Association of High Coronary Heart Disease Risk Status With Circulating Oxidized LDL in the Well-Functioning Elderly
Findings From the Health, Aging, and Body Composition Study
Objective— Although circulating oxidized LDL (oxLDL) is elevated in persons with coronary heart disease (CHD), whether oxLDL is elevated in persons with high CHD risk before any events is unknown. Therefore, we studied the association between high, predicted CHD risk and oxLDL in the Health ABC cohort.
Methods and Results— This cohort included 385 persons with CHD and 1183 persons at high risk; the latter were all persons with CHD risk equivalents: noncoronary forms of clinical atherosclerotic disease, diabetes, and a 10-year risk for CHD >20% by Framingham scoring. The remaining 1535 participants were at low risk. Levels of oxLDL were 1.18±0.61 mg/dL for low-risk persons, 1.50±0.81 mg/dL for high-risk persons without diagnosed CHD, and 1.32±0.83 mg/dL for persons with CHD (P<0.001). The odds ratio for high CHD risk in the highest quintile of oxLDL, compared with the lowest quintile and after adjusting for age, sex, race, LDL cholesterol, smoking status, and C-reactive protein, was 2.79 (P<0.001).
Conclusion— The odds ratio for elevated oxLDL among persons with high CHD risk before any CHD events was higher than that among persons with established CHD. A likely explanation is that once CHD is diagnosed, individuals are frequently treated with a statin, which is associated with lowering of LDL cholesterol and oxLDL levels.
Oxidized LDL (oxLDL) has been shown to play an important role in the pathogenesis of atherosclerosis.1–3⇓⇓ Recently, we, among others, have demonstrated elevated levels of circulating oxLDL in association with coronary heart disease (CHD).4–6⇓⇓ In middle-aged persons, receiver operating characteristic-curve analysis revealed that oxLDL had a higher sensitivity for coronary artery disease than did LDL cholesterol and the total-to-HDL cholesterol ratio. Furthermore, logistic regression analysis revealed that the predictive value of oxLDL was additive to that of the global risk assessment score for cardiovascular risk prediction that is based on Framingham risk factors: age, total cholesterol, HDL cholesterol, systolic blood pressure, diabetes mellitus, and smoking. Ninety-four subjects with high (exceeding the 90th percentile of the distribution in control subjects) circulating oxLDL levels and a high global risk assessment score had angiographically proven coronary artery disease.7 The question remained whether high CHD risk before any cardiovascular event is associated with a higher prevalence of elevated oxLDL. We studied this association in 3033 participants in the Health, Aging, and Body Composition Study.
A useful concept introduced in the third report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (ATP III) to assist in the selection of patients at high CHD risk is that of CHD risk equivalents.8 We have used this scoring system to identify persons at high CHD risk. We have studied the relation of established CHD and CHD risk equivalents with circulating oxLDL levels in the well-functioning elderly. Because CHD risk equivalents were associated with a higher prevalence of elevated oxLDL, we studied the association of the different components of risk categorization: high Framingham score, diabetes, and noncoronary forms of cardiovascular disease. Because a relation between C-reactive protein (CRP), a marker of inflammation, and the development of atherosclerotic disease has been observed,9,10⇓ we also studied the association between circulating oxLDL and CRP.
The study population included 3033 participants in the baseline examination of the Health, Aging, and Body Composition Study, a prospective cohort study of 3075 well-functioning, 70- to 79-year-old persons in the Memphis, Tenn, and Pittsburgh, Pa, vicinities; 42 participants were excluded from analysis because of missing data. Participants were recruited from a random sample of white and all black Medicare-eligible adults living in the 2 study areas from March 1997 to June 1998. “Well-functioning” was determined by self-report and was defined as having no difficulty in either walking one quarter mile or going up 10 steps without resting. Exclusion criteria included difficulties with activities of daily living, obvious cognitive impairment, inability to communicate with the interviewer, intention of moving within 3 years, therapy for cancer within the prior 3 years, or participation in a trial involving a lifestyle intervention.
