Low-Molecular-Weight Heparinoid Orgaran Is More Effective Than Aspirin in the Prevention of Venous Thromboembolism After Surgery for Hip Fracture
Background The study objective was to determine the relative efficacy and safety of a low-molecular-weight heparinoid (Orgaran) compared with aspirin for the prevention of postoperative venous thromboembolism in patients undergoing surgery for fractured hips. A double-blind, randomized, controlled trial was used to study 251 consecutive eligible and consenting patients undergoing surgery for hip fracture in seven participating hospitals.
Methods and Results Patients received either fixed-dose Orgaran by subcutaneous injection every 12 hours in a dose of 750 anti–Factor Xa units or aspirin 100 mg orally twice daily; both regimens were started 12 to 24 hours after surgery and continued for 14 days or until discharge, if sooner. All patients had postoperative 125I-fibrinogen leg scanning and impedance plethysmography. If the results of one or both tests were positive, then venography was performed. Otherwise, venography was done at day 14, or sooner if the patient was ready for discharge. Pulmonary embolism in symptomatic patients was diagnosed on the basis of a high probability perfusion/ventilation lung scan, a positive angiogram, or a clinically significant embolism detected at autopsy. Evaluable venograms were obtained in 90 of the 125 patients randomly assigned to receive Orgaran and in 87 of the 126 patients assigned to receive aspirin. Venous thromboembolism was detected in 25 (27.8%) patients in the Orgaran group and in 39 (44.3%) patients in the aspirin group. Thus, there was a relative risk reduction of 37% with Orgaran (P=.028; 95% confidence interval, 3.7% to 59.7%). Six (6.8%) of 88 patients in the Orgaran group and 12 (14.3%) of 84 patients in the aspirin group developed proximal deep vein thrombosis or pulmonary embolism, a relative risk reduction of 52% with Orgaran (P=.137; 95% confidence interval, −30.7% to 84.6%). Hemorrhagic complications occurred in 2 (1.6%) patients given Orgaran and 8 (6.4%) patients given aspirin (P=.10). There was one major bleed in the Orgaran group compared with four in the aspirin group.
Conclusions This study demonstrates that Orgaran is significantly more efficacious than aspirin in preventing postoperative venous thromboembolism in patients undergoing surgery for fractured hips, with no evidence of any increase in hemorrhagic complications.
Postoperative venous thromboembolism is a common complication in patients after surgery for hip fracture. Without prophylaxis, the incidence of deep vein thrombosis (DVT) is approximately 50%, and for proximal DVT (involving the popliteal or more proximal veins) it is approximately 20%.1 2 A number of prophylactic agents have been reported to reduce the incidence of postoperative venous thrombosis in patients with hip fracture. These include warfarin,3 4 5 6 aspirin,5 dextran,7 8 low-molecular-weight heparin,9 10 11 and the two heparinoids, Orgaran12 13 and dermatan sulfate.14 Of the studies that have evaluated these agents, relatively few have used mandatory venography, interpreted blindly; both of these features are required for the valid assessment of DVT in patients with hip fracture since noninvasive tests such as impedance plethysmography (IPG), 125I-fibrinogen leg scanning, and duplex ultrasonography are relatively insensitive and 125I-fibrinogen scanning is nonspecific when evaluated over the operated thigh.15 16 17
We have previously reported that warfarin, commenced after surgery, reduced the incidence of DVT, including proximal DVT, by over 50% in patients after surgery for hip fracture compared with an untreated group.5 In that study, aspirin (650 mg BID) did not reduce the incidence of calf DVT but, like warfarin, it reduced the incidence of proximal DVT by approximately 50% compared with a matching placebo. The observed effectiveness of aspirin in preventing the more serious proximal DVT is of interest because aspirin, if given in doses of 325 mg daily or less, is relatively free of side effects, is inexpensive, and is easy to use. We considered, therefore, that the use of aspirin as a prophylactic agent warranted further evaluation in patients with hip fracture. The role of aspirin in venous thrombosis prophylaxis is topical because of the results of the recent publication of a meta-analysis by the Antiplatelet Trialists’ Collaboration in which it was concluded that aspirin is effective in preventing DVT after hip fracture surgery.18
Orgaran, a new low-molecular-weight heparinoid, has been shown to be effective in preventing DVT in the following high-risk groups: patients having elective hip surgery, in whom it was shown to be much more effective and as safe as a matching placebo19 ; patients after surgery for hip fracture, in whom it was shown to be more effective than dextran12 and warfarin13 ; and patients with stroke, in whom it was shown to be more effective than placebo20 and low-dose heparin.21
We have performed a randomized, double-blind study to assess the relative efficacy and safety of aspirin (100 mg BID) and Orgaran in the prophylaxis of venous thromboembolism after surgery for hip fracture.
