Saturday 22 February
No 8791
1992
ORIGINAL ARTICLES
Comparison of subcutaneous low-molecular-weight heparin with intravenous standard heparin in proximal deep-vein thrombosis
In view of the
potential of low-molecular-weight heparins (LMWH) to simplify initial therapy and allow outpatient treatment of proximal deep-vein thrombosis, we undertook a randomised comparison of fixed-dose subcutaneous LMWH with adjusteddose intravenous standard heparin in the initial treatment of this disorder. Our main objectives were to compare the efficacy of these regimens for 6 months of follow-up and to assess the risk of clinically important bleeding. Of 170 consecutive symptomatic patients with venographically proven proximal deep-venous thrombosis, 85 received standard heparin (to achieve an activated partial thromboplastin time of 1&mid ot; 5 to 2·0 times the pretreatment value) and 85 LMWH (adjusted only for body weight) for 10 days. Oral coumarin was started on day 7 and continued for at least 3 months. The frequency of recurrent venous thromboembolism diagnosed objectively did not differ significantly between the standard-heparin and LMWH groups (12 [14%] vs 6 [7%]; difference 7% [95% confidence interval -3% to 15%]; p=0·13). Clinically important bleeding was infrequent in both groups (3·5% for standard heparin vs 1·1% for LMWH; p>0·2). We conclude that fixed-dose subcutaneous LMWH is at least as effective and safe as intravenous adjusted-dose heparin in the initial treatment of symptomatic proximal-vein thrombosis. Since there is no need for laboratory monitoring with the LMWH regimen, patients with venous thrombosis can be treated at home.
Introduction
Anticoagulation is the treatment of choice for leg-vein thrombosis to prevent death from pulmonary embolism and to reduce morbidity from the acute event.12 Common practice is to start treatment with 5-10 days of standard heparin, followed by oral anticoagulants for 3 months. Heparin is usually administered by a continuous intravenous infusion and the dose is adjusted on the basis of daily laboratory measurements .2-5 The evidence that heparin is necessary in the initial treatment of established venous thrombosis comes from animal studies and a placebo-controlled, randomised trial that showed a much lower frequency of recurrent venous thromboembolism with an initial course of adjusted-dose intravenous heparin than with oral anticoagulants alone.6,7 In the past decade, low-molecular-weight heparins (LMWH) have been developed.8,9 Compared with standard unfractionated heparin, these compounds have a longer plasma half-life,1O,l1 less variability in the anticoagulant response to fixed doses,12 and a more favourable antithrombotic to haemorrhagic ratio.13 These properties have suggested the possibility that LMWH might be administered subcutaneously in fixed doses for the treatment of established venous thrombosis without the need for laboratory monitoring. Several randomised trials have compared LMWH with unfractionated heparin in the treatment of proven venous ADDRESSES Second Department of Internal Medicine (P Prandoni, MD, M Carta, MD, A. Cogo, MD, Prof A. Ruol, MD); Second Service of Radiology (M. Vigo, MD); and Service of Nuclear Medicine (D. Casara, MD), University Hospital of Padua, Italy; and Centre for Haemostasis, Thrombosis, Atherosclerosis, and Inflammation Research, Academic Medical Centre, F4-237, Meibergdreef 9, 1105 AZ Amsterdam, Netherlands (A. W A. Lensing, MD, H. R. Büller, MD, Prof J W ten Cate, MD, PhD). Correspondence to Dr Anthonie W. A. Lensing
442
