    Moderate 984 (45.9%) 400 (48.5%) 269 (44.8%) 315 (43.8%)
    High 1080 (50.3%) 393 (47.6%) 310 (51.7%) 377 (52.4%)
Non-statin LLT at admission 159 (6.4%) 44 (4.6%) 63 (9.0%) 52 (6.4%) 0.001
Non-statin LLT at discharge 239 (9.7%) 68 (7.0%) 84 (12.1%) 87 (10.8%) < 0.001
Bile Acid Resin at discharge 6 (0.2%) 1 (0.1%) 4 (0.6%) 1 (0.1%) 0.139
Ezetimibe at discharge 141 (5.7%) 47 (4.9%) 57 (8.1%) 37 (4.6%) 0.004
Fibrates at discharge 64 (2.6%) 15 (1.6%) 14 (2.0%) 35 (4.3%) < 0.001
Niacin at discharge 52 (2.1%) 10 (1.0%) 18 (2.6%) 24 (3.0%) 0.010
Omega-3-acid ethyl esters at discharge 4 (0.2%) 1 (0.1%) 1 (0.1%) 2 (0.2%) 0.830
Open in a new tab

Data based on total N of 2476, unless otherwise indicated.

All variables are based on baseline, unless specified otherwise.

Non-statin lipid lowering medication includes bile acid binding resins, ezetimibe, fibrates, niacin, omega-3-acid ethylesters (Lovaza, GlaxoSmithKline)

LDL-C includes cholesterol in “real” LDL, lipoprotein (a) and intermediate density lipoprotein. LDLR-C reflects “real” low-density lipoprotein cholesterol

Continuous variables compared using one-way analysis of variance and categorical variables compared using chi-square or Fisher’s exact test.

Triglycerides and logarithm of LDL ratio are expressed as median (25th, 75th percentile).

HDL-C = high-density lipoprotein cholesterol, LDL(C) = low-density lipoprotein (cholesterol), LDLR-C = real low-density lipoprotein cholesterol, LLT = lipid lowering therapy, Lp(a) = lipoprotein (a)

Overall, there were a total of 458 all-cause deaths and 224 CV deaths over 5 years of follow up (8 patients were censored because of <5 years follow up), with an incident rate of 37 all-cause deaths and 18 CV deaths per 1000 patient-years. The incident rate was 50, 34 and 24 for all-cause deaths and 24, 19 and 10 for CV deaths per 1000 patients per year in patients with LDL pattern A, A/B and B, respectively (Figures 2–3).

Figure 2.

Figure 2

Open in a new tab

Kaplan Meier plot for overall survival by LDL pattern

LDL = low-density lipoprotein

Figure 3.

Figure 3

Open in a new tab

Kaplan Meier plot for survival from CV death by LDL pattern

CV = cardiovascular, LDL = low-density lipoprotein

LDL pattern and death

The hazard ratio (95% confidence interval [HR, 95% CI]) for all-cause mortality in patients with LDL pattern A/B compared with LDL pattern A was 0.64 (0.51–0.80, p<0.01) in unadjusted analysis, and was attenuated but still significant in the multivariable analysis [0.77 (0.61–0.99), p=0.04, Table 3]. When LDL pattern B was compared with LDL pattern A, the association was stronger [0.44 (0.35–0.56), p<0.01 and 0.67 (0.51–0.88), p<0.01 in unadjusted and adjusted analysis, respectively]. For CV mortality, there was no significant association of LDL pattern A/B when compared with LDL pattern A in unadjusted or adjusted analysis (Table 3). However, when LDL pattern B was compared with LDL pattern A, there was significant 60% relative reduction in CV mortality in unadjusted analysis [0.40 (0.28–0.57), p<0.01] and a trend for significant association in the adjusted model [0.69 (0.46–1.03), p=0.07]. Results for non-CV mortality was similar to that for all-cause mortality (Table 3). For all outcomes, there was no significant interaction by use of statin or any LLT.

Table 3.