All patients with a diagnosis of hip fracture admitted to one of seven participating hospitals in Ontario and Quebec were reviewed for eligibility. Patients were excluded from the trial if they had any of the following characteristics: history of bleeding disorder; recent history of gastrointestinal hemorrhage or peptic ulceration; severe hepatic or renal dysfunction; uncontrolled hypertension; requirement for treatment with aspirin, other anti-inflammatory drugs, or anticoagulants; known sensitivity to heparin, aspirin, iodine, or contrast media; or were of childbearing potential. The study protocol was reviewed by the institutional review boards of all participating hospitals, and written informed consent was obtained from each patient or their guardian.
Orgaran (Organon International BV) is a mixture of the sulfated glycosaminoglycans heparan sulfate (84%), dermatan sulfate (12%), and chondroitin sulfate (4%). It has a mean molecular weight of 6500 D and a specific activity of 10 anti–Factor Xa units/mg by chromogenic assay determined against the Third International Heparin Standard. Aspirin was supplied as delayed-release, enteric-coated capsules of 100 mg (Astrix; Faulding Pharma).
Patients were allocated according to a centrally prepared, computer-generated, randomized arrangement, stratified by center in a double-blind manner, to either Orgaran injections and placebo aspirin capsules or to aspirin capsules and placebo injections of saline. Patients, trial nurses, and attending physicians were all unaware of the treatment regimen for individual patients. Treatment was started after surgery as soon as primary hemostasis was achieved. Orgaran was given by subcutaneous injection in a dose of 750 anti–Factor Xa units (0.9 mL) every 12 hours. Aspirin-treated patients received 100 mg orally twice daily. Treatment was continued for 14 days or until discharge from hospital, if earlier.
All patients had 125I-fibrinogen leg scanning as soon as possible after admission to the hospital and then daily after the start of treatment until day 14 or until hospital discharge, if earlier. The fibrinogen was prepared in Hamilton by plasmapheresis of two donors who tested as hepatitis negative and HIV negative. The fibrinogen was stored and not used until repeat testing, 6 months after blood donation, had demonstrated that the donors were not infected with these viruses at the time of donation. Patients also underwent IPG at entry and on days 5, 7, 9, 11, and 13 or until discharge, if earlier.22 Venography was carried out if either leg scanning or IPG results became positive; if both tests remained negative, venography was performed on day 14 or at discharge, if earlier. Thus, bilateral venography was attempted in all patients. DVT was diagnosed when venography revealed a constant intraluminal filling defect in the deep veins.23 Proximal DVT was diagnosed when at least one of the iliac, thigh, or popliteal veins in either leg showed DVT. Pulmonary embolism (PE) in symptomatic patients was diagnosed on the basis of a high probability perfusion/ventilation lung scan, a positive pulmonary angiogram, or a pulmonary embolism detected at autopsy.
If the early venogram result was positive, it was included in the final analysis but not if it was negative. In this latter situation, best efforts were made to get another venogram at day 14±2, or discharge if sooner, and when successful, the result was then used in the final analysis.