Persons with CHD (n=385) were all who had myocardial infarction, angina, coronary angioplasty, or coronary artery bypass surgery (Table 1), based on cardiac procedures and diagnoses reported to the Centers for Medicare and Medicaid Services in the 5 years before enrollment in the study.
According to the ATPIII, persons with CHD risk equivalents have noncoronary forms of clinical atherosclerotic disease, diabetes, and/or a 10-year risk for CHD events >20% by Framingham scoring.8 Noncoronary forms of atherosclerotic disease were peripheral arterial disease, abdominal aortic aneurysm, or carotid artery disease. Peripheral arterial disease was defined as claudication by the Rose questionnaire or by an abnormal ankle-brachial blood pressure index (<0.9)11 (Table 1). Persons with carotid artery disease had a history of carotid endarterectomy, stroke of carotid origin, or transient ischemic attacks (Table 1). Sixty percent of persons with cardiovascular disease were treated with antianginal drugs, 42% with antiarrhythmic drugs, 35% with vasodilators, and 46% with antiplatelet drugs (including aspirin), 90% with blood pressure-lowering drugs, and 21% with lipid-lowering drugs.
Persons in the low-risk group were all persons without diagnosed CHD or CHD equivalents. The mean Framingham score was 9.0±2.7 in men, corresponding to a calculated risk of 13%, and was 9.7±3.8 in women, corresponding to a calculated risk of only 4%.
Hypercholesterolemic persons had a total cholesterol level ≥240 mg/dL and/or an LDL cholesterol level ≥130 mg/dL or were being treated with lipid-lowering drugs. Dyslipidemic persons had HDL cholesterol levels <35 mg/dL, fasting triglyceride levels >200 mg/dL, and/or a triglyceride-to-HDL molar ratio >1.33, irrespective of their LDL cholesterol levels. Hypertensive persons had a systolic pressure ≥140 mm Hg and/or a diastolic pressure ≥90 mm Hg when untreated or were being treated with blood pressure-lowering drugs. Diabetics had fasting glucose levels >126 mg/dL whether they were being treated or not. American Diabetes Association criteria were used for diagnosis of glucose intolerance.
Blood samples were stored at −80°C. Total and HDL cholesterol and serum triglyceride levels were measured on a commercially available analyzer (Vitros 950, Johnson & Johnson). The inter-assay coefficient of variation was 1.5% for total cholesterol, 2.3% for triglycerides, and 2.3% for HDL cholesterol. LDL cholesterol levels were calculated by the Friedewald equation.12 CRP was measured by ELISA, based on purified protein and polyclonal anti-CRP antibodies (Calbiochem).9,13⇓ The CRP assay was standardized according to the World Health Organization First International Reference Standard with a sensitivity of 0.08 μg/mL. Levels of oxLDL were measured (2000 to 2001) blindly at the Center for Experimental Surgery and Anesthesiology, and data were transferred to the Health ABC Study Directory. A monoclonal antibody (4E6)-based competition ELISA was used for measuring plasma levels of oxLDL.5,14,15⇓⇓ The monoclonal antibody 4E6 is directed against a conformational epitope in the apolipoprotein B-100 moiety of LDL that is generated as a consequence of substitution of at least 60 lysine residues of apolipoprotein B-100 with aldehydes. This number of substituted lysines corresponds to the minimal number of substituted lysines required for scavenger-mediated uptake of oxLDL. Substituting aldehydes can be produced by peroxidation of lipids of LDL, resulting in the generation of oxLDL. Aldehydes that are released by endothelial cells under oxidative stress or by activated platelets might also induce the oxidative modification of apolipoprotein B-100 in the absence of peroxidation of lipids of LDL. The specificity of this assay is excellent, because the concentrations that are required to obtain 50% inhibition of antibody binding in the ELISA are 25 mg/dL for native LDL and 0.025 mg/dL for oxLDL with at least 60 aldehyde-substituted lysines per apolipoprotein B-100. Therefore, 160 mg/dL LDL would only contribute <0.2 mg/dL in the oxLDL assay. The interassay coefficient of variation of oxLDL is 12%.