thrombosis.14-17 In three studies, the dose of LMWH was adjusted according to a laboratory test result,14-16 and in the other according to body weight without laboratory control. 17 All four studies used substitute outcomes to evaluate efficacy, such as the reduction in thrombus size on venography, but they suggested that LMWH were more effective than unfractionated heparin and produced similar rates of clinically important bleeding. To evaluate the efficacy and safety of LMWH fully, long-term follow-up must be done and objective diagnostic tests used to the of incidence recurrent venous determine thromboembolism. We have compared adjusted-dose intravenous standard heparin with fixed-dose subcutaneous LMWH for the initial treatment of patients with
venographically proven proximal deep-venous thrombosis; we followed up all patients for 6 months. Our main objectives were to determine the effectiveness of these regimens and the risk of clinically important bleeding. Patients and methods Consecutive patients with clinically suspected deep-vein thrombosis referred by their general practitioners to the department of internal medicine of the University of Padua underwent objective testing.l21 Eligible patients had proximal deep-vein thrombosis (affecting the popliteal or more proximal veins, with or without calf-vein thrombosis) confirmed by contrast venography. Reasons for exclusion were clinically suspected pulmonary embolism at referral; an episode of venous thrombosis in the same leg within the previous 2 years; ongoing anticoagulant treatment at the time of referral; contraindications to heparin treatment; pregnancy; allergy to contrast material; and residence far from the hospital. Eligible patients who gave informed consent joined the study. The study protocol was approved by the institutional review board. We used unfractionated sodium heparin (’Liquemin’; Roche, Basel, Switzerland) and LMWH CY 216 (’Fraxiparine’; Sanofi, Paris, France), which has a mean molecular weight of 4500 Da and specific activities per mg of less than 50 anti-factor IIa IU and more than 220 anti-factor Xa (AXa) Institute Choay (IC) units (1 AXa unit is about 2-4 AXa IC units). Patients were allocated treatment by a prescribed randomisation schedule The standard-heparin group received an intravenous bolus of 100 units/kg unfractionated heparin, followed by a continuous infusion of 35 000 units per 24 h. The activated partial thromboplastin time (APTT) was measured (with actin FS Dade reagent) 6 h after the start of treatment and then once a day. The heparin dose was adjusted to maintain the APTT at 1-5-2-0 times the pretreatment value (APTT ratio). Patients in the LMWH group received subcutaneous injections every 12 h: each injection was 0-5 ml for patients who weighed less than 55 kg; 0-6 ml for those who weighed 55-80 kg; and 0-7 ml for those who weighed more than 80 kg (1 -0 ml contains 25 000 AXa IC units). No laboratory monitoring was done. Oral treatment with coumarin (initial dose 5 mg) was started in all patients on day 7 of heparin treatment. The coumarin dose was adjusted daily to maintain the international normalised ratio (INR) between 2-0 and 3-0. Heparin treatment was discontinued on day 10, or later if the INR was below 2-0. Subsequently, the coumarin dose was adjusted weekly by the patient’s general practitioner. Blood was sampled daily during the initial study period for haemoglobin measurement and platelet counts. Thrombocytopenia was defined as a platelet count of below 1011/1 that had fallen by at least 30% from the pretreatment value. Perfusion lung scans and chest radiographs were done for all patients at baseline. All patients were examined daily during heparin treatment for symptomatic extension of the deep-vein thrombosis, pulmonary embolism, or bleeding. Contrast venography, perfusion lung scanning, and chest radiography were done on day 10, or earlier if signs or symptoms of extending thrombosis, thrombosis in the other leg, or pulmonary embolism
developed.
