Persistent High Levels of Plasma Oxidized Low-Density Lipoprotein After Acute Myocardial Infarction Predict Stent Restenosis
Objective— Recently, elevated levels of plasma oxidized low-density lipoprotein (LDL) have been shown to relate to plaque instability in human atherosclerotic lesions. We investigated prospectively patients admitted with acute myocardial infarction (AMI) who underwent primary coronary stenting to evaluate whether the 6-month outcome could be predicted by measuring plasma oxidized LDL (ox-LDL) levels at the time of hospital discharge.
Methods and Results— Plasma ox-LDL levels were measured in 102 patients with AMI undergoing primary coronary stenting using a highly sensitive ELISA method. Measurements were taken on admission and at discharge, and the findings related to the clinical outcome. At 6-month follow-up, angiographic stent restenosis occurred in 25 (25%) of the 102 AMI patients. Plasma ox-LDL levels at discharge were significantly (P=0.0074) higher in the restenosis group than those in the no-restenosis group (1.03±0.65 versus 0.61±0.34 ng/5 μg LDL protein). Multiple regression analysis showed that only plasma ox-LDL levels at discharge were a statistically significant independent predictor for late lumen loss after stenting (β=0.645; P<0.0001).
Conclusions— This prospective study demonstrates that persistence of an increased level of plasma ox-LDL at discharge is a strong independent predictor of stent restenosis at 6-month follow-up in AMI patients.
Atherosclerosis is a chronic inflammatory disease, and oxidized low-density lipoprotein (LDL) is widely accepted to play an important role in atherogenesis by enhancing the intraplaque inflammatory process.1,2 Oxidized LDL (ox-LDL) is cytotoxic for endothelial cells, acts as a chemoattractant for monocytes, inhibits the motility of tissue macrophages, and triggers thrombosis by inducing platelet adhesion.1,2 These data support the hypothesis that ox-LDL contributes to plaque instability in human atherosclerotic lesions.
Recently, methods have been developed to measure ox-LDL in blood using different antibodies and assay procedures.3–9 Holvoet et al3,4 developed an ELISA method to detect ox-LDL in plasma using a monoclonal antibody 4E6. They demonstrated that plasma levels of ox-LDL correlate with the extent of coronary artery disease in heart transplant patients,3 but in patients with ischemic heart disease, there were no significant differences between those with acute coronary syndromes and those with stable coronary artery disease.4 Our group developed a more sensitive method to measure plasma ox-LDL levels5 using a specific anti–ox-LDL monoclonal antibody DLH3 and demonstrated that plasma ox-LDL levels in patients with acute myocardial infarction (AMI) are significantly higher than in patients with unstable or stable angina pectoris or in controls.6,7 Witzum group also developed an ELISA method to measure circulating minimally ox-LDL levels using an antibody E06 (termed oxLDL-E06)8 and showed a significant increase in oxLDL-E06 levels in AMI patients at &4 days after AMI.9 These findings endorse our observations6 and suggest strongly that elevated plasma levels of ox-LDL bear a relationship with acute coronary syndromes in humans.
However, thus far, no studies have tested a potential prognostic value of plasma ox-LDL levels after stent implantation in patients with AMI. The present study is based on patients admitted with AMI and undergoing primary stent implantation in whom we measured plasma levels of ox-LDL on admission and at discharge and in whom the findings thereafter were related to angiographic outcomes at 6-month follow-up.
All patients provided written informed consent, and the study was approved by the hospital ethics committee.
