A European view on the North American fifth consensus on antithrombotic therapy.

An American-Canadian group of experts have, in the November 1998 issue of CHEST, published for the fifth time their recommendations for antithrombotic therapy. This remarkable consensus document was the result of an extensive review of the literature by an interdisciplinary group. Considering the impact of this document on medical practice, also outside North America, a group of European experts reviewed in detail the fifth report, particularly the sections on clinical indications of antithrombotic treatment. The aim was not to indicate the many areas of agreement and to quote literature that has become available since publication of the last consensus documents, but rather to refer to the gray zones of uncertainty and limited number of divergent opinions.

Are symptomatic endpoints acceptable in venous thromboprophylactic studies?

Chemoprophylaxis with subcutaneous low-molecular-weight heparin has been shown to reduce deep vein thrombosis detected by surrogate endpoints such as fibrinogen scanning and venography. However, there have been few trials which have assessed clinical endpoints attributed to fatal and non-fatal pulmonary embolism following surgery. As these clinical endpoints are rare, large-scale trials using vascular mortality, vascular morbidity and the incidence of haemorrhage as clinical endpoints need to be performed to assess the efficacy of chemical thromboprophylaxis. The benefit of using clinical endpoints against the risk of haemorrhage should also be evaluated.

Aspirin: benefit and risk in thromboprophylaxis.

Aspirin is often perceived either as a harmless panacea or as a useless poison which causes endless, needless trouble. We have carefully reviewed the literature on all aspects of aspirin and find that neither view is justified. Regular use of even low-dose aspirin (150 mg/day or less) may lead to clinically-important adverse events, particularly haemorrhage. The risk of such an event is considerably outweighed by the benefit for patients with a significant risk of a thromboembolic event. For individuals without a clear risk of thrombosis or thromboembolism, the balance is more even: indiscriminate aspirin-taking is to be discouraged.

Von Willebrand factor and risk of ischemic stroke.

We carried out a case-control study to determine whether von Willebrand factor (vWF) antigen (and factor VII and tissue plasminogen activator [tPA] antigens) are associated with ischemic stroke. Ninety-five patients with transient ischemic attack or minor ischemic stroke recruited from the Oxfordshire Community Stroke Project and one neurology clinic were compared with 236 controls, group-matched for age and sex, from the same general practitioners as the incident cases. In crude analyses, concentrations of vWF antigen were significantly higher in cases than in controls (p = 0.004). The age- and sex-adjusted odds ratios from lowest (referent) to highest quartile of vWF antigen were 1.00, 1.15, 2.34, and 2.36 (trend test, p = 0.006). Factor VII antigen and tPA antigen were not significantly different between cases and controls. Although adjustment for other potential risk factors abolished the statistical significance of the association between vWF and disease, this was largely due to the influence of history of ischemic heart disease. We conclude that vWF is a potent and independent risk factor for transient ischemic attack, minor ischemic stroke, and, by extrapolation, ischemic stroke in general. The data also suggest that vWF may be a risk factor for both ischemic stroke and ischemic heart disease. We found no evidence to implicate factor VII and tPA as risk factors for ischemic stroke.

The effect of different microtitre plates on the enzyme-linked immunosorbent assay (ELISA) for anticardiolipin antibodies (ACAs)

Studies have shown an association between the presence of ACAs, measured by an ELISA, and thrombosis, foetal loss and thrombocytopenia. Most patients with the syndrome have systemic lupus erythematosus (SLE) or a related autoimmune disease. In studies of the rates of thrombotic episodes in patients with the lupus anticoagulant (LAC) and, or ACAs associated with SLE and non- SLE diseases there is considerable variation. For instance, investigations of women with a poor obstetric history: 13.1% had ACAs, 42.4%, 81.5%, others have found a lower frequency of habitual aborters with the LAC and, or ACAs arguing that ACAs play only a minor role in spontaneous abortion.

A comparative trial of a low molecular weight heparin (enoxaparin) versus standard heparin for the prophylaxis of postoperative deep vein thrombosis in general surgery.