Association of baseline LDL pattern with mortality over 5 years

Outcomes LDL
pattern		 Unadjusted
model
HR (95% CI)
P value 		Adjusted
model*
HR (95% CI)
P value 		Further
adjusted for
LDL-C
HR (95% CI)
P value 		Further
adjusted for
LDL-C + TG
HR (95% CI)
P value 		P
interaction
by use of
statin		 P interaction by
use of any LLT
All-cause mortality Pattern A/B 0.64 0.77 0.78 0.78 0.93 0.93
(0.51–0.80) (0.61–0.99) (0.61–1.00) (0.61–1.00)
P<0.01 P=0.04 P=0.05 P=0. 049
Pattern B 0.44 0.67 0.68 0.68
(0.35–0.56) (0.51–0.88) (0.52–0.89) (0.52–0.89)
P<0.01 P<0.01 P<0.01 P<0.01
Cardiovascular mortality Pattern A/B 0.78 0.94 0.95 0.95 0.62 0.58
(0.58–1.05) (0.67–1.33) (0.68–1.35) (0.67–1.34)
P=0.11 P=0.74 P=0.79 P=0.78
Pattern B 0.40 0.69 0.69 0.70
(0.28–0.57) (0.46–1.03) (0.46–1.04) (0.46–1.05)
P<0.01 P=0.07 P=0.07 P=0.08
Non-cardiovascular mortality Pattern A/B 0.51 0.64 0.65 0.65 0.89 0.73
(0.37–0.71) (0.45–0.91) (0.46–0.92) (0.46–0.92)
P<0.01 P=0.01 P=0.01 P=0.01
Pattern B 0.48 0.65 0.66 0.66
(0.35–0.66) (0.45–0.93) (0.46–0.95) (0.46–0.95)
P<0.01 P=0.02 P=0.02 P=0.03
Open in a new tab
*

Adjusted for GRACE risk score (age, history of congestive heart failure, history of prior MI, initial heart rate, initial systolic blood pressure, ST segment depression on electrocardiogram at admission, initial creatinine, elevated troponin and in-hospital percutaneous coronary intervention), sex, race, body mass index, history of hypertension, history of diabetes, history of alcohol abuse, current smoking, initial blood glucose, high-sensitivity C-reactive protein, metabolic syndrome, insurance status, presentation (ST segment elevation myocardial infarction vs. non-ST segment elevation myocardial infarction), in-hospital coronary artery bypass grafting and leisure time physical activity. All covariates are based on baseline, unless specified otherwise.

Expressed hazard ratios are compared with LDL pattern A. Interaction P values are based on the adjusted model*.

CI = confidence interval, GRACE = Global Registry of Acute Coronary Events, HR = hazard ratio, LDL = low-density lipoprotein, TG = triglycerides

To understand the relationship of LDL pattern with mortality, in additional analysis, we sequentially adjusted the model with LDL-C and triglycerides. This is because both LDL cholesterol23–25 and triglycerides rich remnant cholesterol levels39 were previously shown to have an inverse relationship with mortality after MI. Such analyses can help understand whether the association of LDL pattern was independent of LDL-C and triglycerides. When we further adjusted LDL-C level in the multivariable model, there was a significantly lower risk associated with pattern B [0.68 (0.52–0.89), p<0.01] and a trend for lower risk among patients with an A/B lipid profile [0.78 (0.61–1.00), p=0.05] for all-cause mortality (Table 3). For CV mortality, there was no significant association of pattern A/B, but there was a trend for lower CV mortality with pattern B [0.69 (0.46–1.04, p=0.07). Further adjustment with triglycerides showed similar results, except that the association of pattern A/B compared with pattern A for all-cause mortality became statistically significant [0.78 (0.61–1.00), p=0.049] (Table 3). Associations of LDL patterns A/B or B with non-CV mortality were still significant when the models were further adjusted for LDL-C or additionally for triglycerides (Table 3).

Analysis in patients with LDL-C <100 mg/dL

There were a total of 1460 patients with LDL-C <100 mg/dL (59% of total). Of these, 595 (40.7%), 424 (29.0%) and 441 (30.3%) had LDL patterns A, A/B and B, respectively. There were a total of 330 all-cause deaths and 162 CV-deaths over 5 years follow up in patients with LDL-C <100 mg/dL with an incident rate of 45 all-cause deaths and 22 CV-deaths per 1000 patients per year. The incident rate was 59, 40 and 31 for all-cause deaths and 28, 23 and 13 for CV deaths per 1000 patients per year in patients with LDL pattern A, A/B and B, respectively. Broadly the association of LDL patterns with outcomes was similar to that of the overall population (Table 4). For example, when LDL pattern B was compared with LDL pattern A, the HR (95% CI) were 0.48 (0.36–0.63) in unadjusted analysis and 0.73 (0.53–1.00) in a fully-adjusted model for all-cause mortality; and 0.42 (0.28–0.67) in unadjusted analysis and 0.71 (0.44–1.13) in fully-adjusted model for CV mortality.

Table 4.