A nonparametric Kruskal-Wallis test was used for comparing continuous variables, and Fisher’s exact test was used for analysis of contingency tables. Stepwise multivariate regression analysis (SPSS, version 10.0 for Windows) was used to study the relation of levels of circulating oxLDL with age, sex, race, and cardiovascular risk factors: hypertension, hypercholesterolemia, dyslipidemia, impaired glucose tolerance and diabetes, inflammation (CRP), and smoking status. Logistic regression analysis was performed to determine odds ratios (ORs) for elevated oxLDL in association with an elevated triglyceride-to-HDL molar ratio and CRP. ORs for the prevalence of CHD risk equivalents and established CHD in the highest compared with the lowest quintile of oxLDL were determined by logistic regression analysis. ORs were adjusted for age, sex, race, LDL cholesterol, smoking status, and CRP. P<0.05 was considered statistically significant.
Table 2 shows the characteristics of 3033 participants according to their CHD status. Compared with persons at low CHD risk, there were more men and smokers among persons with established CHD and persons with CHD risk equivalents, and they reported more impaired glucose tolerance, diabetes, hypercholesterolemia, dyslipidemia (as evidenced by high fasting triglyceride levels and triglyceride-to-HDL molar ratios), and hypertension. They also had higher levels of CRP and of oxLDL (Table 2).
The oxLDL measure varied across categories of CHD risk, with the lowest levels among those with low CHD risk and the highest among those with CHD risk equivalents. Persons with established CHD had lower levels of oxLDL than did persons with CHD risk equivalents with no prior CHD events (P<0.001). The former were more often being treated with a 3-hydroxymethyl 3-methylglutaryl coenzyme A reductase inhibitor (statin). Treatment with this drug class was inversely related to LDL cholesterol (mean difference, −14 mg/dL; P<0.001) and to oxLDL (mean difference, −0.14 mg/dL; P<0.001). CRP levels were 2.50±3.13 mg/L for persons on statin therapy compared with 3.09±4.96 mg/L (P>0.001) among untreated persons.
In stepwise multivariate regression analysis, hypercholesterolemia (P<0.0001), dyslipidemia (P<0.0001), and CRP (P<0.0001) were the strongest independent predictors of levels of circulating oxLDL (Table 3). The same relations were observed in persons without and with CHD. Logistic regression analysis confirmed that higher levels of oxLDL were related to a higher triglyceride-to-HDL molar ratio and higher levels of CRP. Compared with the lowest quintile of the triglyceride-to-HDL ratio, age-, sex-, race-, and LDL cholesterol-adjusted ORs for elevated oxLDL (at a cutoff value exceeding the 90th percentile of the distribution among persons in the lowest quintile) in the next 4 quintiles were as follows: 1.13 (95% confidence interval [CI], 0.86 to 1.46), 1.52 (95% CI, 1.20 to 1.91), 1.78 (95% CI, 1.42 to 2.25), and 3.01 (95% CI, 2.39 to 3.78), respectively. Compared with the lowest quintile of CRP, adjusted ORs for oxLDL were 1.35 (95% CI, 1.08 to 1.68), 1.58 (95% CI, 1.29 to 1.94), 1.79 (95% CI, 1.45 to 2.22), and 1.93 (95% CI, 1.59 to 2.35) respectively.