The study is based on 109 consecutive patients with AMI who underwent primary coronary stenting between December 2001 and December 2003 using a bare metal stent and in whom successful coronary reperfusion was achieved within 24 hours after onset of AMI. We excluded 2 patients with complications of acute or subacute stent thrombosis requiring repeat angioplasty and 3 patients with cardiac arrest or cardiogenic shock attributable to multivessel diseases or left main infarction. We also excluded 2 patients undergoing dialysis. Ultimately, a total of 102 patients (77 men and 25 women; 63±11 years of age; mean±SD) were included for the analysis. There were no patients with concomitant inflammatory diseases or malignant tumors. The diagnosis of AMI was based on a history of prolonged ischemic chest pain, ST segment elevation of >0.2 mV on ≥2 contiguous ECG leads, and elevated creatine kinase (CK; >2× above normal range) within 24 hours after the onset of pain. The culprit vessel was identified on the basis of clinical, ECG, and angiographic data. Primary coronary stenting was considered successful if the residual stenosis of the infarct-related artery was <50% and thrombolysis in myocardial infarction (TIMI) flow grade 3 was present after the procedure. Medications before admission were as follows: angiotensin-converting enzyme inhibitors or angiotensin II type 1 receptor blockers (14 patients), calcium antagonists (16 patients), β-blockers (8 patients), and lipid-lowering agents (8 patients). Antioxidants were not administered to any of the AMI patients before admission. All patients were on aspirin (81 mg) therapy before the procedure and received 200 mg of ticlopidine, which is the standard regimen in Japan. Clopidogrel was not used in this study because this agent is not available in Japan. All patients were followed up for 6 months after admission. Urgent angiography was performed if the patient developed symptoms of angina within the follow-up period. All other patients underwent follow-up angiography at 6 months after onset of AMI.
A total of 86 age- and gender-matched healthy volunteer blood donors served as controls (61 men and 25 women; 63±9 years of age; mean±SD). Among the control subjects, none had hypercholesterolemia or diabetes mellitus, 6 had a history of hypertension, and 15 were smokers. All 6 hypertensives were in stage I according to the criteria established by the Joint National Committee VI; none used antihypertensive medication. Antioxidants were not administered to any controls.
In the AMI patients, the following data were obtained: age, gender, the presence of risk factors (cigarette smoking, hypertension, diabetes mellitus, and hypercholesterolemia [cholesterol level >220 mg/dL]), clinical variables including the presence of previous myocardial infarction, Killip classification, maximum CK, left ventricular ejection fraction at discharge, length of hospital stay, and medications in hospital and after discharge. Venous blood samples from all AMI patients were obtained on admission and at hospital discharge (mean 19±8 days). The following measurements were performed: serum levels of total cholesterol, high-density lipoprotein (HDL) cholesterol, LDL cholesterol, triglycerides, serum high sensitivity C-reactive protein (hs-CRP) levels, a leukocyte count, a neutrophil count, and plasma ox-LDL levels.
In the control subjects, serum levels of total cholesterol, HDL cholesterol, LDL cholesterol, and plasma ox-LDL levels were measured. Cholesterol fractions were measured on a Hitachi 7350 analyzer (Hitachi High Technologies) with reagents from Kyowa Medex. Serum hs-CRP levels were assayed with the use of latex-enhanced immunonephelometric assays on a BN II analyzer (Dade Behring).
Coronary Stenting Procedure
Immediately after the diagnosis of AMI in the emergency room, 5000 IU of intravenous heparin was administered. This was followed by an intravenous bolus injection of 2000 IU every hour during the procedure. The catheterization procedure was performed using the femoral approach in all patients. The infarct-related coronary artery was defined as a major coronary artery perfusing an area compatible with the distribution of ST segment elevation on a 12-lead ECG. The patency of the infarct-related artery was angiographically assessed before stenting according to TIMI perfusion criteria by 2 independent observers who were blinded to the study design. In cases of disagreement, consensus was reached by further joint reading. Subsequent percutaneous coronary intervention was performed for total occlusive lesions or for lesions of >75% diameter stenosis (DS), even with TIMI grade 3 flow. Primary coronary stenting was performed according to established conventions either with (n=71) or without (n=31) balloon predilatation.