Patients were examined daily for evidence of injection-site hematomas, wound bleeding, and other signs of hemorrhage. Bleeding was classified as major if it was overt and was associated with a fall in hemoglobin level of 20 g/L or more; if it led to transfusion of ≥2 units of blood; or if it was retroperitoneal, intra-articular, intraocular, or intracranial. Bleeding was defined as minor if it was overt but did not meet the other criteria for major bleeding. Hemoglobin was measured at baseline and daily thereafter for 14 days or until discharge from the hospital, if earlier. The results of all tests for venous thromboembolism and bleeding episodes were reviewed and classified by a central adjudication committee that was unaware of treatment allocation. These adjudication meetings were held at least monthly in Hamilton and always included two or more of the authors. Study patients were seen or contacted by telephone 3 months after surgery to determine their clinical and outcome event status in the period after completion of the prophylaxis phase of the study.
An estimated sample size requirement of 88 per treatment group was based on an expected incidence of venous thromboembolism of 40% in the aspirin-treated group, a relative risk reduction of 50% with Orgaran, a one-sided α level of 0.05, and a power of 90%. To allow for failure to obtain adequate venography, a total study size of 250 patients was planned.
The primary analysis of efficacy (which was based on an intention to treat) for total venous thromboembolism included patients with evaluable venograms, that is, bilateral venograms with good visualization of the deep venous system or unilateral venograms with visualization of at least one intraluminal filling defect in the iliac, thigh, popliteal, or calf veins, together with those patients who met the criteria for the diagnosis of PE regardless of their venogram status. The efficacy analysis for proximal vein thrombosis included patients with evaluable venograms, that is, bilateral venograms with good visualization of the deep venous system or unilateral venograms with visualization of at least one intraluminal filling defect in the iliac, thigh, or popliteal veins, together with those patients who met the criteria for the diagnosis of PE regardless of their venogram status. For the analyses of efficacy and safety, the decisions of the central adjudication committee were used.
The frequencies of venous thromboembolism and bleeding for the two treatment groups were compared by means of Fisher’s exact test.24 All tests of significance were two sided. Relative risk reductions and associated 95% confidence intervals were calculated using Gart’s method.24
Patient recruitment began in 1989, and the last patient follow-up was completed in 1993. Eight hundred eighty-two consecutive patients admitted to the seven participating hospitals with a diagnosis of hip fracture were assessed for eligibility to enter the trial. Of these, 465 were excluded (Table 1⇓). Of the remaining eligible patients, 166 refused to give consent, and 251 were enrolled into the trial (125 to Orgaran; 126 to aspirin). The patients in the two treatment groups were similar with respect to important baseline characteristics (Table 2⇓). Treatment was started on average within 18 hours in the Orgaran group and within 17 hours in the aspirin group.
The mean duration of treatment was 11.0 days (range, 5 to 15 days) in the Orgaran group and 9.7 days (range, 1 to 17 days) in the aspirin group. The full 14 days of treatment was achieved in 105 patients (59 Orgaran; 46 aspirin), and early discharge from the hospital occurred in 82 patients (47 Orgaran; 35 aspirin). The remaining 64 patients stopped study medication early because of physician or patient withdrawal of consent in 16 patients (6 Orgaran; 10 aspirin) or for suspected/confirmed outcome events in the other 48 patients.
A total of four patients died during the 14-day study period (3 Orgaran; 1 aspirin). Of the patients in the Orgaran group who died, one was due to dehydration secondary to severe diarrhea, one to pneumonia, and the third was an elderly patient found dead in bed. The aspirin-treated patient died as a result of complications of a perforated duodenal ulcer. Autopsies were not obtained on any of these patients.
