Search

SummaryAnionic phospholipid membranes have a dual role in blood coagulation: they are essential for the initiation and propagation as well as for the limitation and termination of the blood coagulation process. Patients with the anti-phospholipid syndrome (APS) carrying antibodies against complexes of anionic phospholipids and plasma proteins, show in vitro inhibited phospholipid dependent coagulation reactions, whereas in vivo the presence of these antibodies is associated with an increased risk of thrombosis. In this study we focussed on the effects of these anti-phospholipid antibodies on the regulation of TF-mediated factor Xa (FXa) generation in plasma. We hypothesized that anti-phospholipid antibodies interfere with the phospholipiddependent inhibition by tissue factor pathway inhibitor (TFPI) of TFinduced coagulation. Indeed, total-IgG, anti-cardiolipin-IgG (aCL) and anti-β2GPI-IgG, isolated from patient plasmas, all stimulated TF-induced FXa generation in normal plasma. This enhanced FXa generation was not observed when the patient’s IgG was depleted of anti-β2GPI-IgG or when normal plasma was depleted of β2GPI or TFPI. Taken together, these data indicate that antibodies to β2GPI, circulating in patients with APS, suppress TFPI-dependent inhibition of TF-induced coagulation, which results in an increased FXa generation.

No Evidence for Enhanced Thrombin Formation through Displacement of Annexin V by Antiphospholipid Antibodies

Generation and inhibition of activated factor IXa was studied in factor XIa-activated plasma containing 4 mmol/L free calcium ions and 20 mumol/L phospholipid (25 mol% phosphatidylserine/75 mol% phosphatidylcholine). Interference of other (activated) clotting factors with the factor IXa activity measurements could be avoided by using a highly specific and sensitive bioassay. Factor IXa generation curves were analyzed according to a model that assumed Michaelis-Menten kinetics of factor XIa-catalyzed factor IXa formation and pseudo first order kinetics of inhibition of factor XIa and factor IXa. In the absence of heparin, factor IXa activity in plasma reached final levels that were found to increase with increasing amounts of factor XIa used to activate the plasma. When the model was fitted to this set of factor IXa generation curves, the analysis yielded a rate constant of inhibition of factor XIa of 0.7 +/- 0.1 min-1 and a kcat/Km ratio of 0.29 +/- 0.01 (nmol/L)-1 min-1. No neutralization of factor IXa activity was observed (the estimated rate constant of inhibition of factor IXa was 0). Thus, in the absence of heparin, the final level of factor IXa in plasma is only dependent on the initial factor XIa concentration. While neutralization of in situ generated factor IXa in normal plasma was negligible, unfractionated heparin dramatically enhanced the rate of inactivation of factor IXa (apparent second order rate constant of inhibition of 5.2 min-1/per microgram heparin/mL). The synthetic pentasaccharide heparin, the smallest heparin chain capable of binding antithrombin III, stimulated the inhibition of in situ generated factor IXa, but sevenfold less than unfractionated heparin (k = 0.76 min-1 per microgram pentasaccharide/mL). We found that free calcium ions were absolutely required to observe an unfractionated heparin and pentasaccharide-stimulated neutralization of factor IXa activity. Factor XIa inhibition (psuedo first order rate constant of 0.7 min-1) was not affected by unfractionated heparin or pentasaccharide in the range of heparin concentrations studied.

Plasma activities of creatine kinase (CK) and alpha-hydroxybutyrate dehydrogenase (HBD) were measured after permanent coronary artery occlusion in the dog. Cumulative release of enzymes in plasma was calculated from these data by using a previously validated two-compartment model for circulating enzymes. Regional myocardial ischemia was measured by injection of radiolabeled microspheres. After 48 hours, the dogs were killed, and a detailed map of left ventricular enzyme activity was obtained from 108 tissue samples. Cumulative release into plasma of CK and HBD was 96 +/- 20% and 112 +/- 26%, respectively, of the total activities depleted from the heart (mean +/- SD, n = 11). The scatter in these values is inherent to the calculations, and it is concluded that both enzymes are recovered completely in plasma and, thus, can be used as quantitative markers of injury. Discrepancies between this result and earlier reports on the recovery of CK are only partly apparent and can be explained partly by underestimation of the elimination rate of CK from plasma, irregardless of tissue edema and incomplete extraction of enzyme activity from tissue.