BACKGROUND: Various studies have been performed in general surgery patients comparing low molecular weight heparin (LMWH) with standard heparin (SH) for the prevention of postoperative deep vein thrombosis (DVT), revealing contradicting results. Therefore, we have compared the efficacy and safety of a LMWH for the prevention of DVT after major general surgery. PATIENTS AND METHODS: Patients received either 20 mg LMWH (enoxaparin) once daily, or 5,000 IU SH TID, starting preoperatively in a prospective, randomized, double-blind international multicenter trial. DVT was diagnosed using fibrinogen I 125 leg scanning. Major and minor bleeding were assessed clinically. RESULTS: A total of 718 patients were randomized to LMWH, and 709 patients to SH. DVT was detected in 58 LMWH-treated patients (8.1%, 95% confidence interval [CI] 6.2% to 10.3%) and in 45 patients allocated to SH (6.3%, 95% CI 4.7% to 8.4%, P > 0.05). Major bleeding complications occurred in 11 LMWH-treated patients (1.5%, 95% CI 0.8% to 2.7%) and in 18 patients to whom standard heparin was administered (2.5%, 95% CI 1.5% to 3.9%, P > 0.05). Four LMWH-treated patients (0.6%) required reoperation for bleeding as compared to 13 patients in the SH group (1.8%, P = 0.03). CONCLUSION: This LMWH appeared as effective and safe as SH. In view of its more convenient way of administration, this LMWH might be preferred for thromboprophylaxis.

Antithrombotic therapy in the secondary prevention of myocardial infarction.

There is now convincing evidence that the majority of myocardial infarcts are caused by acute occlusion of the coronary artery with a thrombus, often originating from a ruptured atherosclerotic plaque. The thrombus is composed of both platelets and fibrin, so it is logical to expect that both antiplatelet and anticoagulant therapy would be effective for the prevention of reinfarction and other thrombotic vascular events following myocardial infarction. A review of clinical trials confirms this suggestion. Aspirin is beneficial in that it reduces vascular mortality by about 12% and total vascular events by 25%. Similarly, oral anticoagulants reduce total mortality by up to 20% and vascular events by about 40%. However, relatively few large scale trials of oral anticoagulants after infarction have been carried out, and the confidence intervals are wider than those for the aspirin trials. In practice aspirin is used more widely than oral anticoagulants as it is easier to administer and control. Only one trial (EPSIM) has directly compared oral anticoagulants and aspirin. This trial showed that both drugs had equal efficacy but that compliance with aspirin was better. Finally, the question of giving antiplatelet agents and anticoagulants together, to block both platelet aggregation and fibrin formation, is considered. A preliminary trial in prosthetic heart-valve patients has been encouraging but more large, long-term trials are required before recommendations can be made, as the use of both drugs together carries an increased risk of severe hemorrhage.

Aprotinin reduces cardiopulmonary bypass-induced blood loss and inhibits fibrinolysis without influencing platelets.

Cardiopulmonary bypass (CPB) induces a bleeding defect which leads to enhanced blood loss. A double-blind study was carried out comparing aprotinin with placebo in patients undergoing re-operation for heart valve replacement. The results confirm that aprotinin is effective at reducing such loss. In the placebo treated group, significant increases were observed, during CPB, in the plasma concentrations of fibrinolytic activity, tissue plasminogen activator antigen, D-dimer, and beta-thromboglobulin. Platelet counts fell within 5-10 min of the patients going onto CPB, but this could be accounted for by the dilutional effect of the extracorporeal circuit. Inhibition of responsiveness of platelets, as judged by aggregometry, was significant only at the end of bypass when collagen was the agonist and after protamine reversal when ristocetin was the agonist. CPB did not enhance the release, into the circulation, of glycocalicin (a proteolytic fragment of glycoprotein Ib). In the aprotinin-treated group, the formation of fibrin degradation products as measured by D-dimer was inhibited. However, aprotinin did not influence the change in platelet count, suppress beta-thromboglobulin release from platelets, prevent the inhibition of platelet function or influence the concentration of plasma glycocalicin during the study period. These observations confirm that CPB leads to a fibrinolytic state and less responsive platelets. This study also indicates that aprotinin-induced reduction in blood loss is associated with inhibition of plasmin-mediated fibrin digestion and that the mechanism by which aprotinin reduces blood loss is not via protection of platelets during CPB.

The fibrinolytic response to ancrod therapy: characterization of fibrinogen and fibrin degradation products.