Venous Thromboembolism and Bleeding
Venography was performed in 203 (80.9%) patients (104 Orgaran; 99 aspirin). The remaining 48 patients did not have venography because of failed access or patient refusal. Of the 203 venograms, 26 (12.8%) were found to be nonevaluable for any DVT by the central adjudication committee (14 Orgaran; 12 aspirin), leaving 177 evaluable venograms (90 Orgaran; 87 aspirin). No patient with an evaluable venogram was investigated for suspected PE. One patient in the aspirin group with a nonevaluable venogram had PE confirmed by ventilation/perfusion lung scan. Similarly, 32 (15.8%) venograms were found to be nonevaluable for proximal DVT by the central adjudication committee (16 Orgaran; 16 aspirin), leaving 171 evaluable venograms (88 Orgaran; 83 aspirin).
Twenty-five (27.8%) of the 90 patients in the Orgaran group and 39 (44.3%) of the 88 patients given aspirin developed venous thromboembolism. Thus, there was a relative risk reduction of 37% with Orgaran (P=.028; 95% confidence interval, 3.7% to 59.7%). Six (6.8%) of 88 patients in the Orgaran group and 12 (14.3%) of 84 patients in the aspirin group developed proximal DVT or PE, a relative risk reduction of 52% with Orgaran (P=.137; 95% confidence interval, −30.7% to 84.6%). The frequencies, probability values, relative risk reductions, and confidence intervals are shown in Table 3⇓.
In the 63 patients with confirmed DVT, thrombosis was detected on the operated side in 20 of 25 (80%) patients in the Orgaran group and in 28 of 38 (74%) in the aspirin group, the overall figures being 48 of 63 (76%). Results of one or both of the noninvasive tests were positive in 29 of 63 patients with DVT (sensitivity, 46%). Results of both leg scanning and IPG were negative in 97 of the 114 patients with negative venogram results (specificity, 85%). Among patients with nonevaluable venograms, 6 of 35 patients (17.1%) in the Orgaran group and 5 of the 39 patients (12.8%) in the aspirin group had positive surveillance tests. Two patients receiving aspirin were investigated for possible PE. One patient with positive IPG and leg scan results and a nonevaluable venogram was diagnosed with PE based on a high-probability lung scan. In the other case, the lung scan was indeterminate for PE, and all noninvasive tests were negative.
As shown in Table 4⇓, overt bleeding complications during the treatment period occurred in 10 patients: one major and one minor bleed (1.6%) in the Orgaran group and four major and four minor bleeds (6.4%) in the aspirin group (P=.10). The number of patients who received blood transfusions and the mean number of units transfused were similar across treatment groups and did not vary appreciably between the designated time periods.
A 3-month poststudy follow-up was carried out on 247 patients available after the 14-day study period; 5 patients in the Orgaran group and 8 patients in the aspirin group died. The causes of death were reported as cardiac (2), congestive heart failure (1), pneumonia (1), and cancer (1) in the Orgaran group and respiratory arrest (3), cardiac arrest (1), acute myocardial infarction (1), aspiration pneumonia (1), bilateral subdural hematomas (1), and unknown cause (1) in the aspirin group. The overall mortality for the entire study was 6.8%.
Assessment for DVT was performed at follow-up only if the patient complained of new symptoms. Two patients in the aspirin group had DVT diagnosed by noninvasive methods, one patient 4 days after study and the other 1 month after study.
The first randomized study evaluating venous thromboembolism prophylaxis was performed by Sevitt and Gallagher4 in 1959 in patients with hip fracture. This landmark study demonstrated that oral anticoagulants are effective in preventing fatal PE in patients who had sustained severe lower limb trauma. Since then, numerous studies have demonstrated the effectiveness of a variety of prophylactic agents in different high-risk patient groups, but there have been few methodologically sound studies performed in patients after surgery for hip fracture.
The results of our study indicate that Orgaran is significantly more effective than and at least as safe as aspirin as a prophylactic agent for postoperative venous thromboembolism after hip fracture. They also confirm our previous observation that the incidence of all venous thrombosis is high when aspirin is the sole means of prophylaxis in these patients. The evaluation of prophylaxis in patients with hip fracture is problematic because venography, the most accurate diagnostic test, can be difficult to perform because many of the patients are old and frail. Even when mandatory venography is used, care must be exercised to avoid bias by ensuring that the venograms are assessed by readers unaware of the treatment to which a patient had been assigned.