SummaryThe activation of prothrombin by prothrombinase was investigated in a continuous flow system at 25 °C. A glass capillary, containing a continuous phospholipid bilayer attached to the interior surface, was first perfused with factor Va. The factor Va bound to the phospholipid surface functioned as sites for the formation of prothrombinase, when subsequently a factor Xa and prothrombin containing solution was perfused. Under the conditions used, steady-state rates of prothrombin activation were attained after 4 to 15 min. The rates of prothrombinase formation increased with increasing factor Xa concentrations and flow rates, which is compatible with the assembly of prothrombinase being dependent on the flux of factor Xa to the phospholipid-bound factor Va. As long as factor Xa and prothrombin were present in the fluid phase the assembly of prothrombinase was apparently irreversible; during at least 20 min no loss of activity occurred. The steady-state rate of prothrombin activation was dependent on the surface concentration of prothrombinase, at 1.0 ΜM prothrombin and a shear rate of 82 s−1 the average rate was 870 mol thrombin/min per mol prothrombinase. In contrast to test tube experiments it was observed that in this flow system, the formation of Α-thrombin is favoured above the formation of meizothrombin (des fragment 1).

Cardiac ischemia causes interstitial leakage of cellular enzymes followed by release of these enzymes into plasma. Quantitative interpretation of these data requires a specific circulatory model, and the performance of such models was investigated. Plasma activities of cardiac enzymes were measured for increasingly abrupt forms of ischemic heart injury in the dog: 1) permanent ligation of the left anterior descending coronary artery (LAD); 2) reperfusion after 2 h of ligation of the LAD; and 3) calcium-free perfusion of the LAD during 10 min (calcium-paradox injury). Release into plasma of a rapidly (41%/h) and a slowly (2.2%/h) catabolized enzyme was calculated from the plasma activities, using a detailed circulatory model with compartments for heart, plasma, muscle, skin, and viscera. The time course of cellular enzyme leakage into interstitial space in the heart was calculated from release into plasma and a range of reported values for transendothelial permeability. Simplification to one- and two-compartment models introduced, respectively, 10 and 2% error in calculated cumulative release. Considering the other sources of error, this implies adequate performance of the two-compartment model. Protein washout from the heart is strongly influenced by expansion of interstitial protein space with dead myocyte volume and depends on the microheterogeneity of necrotic tissue areas. Accelerated release of enzymes into plasma after reperfusion reflects accelerated cellular leakage rather than enhanced washout.

Myocardial activities of several enzymes were measured in infarcted and non-infarcted areas of heart sections obtained from eight patients who died after acute myocardial infarction. Similar data were obtained from four patients with cardiovascular disorders who died from causes other than myocardial infarction and from six patients without previously known heart disease. It was found that both non-infarcted and infarcted tissue samples contained considerably altered enzyme activities. This finding explains the low correlations between enzymatic and histological estimates of infarct size previously reported. However, when the residual myocardial activities of different enzymes were compared with each other, a close correlation was found between creatine kinase, alpha-hydroxybutyrate dehydrogenase, and aspartate aminotransferase. It appears that the pathological changes in the myocardial activities of these enzymes may be explained by the phenomenon of diluted myocardium. This indicates that myocardial injury, as estimated from plasma enzyme activities, may still be expressed meaningfully in gram equivalents of healthy myocardium.