TABLE!—DEMOGRAPHY AND CLIN I CAL CHARACTERISTICS OF STUDY SUBJECTS
*Median
(range)
Clinical follow-up assessments were made at 1, 3, and 6 months. The patients were told to come to the hospital immediately if they noticed signs or symptoms of recurrent deep-vein thrombosis or pulmonary embolism. Patients who did not keep follow-up appointments were assessed at home. Venograms were done with long-leg films and non-ionic contrast materia1,22 The criteria for deep-vein thrombosis were an intraluminal filling defect confirmed in at least two different projections and no filling of a venous segment despite repeated injections with contrast material. A quantitative venographic score was used to assess the extent of the venous thrombosis.23 Six-view perfusion lung scans were done with technetium-99mmacroaggregated albumin. The number of segmental defects was counted with the use of a lung segment reference chart.24 The principal endpoint for the assessment of efficacy was symptomatic recurrent deep-vein thrombosis of the leg (including symptomatic extension) or symptomatic pulmonary embolism. Patients with clinically suspected extension or recurrent deep-vein thrombosis underwent repeat venography, and recurrent deep-vein thrombosis was defined as a constant intraluminal filling defect not present on the day 10 venogram (or that obtained on day 1 in patients with symptomatic extension). If the venogram was inconclusive, recurrent deep-vein thrombosis was diagnosed on the basis of an abnormal iodine-125-labelled fibrinogen leg scan .21 Patients with clinically suspected pulmonary embolism underwent another perfusion lung scan: the diagnosis was based on the presence of at least one segmental defect not seen on the preceding scan and no abnormality on the chest radiograph in that area, or on the presence of pulmonary emboli at necropsy. If the results of
perfusion lung scanning were inconclusive, pulmonary angiography was done and pulmonary embolism was defined as a constant intraluminal filling defect in at least two views or sharp cut-offs in vessels
more
than 2-5
mm
in diameter.
Secondary endpoints for the assessment of efficacy were: the change in the extent of venous thrombosis between the day 10 and day 0 venograms; and the change in the number of segmental defects on the day 10 and day 0 perfusion lung scans. The primary analysis of safety was based on severe bleeding during or within 48 h of the end of heparin treatment. Bleeding was classified as severe if it was associated with a fall in the haemoglobin concentration of at least 2 g/dl; if it was retroperitoneal or intracranial; or if transfusion of 2 or more units of blood was needed.s Bleeding was defined as minor if it was clinically overt but did not meet the other criteria. Minor bleeding was used as a secondary endpoint for the assessment of safety. All clinical endpoints were reviewed by an adjudication committee from the coordinating centre in Amsterdam, unaware of treatment allocation or other details of patients. If necessary, additional information about the endpoints or the cause of death was sought from Padua. The venograms and perfusion lung scans of each patient were scored by a panel of three experienced observers who were unaware of treatment allocation and the sequence in which the tests were done (before or after treatment).
443
TABLE II-THROMBOEMBOLIC COMPLICATIONS DURING INITIAL TREATMENT AND FOLLOW-UP
perioddays 1-10 DVTdeep-vem thrombosis, PE=pulmonary embolism
*initial treatment
TABLE III-BLEEDING COMPLICATIONS DURING OR WITHIN 48 h OF HEPARIN TREATMENT
standard-heparin group happened during the initial treatment period, and 2 of the 4 were associated with heparin-induced thrombocytopenia. 1 of the 6 recurrences in the LMWH group developed during the period of subcutaneous treatment. 3 recurrent events in each group fatal; in all 6 cases necropsy revealed pulmonary emboli. Severe bleeding occurred during or within 48 h of heparin treatment in 3 (4% [0-11%]) of the standard-heparin recipients and 1 (1% [0-7%]) of the LMWH recipients (p>0’2, table ill). Since the 95% CI of the observed difference between the groups is - 3% to 5%, it is unlikely (p < 0-05) that there is a real difference between the groups in this index of safety. There was little difference between the standard-heparin and LMWH groups in the frequency of minor bleeding (6 [7%; 3-15%] vs 2 (2%; 0-9%]; table in). During the period of oral anticoagulant treatment, severe bleeding occurred in 3 standard-heparin recipients and 1 LMWH recipient and minor bleeding in 2 and 4, were