Association of baseline LDL pattern with all-cause and cardiovascular mortality over 5 years in patients with LDL-C <100 mg/dL based on VAP

Outcomes LDL pattern Unadjusted model
HR (95% CI)		 Adjusted model*
HR (95% CI)
All-cause mortality Pattern A/B 0.63 (0.48–0.81) 0.78 (0.58–1.04)
P<0.01 P=0.09
Pattern B 0.48 (0.36–0.63) 0.73 (0.53–1.00)
P<0.01 P=0.05
Cardiovascular mortality Pattern A/B 0.75 (0.53–1.07) 0.93 (0.62–1.40)
P=0.11 P=0.74
Pattern B 0.42 (0.28–0.67) 0.71 (0.44–1.13)
P<0.01 P=0.15
Open in a new tab
*

Adjusted for GRACE risk score (age, history of congestive heart failure, history of prior MI, initial heart rate, initial systolic blood pressure, ST segment depression on electrocardiogram at admission, initial creatinine, elevated troponin and in-hospital percutaneous coronary intervention), sex, race, body mass index, history of hypertension, history of diabetes, history of alcohol abuse, current smoking, initial blood glucose, high-sensitivity C-reactive protein, metabolic syndrome, insurance status, presentation (ST segment elevation myocardial infarction vs. non-ST segment elevation myocardial infarction), in-hospital coronary artery bypass grafting and leisure time physical activity. All covariates are based on baseline, unless specified otherwise.

This analysis is restricted to patients whose LDL-C <100 mg/dL, which includes cholesterol in “real” LDL, lipoprotein (a) and intermediate density lipoprotein.

Expressed hazard ratios are compared with LDL pattern A.

CI = confidence interval, GRACE = Global Registry of Acute Coronary Events, HR = hazard ratio, LDL = low-density lipoprotein

Sensitivity analysis

Sensitivity analysis of LDL particles’ associated cholesterol ratio, cholesterol concentration of each LDL subclass, non-HDL-C, LDL-C and analysis restricted to patients not on any LLT at hospital admission showed broadly similar results (Supplemental Tables 2–6). The strengths of association of non-HDL-C were similar to that of LDL-C.

We included 1286 patients who volunteered to participate at 30 days follow up after the MI hospitalization (Figure 1). Of these, 272 (21.1%) had LDL pattern A, 425 (33.1%) had LDL pattern A/B and 589 (45.8%) had LDL pattern B. In analysis of 30 days LDL pattern, there were no significant associations of LDL pattern A/B when compared with LDL pattern A for all-cause or CV mortality in unadjusted or adjusted models (Supplemental Table 7). However, when LDL pattern B was compared with LDL pattern A, there were significant lower risk for both all-cause and CV deaths in unadjusted models, but not in adjusted models.

Discussion

In patients with recent MI, we found that compared with patients with LDL pattern A, those with LDL pattern B had a significant 32% lower relative risk associated for all-cause mortality and similar reduction in non-CV mortality, independent of LDL-C and triglycerides and a trend for lower CV mortality. We did not find significant interactions with use of either statins or LLT. Furthermore, we also found similar inverse associations of both non-HDL-C and directly measured LDL-C.

Small, dense LDL particles are considered to have greater susceptibility to oxidation,40 longer circulating half-life,41 easier binding to the endothelium,41 greater infiltration into sub-endothelial space,42 and higher inflammatory enzymatic activity.43 Several studies have reported that small, dense LDL particles8–13 as well as small, dense LDL-C33,34 are known to be associated with higher risk for incident CV events compared with large, buoyant LDL particles or cholesterol, while other studies of LDL size showed mixed results.14–16 However, there is paucity of data on LDL pattern after CV disease and prognosis. In a study of 200 patients with acute ischemic stroke, increased level of small, dense LDL particles was associated with higher odds for in-hospital mortality.13 On the other hand, in a nested case-control analysis from the Cholesterol and Recurrent Events trial, increasing LDL size was found to be an independent predictor of recurrent coronary events after MI in the placebo group, but not in those randomized to pravastatin.44 In a study of 1,643 patients not on any LLT referred for coronary angiography (65% had angiographically proven CHD), presence of large LDL was associated with higher risk for all-cause mortality when compared with intermediate sized LDL (HR 1.71, 95% CI: 1.31–2.25), but small LDL particles were not significantly associated with all-cause mortality (HR 1.24, 95% CI: 0.95–1.63).17 Both large and small LDLs were significantly associated with CV mortality, but the association was stronger for larger LDL than smaller LDL particles (HR: 1.89 vs. 1.54, respectively). In case-control analyses, small LDL was shown to have either a null association45 or a lower risk for CAD46 than large LDL. In our study, we did not find significant interaction by use of statin or any LLT, and an analysis restricted to patients not on LLT showed broadly similar results.

