Drugs 13: 46-51 (1977) (CJ ADIS Press 1977
Heparin Therapy: Regimens and Management D.Deykin Boston Veterans Administration Hospital, Boston, Massachusetts
Summary
Heparin remains the most effective antithrombotic drug. It acts by combining with plasma antithrombin. thereby accelerating the neutralisation of thrombin and other activated coagulation factors. Full-dose intravenous heparin is indicated in all cases of pulmonary embolism and established deep venous thrombosis. unless there exist compelling contra indications. Continuous intravenous infusion of heparin appears to be safer than intermittent injection. Low-dose subcutaneous heparin is effective in preventing the initial occurrence of thigh vein thrombi and in reducing the incidence of fatal pulmonary embolism in general surgical patients over the age of 40. The efficacy· of low-dose heparin in preventing pulmonary emboli folloWing hip surgery has not been established. The incidence of severe heparin-induced thrombocytopenia appears to be rising. Platelet counts should be performed in all patients receiving heparin by any mode of administration.
Heparin is the most potent and effective antithrombotic drug in clinical use. The rationale for heparin therapy derives exclusively from its ability to block the coagulation mechanism. Blood coagulation, however, has only a limited role in the pathogenesis of arterial thrombosis, which arises primarily through the interaction of circulating blood platelets with an injured or abnormal vessel wall (Dey kin , 1967). Only when the arterial lumen becomes almost totally occluded by atheroma does the coagulation mechanism become dominant. Therefore, anticoagulant therapy offers little benefit in the prevention of primary or recurrent arterial thrombosis. A different mechanism occurs in veins, since in most venous thrombi there is no evidence for antecedent vascular injury. Although the initiating stimulus for venous thrombosis is unknown, venous thrombi
grow by extension from valve cusps into the vein lumen, propagating through coagulation (Sevitt, 1969). Accordingly, the most uniformly accepted indications for heparin therapy are those in which the coagulation mechanism plays a major role in thrombogenesis: deep thrombophlebitis and pulmonary embolism. An additional indication is the prevention of distal propagation of stasis thrombi following acute arterial' occlusion in the interval before attempted surgical restoration ·of patency.
1. Mode of Action
Heparin acts by binding to a plasma protein, antithrombin, thereby enhancing the ability of antithrombin to neutralise activated clotting factors in the
47
Heparin Therapy
intrinsic limb of the coagulation pathway culminating in the formation of thrombin. The evolution of our understanding of heparin's anticoagulant action has been reviewed by Rosenbert (Rosenberg, 1975) who, with his associates, has clarified the biochemistry of the heparin-antithrombin interaction. Brinkhous (Brinkhous et aI., 1939) and his associates first showed that heparin was an anticoagulant only in the presence of a plasma protein which they called heparin co-factor. Several groups subsequently demonstrated the identity of heparin co-factor with plasma antithrombin. In the absence of heparin, antithrombin combines with the active serine site of thrombin, causing a slow irreversible inhibition of the enzyme. When heparin binds to antithrombin, at a site distinct from that to which heparin binds, it alters antithrombin so that its ability to form a complex with heparin is markedly enhanced. In the presence of heparin the interaction of heparin with antithrombin is virtually instantaneous (Rosenberg, 1973). In addition to prothrombin, four other coagulation proteins are converted to active serine proteases during the process of blood clotting: factor XII (Hageman), XI (PTA), IX (PTe), and X (Stuart). Antithrombin binds to each of these proteases neutralising their activity. Indeed, the affinity of the heparin-antithrombin complex for activated factor X far exceeds that for thrombin itself, suggesting that the true physiological 'target' of antithrombin is activated factor X rather than thrombin (Yin et aI., 1971).
venous thrombi from initially forming (by inhibiting the generation of thrombin) than in prevention of the extension of existing thrombi (after thrombi generation has taken place). We have come to realise, somewhat belatedly, that there are varying degrees of clinical urgency that dictate the initiation of prophylactic heparin therapy. Pulmonary embolism is a recurring, lethal disorder. As a result, once diagnosed, either pulmonary embolism or its precursors, deep venous thrombosis of the pelvis or lower extremity, require prolonged, intensive heparin therapy followed by an extensive period of treatment with oral anticoagulant therapy. Such full dose heparin has been repeatedly shown to be highly effective, but it carries a high morbidity rate. Indeed, between 10 to 20 % of patients receiving conventional intermittent intravenous heparin experience major bleeding during the course of treatment (Gallus and Hirsh, 1976). It is clear that prevention of the initial episode of venous thrombosis may be accomplished by pretreating patients at risk with low doses of heparin, given subcutaneously with little if any risk of haemorrhage. The two modes of heparin therapy - full-dose, intravenous and low-dose, subcutaneous - are quite distinct and should not be confused. Full-dose heparin is indicated once pulmonary embolism or thrombosis of the deep veins of the thigh or pelvis have been documented. Low-dose therapy is useful only to prevent thrombosis from first occurring.