Before the study, we thought that standard heparin and LMWH would have similar efficacy but that LMWH might be associated with a lower frequency of severe haemorrhage. The sample size calculations were based on a 12-5% frequency of recurrent venous thromboembolism during 6 months of follow-up. We calculated that 80 patients in each group would produce an exact confidence interval (CI) of +11% around the observed difference in this frequency if the treatments had equivalent efficacy. Confidence intervals for the true frequencies of recurrent venous thromboembolism and bleeding complications were calculated from the binomial distribution. Exact confidence intervals for differences in the frequencies of efficacy and safety outcomes between the treatment groups were calculated according to Thomas and Gart.26 The X2 test, Fisher’s exact test, and Student’s t tests were used when appropriate. Two-sided p-values of less than 0-05 were considered significant. The comparison of venograms and the occurrence of severe haemorrhage in the first 15 patients in each group has been reported previously as a part of a multicentre
respectively. During the 6 months of follow-up, there were 12 deaths in the standard-heparin group and 6 in the LMWH group (p 0-21 ; table iv). 3 deaths in each group were judged to be related to recurrent venous thromboembolism; 1 death in the standard-heparin group was due to bleeding during oral anticoagulant treatment. The observed difference in =
investigation." Results From May, 1986, to April, 1991, proximal deep-vein thrombosis was confirmed by venography in 211 patients with clinically suspected thrombosis. Of these, 36 were excluded before randomisation because of clinically suspected pulmonary embolism (10), a recent episode of venous thrombosis in the same leg (8), contraindications to
heparin (7), ongoing anticoagulant treatment (6), an allergic reaction to contrast material (1), pregnancy (1), and geographic inaccessibility (3). Of the 175 patients eligible for the study, 5 did not give informed consent. The 85 standard-heparin recipients and 85 LMWH recipients were similar in baseline demographic and clinical characteristics (table I). No patient was lost to follow-up. During the 6 months after the diagnosis of proximal-vein thrombosis, symptomatic extension or recurrent venous thromboembolism confirmed by objective tests (table 11) developed in 12 (14% [95% CI 8-23%]) of the standardheparin recipients and 6 (7% [3-15%]) of the LMWH group. The difference in the frequency of these events was 7% (95% CI -3% to 15%; p=0-13) and it is unlikely (p < 0-05) that there is any real difference in efficacy between standard heparin and LMWH. 4 of the 12 recurrences in the
TABLE IV-CAUSES OF DEATHS
*Conftrmed at necropsy tNot due to cancer
444
TABLE V-CHANGES IN VENOGRAPHIC SCORE
mean daily heparin dose in the standard-heparin during the initial treatment period was 32 463 Uday (see figure). In about 25% of patients the dose required for an APTT ratio of 1-5 or more (value 1-5 or more times the baseline value) was more than 35 000 U/day. The APTT ratio was below 1during the first 48 h in only 6 (7%) patients, 1 of these patients had recurrent
The
group
mortality between the treatment groups was largely due to an imbalance in cancer deaths. Although this is a post-hoc finding, it is interesting that 44% (8 of 18) of the cancer patients in the standard-heparin group died, compared with only 7% (1 of 15) of cancer patients in the LMWH group. This difference is statistically unusual (p 0021; two-tailed Fisher’s exact test) if these treatments had no effect on the likelihood of succumbing to cancer. The changes in the extent of venous thrombosis on venography with an objective scoring system are =