Several studies have shown a direct association of higher serum cholesterol levels with all-cause18–20 and CV mortality21,22 in patients with CHD, but other studies did not find a direct, significant relationship.47 Some studies have even shown inverse associations of LDL-C after MI with all-cause mortality.23–25 Our study further adds to the literature by using directly measured LDL-C, non-HDL-C and LDL pattern B. We found that the event rate was higher in patients with LDL-C <100 mg/dL than the overall population, which further substantiates our findings. A prior analysis from the TRIUMPH registry showed that in patients with MI, higher levels of remnant lipoprotein cholesterol levels, as defined by intermediate density lipoprotein cholesterol and the dense form of very low-density lipoprotein cholesterol, were associated with reduced all-cause mortality at 2 years.39

Although LDL-C was inversely associated with all-cause mortality, we found that LDL pattern B was significantly associated with lower all-cause mortality compared with pattern A, even after accounting for LDL-C and triglycerides, suggesting that the relationship of LDL pattern was beyond that explained by LDL-C and triglyceride levels. It is not clear if statins lower LDL particles more than LDL-C, because if they do, it is possible that statins may preferentially reduce more particles that are present with pattern B than with pattern A, which may explain our findings. However, it is unknown whether an inverse association is present with LDL particle numbers after MI. It is interesting to note that the strength of association of non-HDL-C to mortality was almost similar to that of LDL-C, unlike in primary prevention where non-HDL-C is considered to be more strongly associated with incident CHD than LDL-C.6,48

Despite the interesting finding, as an epidemiological study, our study can’t provide explanations for these results. It is possible that epidemiological biases, including index event bias, lead-time bias, selection bias and collider stratification bias, could explain the observed paradox of LDL pattern.39 Furthermore, it has been postulated that in patients with CHD, residual confounding may partly account for the paradoxical association of small LDL particles,17,44 LDL cholesterol,25 history of hypercholesterolemia,24 and remnant cholesterol39 with mortality, highlighting the limitations of observational study like ours in understanding causal pathways. Whether our finding is because of differences related to outcomes (i.e., initial ischemic CHD in primary prevention literature vs. mortality among MI survivor in the current study) is unknown. Not only did we find a trend for lower CV death, but we also found a lower risk for non-CV death in patients with LDL pattern B compared with pattern A, suggesting that the lower overall all-cause mortality is probably explained by both lower CV and non-CV mortality with LDL pattern B. A prior study showing increased association of large LDL with all-cause and CV mortality also showed an association with the highest levels of inflammatory markers, such as high-sensitive C-reactive protein and interleukin-6 in patients with large LDL particles, suggesting inflammation may partly explain such finding.17 Another possible explanation includes better nutritional reserve, as higher cholesterol level in malnutrition has been associated with lower risk for subsequent death,49,50 and as discussed earlier, confounding by LDL particle numbers. Further studies are needed to understand our finding, and future well-powered studies should also substantiate our findings, particularly for CV death. It should, however, be noted that there is overwhelming evidence that lowering cholesterol in patients with MI significantly reduces CV events,4 and nothing in our study contradicts this benefit. Therefore, CV prevention after MI should be approached according to the prevailing guidelines with aggressive lipid lowering strategy, adopting healthy lifestyle and control of other CV risk factors.

Strengths and Limitations

Since LDL can be an inverse acute phase reactant and can decrease at the time of MI,37 this might have affected our findings, although it is convenient to obtain lipid assessments at the time of an MI and our data facilitates the interpretation of such tests. Furthermore, it has been shown that during acute phase response, lipoproteins tend to be denser.51 However, we found a similar overall trend in sensitivity analyses, including an analysis showing lower event rates with B patterns from blood samples collected 30 days after acute MI. Additionally, although LDL can decrease during MI, it has been shown that acute MI does not affect cholesterol ratios.52 Thus our analysis of LDL particles’ associated cholesterol ratio is less likely to be affected by the acute phase response related to acute MI. Because only a subset of the overall TRIUMPH study participants volunteered to participate in blood work analysis, selection bias is possible. However, we did not find substantial differences in characteristics between patients who participated in optional blood work analysis at baseline from those who did not participate and there is no a priori reason to believe that LDL patterns are associated with patients’ willingness to have participated in the blood collection sub-study. Most importantly, there was no significant difference in either all-cause or CV mortality. We did not have the LDL particle concentrations to examine its association with mortality. This is important because when LDL particles are accounted for, LDL size and density are no longer significantly associated with incident CV disease.6,16 To address this, we used non-HDL-C, which has been shown to be highly correlated with total atherogenic particles,6,36 and we found similar association with non-HDL-C. Several of our sensitivity analyses may not have adequate statistical power. Although LDL pattern assignment by VAP is based on specific LDL peak max time cutoffs, it also depends on a preponderance of LDL subclasses and therefore analyses of LDL pattern as a proxy of size and density may not be very accurate. Despite these limitations, our study is the first prospective cohort study in patients with MI examining association of LDL pattern, as well as non-HDL-C and directly measured LDL-C, with mortality after MI. Other strengths of our study include presence of directly measured lipoproteins and associated cholesterol as well as other rigorously collected detail patient information.

Conclusions

In patients with MI, presence of type B LDL pattern, as compared with pattern A, was associated with lower risk for all-cause and non-CV mortality independent of LDL-C and triglycerides, and there was a trend for lower risk for CV-mortality. Use of either statin or LLT did not modify these associations. Inverse associations were seen also with directly measured LDL-C and non-HDL-C. Further studies are needed to confirm these findings and to illuminate all the mechanisms responsible for these associations.