3. Full-Dose Intravenous Therapy 2. Clinical Action and Dose Regimens
Two generalisations emerge from the foregoing discussion. The first is that heparin has no effect on existing thrombi except to prevent distal propagation and to limit thrombin mediated platelet accretion on the surface of the thrombus. Therefore, all heparin therapy is prophylactic. The second is that a biochemical basis exists for the suggestion that lower doses of heparin may be more effective in preventing
Although full-dose intravenous heparin therapy has been employed for over 35 years, astonishingly few reports provide conclusive evidence for its efficacy. Furthermore, little agreement exists concerning the most satisfactory schedule for heparin administration, the need for control of dosage by laboratory tests, the method of laboratory control, or the intensity and duration of therapy. The only controlled study of the use of heparin was reported by Barritt and Jordan (Barritt and Jordan, 1960) who
48
Heparin Therapy
showed in 1960 that once pulmonary embolism had occurred, there was a high recurrence rate in untreated patients and that of those who experienced recurring pulmonary embolism, over half died. Treatment with heparin followed by oral anticoagulants drastically reduced the recurrence and eliminated deaths from pulmonary embolism. Subsequent reports have confIrmed the high incidence of lethal recurrence of untreated pulmonary embolism and the low incidence of fatal recurrent emboli when patients receive heparin. Although these reports lack the authority of Barritt and Jordan's controlled study, they substantiate their fIndings. Few would now argue against the currently accepted practice that all patients who have experienced proven pulmonary embolism must receive full dose intravenous heparin unless there are compelling contraindications.
3: I Intermittent or Continuous Intravenous Administration We have recently conducted a study of the incidence of bleeding in patients receiving heparin by intravenous injection (Salzman et aI., 1975). Among 100 consecutive patients, major bleeding occurred in 21. Two patients died from bleeding and two had recurrent pulmonary embolism. The incidence of bleeding was the same when therapy was regulated by clotting tests or when heparin was given without clotting tests. In a subsequent prospective trial, we gave patients heparin by one of three regimens: intermittent intravenous injection, with regulation of the dose according to the activated partial thromboplastin time (APTT) (72 patients); intermittent intravenous injection with a fIxed dose of heparin (68 patients); or continuously by intravenous infusion, with the dose regulated by the APTT (table I). Recurrent (non-fatal) thromboembolism occurred once in each group. Major bleeding was 7 times more frequent with intermittent injection (regulated or fIxed dose) than with continuous infusion. We concluded that continuous infusion of heparin (regulated by the APTT) was safer than intermittent injection of heparin and was equally
effective in preventing thromboembolism. In our study, we attempted to maintain the APTT at approximately twice baseline levels. Several reports have suggested that the risk of recurring thromboembolism is correlated with inadequate anticoagulant response to heparin. The correlation between increased risk of bleeding and excessive anticoagulation is less clear-cut, since many other determinants (particularly recent surgery, and concomitant use of drugs which affect haemostasis) may also influence the incidence of bleeding.
3.2 Intravenous Dosage Schedules Current evidence suggests, therefore, that continuous intravenous infusion of heparin is preferable to intermittent injection, put both provide a high degree of protection against recurrent, lethal pulmonary embolism. It is our practice to initiate continuous infusion of heparin with a 'loading dose' of 5,000 units given as a bolus followed by a continuous infusion at an initial rate of 1,000 units/hour, subsequently regulated to keep the APTT between 50 to 80 seconds. In those instances when we give heparin by intermittent injection', we use a standard dose of 5,000 units every 4 hours, adjusting the dose down or up for extremes of body weight. There can be no escaping the bleeding complications of full-dose intravenous heparin, but skillful attention to the details of patient management including scrupulous avoidance of injections and of drugs which impair haemostasis, will reduce the incidence of major bleeding.
4. Low-Dose Subcutaneous Therapy
The introduction of low-dose heparin has been inextricably intertwined with the use of 12SI-labelled fIbrinogen as a method for diagnosing the presence of venous thrombosis. Early studies indicated that the frequency of 'deep venous thrombosis' (DVT) of the calf was extremely high in postoperative patients and
Heparin Therapy
49
Table I. Influence of heparin regimen on bleeding, recurrence, and total daily dose Heparin regimen
Every 4 hours with PTT
Every 4 hours No PTT
Continuous infusion with PTT
Number of patients
72
68
69
Major bleeding
6 (8%)
7 (10%)
1 (1%)
31,700
35,600
24,500
Recurrences Daily heparin dose (units)
in elderly patients subjected to protracted bed rest (21-24). The preoperative administration of low-dose heparin clearly reduced the incidence of 'DVT' in the calf, but there was little substantial evidence to prove that such reduction resulted in prevention of fatal pulmonary embolism. Gallus and Hirsh (Gallus and Hirsh, 1976a) have lucidly summarised both the enthusiasm which attended the introduction of lowdose heparin and the limitations of the first reports of both its usefulness and its ineffectiveness.
