Antibodies Against Cardiolipin and Oxidatively Modified LDL in 50-Year-Old Men Predict Myocardial Infarction
Abstract Autoantibodies against oxidatively modified low-density lipoproteins (oxLDL) and cardiolipin occur in patients with vascular diseases, including atherosclerosis. The ability of such antibodies to predict myocardial infarction (MI) was investigated in a prospective nested case-control study in which healthy 50-year-old men were followed up for 20 years. Raised levels of antibodies against oxLDL and cardiolipin at 50 years of age correlated positively with the incidence of MI and mortality related to MI 10 to 20 years later. IgG and IgA antibodies against cardiolipin were associated with MI between 50 to 60 years of age and IgG and IgA antibodies against oxLDL with MI at 60 to 70 years of age. Moreover, higher antibody levels were noted in those who died from acute MI in comparison to those who survived. The predictive power of IgA and IgG antibodies was strong and largely independent of that of other strong risk factors. In conclusion, raised levels of antibodies against oxLDL and cardiolipin may predict MI and MI-related death.
- Received April 1, 1997.
- Accepted August 13, 1997.
The classical risk factors of coronary heart disease such as hypercholesterolemia, smoking, hypertension, and diabetes explain only a part of the epidemiological features of this disorder. Among other possible factors, autoantibodies against oxidatively modified low-density lipoproteins (oxLDL) and cardiolipin have been implicated in the development of vascular diseases, including myocardial infarction (MI).1–4
Autoantibodies to cardiolipin and other negatively charged phospholipids are the hallmark of the antiphospholipid syndrome, which encompasses an increased risk of arterial and venous thrombosis, repeated spontaneous abortions, and thrombocytopenia. The syndrome occurs in a primary form or secondary to systemic inflammatory diseases.5 The anti-phospholipid syndrome and cardiolipin antibodies as separate entities are encountered in patients with acute MI and in people who are at risk for MI.1,6–8
Serum antibodies against oxLDL are present in a variety of inflammatory and autoimmune conditions as well as in atherosclerosis.2–4,9 A study of patients with carotid artery arteriosclerosis showed the amount of antibodies in serum to be positively correlated to the rate of progression of arteriosclerotic plaques.3 Another study indicated an association between elevated antibody levels and restenosis of coronary arteries after percutaneous transluminal coronary angioplasty.10 In our own investigation of patients with premature atherosclerotic peripheral occlusive vascular disease, the concentrations of antibodies against oxLDL discriminated better than other risk factors between patients and controls.4 Two recent prospective studies have indicated that the presence of high levels of antibodies against oxLDL and cardiolipin in middle-aged dyslipidemic men is an independent risk factor for MI as judged over a 5-year follow-up period.8,11 Other studies, however, have failed to disclose a connection between such antibodies and atherosclerosis.12,13 The importance of these antibodies for the promotion and progression of atherosclerosis is thus still an unresolved issue.
In the present investigation, antibodies against native and oxidatively modified LDL and cardiolipin were determined in samples from healthy 50-year-old males, who were followed for 20 years for development of MI. The levels of antibodies against oxLDL and cardiolipin were found to correlate with the future incidence of MI and death related to MI.
The study was a prospective nested case control study14 comprising 257 individuals: 119 patients with myocardial infarction between 50 to 70 years of age and 138 controls who did not develop MI up to 70 years of age. They were selected as follows: Samples were collected from 2322 males aged 50 years in Uppsala during 1970 to 1972. This cohort made up 82% of all 50-year-old men living in Uppsala at that time. The cohort was then followed for 20 years. Among the 38 men who developed MI between 50 and 60 years of age, 21 died; 81 men contracted MI between 60 and 70 years of age, of whom 18 died. A control group of 138 individuals was randomly selected from those who did not develop MI up to 70 years of age. Thus, in this prospective study, 257 individuals were healthy at 50 years, 219 at 60 years, and 138 at 70 years of age.