summarised in table v. A second venogram was done in all but 2 patients (both in the LMWH group). The distribution of changes in venographic score showed a significant difference (p 0017) in favour of the LMWH group. A repeat perfusion lung scan was done in 81 standardheparin recipients and 79 LMWH recipients. New segmental defects were found in significantly more of the standard-heparin group than of the LMWH group (15 [19%] vs 4 [5%]; difference 13% [3-24%]; p < 0-02). None of the patients had symptoms of pulmonary embolism. Perfusion scans were done for 34 of the 36 patients in the standard-heparin group and 49 of the 50 in the LMWH group who had an improvement in the venographic score. A combination of improved venographic score and new segmental defects on the perfusion lung scans was seen in 9 standard-heparin recipients and 1 LMWH recipient =
(p
No 8791
1992
ORIGINAL ARTICLES
Comparison of subcutaneous low-molecular-weight heparin with intravenous standard heparin in proximal deep-vein thrombosis
In view of the
potential of low-molecular-weight heparins (LMWH) to simplify initial therapy and allow outpatient treatment of proximal deep-vein thrombosis, we undertook a randomised comparison of fixed-dose subcutaneous LMWH with adjusteddose intravenous standard heparin in the initial treatment of this disorder. Our main objectives were to compare the efficacy of these regimens for 6 months of follow-up and to assess the risk of clinically important bleeding. Of 170 consecutive symptomatic patients with venographically proven proximal deep-venous thrombosis, 85 received standard heparin (to achieve an activated partial thromboplastin time of 1&mid ot; 5 to 2·0 times the pretreatment value) and 85 LMWH (adjusted only for body weight) for 10 days. Oral coumarin was started on day 7 and continued for at least 3 months. The frequency of recurrent venous thromboembolism diagnosed objectively did not differ significantly between the standard-heparin and LMWH groups (12 [14%] vs 6 [7%]; difference 7% [95% confidence interval -3% to 15%]; p=0·13). Clinically important bleeding was infrequent in both groups (3·5% for standard heparin vs 1·1% for LMWH; p>0·2). We conclude that fixed-dose subcutaneous LMWH is at least as effective and safe as intravenous adjusted-dose heparin in the initial treatment of symptomatic proximal-vein thrombosis. Since there is no need for laboratory monitoring with the LMWH regimen, patients with venous thrombosis can be treated at home.
Introduction
Anticoagulation is the treatment of choice for leg-vein thrombosis to prevent death from pulmonary embolism and to reduce morbidity from the acute event.12 Common practice is to start treatment with 5-10 days of standard heparin, followed by oral anticoagulants for 3 months. Heparin is usually administered by a continuous intravenous infusion and the dose is adjusted on the basis of daily laboratory measurements .2-5 The evidence that heparin is necessary in the initial treatment of established venous thrombosis comes from animal studies and a placebo-controlled, randomised trial that showed a much lower frequency of recurrent venous thromboembolism with an initial course of adjusted-dose intravenous heparin than with oral anticoagulants alone.6,7 In the past decade, low-molecular-weight heparins (LMWH) have been developed.8,9 Compared with standard unfractionated heparin, these compounds have a longer plasma half-life,1O,l1 less variability in the anticoagulant response to fixed doses,12 and a more favourable antithrombotic to haemorrhagic ratio.13 These properties have suggested the possibility that LMWH might be administered subcutaneously in fixed doses for the treatment of established venous thrombosis without the need for laboratory monitoring. Several randomised trials have compared LMWH with unfractionated heparin in the treatment of proven venous ADDRESSES Second Department of Internal Medicine (P Prandoni, MD, M Carta, MD, A. Cogo, MD, Prof A. Ruol, MD); Second Service of Radiology (M. Vigo, MD); and Service of Nuclear Medicine (D. Casara, MD), University Hospital of Padua, Italy; and Centre for Haemostasis, Thrombosis, Atherosclerosis, and Inflammation Research, Academic Medical Centre, F4-237, Meibergdreef 9, 1105 AZ Amsterdam, Netherlands (A. W A. Lensing, MD, H. R. Büller, MD, Prof J W ten Cate, MD, PhD). Correspondence to Dr Anthonie W. A. Lensing