4.1 Use in General Surgical Patients The recently completed multi-centre trial (International Multicentre Trial, 1975) of the use of low-dose heparin has resolved many of the first lingering doubts over the use of low-dose heparin in general surgical patients over the age of 40. Of the 4,121 patients in the final study group, 2,045 received 5,000 units of heparin subcutaneously before surgery and then every 8 hours after surgery for at least 7 days. The control group was untreated. There were 100 deaths in the control group and 80 in the treated group. Pulmonary embolism was found in 22 of the 72 autopsies in the control group and were judged to be the primary cause of death in 16 .. Pulmonary embolism was found in 5 of 52 autopsies in the treated group and was judged to be the primary cause of death in 2. The difference in incidence of all pulmon-
ary emboli and of 'lethal' pulmonary emboli was sig-
nificantly lower in the treated group. In a subset of patients, 1251-labelled fibrinogen leg scans were performed. Of the 667 control patients tested, 49 had thrombi that extended into the thigh, while of the 625 treated patients tested only 5 had thrombi above the knee. There was no difference in bleeding between the two groups. It is extremely unlikely that another study of such magnitude and excellence will be soon repeated. Therefore, the International Multicentre Trial carries a degree of authority similar to that of the earlier study of Barritt and Jordan and makes it mandatory that low-dose heparin prophylaxis should be routinely employed in general surgical patients over the age of 40. It is our practice to initiate treatment with 5,000 units given subcutaneously 2 hours prior to surgery and every 8 hours thereafter for a minimum of 7 days or until the patient is fully ambulatory.
4.2 Use in Hip and Prostatic Surgery Low-dose heparin may not be effective in preventing post-operative thigh and pelvic thrombi in certain operations which are associated with a much higher risk of fatal pulmonary embolism than the 0.7 to 1.0 % incidence following general abdominal or thoracic surgery. Several authors suggest that lowdose heparin does not reduce the incidence of
Heparin Therapy
venographically proven femoral or pelvic thrombi nor the risk of pulmonary embolism. Sagar and his associates disagree (Sagar et aI., 1976). In a study of 82 patients undergoing elective hip replacement, 52 received low-dose heparin (alone or in conjunction with dihydroergotamine) and 32 received no prophylaxis. In the control group, 22 developed '2SI-labeUed fibrinogen 'OVT of which 17 were shown to be femoral by venography, with extensive thrombi found in 9 cases. 12 of the heparintreated patients developed isotopic evidence for 'OVT of which 4 were found to be femoral by venography. None were judged to be extensive. Sagar and his coworkers conclude that 'in patients undergoing totalhip replacement, low-dose prophylaxis is effective in reducing the frequency of life-threatening extensive thrombi in the femoral vein'. Harris (Harris, 1976) has criticised Sagar's study, pointing out that fibrinogen scanning will not detect most instances of major thrombi in the thigh. Since Sagar performed venography only if 'OVT were detected by scanning, the true incidence of femoral vein thrombi in the treated group may have been underestimated. At the moment, the efficacy of low-dose heparin in hip (and prostatic) surgery remains to be proved, but these should not restrain the use of low-dose heparin as routine prophylaxis in general surgical patients.
5. Complications of Heparin: Thrombocytopenia
The major complication of heparin, bleeding, is a direct extension of its therapeutic action. However, another complication of heparin, thrombocytopenia, is not receiving increased attention. Oa::asional, mild thrombocytopenia has been noted following heparin treatment, and scattered reports have documented rare, profound thrombocytopenia during heparin therapy. Babcock (Babcock, 1976) and his associates have recently reported 5 patients in whom severe depression of the platelet aggregometry (Dey kin and Hellerstein, 1972). In 4, antibodies were also found by the platelet factor 3 availability technique (Kar-
50
patkin, 1969). All 5 patients had either short activated partial thromboplastin time or were within the normal range. Although extensive studies of the coagulation mechanism were not performed, no evidence of disseminated intravascular coagulation were found in the patients studied, and no episodes of major haemorrhage occurred. These patients all developed heparin-induced apparently immune thrombocytopenia within 6 weeks of another at one medical centre. In a parallel study, Bell and his co-workers (Bell et al., 1976) reported the results of their prospective study of the incidence of thrombocytopenia in 52 patients receiving continuous-infusion heparin. Their data were startling: 16 patients (30 %) developed platelet counts less than 100,000/ pI; indeed, in 9, the count was less than 50,000/ pI. In I 0 of the I 6 thrombocytopenic patients, fibrinogen-fibrin degradation products were elevated. In contrast, only 2 of the 32 non-thrombocytopenic patients had high split products. Hypofibrinogenemia was present in 5 thrombocytopenic patients and in none of the nonthrombocytopenic patients. Bell and his associates did not perform immunological assays. In 2 thrombocytopenic patients, the partial thromboplastin times (and prothrombin times as well) were prolonged. Although excessive bleeding from venipuncture sites developed in 2 thrombocytopenic patients, there was no other evidence of untoward bleeding. There seems to be no relationship between the method of administering heparin and the development of thrombocytopenia. Babcock's patients received heparin by intermittent injection, but Bell's patients received it by continuous infusion. Furthermore, Bell reported an additional 2 patients who developed thrombocytopenia while receiving lowdose subcutaneous heparin. What are we to make of this apparent outburst of heparin-induced thrombocytopenia? In both recent studies, there was little evidence of exaggerated bleeding. It is possible, therefore, that a high incidence of heparin-induced, severe thrombocytopenia has hitherto gone undetected. It is also possible that the increasing tendency to polypharmacy or increased use
Heparin Therapy
of more complicated diagnostic procedures (lungscans, pulmonary angiograms, venography) or changes in manufacturing processes may be causing a new phenomenon. Until the true incidence and the cause of heparin-induced thrombocytopenia are established, it seems prudent to observe closely the platelet count in all patients receiving heparin, but not to restrict the use of heparin for those patients for whom heparin is indicated.