For all participants, diagnoses from hospital records were obtained from the National Central Bureau of Statistics until 1983. From 1984, for participants still living in the municipality of Uppsala, these data were requested from the hospitals. Information on causes of death was requested from the National Board of Health and Welfare. Only MI as the main diagnosis was used for classification. At the time of each 10-year follow-up, hospital and death records were scrutinized, and the accuracy of the MI diagnosis was assessed.15 Data on the subjects and control groups regarding hypertension, tobacco use, dyslipoproteinemia, and other conventionally recognized risk factors were presented earlier.16
Determination of Antibodies
IgA, IgG, and IgM antibodies against oxLDL and native LDL were determined by an enzyme-linked immunosorbent assay (ELISA) as described earlier.17 IgG and IgM antibodies against cardiolipin were determined by an ELISA as described earlier.18 For IgA cardiolipin antibodies, the method was essentially the same as described previously, except that serum was used in a dilution of 1/30 and anti-human IgA alkaline phosphatase conjugated (α chain specific, Sigma, St. Louis, Missouri, USA) in a dilution of 1/1000. All the secondary antibodies were analyzed for specificity. The assays for cardiolipin antibodies were standardized by using internationally recognized standard sera19 and sera with previously determined high levels of cardiolipin antibodies of IgG, IgA, and IgM isotypes. The assays for antibodies against oxLDL were standardized by using sera with previously determined high levels of antibodies of IgG, IgA, and IgM isotypes. Each serum was run in triplicate. Both the intraassay and interassay coefficients of variation between triplicate tests were 6% to 7% for all tests. The levels of antibodies were expressed as absorbance values in the ELISA. The cutoff level for the presence of antibodies in patients who developed MI was the mean ±2 SD of the absorbance values in the individuals who remained healthy.
The Mann-Whitney U test was performed to evaluate associations between antibody levels, as determined by the absorbance units in the ELISA, with age and with MI.
A trend test (P value from Spearman rank correlation coefficient) was applied to examine the relation between antibody levels as determined by absorbance units in the ELISA and clinical status at 70 years of age, that is, healthy (coded as 0), survivor of MI (coded as 1), and death by MI (coded as 2). This will therefore be a test of whether raised antibody levels are associated with a worse clinical outcome. This test was also performed with correction for the conventionally recognized risk factors, including supine blood pressure, serum LDL/HDL cholesterol ratio, ratio between arachidonic acid and dihomogammalinoleic acid from serum cholesterol, body mass index, and smoking.16
A multiple logistic regression analysis was used to test the predictive importance of the concentration of the autoantibodies alone and after the adjustment for the above-mentioned risk factors. Odds ratios (OR) for each predictor variable were standardized for standard deviations, which means that an OR shows the relative change in the odds of the event becoming a case when the predictor variable increases by 1 SD
The standardized normal deviate was used for the comparison of two proportions, that is, for comparing the difference in the prevalence of antibodies between patients with MI and healthy individuals and between patients with MI who survived or died. Raised antibody levels were considered to be above the mean +2 SD of values of all individuals who remained healthy up to 70 years of age.
The relative risk for individuals with antibodies to develop MI was calculated as the conditional probability compared to the overall probability of MI in the age group. The incidence of MI in the total Swedish male population between 50 and 70 years of age is 1.75 per 100 individuals, and this value was compared to the incidence of patients with antibodies against cardiolipin and oxLDL. Pearson correlation coefficients were used to analyze the correlation between plasma LDL values and antibodies against oxLDL. Simple regression was used to analyze the correlation between the levels of antibodies against oxLDL and cardiolipin. A value of P<0.05 was considered to be significant.
Antibodies Against oxLDL and Cardiolipin
Men who developed MI had higher levels of IgG antibodies against oxLDL than did those who remained healthy (P<.05). The concentrations of antibodies against LDL were low and did not differ between the groups (data not shown). IgA and IgG antibody levels in 50-year-old men who developed MI between 60 and 70 years of age were higher than those in men who remained healthy (P<.05; Table 1⇓). Moreover, IgG and IgA antibody levels were higher in individuals who died of MI than in those who remained healthy (Table 2⇓) (IgG P<.001, IgA P<.05). IgG antibodies were higher in individuals who died of MI than in survivors (P<.01).