Logistic regression analysis was performed to study the relation of high CHD risk and established CHD with oxLDL. Table 4 shows that the OR for high CHD risk in the highest quintile of oxLDL, compared with the lowest quintile and after adjusting for age, sex, and race, was 4.37 (P<0.001). The OR for established CHD was 2.02. After additional adjustment for smoking and LDL cholesterol, these values were 3.04 (P<0.001) and 2.36 (P<0.001), respectively. After additional adjustment for CRP, the values were 2.99 (P<0.001) and 2.33 (P<0.001), respectively. The association of both high CHD risk and established CHD with elevated oxLDL was consistent across both race and sex. Corresponding ORs for LDL cholesterol were 1.007 (95% CI, 1.004 to 1.010) and 0.995 (95% CI, 0.991 to 0.999), respectively.
In persons without prior CHD events, a high Framingham score (corresponding to a predicted 10-year risk for CHD >20%) was a better predictor of elevated oxLDL than was diabetes and the occurrence of noncoronary forms of cardiovascular disease. The OR for elevated oxLDL in persons with a high Framingham score, compared with persons with a low Framingham score (corresponding to a predicted 10-year risk for CHD <20%) and after adjusting for age, sex, race, smoking, and LDL cholesterol, was 3.00 (95% CI, 2.09 to 4.32). A high Framingham score was associated with a higher age-, sex-, and race-adjusted prevalence of noncoronary forms of cardiovascular disease (OR, 2.08; 95% CI, 1.47 to 2.93) and of CHD (OR, 1.45; 95% CI, 1.24 to 1.71).
In the well-functioning elderly, a high CHD risk before the occurrence of any CHD event was associated with higher levels of circulating oxLDL. This association persisted after adjustment for LDL cholesterol levels, CRP, and body mass index. A high Framingham score (corresponding to a predicted 10-year risk for CHD >20%), which was associated with a higher prevalence of both noncoronary forms of cardiovascular disease and CHD, was also associated with a higher prevalence of elevated oxLDL. The weak association between LDL cholesterol and cardiovascular risk status in our cross-sectional study is in agreement with previous findings that LDL cholesterol is a poor predictor of vascular risk in elderly persons.16 The OR for elevated oxLDL among persons with high CHD risk before any CHD events was higher than that among persons with established CHD. A likely explanation is that once CHD is diagnosed, persons are more likely to be treated with a statin, which is associated with a lowering of LDL cholesterol and oxLDL levels. Recently, a relation between CRP and the development of atherosclerotic disease has been observed in experimental17,18⇓ and epidemiological9,10,19⇓⇓ studies. The association between high CHD risk and elevated oxLDL persisted after additional adjustment for CRP. The strongest independent predictors of levels of circulating oxLDL were hypercholesterolemia, followed by dyslipidemia and inflammation (CRP). Similar associations were observed in persons without and with CHD.
Hypercholesterolemia, Dyslipidemia, and Oxidation of LDL
Because LDL is the oxidation substrate, the relation between hypercholesterolemia and increased oxLDL was expected. An important and new finding is that the association between high cardiovascular risk status and circulating oxLDL persisted after adjustment for LDL cholesterol and that this association was stronger than that for LDL cholesterol. Atherogenic dyslipidemia is characterized by 3 lipid abnormalities: elevated triglycerides, reduced HDL cholesterol, and small LDL particle size.20 LDL is heterogeneous in terms of size and density.21 Cross-sectional and prospective studies have shown that small, dense LDL particles are more atherogenic than are larger LDL particles.22–24⇓⇓ Small, dense LDL particles are also particularly prone to oxidation, providing a possible mechanism for their atherogenicity.25 In the Health ABC study, the prevalence of small, dense LDL was not studied directly. Because of the complexity of the techniques for analyzing LDL subfractions, they are not likely to be used in large population studies. Therefore, attempts have been made to predict LDL size on the basis of a regular lipid profile. Recently, the triglyceride-to-HDL molar ratio was identified as a valid predictor of the small, dense LDL phenotype.26 The strong association between high triglyceride-to-HDL molar ratios and high levels of circulating oxLDL, even after adjustment for LDL cholesterol levels, thus suggests a strong association between the small, dense LDL phenotype and oxidation of LDL.