References Babcock. R.B; Dumper. C.W. and Scharfman. W .B.: Heparin-induced thrombocytopenia. New England Journal of Medicine 295: 237-241 (( 976). Barritt. D.W. and Jordan. S.c.: Anticoagulant Drugs in the treatment of pulmonary embolism. A controlled trial. Lancet I: 1309-1312 (( 960). Bell. W .; Tomasulo. P.A.; Alving. B.M. and Duffy. T.P . Thrombocytopenia occurring during the administration of heparin. A prospective study in 52 patients. Annals of Internal Medicine 85: 155-160 (( 976). Brinkhous. K.; Smith, H.P.; Warner, E.D. and Seegers. W.H.: The inhibition of blood clotting: an unidentified substance which acts in conjunction with heparin to prevent the conversion of prothrombin into thrombin. American Journal of Physiology 125: 683 (( 939). Deykin. D.: Thrombogenesis. New England Journal of Medicine 276: 622-628 (( 967). Deykin. D. and Hellerstein. L.J.: The assessment of drug-dependent and isoimmune antiplatelet antibodies by the use of platelet aggregometry. Journal of Clinical Investigation 51: 3142-3153 (( 972). Gallus. A.S. and' Hirsh. J .: Prevention of venous thromboembolism. Seminars in Thrombosis and Hemostasis 2: 232-290
51
((976a). See also Gallus. A.S. and Hirsh. J .: Antithrombotic Drugs. Drugs 12: 41-68 (( 976). Gallus. A.S. and Hirsh. J .: Treatment of venous thromboembolic disease. Seminars in Thrombosis and Hemostasis 2: 291-331 ((976b). See also Gallus. A .S. and Hirsh. J.: Antithrombotic drugs. Drugs 12: 41-68 (( 976). Harris. W.H.: Low-dose heparin in total-hip replacement. Lancet 2: 423 (( 976). International Multicentre Trial: Prevention of fatal post-operative pulmonary embolism by low doses of heparin. Lancet 2: 45-51 (( 975). Karpatkin. S. and Siskind. G .W.: In vitro detection of platelet antibody in patients with idiopathic thrombocytopenic purpura and systemic lupus erythematosus. Blood 33: 795-812 (( 969). Rosenberg. R.D.: Actions and interactions of heparin. New England Jurnal of Medicine .292: 146-151 ((975). Rosenberg, R.D. and Damus, P.S.: The purification and mechanism of action of human Antithrombin - heparin co-factor. Journal of Biological Chemistry 248: 649~505 (1973). Sagar. S.; Stamatakis. J .D.; Higgins, A.F.; Nairn. D.; Maffei. F.H.; Thomas. D.P. and Kakkar. V.V.: Efficacy of low-dose heparin in prevention of extensive deep-vein thrombosis in patients undergoing total-hip replacement. Lancet I: 1151-1154 (1976). Salzman. E.W .; Deykin. D.; Shapiro. R.M . and Rosenberg. R.: Management of heparin therapy. Controlled prospective trial. New England Journal of Medicine 292: 1046-1050 (( 975). Sevin. S.: Venous thrombosis in injured patients (with some observations on pathogenesis); in Sherry. Brinkhous. Genton and Stengle (Eds) Thrombosis. p.29-49 (National Academy of Sciences. Washington DC 1969). Yin. E.T.; Wessler. S. and Stoll. P.J.: Biological properties of the naturally occurring plasma inhibitor to activated factor X. Journal of Biological Chemistry 246: 3703-3711 (( 971).
Author's address: Dr Daniel Deykin. Boston Veterans Administration Hospital. 150 South Huntington Avenue. Boston. Mass. 02130 (USA).