Supplementary Material

  • LDL pattern B is strongly associated with incident CHD than A, but association for death after MI is unknown

  • We assessed prognosis for all-cause and CV-death within LDL patterns after MI

  • Pattern B was independently associated with lower risk for all-cause death than A

  • This was independent of LDL-C and triglycerides

  • Results were similar for non-CV death with similar trend for CV-death

Acknowledgments

Study support: The TRIUMPH study was funded by National Heart, Lung, and Blood Institute grant (P50 HL077113) and CV Outcomes, Kansas City, Missouri.

Drs. Pokharel, Patel and Qintar are supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under Award Number T32HL110837.

Abbreviations

CHD

coronary heart disease

CV

cardiovascular

GRACE

Global Registry of Acute Coronary Events

LDL (C)

low density lipoprotein (cholesterol)

LLT

lipid lowering therapy

Lp (a)

lipoprotein (a)

MI

myocardial infarction

Non-HDL-C

non-high density lipoprotein cholesterol

TRIUMPH

Translational Research Investigating Underlying disparities in acute Myocardial infarction Patients’ Health Status

VAP

vertical auto profile

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Disclosure of potential conflicts of interest.

Dr. O’Keefe is on the speaker’s bureau for Amgen and Regeneron, and has a major ownership interest in Cardiotabs.

Dr. Kulkarni was a former employee of Atherotech Diagnostics Lab

Dr. Jones is a chief scientific officer of the National Lipid Association

Dr. Martin is a co-inventor on a pending patent filed by Johns Hopkins University for novel LDL-C estimation. He has served as a consultant to Quest Diagnostics, Sanofi-Regeneron, Amgen, Pressed Juicery, Abbott Nutrition, and Pew Institute. He has received honoraria from the American College of Cardiology for lipid related activities. He has received research support from the American Heart Association, Aetna Foundation, CASCADE FH, Google, and Apple.

Dr. Virani is supported by the Department of Veterans Affairs Health Services Research and Development Service (HSR&D), American Heart Association Beginning-Grant-in-Aid, the American Diabetes Association Clinical Science and Epidemiology Award, and Baylor College of Medicine’s Global Initiatives (Paid to institution and not individual). He serves on the steering committee (no financial remuneration) for the Patient and Provider Assessment of Lipid Management (PALM) Registry at the Duke Clinical Research Institute. Honorarium from the American College of Cardiology for service as an Associate Editor for Innovations, acc.org

Dr. Spertus has received grant funding from Patient-Centered Outcomes Research Institute (PCORI), Abbott Vascular, Lilly; has an equity interest in Health Outcomes Sciences.

The remaining authors have no relevant relationships to disclose.

Author Contributions: Drs. Pokharel and Tang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Pokharel, Tang, Bhardwaj, Kulkarni, Martin, Virani, Spertus

Acquisition, analysis, or interpretation of data: All authors

Drafting of the manuscript: Pokharel

Critical revision of the manuscript for important intellectual content: All authors