Quantitative Coronary Angiography
All preprocedure, postprocedure, and follow-up angiography was conducted immediately after the administration of 0.25 mg of intracoronary nitroglycerin. Follow-up angiography was performed with guiding catheters of ≥5F in diameter. Angiography was performed so that each lesion could be viewed from ≥2 angles. In 102 patients after stenting, off-line quantitative coronary angiography (QCA) was conducted with the view revealing the highest degree of stenosis. Calculation was performed with the use of the Cardiovascular Measurement System (CMS-MEDIS Medical Imaging System) by 1 investigator who was unaware of the study design. The reference diameter, DS, and minimal lumen diameter (MLD) were measured before and after stenting and at the time of the follow-up coronary angiography. On the basis of these measurements, we obtained the value of acute gain (MLD after stenting minus MLD before stenting) and late lumen loss (MLD after stenting minus MLD at follow-up angiography) for the lesions. Angiographic restenosis was defined as >50% DS at follow-up angiography.
Measurement of Plasma Ox-LDL Levels
Plasma ox-LDL levels were measured using a highly sensitive sandwich ELISA method, described previously,5 and applied.6,7 The LDL fraction was separated from blood plasma before the ELISA procedure to minimize potential interferences with other plasma constituents, such as ox-VLDL, anti–ox-LDL autoantibodies, and antiphospholipid antibodies. The LDL fractions were obtained from the samples by sequential ultracentrifugation. Diluted LDL fractions (5 μg/well) were added to the microtiter wells that were precoated with 0.5 μg of the anti–ox-LDL monoclonal antibody DLH3. After extensive washing, the remaining ox-LDL was detected with a sheep anti-human apolipoprotein B antibody and an alkaline phosphatase-conjugated anti-sheep IgG antibody. In each ELISA plate, various concentrations of standard ox-LDL, which was prepared by incubating LDL with 5 μmol/L CuSO4 at 37°C for 3 hours, were run simultaneously to determine a standard curve.
The results are expressed as mean±SD. The 2 groups were compared with an unpaired Student t test or with a Mann–Whitney U test when the variance was heterogeneous. Comparisons of variables between >3 groups were performed by 1-way ANOVA and post hoc multiple comparison using Newman–Keuls multiple test or Scheffe test. Categorical variables were compared by use of χ2 test. Correlation between plasma ox-LDL levels and late lumen loss were assessed using simple linear regression. Multiple regression analysis was performed for various parameters, possibly affecting restenosis, predicting the late lumen loss. Values of P<0.05 were considered statistically significant.
There was no in-hospital mortality, and all patients were discharged without recurrent myocardial infarction or coronary artery bypass surgery. In 5 patients, urgent angiography was performed within 6 months: in 4 patients because of symptoms of effort angina, whereas the remaining patient developed restenosis with heart failure at 47 days after the onset of AMI.
The clinical and angiographic follow-up was accomplished in all patients (100%), with a mean time interval of 168±24 days. Stent restenosis occurred in 25 of the 102 AMI patients (25%).
Time Course of Plasma Ox-LDL Levels in Patients With AMI
Figure 1 shows that the plasma levels of ox-LDL at hospital discharge had decreased significantly (P<0.01) compared with levels on admission (admission, 1.55±1.21; discharge, 0.71±0.47 ng/5 μg LDL protein). However, plasma ox-LDL levels at discharge were still significantly higher (P<0.05) than in control subjects (discharge, 0.71±0.47; control, 0.48±0.19 ng/5 μg LDL protein).
Evaluation of the relationship between plasma ox-LDL levels and the in-hospital stay (mean 19±8 days) revealed no significant differences in plasma ox-LDL levels at discharge between those with a hospital stay of ≤14 days (n=29; 0.72±0.46), 15 to 28 days (n=60; 0.70±0.48), or ≥29 days (n=13; 0.72±0.51 ng/5 μg LDL protein).
Correlation With Restenosis
Patients were divided into 2 groups, no-restenosis and restenosis, according to the results of QCA. Baseline clinical and angiographic characteristics of both groups are listed in Table 1. There were no significant differences between the 2 groups with respect to age, gender, risk factors, serum levels of total cholesterol, HDL cholesterol, LDL cholesterol, clinical variables, duration of hospital stay, medications in hospital and after discharge, balloon predilation, stent length, maximum inflation pressure, or stent/artery ratio. Regarding QCA analysis, there were no significant differences in postprocedure MLD, acute gain, or DS between the 2 groups. Antioxidants were not administered to any AMI patients in the 2 groups either in hospital nor after discharge.