Myocardial injury after aorto-coronary bypass surgery was estimated in 72 patients from total release into plasma of cardiac creatine kinase (CK-MB) and alpha-hydroxybutyrate dehydrogenase (HBD). Activities of CK-MB were determined both by immuno-inhibition of CK-M units and by ion-exchange chromatography. After correction for per-operative hemolysis, the estimates based on HBD were in agreement with the estimates based on CK-MB as determined by the ion-exchange method. Both enzymes indicated a mean loss of only about 2 gramequivalents of myocardium. Such minimal injury was also found in metabolic and ultrastructural studies of myocardial biopsies in the same patients, as reported earlier [1,2]. However, approximately two-fold larger estimates of injury were obtained from plasma CK-MB activities determined by immuno-inhibition. This apparent extra release of CK-MB runs parallel with massive release of CK-activity from skeletal muscle damaged by surgery. Taking also into account the various calculation methods used by different authors, overestimates as large as 10–20 gram-equivalents of lost myocardium after uncomplicated bypass surgery, as published in the literature [3, 4], can be explained.

The changes induced by continuous aortic cross-clamping in combination with multidose ice-cold St. Thomas Hospital cardioplegia (myocardial temperature below 16 degrees C), or intermittent aortic cross-clamping at 34 or 25 degrees C were evaluated in a randomized study on 72 patients undergoing extensive aorto-coronary bypass surgery. The cumulative release of heart-specific enzymes was very small and no marked ultrastructural changes in mitochondria of both the subepi- and the subendocardial layer of the left ventricular free wall occurred. No differences between the three operation techniques could be observed on the basis of the above-mentioned parameters. Myocardial ATP and glycogen contents were decreased in post-ischaemic tissue in both the normothermic and hypothermic intermittent aortic cross-clamp groups. This decrease was associated with a release of lactate and inorganic phosphate during the repetitive periods of reperfusion. No change in myocardial ATP and glycogen content could be observed in the cardioplegia-treated hearts. St. Thomas Hospital cardioplegia is obviously most effective in preventing changes in myocardial metabolism such as reduction of ATP and carbohydrate stores during the reversible phase of ischaemic insult.

Soon after the discovery of myocardial enzyme release in patients with acute myocardial infarction (AMI) [269], the relationships were studied between maximal levels of aspartate aminotransferase (AST) or latate dehydrogenase (LDH) and histologically assessed infarct size after experimental infarction in dogs [334,399,82]. Although significant correlations were found, these were insufficiently high to allow accurate estimation of infarct size from maximal enzyme levels.

Estimation of enzyme release in plasma requires knowledge of the fractional catabolic rate constant (FCR) for the elimination enzyme activity from plasma. However, the total plasma content of such enzymes usually consists of several isoenzymes with different values of FCR. Thus, the use of a single overall value for FCR may cause error. This problem was studied by determination of the plasma isoenzyme activities of creatine kinase, lactate dehydrogenase, aspartate aminotransferase and alpha-hydroxybutyrate dehydrogenase in patients after cardiac surgery and after acute myocardial infarction. Values of FCR and the cumulative release of activity in plasma are estimated for separate isoenzymes and for total enzyme activity. Results are compared with the enzyme content of myocardium, skeletal muscle and blood cells. It is concluded that isoenzyme separation is not required for the quantitative use of such data. The implications for the validation of enzymatic estimation of cardiac injury are discussed. The results indicate that local inactivation of enzymes after cardiac injury must be limited.

The effects of early intracoronary streptokinase (SK) on enzymatic infarct size and rate of enzyme release were studied in a randomized multicenter trial. A total of 533 patients with acute myocardial infarction (AMI) were allocated to either the SK treatment group (n = 269) or the conventional (control) treatment group (n = 264). Enzymatic infarct size was represented by the cumulative quantity of alpha-hydroxybutyrate dehydrogenase (HBDH) released by the heart per liter of plasma in the first 72 hours. Rate of enzyme release was represented by the ratio of HBDH quantities released in 24 hours and 72 hours. On an "intention to treat" basis, the SK group had a smaller (by 30%; p = 0.0001) median enzymatic infarct size and a higher (by 35%; p = 0.0001) median rate of enzyme release than the control group. Limitation of infarct size was less apparent in patients treated with intracoronary SK only (25%) than in patients treated with intravenous plus intracoronary SK (34%). Compared to the control group, the enzyme release rate in patients treated with intracoronary SK only was slightly less (34%) than that in patients treated with intravenous plus intracoronary SK (38%). Patients with a patent infarct-related coronary artery at acute angiography had a median infarct size which was 55% (p = 0.0001) smaller than the median infarct size of the control group, and the median rate of enzyme release was 38% (p = 0.001) higher than the median release rate of the control group. Patients with successful recanalization during intracoronary SK infusion had a median infarct size which was 31% (p = 0.002) smaller than the median infarct size of the control group and a median rate of enzyme release which was 42% (p = 0.0001) higher than the median release rate of the control group. Patients with persistent coronary occlusion in spite of thrombolytic therapy had a median infarct size which was 11% (NS) higher than the median infarct size of the control group, although the median rate of enzyme release was still 23% (p = 0.02) higher than the median release rate of the control group. It is concluded that thrombolysis in the early phase of AMI limits infarct size and that intracoronary SK treatment itself accelerates the process of enzyme release from infarcted myocardium, independent of the angiographic result.

