Lipoprotein (a) Blood l+vels in Unstable Angina Pectoris, Acute Myocardial Infarction, and After Thrombolytic Therapy Shiqiang Qiu, MD, Pierre Thkroux, MD, Jacques Genest, Jr., MD, B. Charles Solymoss, MD, Danielle Robitaille, MD, and Michel Marcil, BSC
Lipoprotein (a) [Lp(a)] appears to be involved in atherogenesis and in vitro studies have suggested that it may interfere with thrombolysis. In this study, Lp(a) serum levels were determined by radioimmunoassay in 124 patients with ischemic heart disease. Of these, 47 had acute myocardial infarction, 13 had unstable angina, and 64 were age-matched patients with stable angina. Of the 60 patients with acute coronary artery disease, 34 received thrombolysis and 26 did not. In addition to Lp(a), serum plasminogen, CY~antiplasmin, fibrinogen, and D-dimer (cross-linked fibrin degradation products) levels were measured. These tests were repeated after 6 hours in patients with myocardial infarction and unstable angina. No significant difference was found for admission Lp(a) levels among patients with myocardial infarction (0.324 f 0.047 g/liter), unstable angina (0.435 f 0.123 g/liter) and stable angina (0.431 f 0.023 g/liter), between patients with myocardial infarction with or without thrombolytic treatment, nor between late and early measurements in patients with unstable angina and acute myocardial infarction. Plasminogen, a2 antiplasmin and fibrinogen values decreased significantly after thrombolytic treatment. The size of this decrease correlated positively with higher Lp(a) blood levels (p 0.25 g/liter had a 66% decrease in fibrinogen and a 53% decrease in antiplasmin, compared with 35 and 32%, respectively, in patients with Lp(a) level 10.25 g/liter (p 1 molecule of apo(a).5,6 Ape(a) is a glycosylated protein that shares considerable homology with plasminogen.7 The gene for ape(a) is adjacent to that of plasminogen on chromosome 6 (6827) and is composed of multiple repeats of kringle IV-like domains.* Although the role of Lp(a) is poorly understood, it interferes with thrombolysis by decreasing the rate of plasminogen activation by streptokinase or tissue plasminogen activator by binding to fibrin and by competing with plasminogen for its receptor on vascular endothelial cells.9-13 In this study, we assessedserum levels of Lp(a) during the acute process of unstable angina and acute myocardial infarction and examined the interaction of these levels with markers of fibrinolytic activity during thrombolysis.
U
METHODS Study patients: The study group comprised 124 hospitalized patients, 60 with an acute manifestation of coronary artery disease and 64 with stable angina. Patients with acute disease were a consecutive series of patients admitted for 2 1 episode of chest pain ~30 minutes in duration, not relieved by sublingual nitroLP(A)
BLOOD
LEVELS
IN THROMBOTIC
STATES
1175
glycerin, and accompanied by an ST-segment shift or T-wave inversion on the 12-lead electrocardiogram. Serial creatine kinase and MB-isoenzyme blood levels were obtained, and acute myocardial infarction was diagnosed when total creatine kinase values doubled with the MB fraction present; unstable angina was diagnosed when no such enzyme elevation was observed. Myocardial infarction occurred in 47 patients and unstable angina in 13. Patients were further divided into a group of 34 patients who received thrombolytic therapy and a group of 26 who did not. The thrombolytic agents used were streptokinase in 29 patients at a dose of 1.5 million U administered intravenously during 30 to 45 minutes or single-chain, recombinant tissue-type plasminogen activator (Activase@, Genentech Corporation) in 5 patients at a total dose of 100 mg administered over 4 hours with the step-down infusion protocol recommended by the manufacturer. All patients who received thrombolysis had ST-segment elevation on their initial electrocardiogram; one-third of those who did not also had ST elevation. The patients with stable angina were selected by computer in a random order from a pool of 900 patients hospitalized for elective cardiac catheterization requested by the treating physician. Criteria for their selection included the presence of angiographically documented coronary artery disease with percent luminal diameter reduction of 250% in the right or left coronary artery. They were further selected by computer to match the age and sex of the group of patients with acute coronary artery disease. Blood sampling and analyses: Blood analyses performed included total creatine kinase and its MB fraction, total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, Lp(a), plasminogen, fibrinogen, a~ antiplasmin and D-dimers levels. In patients with unstable coronary artery disease, cardiac enzymes were determined at admission, every 2 hours for 6 hours and every 4 hours for the following 24 hours. Lp(a) values were obtained at admission before the administration of any specific treatment for acute myocardial infarction or unstable angina and 6 hours later. Fibrinolytic activity measured by levels of plasminogen, fibrinogen, antiplasmin and D-dimers were similarly assessed at admission and 6 hours later. The lipid profile, excluding Lp(a), was obtained in the fasting state between 8 to 18 hours after admission. In patients with stable angina, the lipid measurements were performed the morning after admission, after a 12-hour fasting period and before angiography. Creatine kinase blood levels were measured with optimized reagents from Boehringer-Mannheim and its