442
thrombosis.14-17 In three studies, the dose of LMWH was adjusted according to a laboratory test result,14-16 and in the other according to body weight without laboratory control. 17 All four studies used substitute outcomes to evaluate efficacy, such as the reduction in thrombus size on venography, but they suggested that LMWH were more effective than unfractionated heparin and produced similar rates of clinically important bleeding. To evaluate the efficacy and safety of LMWH fully, long-term follow-up must be done and objective diagnostic tests used to the of incidence recurrent venous determine thromboembolism. We have compared adjusted-dose intravenous standard heparin with fixed-dose subcutaneous LMWH for the initial treatment of patients with
venographically proven proximal deep-venous thrombosis; we followed up all patients for 6 months. Our main objectives were to determine the effectiveness of these regimens and the risk of clinically important bleeding. Patients and methods Consecutive patients with clinically suspected deep-vein thrombosis referred by their general practitioners to the department of internal medicine of the University of Padua underwent objective testing.l21 Eligible patients had proximal deep-vein thrombosis (affecting the popliteal or more proximal veins, with or without calf-vein thrombosis) confirmed by contrast venography. Reasons for exclusion were clinically suspected pulmonary embolism at referral; an episode of venous thrombosis in the same leg within the previous 2 years; ongoing anticoagulant treatment at the time of referral; contraindications to heparin treatment; pregnancy; allergy to contrast material; and residence far from the hospital. Eligible patients who gave informed consent joined the study. The study protocol was approved by the institutional review board. We used unfractionated sodium heparin (’Liquemin’; Roche, Basel, Switzerland) and LMWH CY 216 (’Fraxiparine’; Sanofi, Paris, France), which has a mean molecular weight of 4500 Da and specific activities per mg of less than 50 anti-factor IIa IU and more than 220 anti-factor Xa (AXa) Institute Choay (IC) units (1 AXa unit is about 2-4 AXa IC units). Patients were allocated treatment by a prescribed randomisation schedule The standard-heparin group received an intravenous bolus of 100 units/kg unfractionated heparin, followed by a continuous infusion of 35 000 units per 24 h. The activated partial thromboplastin time (APTT) was measured (with actin FS Dade reagent) 6 h after the start of treatment and then once a day. The heparin dose was adjusted to maintain the APTT at 1-5-2-0 times the pretreatment value (APTT ratio). Patients in the LMWH group received subcutaneous injections every 12 h: each injection was 0-5 ml for patients who weighed less than 55 kg; 0-6 ml for those who weighed 55-80 kg; and 0-7 ml for those who weighed more than 80 kg (1 -0 ml contains 25 000 AXa IC units). No laboratory monitoring was done. Oral treatment with coumarin (initial dose 5 mg) was started in all patients on day 7 of heparin treatment. The coumarin dose was adjusted daily to maintain the international normalised ratio (INR) between 2-0 and 3-0. Heparin treatment was discontinued on day 10, or later if the INR was below 2-0. Subsequently, the coumarin dose was adjusted weekly by the patient’s general practitioner. Blood was sampled daily during the initial study period for haemoglobin measurement and platelet counts. Thrombocytopenia was defined as a platelet count of below 1011/1 that had fallen by at least 30% from the pretreatment value. Perfusion lung scans and chest radiographs were done for all patients at baseline. All patients were examined daily during heparin treatment for symptomatic extension of the deep-vein thrombosis, pulmonary embolism, or bleeding. Contrast venography, perfusion lung scanning, and chest radiography were done on day 10, or earlier if signs or symptoms of extending thrombosis, thrombosis in the other leg, or pulmonary embolism
developed.