Heparin Therapy: Regimens and Management D.Deykin Boston Veterans Administration Hospital, Boston, Massachusetts
Summary
Heparin remains the most effective antithrombotic drug. It acts by combining with plasma antithrombin. thereby accelerating the neutralisation of thrombin and other activated coagulation factors. Full-dose intravenous heparin is indicated in all cases of pulmonary embolism and established deep venous thrombosis. unless there exist compelling contra indications. Continuous intravenous infusion of heparin appears to be safer than intermittent injection. Low-dose subcutaneous heparin is effective in preventing the initial occurrence of thigh vein thrombi and in reducing the incidence of fatal pulmonary embolism in general surgical patients over the age of 40. The efficacy· of low-dose heparin in preventing pulmonary emboli folloWing hip surgery has not been established. The incidence of severe heparin-induced thrombocytopenia appears to be rising. Platelet counts should be performed in all patients receiving heparin by any mode of administration.
Heparin is the most potent and effective antithrombotic drug in clinical use. The rationale for heparin therapy derives exclusively from its ability to block the coagulation mechanism. Blood coagulation, however, has only a limited role in the pathogenesis of arterial thrombosis, which arises primarily through the interaction of circulating blood platelets with an injured or abnormal vessel wall (Dey kin , 1967). Only when the arterial lumen becomes almost totally occluded by atheroma does the coagulation mechanism become dominant. Therefore, anticoagulant therapy offers little benefit in the prevention of primary or recurrent arterial thrombosis. A different mechanism occurs in veins, since in most venous thrombi there is no evidence for antecedent vascular injury. Although the initiating stimulus for venous thrombosis is unknown, venous thrombi
grow by extension from valve cusps into the vein lumen, propagating through coagulation (Sevitt, 1969). Accordingly, the most uniformly accepted indications for heparin therapy are those in which the coagulation mechanism plays a major role in thrombogenesis: deep thrombophlebitis and pulmonary embolism. An additional indication is the prevention of distal propagation of stasis thrombi following acute arterial' occlusion in the interval before attempted surgical restoration ·of patency.
1. Mode of Action
Heparin acts by binding to a plasma protein, antithrombin, thereby enhancing the ability of antithrombin to neutralise activated clotting factors in the
47
Heparin Therapy
intrinsic limb of the coagulation pathway culminating in the formation of thrombin. The evolution of our understanding of heparin's anticoagulant action has been reviewed by Rosenbert (Rosenberg, 1975) who, with his associates, has clarified the biochemistry of the heparin-antithrombin interaction. Brinkhous (Brinkhous et aI., 1939) and his associates first showed that heparin was an anticoagulant only in the presence of a plasma protein which they called heparin co-factor. Several groups subsequently demonstrated the identity of heparin co-factor with plasma antithrombin. In the absence of heparin, antithrombin combines with the active serine site of thrombin, causing a slow irreversible inhibition of the enzyme. When heparin binds to antithrombin, at a site distinct from that to which heparin binds, it alters antithrombin so that its ability to form a complex with heparin is markedly enhanced. In the presence of heparin the interaction of heparin with antithrombin is virtually instantaneous (Rosenberg, 1973). In addition to prothrombin, four other coagulation proteins are converted to active serine proteases during the process of blood clotting: factor XII (Hageman), XI (PTA), IX (PTe), and X (Stuart). Antithrombin binds to each of these proteases neutralising their activity. Indeed, the affinity of the heparin-antithrombin complex for activated factor X far exceeds that for thrombin itself, suggesting that the true physiological 'target' of antithrombin is activated factor X rather than thrombin (Yin et aI., 1971).
venous thrombi from initially forming (by inhibiting the generation of thrombin) than in prevention of the extension of existing thrombi (after thrombi generation has taken place). We have come to realise, somewhat belatedly, that there are varying degrees of clinical urgency that dictate the initiation of prophylactic heparin therapy. Pulmonary embolism is a recurring, lethal disorder. As a result, once diagnosed, either pulmonary embolism or its precursors, deep venous thrombosis of the pelvis or lower extremity, require prolonged, intensive heparin therapy followed by an extensive period of treatment with oral anticoagulant therapy. Such full dose heparin has been repeatedly shown to be highly effective, but it carries a high morbidity rate. Indeed, between 10 to 20 % of patients receiving conventional intermittent intravenous heparin experience major bleeding during the course of treatment (Gallus and Hirsh, 1976). It is clear that prevention of the initial episode of venous thrombosis may be accomplished by pretreating patients at risk with low doses of heparin, given subcutaneously with little if any risk of haemorrhage. The two modes of heparin therapy - full-dose, intravenous and low-dose, subcutaneous - are quite distinct and should not be confused. Full-dose heparin is indicated once pulmonary embolism or thrombosis of the deep veins of the thigh or pelvis have been documented. Low-dose therapy is useful only to prevent thrombosis from first occurring.