Men who developed MI had higher levels of IgG and IgA antibodies against cardiolipin than did those who remained healthy (P<.05; data not shown). IgG and IgA cardiolipin antibodies were higher in patients who developed MI between 50 and 60 years of age than in those who remained healthy (P<.05).
Individuals who died of MI had higher IgG and IgA antibody levels than did those who survived the MI and those who remained healthy (P<.05; Table 2⇑).
Men who acquired MI had higher prevalence of IgG antibodies against cardiolipin and of IgA antibodies against oxLDL (P<.05). The prevalence of IgA antibodies against cardiolipin was very close to being significantly higher for those who developed MI. The risk for individuals with IgG and IgA antibodies against cardiolipin at 50 years of age to develop MI within a 20-year period was 2.7 and 3.0, respectively, and the risk in the presence of antibodies against oxLDL was 2.6 when compared to the overall risk of MI in the same age group in the Swedish male population.
Antibodies Against oxLDL and Cardiolipin: Relation to Other Predictive Factors
There was no relation between the plasma levels of LDL and the levels of antibodies against oxLDL. The levels of IgG and IgA antibodies against oxLDL or cardiolipin all correlated to the clinical outcome (remained healthy, survived MI, or suffered MI-related death; Table 3⇓). This correlation was even more marked when the data had been adjusted for other risk factors.
In a logistic regression analysis, the concentration of IgG and IgA antibodies against oxLDL and cardiolipin at age 50 was associated with an increased risk for future MI (Table 4⇓). This increased risk was of the same order of magnitude as that for serum LDL/HDL cholesterol ratio.16 The increased risk for MI associated with IgG and IgA antibodies against cardiolipin was independent of other risk factors. The risk for MI associated with IgG and IgA antibodies against oxLDL was not significantly independent of that of other risk factors, since the P values for the corrected odds ratios were 0.08 and 0.06, respectively, and higher than the P values for the uncorrected odds ratios (0.04 and 0.04, respectively; Table 5⇓).
Correlation Between Antibodies Against oxLDL and Cardiolipin
There was a positive correlation between IgA antibodies against cardiolipin and oxLDL (r=.435, P=.0001). There was a weaker correlation between IgM antibodies (r=.201, P=.0012), and there was no correlation between the IgG antibodies (r=.085, P=.1736).
This prospective study suggests that antibodies against cardiolipin and oxLDL in healthy males at 50 years of age can predict MI and MI-related death before the age of 70. The risk for MI connected with these antibodies was independent of other well-established predictors, such as supine blood pressure, serum LDL/HDL cholesterol ratio, ratio between arachidonic acid and dihomogammalinoleic acid from serum cholesterol, body mass index, and smoking. An important observation was that antibodies of IgG and IgA class were lowest in the group of men who remained healthy, higher in those who survived a MI, and highest in individuals who died from MI. Thus, the presence and concentration of such autoantibodies may constitute heretofore unrecognized risk factors for MI. Our observations may also imply that these autoantibodies are involved in processes leading to MI in men. So far, there are no comparable prospective studies in women.
The involvement of immune-inflammatory mechanisms in the development and regulation of atherosclerosis and its complications is supported by numerous observations.20 Patients with premature MI before the age of 45 had a high prevalence of persisting circulating immune complexes and this was negatively correlated to the plasma LDL concentration.21 A number of retrospective studies have pointed to an increased prevalence of antibodies against oxLDL and/or cardiolipin after acute MI.1,6,7,10,22 The presence of cardiolipin antibodies was even indicative of risk for reinfarction.1 There are also reports that fail to disclose associations between antibodies against oxLDL and/or cardiolipin and MI/ischemic heart disease.12,13,23,24,25 Prospective controlled trials should shed light on this controversial subject. Two earlier prospective studies of dyslipidemic men reported that antibodies against cardiolipin and oxLDL are associated with an increased risk for MI during a 5-year follow-up.8,11 The relationship between antibody levels and the MI-related mortality was not studied. Moreover, dyslipidemic individuals have other risk factors for the development of coronary artery disease that may confound the study.