Ancrod is a purified coagulant venom which renders blood incoagulable by cleaving fibrinopeptide A (FPA) from fibrinogen, but the mechanism involved in the clearance of fibrin from the circulation is unknown. To investigate the fibrinolytic response to ancrod, and to increase understanding of clearance mechanisms, six patients with peripheral vascular disease causing claudication were infused with ancrod at 2 u/kg over 6 h followed by 2 u/kg at 12 h intervals for 38 h. Venous blood samples were taken at time 0, 3, 6, 25 and 49 h for assay of fibrinogen (Fbg), fibrinopeptide A (FPA), total fibrin(ogen) degradation products (TDP), fibrin degradation products (FbDP), fibrinogen degradation products (FgDP), cross-linked fibrin degradation products (XL-FDP), tissue plasminogen activator (tPA), urinary type plasminogen activator (u-PA), plasminogen, alpha 2 antiplasmin (alpha 2 AP) and plasminogen activator inhibitor-1 (PAI-1). Fibrinogen (median and range) was 2.3 (1.4-3.90) g/l at time 0 and thereafter was undetectable. FPA rose from 2.5 (1.8-3.6) to 600 and 188 pmol/l at 3 h and 6 h and remained elevated. TDP, FbDP and FgDP increased greatly following ancrod while there was no evidence of XL-FDP. The surprising increase in FgDP during defibrination suggests either that fibrinogen is digested following its incorporation into circulating fibrin protofibrils or that some of the fibrin subunits in the photofibril retain one of the two fibrinopeptide A's. tPA and uPA remained unchanged. Plasminogen fell from 125 (100-155)% to 79 (40-118)% at 49 h and alpha 2 AP fell from 91 (75-107)% to 24 (10-35)% at 49 h. The level of PAI-1 was depressed during defibrination, with the exception of the 6 h data. The results demonstrate that ancrod removes FPA from fibrinogen to produce non-cross-linked (soluble) fibrin. This is cleared from the circulation without evidence of an increase in the circulating activities of the plasminogen activators, tPA or UK, but with evidence of plasminogen activation and consumption.

The investigation of a heart valve patient with suspected hypercoagulability.

A major complication associated with prosthetic heart valves is valve thrombosis and embolization. Depending on the valve design these complications should be almost eliminated by anticoagulants or aspirin. However, valve thrombosis and embolism may occur despite apparently adequate anticoagulation. The question is whether these thrombi arise as a result of local factors such as the foreign surface and disturbance of flow at the valve site, or as a result of alteration of the hemostatic balance within the blood favoring thrombus formation. Such a disturbance, known as hypercoagulability, is the result of increased platelet activation or coagulation, or diminished fibrinolytic activity. In this article the underlying mechanism of the coagulation and fibrinolytic mechanisms is discussed, with particular emphasis on the pathways, by which activation of the coagulation system or an increase in the inhibition of fibrinolysis might occur. The laboratory tests which might be useful in detecting platelet activation, activation of coagulation or fibrinolytic activity are discussed. The potential significance of raised blood viscosity associated with a rise in fibrinogen is emphasized, in particular the fact that elevated fibrinogen is a risk marker of many types of occlusive vascular disease. Whether raised fibrinogen is causal in the pathogenesis of thrombosis at the valve site, or merely is a consequence of the prosthesis, is not known. There is some information on the fact that patients with intracardiac, and left atrial thrombi in atrial fibrillation, have activation of coagulation as shown by elevated levels of markers and also raised von Willebrand factor levels. However, it is emphasized that large scale prospective studies, in which the risk markers are measured, and then the patients with prosthetic valves followed to determine whether or not they experience clinical embolic events has not been carried out. It is unlikely that a single hemostatic test, or indeed, a number of tests, will be able to predict with accuracy the individual who is at particular risk of embolism. Nevertheless, it is important that we seek to gain further information of the mechanisms of prosthetic valve thrombus formation, and then apply this knowledge in controlled clinical trials. Only this way will better anticoagulant and antithrombotic control of these devastating embolic events be achieved.

Levels of plasma fibrinopeptide-A during surgery and the severity of the operation.

There is evidence to suggest that the rise of fibrinopeptide A (FPA) during surgery is influenced by tissue thromboplastin released during tissue damage. To investigate whether FPA correlates with the severity of the damage of the operation, 39 patients were recruited and venous blood samples were taken pre-operatively, during skin incision, during bowel manipulation and post-operatively for the assay of FPA. The operations are grouped as minor (A), moderate (B), major (C) and very major (D). The peak FPA levels occurred during bowel manipulation in every degree of operations, and ranged between 6.0 to 22.2 pmol/ml. There was a tendency that peak FPA values rose according to the severity of the surgery, however only very major operations (D) are significantly higher when compared with minor operations (A) (p < 0.01). There was no good correlation between peak FPA levels and length of skin incision (p = 0.83, r = 0.04) as well as peak FPA levels and duration of operation (p = 0.90, r = 0.03). Significantly higher levels of FPA in very major operation (D) was due to more excessive tissue damage during surgery, while due to relatively minimal tissue injury, size of skin incision correlated poorly with FPA.

Treatment of hypertension induces a fall in platelet basal cytoplasmic calcium concentration without influencing platelet aggregation.