3. Full-Dose Intravenous Therapy 2. Clinical Action and Dose Regimens
Two generalisations emerge from the foregoing discussion. The first is that heparin has no effect on existing thrombi except to prevent distal propagation and to limit thrombin mediated platelet accretion on the surface of the thrombus. Therefore, all heparin therapy is prophylactic. The second is that a biochemical basis exists for the suggestion that lower doses of heparin may be more effective in preventing
Although full-dose intravenous heparin therapy has been employed for over 35 years, astonishingly few reports provide conclusive evidence for its efficacy. Furthermore, little agreement exists concerning the most satisfactory schedule for heparin administration, the need for control of dosage by laboratory tests, the method of laboratory control, or the intensity and duration of therapy. The only controlled study of the use of heparin was reported by Barritt and Jordan (Barritt and Jordan, 1960) who
48
Heparin Therapy
showed in 1960 that once pulmonary embolism had occurred, there was a high recurrence rate in untreated patients and that of those who experienced recurring pulmonary embolism, over half died. Treatment with heparin followed by oral anticoagulants drastically reduced the recurrence and eliminated deaths from pulmonary embolism. Subsequent reports have confIrmed the high incidence of lethal recurrence of untreated pulmonary embolism and the low incidence of fatal recurrent emboli when patients receive heparin. Although these reports lack the authority of Barritt and Jordan's controlled study, they substantiate their fIndings. Few would now argue against the currently accepted practice that all patients who have experienced proven pulmonary embolism must receive full dose intravenous heparin unless there are compelling contraindications.
3: I Intermittent or Continuous Intravenous Administration We have recently conducted a study of the incidence of bleeding in patients receiving heparin by intravenous injection (Salzman et aI., 1975). Among 100 consecutive patients, major bleeding occurred in 21. Two patients died from bleeding and two had recurrent pulmonary embolism. The incidence of bleeding was the same when therapy was regulated by clotting tests or when heparin was given without clotting tests. In a subsequent prospective trial, we gave patients heparin by one of three regimens: intermittent intravenous injection, with regulation of the dose according to the activated partial thromboplastin time (APTT) (72 patients); intermittent intravenous injection with a fIxed dose of heparin (68 patients); or continuously by intravenous infusion, with the dose regulated by the APTT (table I). Recurrent (non-fatal) thromboembolism occurred once in each group. Major bleeding was 7 times more frequent with intermittent injection (regulated or fIxed dose) than with continuous infusion. We concluded that continuous infusion of heparin (regulated by the APTT) was safer than intermittent injection of heparin and was equally
effective in preventing thromboembolism. In our study, we attempted to maintain the APTT at approximately twice baseline levels. Several reports have suggested that the risk of recurring thromboembolism is correlated with inadequate anticoagulant response to heparin. The correlation between increased risk of bleeding and excessive anticoagulation is less clear-cut, since many other determinants (particularly recent surgery, and concomitant use of drugs which affect haemostasis) may also influence the incidence of bleeding.
3.2 Intravenous Dosage Schedules Current evidence suggests, therefore, that continuous intravenous infusion of heparin is preferable to intermittent injection, put both provide a high degree of protection against recurrent, lethal pulmonary embolism. It is our practice to initiate continuous infusion of heparin with a 'loading dose' of 5,000 units given as a bolus followed by a continuous infusion at an initial rate of 1,000 units/hour, subsequently regulated to keep the APTT between 50 to 80 seconds. In those instances when we give heparin by intermittent injection', we use a standard dose of 5,000 units every 4 hours, adjusting the dose down or up for extremes of body weight. There can be no escaping the bleeding complications of full-dose intravenous heparin, but skillful attention to the details of patient management including scrupulous avoidance of injections and of drugs which impair haemostasis, will reduce the incidence of major bleeding.
4. Low-Dose Subcutaneous Therapy
The introduction of low-dose heparin has been inextricably intertwined with the use of 12SI-labelled fIbrinogen as a method for diagnosing the presence of venous thrombosis. Early studies indicated that the frequency of 'deep venous thrombosis' (DVT) of the calf was extremely high in postoperative patients and
Heparin Therapy
49
Table I. Influence of heparin regimen on bleeding, recurrence, and total daily dose Heparin regimen
Every 4 hours with PTT
Every 4 hours No PTT
Continuous infusion with PTT
Number of patients
72
68
69
Major bleeding
6 (8%)
7 (10%)
1 (1%)
31,700
35,600
24,500
Recurrences Daily heparin dose (units)
in elderly patients subjected to protracted bed rest (21-24). The preoperative administration of low-dose heparin clearly reduced the incidence of 'DVT' in the calf, but there was little substantial evidence to prove that such reduction resulted in prevention of fatal pulmonary embolism. Gallus and Hirsh (Gallus and Hirsh, 1976a) have lucidly summarised both the enthusiasm which attended the introduction of lowdose heparin and the limitations of the first reports of both its usefulness and its ineffectiveness.