An important question is how antibodies to cardiolipin and oxLDL are linked to the disease process. Several alternative possibilities can be envisaged. The fact that the antibodies were present in healthy males 10 to 20 years before the men developed a MI and that these antibodies are linked to the atherosclerotic process and coagulation disturbances, two key events in MI, would support causal connections.
The causes of formation of antibodies to cardiolipin and oxLDL are not known. The immune response against cardiolipin and oxLDL may be induced by autoantigens and/or exogenous antigens. Cardiolipin and other phospholipids are present in many cells in the body. Certain microorganisms have phospholipids on their surface and may well induce an immune response against cardiolipin and also against oxLDL.26,27 In this regard, the antibodies of IgA isotype are of special interest, since the initiating event for these antibodies often occurs at the mucosal surface, in the gastrointestinal tract or in the respiratory system. It has recently been shown that one important epitope for cardiolipin antibodies consists of oxidatively modified cardiolipin.28 OxLDL contains phospholipids that are exposed and oxidatively modified during oxidation of LDL. Oxidatively modified cardiolipin might thus constitute a common epitope for both anti-cardiolipin and anti-oxLDL antibodies. However, antibodies to oxLDL may be directed to various other epitopes created during the oxidative modification of LDL. Thus, though antibodies to cardiolipin and oxLDL may represent cross-reacting species, they may also represent the results of diverging triggering mechanisms. This is in line with our observation that there was no strong correlation between the presence of these two antibody species in the same individual.
The mechanism of action for the thromboembolic events related to cardiolipin antibodies is incompletely known.5 Among the mechanisms discussed is antibody binding to platelets and endothelial cells. This may lead to changes of cell surface structures and activation of the complement and coagulation systems with an inflammation of the vessel surface as the result. Interference with the function of a cofactor for cardiolipin antibodies, β2-glycoprotein 1, an inhibitor of the intrinsic pathway of coagulation, is another proposed mechanism of action.29 It is of interest to note that antibodies against cardiolipin were predictive of MI approximately one decade earlier that antibodies against oxLDL. Thus, individuals with antibodies against cardiolipin seem to have a more unfavorable outcome with earlier disease.
The mechanism of action of antibodies against oxLDL is even more incompletely elucidated. OxLDL has direct toxic effects on endothelial cells and causes dysfunction and increased permeability of the vessel wall.30 The antibodies might have beneficial effects by combining with oxLDL and promoting its clearance. This notion is supported by studies of LDL-receptor-deficient rabbits in which immunization with malondialdehyde-modified LDL resulted in a less rapid progression of atherosclerosis.31 On the other hand, antibodies against oxLDL might increase the formation of foam cells that might accelerate the atherosclerotic process. In a previous study by us, human IgG and IgM antibodies markedly enhanced the rate of uptake of oxLDL by a monocytelike cell line.17 OxLDL that is present in immune complex form together with its antibody might also induce immune complex–mediated damage to the vessel wall.20 Whether these autoantibodies really are involved in the disease process or merely represent markers of the inflammatory/atherosclerotic disease is therefore still a question to be solved.
Only IgG and IgA antibodies against cardiolipin and oxLDL were associated with future MI, whereas IgM antibodies were not. In the earlier studies, only IgG antibodies were determined.8,11 Antibodies of IgG and IgA type are present at a mature and/or secondary immune response. IgG antibodies mediate tissue damage by efficiently activating complement, mainly by the classical pathway, and by binding to Fc receptors on cells, whereas IgA antibodies have weaker effects. These properties may therefore be of importance for the pathogenic potentials of the antibodies.