Relatively few studies of patients with fractured hip have been published that have both used venography and provided assurance that bias was avoided in interpreting the outcome measure(s). Of the prophylactic agents evaluated in studies satisfying these essential criteria, only warfarin and Orgaran have been shown to significantly reduce the incidence of all DVT and proximal DVT. In a study of 194 patients comparing the relative efficacy and safety of warfarin, aspirin, and an aspirin placebo, the incidence of DVT was 20.0% in the warfarin group, 40.9% in the aspirin group, and 46.0% in the placebo group; the corresponding rates for proximal DVT were 9.2%, 10.6%, and 30.2%, respectively.5 Our latest results with aspirin are consistent with these findings and with the meta-analysis from the Antiplatelet Trialists’ Collaboration,18 in which the pooled incidence of venous thrombosis in hip fracture patients was 35.9% in the groups receiving antiplatelet therapy. We used a relatively low dose of aspirin (200 mg per day), which, however, is well above the minimum dose required for an optimal antithrombotic effect of aspirin.
Two randomized trials have evaluated Orgaran after surgery for hip fracture.12 13 In the first, 308 patients were randomly allocated to receive subcutaneous Orgaran, 750 units twice daily, or dextran; both regimens commenced before surgery.12 Deep venous thrombosis was diagnosed by venography, which was undertaken in all patients. The incidence of DVT was 10% in the Orgaran group and 30% in the dextran group (P<.01). The number of units of blood transfused was significantly higher in the dextran group.
In the second study, patients were randomly allocated to receive subcutaneous Orgaran, 750 units twice daily, or warfarin in a dose that prolonged the prothrombin time to 1.5 times the control level; both regimens were started before surgery.13 DVT was diagnosed by a combination of 125I-fibrinogen leg scanning and IPG with a positive result confirmed by venography or compression ultrasonography. Using these criteria (which would fail to detect about 50% of the thrombi), DVT occurred in 5 of 132 patients (3.8%) in the Orgaran group and in 11 of 131 patients (8.4%) in the warfarin group (P=.13).
In the present study, Orgaran was started after surgery and was shown to be significantly more efficacious than aspirin. The relative risk reduction was 37% for all venous thromboembolism (distal and proximal) and 52% for proximal DVT/PE. This observed incidence of venous thromboembolism in the Orgaran group is higher than in the previous report of Orgaran in patients with hip fracture in which venography was performed in all patients; whether this difference is a chance finding or due to superior efficacy resulting from the use of a preoperative treatment regimen in the study reported by Bergqvist et al12 is uncertain. The observed incidence of venous thromboembolism in the Orgaran group in our study was also higher than that reported by Gerhardt et al.13 However, in the latter study, mandatory venography was not performed in all patients. Rather, DVT was diagnosed only in patients with positive 125I-fibrinogen leg scanning or IPG, which was confirmed by venography or compression ultrasonography.13 Since the sensitivity of leg scanning plus IPG is only about 50%,15 their reported rates of DVT would be expected to be lower than the rates obtained when venography is used in all patients.
Diagnostic suspicion bias for venous thrombosis was avoided because the trial design was double-blind. To provide reliable estimates of event rates, our analysis was based on the adjudicated results of venography, either performed early because noninvasive surveillance tests were positive or performed on day 14 or at discharge, when the noninvasive tests were negative. Therefore, our conclusion that Orgaran is more effective than aspirin (100 mg BID) should be valid.
Our study shows that Orgaran is significantly more effective than aspirin in reducing the incidence of postoperative venous thrombosis after surgery for hip fracture with no evidence of any increase in hemorrhagic complications. While aspirin may have a modest effect in reducing the incidence of venous thrombosis in patients after surgery for hip fracture, it should not be used as the sole prophylactic agent.