A method is presented for routine estimation of ultimate infarct size in patients with acute myocardial infarction (AMI). This method implies the calculation of the total quantity of alpha-hydroxybutyrate dehydrogenase (HBDH) released by the heart per liter of plasma, with the use of five to seven plasma samples per patient during the first 4 days after onset of infarction. The choice of HBDH for this purpose was motivated by a relatively small error in estimated enzyme release for slowly eliminated enzymes. The practicality of this method was studied in the coronary care unit at Leiden University Hospital where, in 1979, 146 patients with AMI were included in the study. In 100 patients (68%) HBDH-infarct size could be calculated precisely, and in 12 other patients (8%) the assessment of HBDH-infarct size was less accurate. In 34 patients (23%), HBDH-infarct size was unobtainable because of early death (eight cases), infarctions too small in size to assess or nonexistent (six cases), too much time elapsed since infarction (six cases), and incomplete plasma sampling (14 cases). Analysis of data shows that the larger HBDH-infarct sizes were found to be associated significantly with signs of heart failure, low stroke index and low ejection fraction, presence of tachycardias and interventricular conduction disturbances, high score of left ventricular wall motion abnormalities, large myocardial perfusion defect measured with thallium-201, in-hospital death, and death in the first year after infarction. A low but significant correlation was observed between HBDH-infarct size and the severity of coronary arterial lesions.

Pathological elevations of the plasma activities of liver enzymes are not simply related to the quantitative release of such enzymes from the liver. Several enzymatic indices, such as the well-known de Ritis quotient, may be determined by differences in the time course of hepatic enzyme release, rather than reflecting true differences in the released quantities of various enzymes. A more quantitative use of enzymatic data is hampered by the fact that the fractional catabolic rate constants for the elimination of enzyme activities from plasma are unknown. In the present study, three of these constants are estimated by comparison of the time-activity curves in plasma with the corresponding curve of a simultaneously released, more slowly eliminated reference enzyme. This method can be applied in patients with an acute short period of hepatic enzyme release. Values obtained for the cytosolic isoforms of lactate dehydrogenase, AST and ALT are: fractional catabolic rate constant (lactate dehydrogenase isoenzyme 5) = 0.13 +/- 0.01 hr-1, fractional catabolic rate constant (cytosolic AST) = 0.088 +/- 0.016 hr-1 and fractional catabolic rate constant (cytosolic ALT) = 0.034 +/- 0.004 hr-1 (mean +/- S.E., n = 10). These values are much higher than the apparent disappearance rate constants, because of extravascular return of activity and tailing release of enzymes during the major part of the elimination phase. It is shown that these results are consistent with earlier published data on the disappearance rates from plasma of lactate dehydrogenase, AST and ALT after acute liver injury. Cumulative release of various cytosolic enzymes occurred in proportion to the corresponding activities in human control livers.