A correlation between the concentrations of antibodies against cardiolipin and oxLDL, respectively, has earlier been described for IgG antibodies.8,9 Such correlations, although weak, were present also in our patients but were significant only for IgA and IgM antibodies. IgG antibodies showed no such correlation, a finding that implies that these antibodies have mutated during the development of the immune response and are mainly directed against antigenic sites that are not common for oxLDL and cardiolipin.
In conclusion, raised levels of antibodies against cardiolipin and oxLDL in a cohort of healthy 50-year-old Swedish men were associated with MI and MI-related death 10 to 20 years later. The predictive power of such antibodies was strong and largely independent of that of other strong risk factors.
This work was supported by grants from the Nanna Svartz Foundation, the Alice and Knut Wallenberg Foundation, the Petrus and Augusta Hedlund Foundation, Sweden, and the Swedish Heart-Lung Foundation.
Hamsten A, Norberg R, Björkholm M, de Faire U, Holm G. Antibodies to cardiolipin in young survivors of myocardial infarction: an association with recurrent cardiovascular events. Lancet. 1986;18:113–115.
Bergmark C, Wu R, de Faire U, Lefvert AK, Swedenborg J. Patients with early-onset peripheral vascular disease have increased levels of autoantibodies against oxidized LDL. Arterioscler Thromb Vasc Biol. 1995;15:441–445.
Ferlazzo B, Bonanno D, Quattrocchi P, Paino AM, Arrigo EF. Anticardiolipin antibodies and ischaemic heart disease. Minerva Cardiol. 1993;41:113–117.
Vaarala O, Mänttäri M, Manninen V, Tenkanen L, Puurunen M, Aho K, Palosuo T. Anti-cardiolipin antibodies and risk of myocardial infarction in a prospective cohort of middle-aged men. Circulation.. 1995;91:23–27.
Schumacher M, Eber B, Tatzber F, Kaufmann P, Esterbauer H, Klein W. Letter. Lancet. 1992;340:123.
Clayton D, Hills M. Statistical methods in Epidemiology. Oxford University Press, Oxford,1994;329–335.
Öhrvall M, Berglund L, Salminen I, Lithell H, Aro A, Vessby B. The serum cholesterol ester fatty acid composition but not the serum concentration of alpha tocopherol predicts the development of myocardial infarction in 50-year-old men: 19 years follow-up. Atheroscler. 1996;127:65–71.
Lefvert AK, Hamsten A, Holm G. Association between circulating immune complexes, complement C4 and myocardial infarction before age 45. Arterioscler Thromb Vasc Biol. 1995;15:665–668.
Phadke KV, Phillips RA, Clarke DT, Jones M, Naish P, Carson P. Anticardiolipin antibodies in ischaemic heart disease: marker or myth? Br Heart J. 1993;69:391–394.
Antiphospholipids-A: ‘guilty ghost’ syndrome [editorial]. Clin Rev Allerg Immunol.. 1995;13:1–4.
Hörkkö S, Miller E, Dudl E, Reaven P, Curtiss LK, Zvaifler NJ, Terkeltaub R, Pierangeli SS, Branch DW, Palinski W, Witztum JL. Antiphospholipid antibodies are directed against epitopes of oxidized phospholipids: recognition of cardiolipin by monoclonal antibodies to epitopes of oxidized low density lipoprotein. J Clin Invest. 1996;98:815–825.
McNeil HP, Simpson RJ, Chesterman CN, Krilis SA. Anti-phospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibitor of coagulation: β2-glycoprotein I (apolipoprotein H). Proc Natl Acad Sci. U S A.. 1990;87:4120–4124.
Hessler JR, Morel DW, Lewis LJ, Chisolm GM. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis. 1983;3:215–222.
Palinski W, Miller E, Witztum JL. Immunization of low density lipopriotein (LDL) receptor-deficient rabbits with homologous malondialdehyde modified LDL reduces atherogenesis. Proc Natl Acad Sci. U S A.. 1995;92:821–825.