Inflammation and Oxidation of LDL
Here we demonstrate a positive association between CRP and the oxidation of LDL. Recently, an association between acute inflammation and increased oxidation of LDL has been demonstrated in 3 distinct, experimental models of infection and inflammation, suggesting that oxidation of LDL might be an important mechanism by which inflammation promotes atherosclerotic cardiovascular diseases.27 OxLDL was found to reduce the motility of macrophages and impair the platelet-activating factor-mediated acute inflammatory response in the arterial wall, thereby leading to development of a state of chronic, low-level inflammation that is consistent with the prolonged period required for the conversion of fatty streaks to more advanced atheromatous plaques.
In this study, intima-media thickness was not assessed by carotid ultrasound. This technique might have identified more persons with noncoronary forms of atherosclerotic disease. Because this was a cross-sectional study, we cannot determine whether the increased oxLDL is a cause or a consequence of atherosclerotic cardiovascular disease. At least our data suggest that oxLDL can be used to identify high-risk persons, even before events do occur. Upcoming longitudinal studies in the same population will help to clarify whether elevated oxLDL predicts definite CHD events. We also cannot determine the origin of the circulating oxLDL. We and others have shown a transient increase in circulating oxidatively modified LDL in patients with acute coronary syndromes, suggesting a relation between plaque instability and release of oxLDL into the blood.28,29⇓ In this cohort population study, baseline samples were analyzed. For participants with a history of an acute coronary event, it was not possible to determine with great accuracy the time between baseline blood sampling and the last acute coronary syndrome; therefore, it was not possible to determine this relation. The Health ABC study group had only limited resources for laboratory analyses and tried to focus on those measures that were known to be active in multiorgan models of risk factors and disease. Levels of apolipoprotein B100 and lipoprotein(a) were not measured.
In aggregate, this study shows for the first time in the general population that high CHD risk before the occurrence of CHD events is associated with high levels of circulating oxLDL, even after adjustment for LDL cholesterol. Our data suggest that circulating oxLDL can be a useful marker for identifying older persons at high cardiovascular risk.
The National Institutes of Health (N01-AG-6[hyphen]2103, N01-AG-6[hyphen]210, and N01-AG-6[hyphen]2106; Bethesda, Md), the Interuniversitaire Attractiepolen Programma of the Belgian Federal Government (P05/02), and the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (G.0263.01) supported this work. Peter Verhamme is an FWO Research Fellow. We thank M. Landeloos and Els Deridder for excellent technical support.
- Received January 28, 2003.
- Accepted May 13, 2003.
- ↵Goldstein JL, Brown MS. Molecular medicine: the cholesterol quartet. Science. 2001; 292: 1310–1312.
- ↵Mertens A, Holvoet P. Oxidized LDL and HDL: antagonists in atherothrombosis. FASEB J. 2001; 15: 2073–2084.
- ↵Ehara S, Ueda M, Naruko T, Haze K, Itoh A, Otsuka M, Komatsu R, Matsuo T, Itabe H, Takano T, Tsukamoto Y, Yoshiyama M, Takeuchi K, Yoshikawa J, Becker AE. Elevated levels of oxidized low density lipoprotein show a positive relationship with the severity of acute coronary syndromes. Circulation. 2001; 103: 1955–1960.
- ↵Holvoet P, Vanhaecke J, Janssens S, Van de WF, Collen D. Oxidized LDL and malondialdehyde-modified LDL in patients with acute coronary syndromes and stable coronary artery disease. Circulation. 1998; 98: 1487–1494.
- ↵Toshima S, Hasegawa A, Kurabayashi M, Itabe H, Takano T, Sugano J, Shimamura K, Kimura J, Michishita I, Suzuki T, Nagai R. Circulating oxidized low density lipoprotein levels: a biochemical risk marker for coronary heart disease. Arterioscler Thromb Vasc Biol. 2000; 20: 2243–2247.