The relationship between blood pressure and platelet basal cytoplasmic calcium concentration ([Ca2+]i) and platelet sensitivity to aggregating agents in hypertension has been investigated in hypertensive patients and normotensive subjects. Ten severely hypertensive patients whose blood pressures were poorly controlled with metoprolol, were given calcium antagonist (either nifedipine or felodipine) as a second line agent. Venous blood samples were collected at each treatment phase for measurement, in whole blood, of platelet aggregation in response to ADP and collagen, and of basal [Ca2+]i using fura-2. Control of blood pressure by the combination of metroprolol and a calcium antagonist induced a significant decrease in median [Ca2+]i from 116 (76-181) to 73 (60-83) nM, which was similar to the median value of 70 (61-80) nM obtained in 14 normotensive subjects. Overall [Ca2+]i correlated with mean blood pressure (r = 0.51). Treatment of hypertension with calcium antagonist did not change the response of platelets to collagen or ADP. The results confirm that effective treatment of hypertension significantly reduced basal [Ca2+]i in platelets but raise doubts whether elevated basal [Ca2+]i is necessarily the sole mechanism by which the sensitivity of platelets to aggregatory agents is increased in hypertension.

Reduction in factor VII, fibrinogen and plasminogen activator inhibitor-1 activity after surgical treatment of morbid obesity.

The aim of this study was to determine the effects of the surgical treatment of morbid obesity on some aspects of haemostatic and fibrinolytic function. Measurement of haemostatic and fibrinolytic factors was performed before and again 6 and 12 months after operation in 19 patients suffering from morbid obesity. Surgical treatment resulted in a mean decrease in body weight of 50 kg at 6 months and 64 kg at 12 months. Weight loss was accompanied at 12 months by significant reductions in median (interquartile range) concentrations of serum cholesterol from 5.3 (4.5-6.2) mmol/l to 3.6 (2.9-4.6) mmol/l; factor VII from 113 (92-145)% of normal to 99 (85-107)%; of fibrinogen from 3.5 (3-9.3) g/l to 2.8 (2.4-3.8) g/l; and of plasminogen activator inhibitor-1 (PAI-1) activity from 21 (11-30) IU/ml to 6.3 (5-10) IU/ml. The decrease in PAI-1 activity probably accounted for a significant reduction in euglobulin clot lysis time. Tissue plasminogen activator activity was undetectable in most patients pre-operatively but increased slightly after 1 year to 110 (100-204) mIU/ml. There were no significant changes in plasma levels of KCCT, factor VIII, von Willebrand factor antigen, alpha-2-antiplasmin, antithrombin III, protein C antigen, beta thromboglobulin, platelet factor 4, fibrinopeptide A or platelet count. These findings provide support for the hypothesis that the surgical treatment of morbid obesity may have a long-term beneficial effect on mortality from cardiovascular and thromboembolic disease.

Effects of pimobendan (UDCG 115) on renal function in healthy volunteers.

A phase I, double-blind, single-dose, randomized, two-period crossover study was conducted to investigate the effects of pimobendan on renal function to assess whether renal events were a contra-indication to its administration. Results in eight healthy volunteers indicated no significant adverse events on renal plasma flow, glomerular filtration rate, or laboratory safety screens; side-effects were also found to be minimal. Further studies are indicated to assess whether, in the proposed treatment group, i.e. patients with heart failure (in whom compromised renal function is common), pimobendan similarly elicits no serious adverse renal effects.

Vasopressin and haemostatic responses to inguinal hernia repair under local anaesthesia.

Major abdominal surgery is accompanied by intra-operative increases in factor VIII (FVIII), plasminogen activator activity (PAA) and fibrinopeptide A (FPA). Vasopressin (aVP) released during surgery mediates some of the effects but the mechanisms involved in this response are unclear. To investigate the role of the operative procedure, 20 subjects were studied during inguinal hernia operation under local anaesthesia. Venous blood samples were taken for FVIII coagulant activity (FVIII:C), euglobulin clot lysis time (ECLT), FPA, crosslinked FDPs (XL-FDP) and a VP. In six patients, aVP rose from (median) 0.5 to 38.3 pg/ml at bowel manipulation and fell to 4.1 pg/ml post-operatively. PAA rose from 33 units to 377 and 316 units (P less than 0.01), FVIII:C from 1.58 to 2.4 IU/ml (P less than 0.01) and FPA from 5.0 to 6.8 and 11.0 pmol/ml intra-operatively (P less than 0.002). XL-FDP rose from a median value of 34 ng/ml pre-operatively to 230 ng/ml post-operatively. In 14 patients plasma aVP levels remained constant and both FVIII:C and PAA remained unchanged. FPA rose from 2.6 pmol/ml to 5.9 pmol/ml intra-operatively (P less than 0.05) and XL-FDP fell from 110 to 60 ng/ml. Between groups, the changes were significantly different for FVIII:C (P less than 0.05) and PAA (P less than 0.03) with no differences in blood pressure, pulse or symptoms. These results support the hypothesis that aVP secretion during surgery mediates increases in FVIII and PAA. FPA tended to be higher in the aVP secreting group which indicates that aVP mediated activation of coagulation results in a hypercoagulable state.