1176
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
67
MB fractions by immunoinhibition reagent from Boehringer-Mannheim, verified if necessary by agarose electrophoresis. Blood samples for lipid determination were obtained in tubes with no anticoagulant, and for coagulation factor determination in tubes with buffered sodium citrate. These were immediately centrifuged and the plasma stored at -7O“C for subsequent batch analyses. Total blood cholesterol, HDL cholesterol and triglyceride levels were measured by enzymatic techniques with standards and reagents from Boehringer-Mannheim. The precipitating agent used for HDL cholesterol was low-molecular-weight dextran sulfate with magnesium chloride. LDL cholesterol levels were calculated using the Friedewald equation. l4 Lp(a) blood levels were analyzed by radioimmunoassay with antisera standards and control materials from Pharmacia Diagnostic AB (Uppsala, Sweden). Cross reactivity with plasminogen and apo B is absent up to concentrations of 8.5 g/liter of plasminogen and 7.0 g/liter of apo B in this assay.15 Serum fibrinogen, plasminogen and cy2antiplasmin were measured by the Automated Coagulation Laboratory (Fisher Scientific Ltd, Milan, Italy). Thromboplastin from Baxter Co. Ltd (Miami, Florida) was used for the fibrinogen assay. The plasminogen assay was performed with the samples incubated in an excess of streptokinase and the plasmin-like activity detected on a synthetic chromogenic substrate. Reagents for plasminogen and a!2 antiplasmin were from Instrumentation Laboratory SPA (Milan, Italy). The samples were incubated with an excess of plasmin in biologic conditions minimizing the influence of other plasmin inhibitors, and residual plasmin activity was measured on the synthetic chromogenic substrate. Determination of Ddimers was made semiquantitatively with the Latex slide test (Ortho Dimer test, Ortho Diagnostic Systems, Beerse, Belgium). Statistical analysis: Data were analyzed using the BMDP statistical software package (Los Angeles, California). Baseline characteristics of the various study groups were compared using the t test for continuous data and chi-square analysis for discrete data after having assessed any significant departure of the data from the mean. Paired t tests were used when comparing changes in values within the same subjects. Because Lp(a) blood levels followed a non-normal distribution, they were analyzed using the nonparametric MannWhitney analysis for nonmatched comparisons of 2 populations. An Lp(a) cut-off value of 0.25 g/liter served to discriminate normal or abnormally high values. The semiquantitative data of D-dimers levels were analyzed by a significant test of rate of 2 samples. The level of a p value
Lipoprotein (a) [Lp(a)] appears to be involved in atherogenesis and in vitro studies have suggested that it may interfere with thrombolysis. In this study, Lp(a) serum levels were determined by radioimmunoassay in 124 patients with ischemic heart disease. Of these, 47 had acute myocardial infarction, 13 had unstable angina, and 64 were age-matched patients with stable angina. Of the 60 patients with acute coronary artery disease, 34 received thrombolysis and 26 did not. In addition to Lp(a), serum plasminogen, CY~antiplasmin, fibrinogen, and D-dimer (cross-linked fibrin degradation products) levels were measured. These tests were repeated after 6 hours in patients with myocardial infarction and unstable angina. No significant difference was found for admission Lp(a) levels among patients with myocardial infarction (0.324 f 0.047 g/liter), unstable angina (0.435 f 0.123 g/liter) and stable angina (0.431 f 0.023 g/liter), between patients with myocardial infarction with or without thrombolytic treatment, nor between late and early measurements in patients with unstable angina and acute myocardial infarction. Plasminogen, a2 antiplasmin and fibrinogen values decreased significantly after thrombolytic treatment. The size of this decrease correlated positively with higher Lp(a) blood levels (p 0.25 g/liter had a 66% decrease in fibrinogen and a 53% decrease in antiplasmin, compared with 35 and 32%, respectively, in patients with Lp(a) level 10.25 g/liter (p 1 molecule of apo(a).5,6 Ape(a) is a glycosylated protein that shares considerable homology with plasminogen.7 The gene for ape(a) is adjacent to that of plasminogen on chromosome 6 (6827) and is composed of multiple repeats of kringle IV-like domains.* Although the role of Lp(a) is poorly understood, it interferes with thrombolysis by decreasing the rate of plasminogen activation by streptokinase or tissue plasminogen activator by binding to fibrin and by competing with plasminogen for its receptor on vascular endothelial cells.9-13 In this study, we assessedserum levels of Lp(a) during the acute process of unstable angina and acute myocardial infarction and examined the interaction of these levels with markers of fibrinolytic activity during thrombolysis.