4.1 Use in General Surgical Patients The recently completed multi-centre trial (International Multicentre Trial, 1975) of the use of low-dose heparin has resolved many of the first lingering doubts over the use of low-dose heparin in general surgical patients over the age of 40. Of the 4,121 patients in the final study group, 2,045 received 5,000 units of heparin subcutaneously before surgery and then every 8 hours after surgery for at least 7 days. The control group was untreated. There were 100 deaths in the control group and 80 in the treated group. Pulmonary embolism was found in 22 of the 72 autopsies in the control group and were judged to be the primary cause of death in 16 .. Pulmonary embolism was found in 5 of 52 autopsies in the treated group and was judged to be the primary cause of death in 2. The difference in incidence of all pulmon-
ary emboli and of 'lethal' pulmonary emboli was sig-
nificantly lower in the treated group. In a subset of patients, 1251-labelled fibrinogen leg scans were performed. Of the 667 control patients tested, 49 had thrombi that extended into the thigh, while of the 625 treated patients tested only 5 had thrombi above the knee. There was no difference in bleeding between the two groups. It is extremely unlikely that another study of such magnitude and excellence will be soon repeated. Therefore, the International Multicentre Trial carries a degree of authority similar to that of the earlier study of Barritt and Jordan and makes it mandatory that low-dose heparin prophylaxis should be routinely employed in general surgical patients over the age of 40. It is our practice to initiate treatment with 5,000 units given subcutaneously 2 hours prior to surgery and every 8 hours thereafter for a minimum of 7 days or until the patient is fully ambulatory.
4.2 Use in Hip and Prostatic Surgery Low-dose heparin may not be effective in preventing post-operative thigh and pelvic thrombi in certain operations which are associated with a much higher risk of fatal pulmonary embolism than the 0.7 to 1.0 % incidence following general abdominal or thoracic surgery. Several authors suggest that lowdose heparin does not reduce the incidence of
Heparin Therapy
venographically proven femoral or pelvic thrombi nor the risk of pulmonary embolism. Sagar and his associates disagree (Sagar et aI., 1976). In a study of 82 patients undergoing elective hip replacement, 52 received low-dose heparin (alone or in conjunction with dihydroergotamine) and 32 received no prophylaxis. In the control group, 22 developed '2SI-labeUed fibrinogen 'OVT of which 17 were shown to be femoral by venography, with extensive thrombi found in 9 cases. 12 of the heparintreated patients developed isotopic evidence for 'OVT of which 4 were found to be femoral by venography. None were judged to be extensive. Sagar and his coworkers conclude that 'in patients undergoing totalhip replacement, low-dose prophylaxis is effective in reducing the frequency of life-threatening extensive thrombi in the femoral vein'. Harris (Harris, 1976) has criticised Sagar's study, pointing out that fibrinogen scanning will not detect most instances of major thrombi in the thigh. Since Sagar performed venography only if 'OVT were detected by scanning, the true incidence of femoral vein thrombi in the treated group may have been underestimated. At the moment, the efficacy of low-dose heparin in hip (and prostatic) surgery remains to be proved, but these should not restrain the use of low-dose heparin as routine prophylaxis in general surgical patients.
5. Complications of Heparin: Thrombocytopenia
The major complication of heparin, bleeding, is a direct extension of its therapeutic action. However, another complication of heparin, thrombocytopenia, is not receiving increased attention. Oa::asional, mild thrombocytopenia has been noted following heparin treatment, and scattered reports have documented rare, profound thrombocytopenia during heparin therapy. Babcock (Babcock, 1976) and his associates have recently reported 5 patients in whom severe depression of the platelet aggregometry (Dey kin and Hellerstein, 1972). In 4, antibodies were also found by the platelet factor 3 availability technique (Kar-
50
patkin, 1969). All 5 patients had either short activated partial thromboplastin time or were within the normal range. Although extensive studies of the coagulation mechanism were not performed, no evidence of disseminated intravascular coagulation were found in the patients studied, and no episodes of major haemorrhage occurred. These patients all developed heparin-induced apparently immune thrombocytopenia within 6 weeks of another at one medical centre. In a parallel study, Bell and his co-workers (Bell et al., 1976) reported the results of their prospective study of the incidence of thrombocytopenia in 52 patients receiving continuous-infusion heparin. Their data were startling: 16 patients (30 %) developed platelet counts less than 100,000/ pI; indeed, in 9, the count was less than 50,000/ pI. In I 0 of the I 6 thrombocytopenic patients, fibrinogen-fibrin degradation products were elevated. In contrast, only 2 of the 32 non-thrombocytopenic patients had high split products. Hypofibrinogenemia was present in 5 thrombocytopenic patients and in none of the nonthrombocytopenic patients. Bell and his associates did not perform immunological assays. In 2 thrombocytopenic patients, the partial thromboplastin times (and prothrombin times as well) were prolonged. Although excessive bleeding from venipuncture sites developed in 2 thrombocytopenic patients, there was no other evidence of untoward bleeding. There seems to be no relationship between the method of administering heparin and the development of thrombocytopenia. Babcock's patients received heparin by intermittent injection, but Bell's patients received it by continuous infusion. Furthermore, Bell reported an additional 2 patients who developed thrombocytopenia while receiving lowdose subcutaneous heparin. What are we to make of this apparent outburst of heparin-induced thrombocytopenia? In both recent studies, there was little evidence of exaggerated bleeding. It is possible, therefore, that a high incidence of heparin-induced, severe thrombocytopenia has hitherto gone undetected. It is also possible that the increasing tendency to polypharmacy or increased use
Heparin Therapy
of more complicated diagnostic procedures (lungscans, pulmonary angiograms, venography) or changes in manufacturing processes may be causing a new phenomenon. Until the true incidence and the cause of heparin-induced thrombocytopenia are established, it seems prudent to observe closely the platelet count in all patients receiving heparin, but not to restrict the use of heparin for those patients for whom heparin is indicated.