TABLE!—DEMOGRAPHY AND CLIN I CAL CHARACTERISTICS OF STUDY SUBJECTS
*Median
(range)
Clinical follow-up assessments were made at 1, 3, and 6 months. The patients were told to come to the hospital immediately if they noticed signs or symptoms of recurrent deep-vein thrombosis or pulmonary embolism. Patients who did not keep follow-up appointments were assessed at home. Venograms were done with long-leg films and non-ionic contrast materia1,22 The criteria for deep-vein thrombosis were an intraluminal filling defect confirmed in at least two different projections and no filling of a venous segment despite repeated injections with contrast material. A quantitative venographic score was used to assess the extent of the venous thrombosis.23 Six-view perfusion lung scans were done with technetium-99mmacroaggregated albumin. The number of segmental defects was counted with the use of a lung segment reference chart.24 The principal endpoint for the assessment of efficacy was symptomatic recurrent deep-vein thrombosis of the leg (including symptomatic extension) or symptomatic pulmonary embolism. Patients with clinically suspected extension or recurrent deep-vein thrombosis underwent repeat venography, and recurrent deep-vein thrombosis was defined as a constant intraluminal filling defect not present on the day 10 venogram (or that obtained on day 1 in patients with symptomatic extension). If the venogram was inconclusive, recurrent deep-vein thrombosis was diagnosed on the basis of an abnormal iodine-125-labelled fibrinogen leg scan .21 Patients with clinically suspected pulmonary embolism underwent another perfusion lung scan: the diagnosis was based on the presence of at least one segmental defect not seen on the preceding scan and no abnormality on the chest radiograph in that area, or on the presence of pulmonary emboli at necropsy. If the results of
perfusion lung scanning were inconclusive, pulmonary angiography was done and pulmonary embolism was defined as a constant intraluminal filling defect in at least two views or sharp cut-offs in vessels
more
than 2-5
mm
in diameter.
Secondary endpoints for the assessment of efficacy were: the change in the extent of venous thrombosis between the day 10 and day 0 venograms; and the change in the number of segmental defects on the day 10 and day 0 perfusion lung scans. The primary analysis of safety was based on severe bleeding during or within 48 h of the end of heparin treatment. Bleeding was classified as severe if it was associated with a fall in the haemoglobin concentration of at least 2 g/dl; if it was retroperitoneal or intracranial; or if transfusion of 2 or more units of blood was needed.s Bleeding was defined as minor if it was clinically overt but did not meet the other criteria. Minor bleeding was used as a secondary endpoint for the assessment of safety. All clinical endpoints were reviewed by an adjudication committee from the coordinating centre in Amsterdam, unaware of treatment allocation or other details of patients. If necessary, additional information about the endpoints or the cause of death was sought from Padua. The venograms and perfusion lung scans of each patient were scored by a panel of three experienced observers who were unaware of treatment allocation and the sequence in which the tests were done (before or after treatment).
443
TABLE II-THROMBOEMBOLIC COMPLICATIONS DURING INITIAL TREATMENT AND FOLLOW-UP
perioddays 1-10 DVTdeep-vem thrombosis, PE=pulmonary embolism
*initial treatment
TABLE III-BLEEDING COMPLICATIONS DURING OR WITHIN 48 h OF HEPARIN TREATMENT
standard-heparin group happened during the initial treatment period, and 2 of the 4 were associated with heparin-induced thrombocytopenia. 1 of the 6 recurrences in the LMWH group developed during the period of subcutaneous treatment. 3 recurrent events in each group fatal; in all 6 cases necropsy revealed pulmonary emboli. Severe bleeding occurred during or within 48 h of heparin treatment in 3 (4% [0-11%]) of the standard-heparin recipients and 1 (1% [0-7%]) of the LMWH recipients (p>0’2, table ill). Since the 95% CI of the observed difference between the groups is - 3% to 5%, it is unlikely (p < 0-05) that there is a real difference between the groups in this index of safety. There was little difference between the standard-heparin and LMWH groups in the frequency of minor bleeding (6 [7%; 3-15%] vs 2 (2%; 0-9%]; table in). During the period of oral anticoagulant treatment, severe bleeding occurred in 3 standard-heparin recipients and 1 LMWH recipient and minor bleeding in 2 and 4, were