Statistical analysis: Tang

Administrative, technical, or material support: Spertus,

Study supervision: Pokharel

Approval of the final version: All authors

References

  • 1.Cohen JC, Boerwinkle E, Mosley TH, Jr, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med. 2006;354(12):1264–1272. doi: 10.1056/NEJMoa054013. [DOI] [PubMed] [Google Scholar]
  • 2.Ference BA, Majeed F, Penumetcha R, Flack JM, Brook RD. Effect of naturally random allocation to lower low-density lipoprotein cholesterol on the risk of coronary heart disease mediated by polymorphisms in NPC1L1, HMGCR, or both: a 2 × 2 factorial Mendelian randomization study. J Am Coll Cardiol. 2015;65(15):1552–1561. doi: 10.1016/j.jacc.2015.02.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Ford ES, Ajani UA, Croft JB, et al. Explaining the decrease in U.S. deaths from coronary disease, 1980–2000. N Engl J Med. 2007;356(23):2388–2398. doi: 10.1056/NEJMsa053935. [DOI] [PubMed] [Google Scholar]
  • 4.Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25 Pt B):2889–2934. doi: 10.1016/j.jacc.2013.11.002. [DOI] [PubMed] [Google Scholar]
  • 5.Libby P. The forgotten majority: unfinished business in cardiovascular risk reduction. J Am Coll Cardiol. 2005;46(7):1225–1228. doi: 10.1016/j.jacc.2005.07.006. [DOI] [PubMed] [Google Scholar]
  • 6.Pokharel Y, Negi S, Virani SS. Cholesterol: Concentration, Ratio, and Particle Number. 2. Philadelphia: Saunders; 2014. Clinical Lipidology: A Companion to Braunwald’s Heart Disease. Chapter 8. [Google Scholar]
  • 7.Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. The Journal of clinical investigation. 1991;88(6):1785–1792. doi: 10.1172/JCI115499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Lamarche B, Tchernof A, Moorjani S, et al. Small, dense low-density lipoprotein particles as a predictor of the risk of ischemic heart disease in men. Prospective results from the Quebec Cardiovascular Study. Circulation. 1997;95(1):69–75. doi: 10.1161/01.cir.95.1.69. [DOI] [PubMed] [Google Scholar]
  • 9.Gardner CD, Fortmann SP, Krauss RM. Association of small low-density lipoprotein particles with the incidence of coronary artery disease in men and women. Jama. 1996;276(11):875–881. [PubMed] [Google Scholar]
  • 10.El Harchaoui K, van der Steeg WA, Stroes ES, et al. Value of low-density lipoprotein particle number and size as predictors of coronary artery disease in apparently healthy men and women: the EPIC-Norfolk Prospective Population Study. J Am Coll Cardiol. 2007;49(5):547–553. doi: 10.1016/j.jacc.2006.09.043. [DOI] [PubMed] [Google Scholar]
  • 11.Kamigaki AS, Siscovick DS, Schwartz SM, et al. Low density lipoprotein particle size and risk of early-onset myocardial infarction in women. American journal of epidemiology. 2001;153(10):939–945. doi: 10.1093/aje/153.10.939. [DOI] [PubMed] [Google Scholar]
  • 12.Blake GJ, Otvos JD, Rifai N, Ridker PM. Low-density lipoprotein particle concentration and size as determined by nuclear magnetic resonance spectroscopy as predictors of cardiovascular disease in women. Circulation. 2002;106(15):1930–1937. doi: 10.1161/01.cir.0000033222.75187.b9. [DOI] [PubMed] [Google Scholar]
  • 13.Zeljkovic A, Vekic J, Spasojevic-Kalimanovska V, et al. LDL and HDL subclasses in acute ischemic stroke: prediction of risk and short-term mortality. Atherosclerosis. 2010;210(2):548–554. doi: 10.1016/j.atherosclerosis.2009.11.040. [DOI] [PubMed] [Google Scholar]
  • 14.Mykkanen L, Kuusisto J, Haffner SM, Laakso M, Austin MA. LDL size and risk of coronary heart disease in elderly men and women. Arteriosclerosis, thrombosis, and vascular biology. 1999;19(11):2742–2748. doi: 10.1161/01.atv.19.11.2742. [DOI] [PubMed] [Google Scholar]
  • 15.Otvos JD, Collins D, Freedman DS, et al. Low-density lipoprotein and high-density lipoprotein particle subclasses predict coronary events and are favorably changed by gemfibrozil therapy in the Veterans Affairs High-Density Lipoprotein Intervention Trial. Circulation. 2006;113(12):1556–1563. doi: 10.1161/CIRCULATIONAHA.105.565135. [DOI] [PubMed] [Google Scholar]
  • 16.Mora S, Otvos JD, Rifai N, Rosenson RS, Buring JE, Ridker PM. Lipoprotein particle profiles by nuclear magnetic resonance compared with standard lipids and apolipoproteins in predicting incident cardiovascular disease in women. Circulation. 2009;119(7):931–939. doi: 10.1161/CIRCULATIONAHA.108.816181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Grammer TB, Kleber ME, Marz W, et al. Low-density lipoprotein particle diameter and mortality: the Ludwigshafen Risk and Cardiovascular Health Study. Eur Heart J. 2015;36(1):31–38. doi: 10.1093/eurheartj/ehu055. [DOI] [PubMed] [Google Scholar]