The activity of alpha-hydroxybutyrate dehydrogenase, creatine kinase, creatine kinase MB and aspartate aminotransferase was measured on serial plasma samples from patients with acute myocardial infarction. The study was part of a multicentre randomised trial of the effect of thrombolytic treatment in the acute phase of acute myocardial infarction. The applicability and comparability of enzyme tests for the estimation of myocardial injury were studied in 76 control patients and 74 patients treated with streptokinase. Treatment with streptokinase caused a considerable acceleration of enzyme release after acute myocardial infarction, both in patients with persistent coronary occlusion and in those with successful reperfusion. But this changed pattern of enzyme release did not affect the rate of enzyme elimination from plasma or the released proportions of different enzymes. Thus the assessment of infarct size by measurement of these enzyme activities can also be applied to patients treated with streptokinase. Moreover, the enzymes measured in the present study are all equally valid markers of myocardial injury.

Plasma activities of various (iso)enzymes were measured in patients after cardiac surgery (n = 114) and after acute myocardial infarction (n = 40). From these activities, the cumulative release of enzymes in plasma was calculated with a two-compartment circulatory model. This model was adapted to transient postoperative changes in plasma volume and similar changes in the transcapillary escape rate of proteins, observed after cardiac surgery and verified in dogs after cardiopulmonary bypass (CPB). Comparison of cumulative release of enzymes with the enzyme content of myocardium, skeletal muscle, and blood cells allows identification of the various sources of enzyme release. Cardiac injury after uncomplicated bypass surgery is only 1.5 +/- 1.5 (mean +/- SD) gram equivalents (gmEq) of myocardium, compared to a loss of 31 +/- 13 gmEq of myocardium after AMI. Peroperative hemolysis is estimated at 68 +/- 15 ml of blood. Total loss of skeletal muscle amounts to 13 +/- 10 gmEq. Some hepatic enzyme release is observed after AMI but not after surgery. Large differences in time course exist between the release of enzymes from myocardium and skeletal muscle and also between myocardial release in the surgery group and in the AMI group. The accuracy of estimations is discussed and indicated as a function of the extent of cardiac injury.

This study concerns the existing controversy about the validity of one-vs two-compartment models for the distribution of intravenously injected enzymes in the dog. It is shown that infusion of enzyme at a constant rate permits direct calculation of enzyme clearance constants, independently of the specific model used. Enzyme preparations were obtained by leakage of enzymes from heart tissue under hypoxic conditions. The values for the plasma clearance constants k of creatine kinase (CK) and aspartate aminotransferase (AST) are 0.36 ± 0.030 h−1 and 0.21 ± 0.014 h−1 respectively (mean ± SE, n = 8). These values are compared with clearance rates as estimated from a bolus injection of enzyme given 90 min before the infusion was started. For both enzymes the values of k as estimated from the infusion equal the apparent disappearance constants kd obtained from the preceding bolus injections. From these experiments it is concluded that plasma clearance of CK and AST is adequately described by a one-compartment model. In 10 experiments plasma clearance curves after bolus injections were sampled over 8 h. Disappearance of AST was strictly mono-exponential, while for CK a slight biphasic effect was observed in five out of ten cases resulting into improved double-exponential fits. It is concluded that this effect is not due to exchange of enzyme with an extravascular pool but is caused by a fraction of enzyme with a slightly higher elimination rate. This phenomenon is illustrated by showing that AST also shows biphasic disappearance upon injection of an enzyme preparation obtained by mechanical disruption of cells. Individual and day-to-day variations in k values obtained for CK and AST do not show any correlation. This suggests independent clearance mechanisms for both enzymes.

The concentrations in plasma of several proteins show major changes in response to tissue damage such as caused by surgery, bone fractures or myocardial infarction. This so-called acute phase response consists of a prolonged increase of plasma concentrations for fibrinogen, alpha1-antitrypsin, haptoglobin and alpha1-acid glycoprotein. Moreover, there appears in the plasma a considerable quantity of a protein, called C-reactive protein, which normally is absent or only occurs in trace amounts. Such proteins, defined as trauma-induced liver-produced glycoproteins [262], are called acute phase reactants. The proteinase inhibitor alpha2-macroglobulin has also been described as an acute phase reactant by some authors, but this was denied by others [262].