U
METHODS Study patients: The study group comprised 124 hospitalized patients, 60 with an acute manifestation of coronary artery disease and 64 with stable angina. Patients with acute disease were a consecutive series of patients admitted for 2 1 episode of chest pain ~30 minutes in duration, not relieved by sublingual nitroLP(A)
BLOOD
LEVELS
IN THROMBOTIC
STATES
1175
glycerin, and accompanied by an ST-segment shift or T-wave inversion on the 12-lead electrocardiogram. Serial creatine kinase and MB-isoenzyme blood levels were obtained, and acute myocardial infarction was diagnosed when total creatine kinase values doubled with the MB fraction present; unstable angina was diagnosed when no such enzyme elevation was observed. Myocardial infarction occurred in 47 patients and unstable angina in 13. Patients were further divided into a group of 34 patients who received thrombolytic therapy and a group of 26 who did not. The thrombolytic agents used were streptokinase in 29 patients at a dose of 1.5 million U administered intravenously during 30 to 45 minutes or single-chain, recombinant tissue-type plasminogen activator (Activase@, Genentech Corporation) in 5 patients at a total dose of 100 mg administered over 4 hours with the step-down infusion protocol recommended by the manufacturer. All patients who received thrombolysis had ST-segment elevation on their initial electrocardiogram; one-third of those who did not also had ST elevation. The patients with stable angina were selected by computer in a random order from a pool of 900 patients hospitalized for elective cardiac catheterization requested by the treating physician. Criteria for their selection included the presence of angiographically documented coronary artery disease with percent luminal diameter reduction of 250% in the right or left coronary artery. They were further selected by computer to match the age and sex of the group of patients with acute coronary artery disease. Blood sampling and analyses: Blood analyses performed included total creatine kinase and its MB fraction, total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides, Lp(a), plasminogen, fibrinogen, a~ antiplasmin and D-dimers levels. In patients with unstable coronary artery disease, cardiac enzymes were determined at admission, every 2 hours for 6 hours and every 4 hours for the following 24 hours. Lp(a) values were obtained at admission before the administration of any specific treatment for acute myocardial infarction or unstable angina and 6 hours later. Fibrinolytic activity measured by levels of plasminogen, fibrinogen, antiplasmin and D-dimers were similarly assessed at admission and 6 hours later. The lipid profile, excluding Lp(a), was obtained in the fasting state between 8 to 18 hours after admission. In patients with stable angina, the lipid measurements were performed the morning after admission, after a 12-hour fasting period and before angiography. Creatine kinase blood levels were measured with optimized reagents from Boehringer-Mannheim and its
1176
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
67
MB fractions by immunoinhibition reagent from Boehringer-Mannheim, verified if necessary by agarose electrophoresis. Blood samples for lipid determination were obtained in tubes with no anticoagulant, and for coagulation factor determination in tubes with buffered sodium citrate. These were immediately centrifuged and the plasma stored at -7O“C for subsequent batch analyses. Total blood cholesterol, HDL cholesterol and triglyceride levels were measured by enzymatic techniques with standards and reagents from Boehringer-Mannheim. The precipitating agent used for HDL cholesterol was low-molecular-weight dextran sulfate with magnesium chloride. LDL cholesterol levels were calculated using the Friedewald equation. l4 Lp(a) blood levels were analyzed by radioimmunoassay with antisera standards and control materials from Pharmacia Diagnostic AB (Uppsala, Sweden). Cross reactivity with plasminogen and apo B is absent up to concentrations of 8.5 g/liter of plasminogen and 7.0 g/liter of apo B in this assay.15 Serum fibrinogen, plasminogen and cy2antiplasmin were measured by the Automated Coagulation Laboratory (Fisher Scientific Ltd, Milan, Italy). Thromboplastin from Baxter Co. Ltd (Miami, Florida) was used for the fibrinogen assay. The plasminogen assay was performed with the samples incubated in an excess of streptokinase and the plasmin-like activity detected on a synthetic chromogenic substrate. Reagents for plasminogen and a!2 antiplasmin were from Instrumentation Laboratory SPA (Milan, Italy). The samples were incubated with an excess of plasmin in biologic conditions minimizing the influence of other plasmin inhibitors, and residual plasmin activity was measured on the synthetic chromogenic substrate. Determination of Ddimers was made semiquantitatively with the Latex slide test (Ortho Dimer test, Ortho Diagnostic Systems, Beerse, Belgium). Statistical analysis: Data were analyzed using the BMDP statistical software package (Los Angeles, California). Baseline characteristics of the various study groups were compared using the t test for continuous data and chi-square analysis for discrete data after having assessed any significant departure of the data from the mean. Paired t tests were used when comparing changes in values within the same subjects. Because Lp(a) blood levels followed a non-normal distribution, they were analyzed using the nonparametric MannWhitney analysis for nonmatched comparisons of 2 populations. An Lp(a) cut-off value of 0.25 g/liter served to discriminate normal or abnormally high values. The semiquantitative data of D-dimers levels were analyzed by a significant test of rate of 2 samples. The level of a p value