  • 18.Schlant RC, Forman S, Stamler J, Canner PL. The natural history of coronary heart disease: prognostic factors after recovery from myocardial infarction in 2789 men. The 5-year findings of the coronary drug project. Circulation. 1982;66(2):401–414. doi: 10.1161/01.cir.66.2.401. [DOI] [PubMed] [Google Scholar]
  • 19.Ulvenstam G, Bergstrand R, Johansson S, et al. Prognostic importance of cholesterol levels after myocardial infarction. Preventive medicine. 1984;13(4):355–366. doi: 10.1016/0091-7435(84)90027-6. [DOI] [PubMed] [Google Scholar]
  • 20.Wong ND, Wilson PW, Kannel WB. Serum cholesterol as a prognostic factor after myocardial infarction: the Framingham Study. Annals of internal medicine. 1991;115(9):687–693. doi: 10.7326/0003-4819-115-9-687. [DOI] [PubMed] [Google Scholar]
  • 21.Frost PH, Verter J, Miller D. Serum lipids and lipoproteins after myocardial infarction: associations with cardiovascular mortality and experience in the Aspirin Myocardial Infarction Study. Am Heart J. 1987;113(6):1356–1364. doi: 10.1016/0002-8703(87)90648-x. [DOI] [PubMed] [Google Scholar]
  • 22.Pekkanen J, Linn S, Heiss G, et al. Ten-year mortality from cardiovascular disease in relation to cholesterol level among men with and without preexisting cardiovascular disease. N Engl J Med. 1990;322(24):1700–1707. doi: 10.1056/NEJM199006143222403. [DOI] [PubMed] [Google Scholar]
  • 23.Reddy VS, Bui QT, Jacobs JR, Begelman SM, Miller DP, French WJ. Relationship between serum low-density lipoprotein cholesterol and in-hospital mortality following acute myocardial infarction (the lipid paradox) Am J Cardiol. 2015;115(5):557–562. doi: 10.1016/j.amjcard.2014.12.006. [DOI] [PubMed] [Google Scholar]
  • 24.Wang TY, Newby LK, Chen AY, et al. Hypercholesterolemia paradox in relation to mortality in acute coronary syndrome. Clin Cardiol. 2009;32(9):E22–28. doi: 10.1002/clc.20518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cho KH, Jeong MH, Ahn Y, et al. Low-density lipoprotein cholesterol level in patients with acute myocardial infarction having percutaneous coronary intervention (the cholesterol paradox) Am J Cardiol. 2010;106(8):1061–1068. doi: 10.1016/j.amjcard.2010.06.009. [DOI] [PubMed] [Google Scholar]
  • 26.Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III) JAMA. 2001;285(19):2486–2497. doi: 10.1001/jama.285.19.2486. [DOI] [PubMed] [Google Scholar]
  • 27.Arnold SV, Chan PS, Jones PG, et al. Translational Research Investigating Underlying Disparities in Acute Myocardial Infarction Patients’ Health Status (TRIUMPH): design and rationale of a prospective multicenter registry. Circ Cardiovasc Qual Outcomes. 2011;4(4):467–476. doi: 10.1161/CIRCOUTCOMES.110.960468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Kulkarni KR, Garber DW, Marcovina SM, Segrest JP. Quantification of cholesterol in all lipoprotein classes by the VAP-II method. Journal of lipid research. 1994;35(1):159–168. [PubMed] [Google Scholar]
  • 29.Kulkarni KR. Cholesterol profile measurement by vertical auto profile method. Clinics in laboratory medicine. 2006;26(4):787–802. doi: 10.1016/j.cll.2006.07.004. [DOI] [PubMed] [Google Scholar]
  • 30.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clinical chemistry. 1972;18(6):499–502. [PubMed] [Google Scholar]
  • 31.Instruction Manual, Centers for Disease Control and Prevention: Part 9. [Accessed March 3, 2017];ICD-10 Cause-of-Death Lists for Tabulating Mortality Statistics. https://www.cdc.gov/nchs/data/dvs/im9_2002.pdf.pdf.
  • 32.Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. Jama. 2004;291(22):2727–2733. doi: 10.1001/jama.291.22.2727. [DOI] [PubMed] [Google Scholar]
  • 33.Hoogeveen RC, Gaubatz JW, Sun W, et al. Small dense low-density lipoprotein-cholesterol concentrations predict risk for coronary heart disease: the Atherosclerosis Risk In Communities (ARIC) study. Arteriosclerosis, thrombosis, and vascular biology. 2014;34(5):1069–1077. doi: 10.1161/ATVBAHA.114.303284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Tsai MY, Steffen BT, Guan W, et al. New automated assay of small dense low-density lipoprotein cholesterol identifies risk of coronary heart disease: the Multi-ethnic Study of Atherosclerosis. Arteriosclerosis, thrombosis, and vascular biology. 2014;34(1):196–201. doi: 10.1161/ATVBAHA.113.302401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ahmed HM, Elshazly MB, Martin SS, Blaha MJ, Kulkarni KR, Jones SR. Ratio of dense to buoyant LDL subclass is associated with LDL density phenotype (VLDL-5) Open Chem Biomed Methods J. 2013;6:1–5. [Google Scholar]