References Babcock. R.B; Dumper. C.W. and Scharfman. W .B.: Heparin-induced thrombocytopenia. New England Journal of Medicine 295: 237-241 (( 976). Barritt. D.W. and Jordan. S.c.: Anticoagulant Drugs in the treatment of pulmonary embolism. A controlled trial. Lancet I: 1309-1312 (( 960). Bell. W .; Tomasulo. P.A.; Alving. B.M. and Duffy. T.P . Thrombocytopenia occurring during the administration of heparin. A prospective study in 52 patients. Annals of Internal Medicine 85: 155-160 (( 976). Brinkhous. K.; Smith, H.P.; Warner, E.D. and Seegers. W.H.: The inhibition of blood clotting: an unidentified substance which acts in conjunction with heparin to prevent the conversion of prothrombin into thrombin. American Journal of Physiology 125: 683 (( 939). Deykin. D.: Thrombogenesis. New England Journal of Medicine 276: 622-628 (( 967). Deykin. D. and Hellerstein. L.J.: The assessment of drug-dependent and isoimmune antiplatelet antibodies by the use of platelet aggregometry. Journal of Clinical Investigation 51: 3142-3153 (( 972). Gallus. A.S. and' Hirsh. J .: Prevention of venous thromboembolism. Seminars in Thrombosis and Hemostasis 2: 232-290
51
((976a). See also Gallus. A.S. and Hirsh. J .: Antithrombotic Drugs. Drugs 12: 41-68 (( 976). Gallus. A.S. and Hirsh. J .: Treatment of venous thromboembolic disease. Seminars in Thrombosis and Hemostasis 2: 291-331 ((976b). See also Gallus. A .S. and Hirsh. J.: Antithrombotic drugs. Drugs 12: 41-68 (( 976). Harris. W.H.: Low-dose heparin in total-hip replacement. Lancet 2: 423 (( 976). International Multicentre Trial: Prevention of fatal post-operative pulmonary embolism by low doses of heparin. Lancet 2: 45-51 (( 975). Karpatkin. S. and Siskind. G .W.: In vitro detection of platelet antibody in patients with idiopathic thrombocytopenic purpura and systemic lupus erythematosus. Blood 33: 795-812 (( 969). Rosenberg. R.D.: Actions and interactions of heparin. New England Jurnal of Medicine .292: 146-151 ((975). Rosenberg, R.D. and Damus, P.S.: The purification and mechanism of action of human Antithrombin - heparin co-factor. Journal of Biological Chemistry 248: 649~505 (1973). Sagar. S.; Stamatakis. J .D.; Higgins, A.F.; Nairn. D.; Maffei. F.H.; Thomas. D.P. and Kakkar. V.V.: Efficacy of low-dose heparin in prevention of extensive deep-vein thrombosis in patients undergoing total-hip replacement. Lancet I: 1151-1154 (1976). Salzman. E.W .; Deykin. D.; Shapiro. R.M . and Rosenberg. R.: Management of heparin therapy. Controlled prospective trial. New England Journal of Medicine 292: 1046-1050 (( 975). Sevin. S.: Venous thrombosis in injured patients (with some observations on pathogenesis); in Sherry. Brinkhous. Genton and Stengle (Eds) Thrombosis. p.29-49 (National Academy of Sciences. Washington DC 1969). Yin. E.T.; Wessler. S. and Stoll. P.J.: Biological properties of the naturally occurring plasma inhibitor to activated factor X. Journal of Biological Chemistry 246: 3703-3711 (( 971).
Author's address: Dr Daniel Deykin. Boston Veterans Administration Hospital. 150 South Huntington Avenue. Boston. Mass. 02130 (USA).