With regard to mean leukocyte counts, mean neutrophil counts, and serum levels of hs-CRP, there were no significant differences between the 2 groups both on admission and at discharge (Table 2). The plasma ox-LDL levels on admission were not significantly different between the 2 groups. However, plasma ox-LDL levels at discharge were significantly (P=0.0074) higher in the restenosis group than in the no-restenosis group (restenosis group 1.03±0.65 versus no-restenosis group 0.61±0.34 ng/5 μg LDL protein; Table 2).
In addition, plasma ox-LDL levels at discharge were divided into tertiles (I, <0.44 ng/5 mg LDL protein, n=34; II, 0.44 to 0.80 ng/5 mg LDL protein, n=34; III, >0.80 ng/5 mg LDL protein, n=34) for evaluating MLD, late lumen loss, DS, and restenosis rate (Figure 2). MLD in tertile III was significantly smaller than in either tertile I or tertile II (I, 2.15±0.71; II, 1.97±0.66; and III, 1.46±0.83 mm; I versus III, P<0.005; and II versus III, P<0.05), and late lumen loss in tertile III was significantly higher than in either tertile I or II (I, 0.54±0.48; II, 0.85±0.61; and III, 1.44±0.64 mm; I versus III, P<0.0001; and II versus III, P<0.0005). DS in tertile III was significantly higher than in either tertile I or II (I, 29±16; II, 35±23; and III, 49±26%; I versus III, P<0.005; and II versus III, P<0.05). Moreover, restenosis rates were significantly different among the 3 groups (P<0.005). Restenosis rates in tertile III were the highest (44%).
Relationship Between Plasma Ox-LDL Levels at Discharge and Neointimal Proliferation After Stenting
To investigate whether plasma ox-LDL levels at discharge could relate to neointimal proliferation after stenting, we assessed the association between plasma ox-LDL levels at discharge and the degree of late lumen loss at 6 months after stenting. Plasma ox-LDL levels at discharge showed a positive correlation (R=0.62; P<0.0001; Figure 3) with the late lumen loss after stenting. Other fractions had no correlation with late lumen loss.
To the best of our knowledge, this is the first study to demonstrate that elevated plasma ox-LDL levels at discharge serve as a strong predictor of restenosis in AMI patients undergoing primary stent implantation. Our present study clearly showed a positive correlation between plasma ox-LDL levels at discharge and the degree of neointimal proliferation after coronary stenting in AMI.
It was believed previously that ox-LDL was not present in the circulation because blood plasma has strong antioxidative abilities and because experimental studies had shown that ox-LDL injected intravenously was rapidly cleared from the circulation by the liver, in particular by Kupffer cells and sinusoidal endothelial cells.10 However, recently, it was shown that ox-LDL does occur in blood and that plasma ox-LDL levels were elevated in patients with transplant-associated coronary artery disease.3 We developed a new sandwich ELISA method to measure plasma ox-LDL levels, using only the LDL fraction from blood plasma and not using whole blood plasma, to minimize potential interferences with other plasma substances such as ox-VLDL, anti–ox-LDL autoantibodies, and antiphospholipid antibodies.5 This method, which takes 4 days for each measurement, is laborious but highly sensitive in detecting minute amounts of ox-LDL particles in blood. Using this method, we demonstrated previously for the first time that plasma ox-LDL levels in AMI patients on admission were significantly higher than those in control subjects.6 Moreover, our recent study7 shows that elevated plasma levels of ox-LDL in AMI patients relate to coronary plaque instability associated with intraplaque LDL retention, plaque inflammation with macrophage accumulation and neutrophil infiltration, and plaque disruption with thrombosis in coronary atherosclerotic lesions. The present study with 102 AMI patients endorses these observations but, in addition, provides novel data as to the time course of ox-LDL plasma levels during follow-up and its clinical relevance. From our prospective study, it appeared that in the majority of AMI patients, plasma ox-LDL levels decreased significantly by the time of hospital discharge. Tsimikas et al9 recently demonstrated a similar gradual decline of plasma oxLDL-E06 levels in AMI patients, although their results were based on a small number of patients (n=10). Uno et al11 found in stroke patients with elevated plasma ox-LDL levels a gradual decrease to normal at 30 days after the onset of stroke. The phenomenon of gradual decline in plasma ox-LDL levels in patients with AMI or stroke suggests that in these patients, after the acute event, the balance between continued release or generation of ox-LDL and the clearance or protection systems of ox-LDL have shifted in favor of the latter. This appears clinically relevant because apparently in some patients, this is not the case. Those AMI patients who presented prolonged elevation of plasma ox-LDL levels at discharge had an increased incidence of restenosis. The impact of plasma ox-LDL levels on restenosis was also obtained by a subgroup analysis based on the tertiles of plasma ox-LDL levels at discharge. Moreover, multiple regression analysis identified that the level of plasma ox-LDL at discharge was the only independent predictor for restenosis after primary stenting during 6-month follow-up.