  • 36.Ballantyne CM, Andrews TC, Hsia JA, Kramer JH, Shear C. Correlation of non-high-density lipoprotein cholesterol with apolipoprotein B: effect of 5 hydroxymethylglutaryl coenzyme A reductase inhibitors on non-high-density lipoprotein cholesterol levels. Am J Cardiol. 2001;88(3):265–269. doi: 10.1016/s0002-9149(01)01638-1. [DOI] [PubMed] [Google Scholar]
  • 37.Pitt B, Loscalzo J, Ycas J, Raichlen JS. Lipid levels after acute coronary syndromes. J Am Coll Cardiol. 2008;51(15):1440–1445. doi: 10.1016/j.jacc.2007.11.075. [DOI] [PubMed] [Google Scholar]
  • 38.Austin PC, Lee DS, Fine JP. Introduction to the Analysis of Survival Data in the Presence of Competing Risks. Circulation. 2016;133(6):601–609. doi: 10.1161/CIRCULATIONAHA.115.017719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Martin SS, Faridi KF, Joshi PH, et al. Remnant Lipoprotein Cholesterol and Mortality After Acute Myocardial Infarction: Further Evidence for a Hypercholesterolemia Paradox From the TRIUMPH Registry. Clin Cardiol. 2015;38(11):660–667. doi: 10.1002/clc.22470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Chait A, Brazg RL, Tribble DL, Krauss RM. Susceptibility of small, dense, low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med. 1993;94(4):350–356. doi: 10.1016/0002-9343(93)90144-e. [DOI] [PubMed] [Google Scholar]
  • 41.Galeano NF, Al-Haideri M, Keyserman F, Rumsey SC, Deckelbaum RJ. Small dense low density lipoprotein has increased affinity for LDL receptor-independent cell surface binding sites: a potential mechanism for increased atherogenicity. Journal of lipid research. 1998;39(6):1263–1273. [PubMed] [Google Scholar]
  • 42.Nordestgaard BG, Nielsen LB. Atherosclerosis and arterial influx of lipoproteins. Current opinion in lipidology. 1994;5(4):252–257. doi: 10.1097/00041433-199408000-00002. [DOI] [PubMed] [Google Scholar]
  • 43.Karabina SA, Liapikos TA, Grekas G, Goudevenos J, Tselepis AD. Distribution of PAF-acetylhydrolase activity in human plasma low-density lipoprotein subfractions. Biochimica et biophysica acta. 1994;1213(1):34–38. doi: 10.1016/0005-2760(94)90219-4. [DOI] [PubMed] [Google Scholar]
  • 44.Campos H, Moye LA, Glasser SP, Stampfer MJ, Sacks FM. Low-density lipoprotein size, pravastatin treatment, and coronary events. Jama. 2001;286(12):1468–1474. doi: 10.1001/jama.286.12.1468. [DOI] [PubMed] [Google Scholar]
  • 45.Sherrard B, Simpson H, Cameron J, Wahi S, Jennings G, Dart A. LDL particle size in subjects with previously unsuspected coronary heart disease: relationship with other cardiovascular risk markers. Atherosclerosis. 1996;126(2):277–287. doi: 10.1016/0021-9150(96)05920-5. [DOI] [PubMed] [Google Scholar]
  • 46.Campos H, Roederer GO, Lussier-Cacan S, Davignon J, Krauss RM. Predominance of large LDL and reduced HDL2 cholesterol in normolipidemic men with coronary artery disease. Arteriosclerosis, thrombosis, and vascular biology. 1995;15(8):1043–1048. doi: 10.1161/01.atv.15.8.1043. [DOI] [PubMed] [Google Scholar]
  • 47.Kronmal RA, Kosinski AS, Mock MB. The relationship between cholesterol level and myocardial infarction or mortality risk in patients with coronary artery disease A report from the Coronary Artery Surgery Study (CASS) registry. Annals of epidemiology. 1992;2(1–2):129–136. doi: 10.1016/1047-2797(92)90047-t. [DOI] [PubMed] [Google Scholar]
  • 48.Di Angelantonio E, Sarwar N, Perry P, et al. Major lipids, apolipoproteins, and risk of vascular disease. Jama. 2009;302(18):1993–2000. doi: 10.1001/jama.2009.1619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Liu Y, Coresh J, Eustace JA, et al. Association between cholesterol level and mortality in dialysis patients: role of inflammation and malnutrition. Jama. 2004;291(4):451–459. doi: 10.1001/jama.291.4.451. [DOI] [PubMed] [Google Scholar]
  • 50.Iseki K, Yamazato M, Tozawa M, Takishita S. Hypocholesterolemia is a significant predictor of death in a cohort of chronic hemodialysis patients. Kidney international. 2002;61(5):1887–1893. doi: 10.1046/j.1523-1755.2002.00324.x. [DOI] [PubMed] [Google Scholar]
  • 51.Cabana VG, Siegel JN, Sabesin SM. Effects of the acute phase response on the concentration and density distribution of plasma lipids and apolipoproteins. Journal of lipid research. 1989;30(1):39–49. [PubMed] [Google Scholar]
  • 52.Wattanasuwan N, Khan IA, Gowda RM, Vasavada BC, Sacchi TJ. Effect of acute myocardial infarction on cholesterol ratios. Chest. 2001;120(4):1196–1199. doi: 10.1378/chest.120.4.1196. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

NIHMS910549-supplement-supplement_1.pdf (59.8KB, pdf)

RESOURCES