Dr J. Hirsh is a distinguished research professor of the Heart and Stroke Foundation of Ontario.
- Received March 16, 1995.
- Revision received August 14, 1995.
- Accepted August 15, 1995.
- Copyright © 1996 by American Heart Association
Morris GK, Mitchell JRA. Preventing venous thromboembolism in elderly patients with hip fractures: studies of low-dose heparin, dipyridamole, aspirin, and flurbiprofen. Br Med J. 1977;1:535-537.
Bergqvist D. Dextran. In: Goldhaber S, ed. Prevention of Venous Thromboembolism. New York, NY: Marcel Dekker; 1993:167-215.
Lassen MR, Borris LC, Christiansen HM, Moller-Larsen F, Knudsen VE, Boris P, Nehen AM, Jurik AG, de Carvalho A, Nielsen BW. Prevention of thromboembolism in hip-fracture patients: comparison of low-dose heparin and low-molecular-weight heparin combined with dihydro-ergotamine. Arch Orthop Trauma Surg. 1989;108:10-13.
Monreal M, Lafoz E, Navarro A, Granero X, Caja V, Caceres E, Salvador R, Ruiz J. A prospective double-blind trial of a low molecular weight heparin once daily compared with conventional low-dose heparin three times daily to prevent pulmonary embolism and venous thrombosis in patients with hip fracture. J Trauma. 1989;29:873-875.
Bergqvist D, Kettunen K, Fredin H, Fauno P, Suomalainen O, Soimakallio S, Karjalainen P, Cederholm C, Jensen LJ, Justesen T, Stiekema JCJ. Thromboprophylaxis in patients with hip fractures: a prospective, randomized, comparative study between Org 10172 and dextran 70. Surgery. 1991;109:617-622.
Gerhart TN, Yett HS, Robertson LK, Lee MA, Smith M, Salzman E. Low molecular weight heparinoid compared with warfarin for prophylaxis of deep-vein thrombosis in patients who are operated on for fracture of the hip. J Bone Joint Surg. 1991;73A:494-502.
Agnelli G, Cosmi B, DiFilippo DP, Ranucci V, Veschi F, Longetti M, Renga C, Barzi F, Gianese F, Lupattelli L, Rinonapoli E, Nenci GG. A randomised, double-blind, placebo-controlled trial of dermatan sulphate for prevention of deep vein thrombosis in hip fracture. Thromb Haemost. 1992;67:203-208.
Ginsberg JS, Caco C, Brill-Edwards P, Panju A, Bona R, Demers C, Tuters L, Nugent P, McGinnis J, Grant B, Vander Lander Vries M. Venous thrombosis in patients who have undergone major hip or knee surgery: detection with compression US and impedance plethysmography. Radiology. 1991;181:651-654.
Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy, III: reduction in venous thrombosis and pulmonary embolism by antiplatelet prophylaxis among surgical and medical patients. BMJ. 1994;308:235-246.
Turpie AGG, Gent M, Cote R, Levine LM, Ginsberg JS, Powers PJ, Leclerc J, Geerts W, Jay R, Neemeh J, Klimek M, Hirsh J. A low-molecular-weight heparinoid compared with unfractionated heparin in the prevention of deep vein thrombosis in patients with acute ischemic stroke: a randomized, double-blind study. Ann Intern Med. 1992;117:353-357.
Hull RD, Hirsh J, Sackett DL, Powers PJ, Turpie AGG, Walker IR, McBride JA. The value of adding impedance plethysmography to 125I-fibrinogen leg scanning for the detection of deep vein thrombosis in high risk surgical patients: a comparative study between patients undergoing general surgery and hip surgery. Thromb Res. 1979;15:227-234.
Fleiss JL. Statistical Methods for Rates and Proportions. 3rd ed. New York, NY: Wiley; 1981.