The background of this finding remains unclear. However, one could hypothesize that persistent high levels of plasma ox-LDL at discharge, approximately at 2 to 4 weeks after coronary stenting, result from a continuing imbalance between pro-oxidant and antioxidant forces in the blood. This could be the reflection of the presence of marked plaque injuries with extensive thrombosis and inflammatory reactions at the site of stenting, generating excessive pro-oxidants outmatching the antioxidant systems. This line of thought fits with previous studies of human specimens after coronary stenting from our group12 and others,13 which showed a crucial role for mural thrombosis and infiltration of monocytes and neutrophils in the development of exuberant neointimal proliferation of in-stent restenosis. Recently, Farb et al14 demonstrated a strong link between increased neointimal macrophage content and in-stent restenosis. Infiltrated monocytes and neutrophils and activated platelets have been shown to promote LDL oxidation.1,2 In turn, ox-LDL stimulates proliferation and migration of smooth muscle cells via induction of platelet-derived growth factor.15 Together with our present findings, one could hypothesize that once excessive generation of pro-oxidants occurs because of marked inflammatory reactions and extensive thrombus formation at the site of stenting, this could lead to an imbalance between pro-oxidants and antioxidants in the blood and could cause persistent high levels of plasma ox-LDL. Subsequently, increased ox-LDL levels may stimulate excessive proliferation and migration of smooth muscle cells, which could contribute to exuberant neointimal proliferation, leading to the development of stent restenosis at later stages.
Because there is potentially a continuous spectrum of degrees of oxidation in what we call “ox-LDL,” it is assumed that heterogeneous types of ox-LDL could be present in vivo. Minimally modified LDL (MM-LDL), in which oxidative modification has not been sufficient to cause changes recognized by scavenger receptors, is known to have a variety of proinflammatory functions.2 Our recent study has demonstrated that MM-LDL is a type of ox-LDL enriched with oxidized phosphatidycholine (ox-PC), and our DLH3 antibody, which recognizes specifically ox-PC, binds not only to heavily ox-LDL but also to MM-LDL.16 On the basis of these data, one could hypothesize that our measuring method may detect ox-PC particles as part of ox-LDL and MM-LDL, released or generated at the sites of coronary stenting associated with mural thrombosis and inflammatory reactions.