Before the study, we thought that standard heparin and LMWH would have similar efficacy but that LMWH might be associated with a lower frequency of severe haemorrhage. The sample size calculations were based on a 12-5% frequency of recurrent venous thromboembolism during 6 months of follow-up. We calculated that 80 patients in each group would produce an exact confidence interval (CI) of +11% around the observed difference in this frequency if the treatments had equivalent efficacy. Confidence intervals for the true frequencies of recurrent venous thromboembolism and bleeding complications were calculated from the binomial distribution. Exact confidence intervals for differences in the frequencies of efficacy and safety outcomes between the treatment groups were calculated according to Thomas and Gart.26 The X2 test, Fisher’s exact test, and Student’s t tests were used when appropriate. Two-sided p-values of less than 0-05 were considered significant. The comparison of venograms and the occurrence of severe haemorrhage in the first 15 patients in each group has been reported previously as a part of a multicentre
respectively. During the 6 months of follow-up, there were 12 deaths in the standard-heparin group and 6 in the LMWH group (p 0-21 ; table iv). 3 deaths in each group were judged to be related to recurrent venous thromboembolism; 1 death in the standard-heparin group was due to bleeding during oral anticoagulant treatment. The observed difference in =
investigation." Results From May, 1986, to April, 1991, proximal deep-vein thrombosis was confirmed by venography in 211 patients with clinically suspected thrombosis. Of these, 36 were excluded before randomisation because of clinically suspected pulmonary embolism (10), a recent episode of venous thrombosis in the same leg (8), contraindications to
heparin (7), ongoing anticoagulant treatment (6), an allergic reaction to contrast material (1), pregnancy (1), and geographic inaccessibility (3). Of the 175 patients eligible for the study, 5 did not give informed consent. The 85 standard-heparin recipients and 85 LMWH recipients were similar in baseline demographic and clinical characteristics (table I). No patient was lost to follow-up. During the 6 months after the diagnosis of proximal-vein thrombosis, symptomatic extension or recurrent venous thromboembolism confirmed by objective tests (table 11) developed in 12 (14% [95% CI 8-23%]) of the standardheparin recipients and 6 (7% [3-15%]) of the LMWH group. The difference in the frequency of these events was 7% (95% CI -3% to 15%; p=0-13) and it is unlikely (p < 0-05) that there is any real difference in efficacy between standard heparin and LMWH. 4 of the 12 recurrences in the
TABLE IV-CAUSES OF DEATHS
*Conftrmed at necropsy tNot due to cancer
444
TABLE V-CHANGES IN VENOGRAPHIC SCORE
mean daily heparin dose in the standard-heparin during the initial treatment period was 32 463 Uday (see figure). In about 25% of patients the dose required for an APTT ratio of 1-5 or more (value 1-5 or more times the baseline value) was more than 35 000 U/day. The APTT ratio was below 1during the first 48 h in only 6 (7%) patients, 1 of these patients had recurrent
The
group
mortality between the treatment groups was largely due to an imbalance in cancer deaths. Although this is a post-hoc finding, it is interesting that 44% (8 of 18) of the cancer patients in the standard-heparin group died, compared with only 7% (1 of 15) of cancer patients in the LMWH group. This difference is statistically unusual (p 0021; two-tailed Fisher’s exact test) if these treatments had no effect on the likelihood of succumbing to cancer. The changes in the extent of venous thrombosis on venography with an objective scoring system are =
summarised in table v. A second venogram was done in all but 2 patients (both in the LMWH group). The distribution of changes in venographic score showed a significant difference (p 0017) in favour of the LMWH group. A repeat perfusion lung scan was done in 81 standardheparin recipients and 79 LMWH recipients. New segmental defects were found in significantly more of the standard-heparin group than of the LMWH group (15 [19%] vs 4 [5%]; difference 13% [3-24%]; p < 0-02). None of the patients had symptoms of pulmonary embolism. Perfusion scans were done for 34 of the 36 patients in the standard-heparin group and 49 of the 50 in the LMWH group who had an improvement in the venographic score. A combination of improved venographic score and new segmental defects on the perfusion lung scans was seen in 9 standard-heparin recipients and 1 LMWH recipient =
(p