- ↵Holvoet P, Mertens A, Verhamme P, Bogaerts K, Beyens G, Verhaeghe R, Collen D, Muls E, Van de WF. Circulating oxidized LDL is a useful marker for identifying patients with coronary artery disease. Arterioscler Thromb Vasc Biol. 2001; 21: 844–848.
- ↵Tracy RP, Lemaitre RN, Psaty BM, Ives DG, Evans RW, Cushman M, Meilahn EN, Kuller LH. Relationship of C-reactive protein to risk of cardiovascular disease in the elderly: results from the Cardiovascular Health Study and the Rural Health Promotion Project. Arterioscler Thromb Vasc Biol. 1997; 17: 1121–1127.
- ↵Newman AB, Shemanski L, Manolio TA, Cushman M, Mittelmark M, Polak JF, Powe NR, Siscovick D. Ankle-arm index as a predictor of cardiovascular disease and mortality in the Cardiovascular Health Study: the Cardiovascular Health Study Group. Arterioscler Thromb Vasc Biol. 1999; 19: 538–545.
- ↵Friedewald W, Levy R, Fredrickson D. Estimation of the concentration of low-density-lipoprotein cholesterol in plasma without the use of the ultracentrifuge. Clin Chem. 1972; 18: 449–502.
- ↵Macy EM, Hayes TE, Tracy RP. Variability in the measurement of C-reactive protein in healthy subjects: implications for reference intervals and epidemiological applications. Clin Chem. 1997; 43: 52–58.
- ↵Holvoet P, Stassen JM, Van Cleemput J, Collen D, Vanhaecke J. Oxidized low density lipoproteins in patients with transplant-associated coronary artery disease. Arterioscler Thromb Vasc Biol. 1998; 18: 100–107.
- ↵Cermak J, Key NS, Bach RR, Balla J, Jacob HS, Vercellotti GM. C-reactive protein induces human peripheral blood monocytes to synthesize tissue factor. Blood. 1993; 82: 513–520.
- ↵Torzewski J, Torzewski M, Bowyer DE, Frohlich M, Koenig W, Waltenberger J, Fitzsimmons C, Hombach V. C-reactive protein frequently colocalizes with the terminal complement complex in the intima of early atherosclerotic lesions of human coronary arteries. Arterioscler Thromb Vasc Biol. 1998; 18: 1386–1392.
- ↵Krauss RM, Burke DJ. Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res. 1982; 23: 97–104.
- ↵Campos H, Genest JJ Jr, Blijlevens E, McNamara JR, Jenner JL, Ordovas JM, Wilson PW, Schaefer EJ. Low density lipoprotein particle size and coronary artery disease. Arterioscler Thromb. 1992; 12: 187–195.
- ↵Lamarche B, Tchernof A, Moorjani S, Cantin B, Dagenais GR, Lupien PJ, Despres JP. 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: 69–75.
- ↵Chancharme L, Therond P, Nigon F, Lepage S, Couturier M, Chapman MJ. Cholesteryl ester hydroperoxide lability is a key feature of the oxidative susceptibility of small, dense LDL. Arterioscler Thromb Vasc Biol. 1999; 19: 810–820.
- ↵Boizel R, Benhamou PY, Lardy B, Laporte F, Foulon T, Halimi S. Ratio of triglycerides to HDL cholesterol is an indicator of LDL particle size in patients with type 2 diabetes and normal HDL cholesterol levels. Diabetes Care. 2000; 23: 1679–1685.
- ↵Memon RA, Staprans I, Noor M, Holleran WM, Uchida Y, Moser AH, Feingold KR, Grunfeld C. Infection and inflammation induce LDL oxidation in vivo. Arterioscler Thromb Vasc Biol. 2000; 20: 1536–1542.