Our study population (n=102) was relatively small, and this may have limited the statistical power for detecting the predictors of restenosis after stenting. On the other hand, the study was designed as a prospective investigation, and the clinical and angiographic follow-up data were available in all 102 patients. Moreover, our ox-LDL–measuring method is a highly sensitive method that allows the detection of minute amounts of the ox-LDL particles in blood. Altogether, we consider the quality of data sufficiently high to warrant our conclusion. It is worth commenting also on the length of the in-hospital stay of our patient cohort because the measurements of ox-LDL were taken at the time of hospital discharge. Compared with most modern standards, an in-hospital stay of 19±8 days is rather long. However, recent study17 has demonstrated that Japanese patients with AMI were hospitalized for an average of 29.8 days in 2002. Hence, our average of 19±8 days is in accordance with the situation as it exists in Japan. These considerations should be taken into account when implementing our observations in other societies, in particular because in our experience, patients with an AMI have an elevated plasma level of ox-LDL on admission,6 with a gradual decline in the days afterward. However, these reflections should not detract from the observation that persistent high levels of plasma ox-LDL have predictive value for the development of stent restenosis. Our ox-LDL–measuring method is laborious and time consuming, and from that point of view, its clinical use appears limited. However, our DLH3 antibody recognizes specifically ox-PC and binds not only to fully oxidized LDL but also to MM-LDL.16 Therefore, our ox-LDL–measuring method is a sensitive method to analyze the behavior of ox-PC particles as part of ox-LDL and MM-LDL in humans, leading to identify the specific pathophysiological roles of ox-LDL in the development of restenosis in humans.
In conclusion, this prospective study demonstrates that persistence of an increased level of plasma ox-LDL after an AMI is a strong independent predictor of stent restenosis at 6-month follow-up.
- Received April 23, 2005.
- Accepted January 19, 2006.
Steinberg D. Low density lipoprotein oxidation and its pathobiological significance. J Biol Chem. 1997; 272: 20963–20966.
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.
Holvoet P, Vanhaecke J, Janssens S, Van de Werf F, Collen D. Oxidized LDL and malondialdehyde-modified LDL in patients with acute coronary syndromes and stable coronary artery disease. Circulation. 1998; 98: 1487–1494.
Itabe H, Yamamoto H, Imanaka T, Shimamura K, Uchiyama H, Kimura J, Sanaka T, Hata Y, Takano T. Sensitive detection of oxidatively modified low density lipoprotein using a monoclonal antibody. J Lipid Res. 1996; 37: 45–53.
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 lipoproteinshow a positive relationship with the severity of acute coronary syndromes. Circulation. 2001; 103: 1955–1960.
Kayo S, Ohsawa M, Ehara S, Naruko T, Ikura Y, Hai E, Yoshimi N, Shirai N, Tsukamoto Y, Itabe H, Higuchi K, Arakawa T, Ueda M. Oxidized low density lipoprotein levels circulating in plasma and deposited in the tissues: comparison between Helicobacter pylori-associated gastritis and acute myocardial infarction. Am Heart J. 2004; 148: 818–825.
Wu R, de Faire U, Lemne C, Witztum JL, Frostegard J. Autoantibodies to OxLDL are decreased in individuals with borderline hypertension. Hypertension. 1999; 33: 53–59.
Van Berkel TJ, De Rijke YB, Kruijt JK. Different fate in vivo of oxidatively modified low density lipoprotein and acetylated low density lipoprotein in rats. Recognition by various scavenger receptors on Kupffer and endothelial liver cells. J Biol Chem. 1991; 266: 2282–2289.
Uno M, Kitazato KT, Nishi K, Itabe H, Nagahiro S. Raised plasma oxidized LDL in acute cerebral infarction. J Neurol Neurosurg Psychiatry. 2003; 74: 312–316.
Komatsu R, Ueda M, Naruko T, Kojima A, Becker AE. Neointimal tissue response at sites of coronary stenting in humans: macroscopic, histological, and immunohistochemical analyses. Circulation. 1998; 98: 224–233.
Farb A, Weber DK, Kolodgie FD, Burke AP, Virmani R. Morphological predictors of restenosis after coronary stenting in humans. Circulation. 2002; 105: 2974–2980.
Stiko-Rahm A, Hultgardh-Nilsson A, Regnstrom J, Hamsten A, Nilsson J. Native and oxidized LDL enhances production of PDGF AA and the surface expression of PDGF receptors in cultured human smooth muscle cells. Arterioscler Thromb. 1992; 12: 1099–1109.
Itabe H, Mori M, Fujimoto Y, Higashi Y, Takano T. Minimally modified LDL is an oxidized LDL enriched with oxidized phosphatidylcholines. J Biochem. 2003; 134: 459–465.