The New England Journal of Medicine Original Article Volume 345:1298-1304 November 1, 2001 Number 18 Fondaparinux Compared with Enoxaparin for the Prevention of Venous Thromboembolism after Hip-Fracture Surgery Bengt I. Eriksson, M.D., Kenneth A. Bauer, M.D., Michael R. Lassen, M.D., Alexander G.G. Turpie, F.R.C.P., for the Steering Committee of the Pentasaccharide in Hip-Fracture Surgery Study ABSTRACT Background Surgery for hip fracture carries a high risk of venous thromboembolism, despite the use of current thromboprophylactic treatments. Fondaparinux, a synthetic pentasaccharide, is a new antithrombotic agent that may reduce this risk. Methods In a double-blind study, we randomly assigned 1711 consecutive patients undergoing surgery for fracture of the upper third of the femur to receive subcutaneous doses of either 2.5 mg of fondaparinux once daily, initiated postoperatively, or 40 mg of enoxaparin once daily, initiated preoperatively, for at least five days. The primary efficacy outcome was venous thromboembolism up to postoperative day 11. Venous thromboembolism was defined as deep-vein thrombosis detected by mandatory bilateral venography, documented symptomatic deep-vein thrombosis, or documented symptomatic pulmonary embolism. The main safety outcomes were major bleeding and mortality from all causes. The duration of follow-up was six weeks. Results The incidence of venous thromboembolism by day 11 was 8.3 percent (52 of 626 patients) in the fondaparinux group and 19.1 percent (119 of 624 patients) in the enoxaparin group (P<0.001). The reduction in risk with fondaparinux was 56.4 percent (95 percent confidence interval, 39.0 to 70.3 percent). There were no significant differences between the two groups in the incidence of death or clinically relevant bleeding. Conclusions In patients undergoing surgery for hip fracture, fondaparinux was more effective than enoxaparin in preventing venous thromboembolism and was equally safe. _____ Patients undergoing surgery for hip fracture are in the highest category of risk for postoperative venous thromboembolism. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 2 <http://content.nejm.org/cgi/content/full/345/18/#R2> Fatal pulmonary embolism occurs in 3.6 to 12.9 percent of patients who have not received prophylaxis against thromboembolism. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> There are few data on thromboprophylaxis after surgery for hip fracture, and recommendations are based mainly on expert opinion. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> Even with current methods of thromboprophylaxis, the incidence of venographically confirmed deep-vein thrombosis is 24 to 34 percent. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 3 <http://content.nejm.org/cgi/content/full/345/18/#R3> , 4 <http://content.nejm.org/cgi/content/full/345/18/#R4> , 5 <http://content.nejm.org/cgi/content/full/345/18/#R5> , 6 <http://content.nejm.org/cgi/content/full/345/18/#R6> , 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> , 8 <http://content.nejm.org/cgi/content/full/345/18/#R8> , 9 <http://content.nejm.org/cgi/content/full/345/18/#R9> , 10 <http://content.nejm.org/cgi/content/full/345/18/#R10> , 11 <http://content.nejm.org/cgi/content/full/345/18/#R11> , 12 <http://content.nejm.org/cgi/content/full/345/18/#R12> , 13 <http://content.nejm.org/cgi/content/full/345/18/#R13> , 14 <http://content.nejm.org/cgi/content/full/345/18/#R14> , 15 <http://content.nejm.org/cgi/content/full/345/18/#R15> , 16 <http://content.nejm.org/cgi/content/full/345/18/#R16> , 17 <http://content.nejm.org/cgi/content/full/345/18/#R17> , 18 <http://content.nejm.org/cgi/content/full/345/18/#R18> , 19 <http://content.nejm.org/cgi/content/full/345/18/#R19> , 20 <http://content.nejm.org/cgi/content/full/345/18/#R20> , 21 <http://content.nejm.org/cgi/content/full/345/18/#R21> Fondaparinux is a new synthetic pentasaccharide that causes selective inhibition of activated factor X (factor Xa). 22 <http://content.nejm.org/cgi/content/full/345/18/#R22> , 23 <http://content.nejm.org/cgi/content/full/345/18/#R23> , 24 <http://content.nejm.org/cgi/content/full/345/18/#R24> , 25 <http://content.nejm.org/cgi/content/full/345/18/#R25> A recent study of patients undergoing major orthopedic procedures suggested that a once-daily subcutaneous injection of fondaparinux reduces the risk of venous thromboembolism more than does low-molecular-weight heparin. 26 <http://content.nejm.org/cgi/content/full/345/18/#R26> We conducted a multicenter, randomized, double-blind trial to compare two types of thromboprophylaxis after hip-fracture surgery: a once-daily subcutaneous injection of fondaparinux, initiated postoperatively, and a once-daily subcutaneous injection of enoxaparin, initiated preoperatively. Methods Patients Patients were considered for inclusion if they were at least 18 years of age and were scheduled to undergo standard surgery for fracture of the upper third of the femur, including the femoral head and neck, within 48 hours after admission. The main reasons for exclusion were multiple trauma affecting more than one organ system; an interval of more than 24 hours between the injury and hospital admission; pregnancy; active bleeding; a documented congenital or acquired bleeding disorder; current ulcerative or angiodysplastic gastrointestinal disease; a history of hemorrhagic stroke or brain, spinal, or ophthalmologic surgery within the previous three months; planned use of an indwelling intrathecal or epidural catheter for more than six hours after surgery; hypersensitivity to heparin, low-molecular-weight heparins, porcine products, or iodinated contrast medium; a contraindication to anticoagulant therapy; a current addictive disorder; a serum creatinine concentration above 2 mg per deciliter (177 µmol per liter) in a well-hydrated patient; and a platelet count below 100,000 per cubic millimeter. Patients who required anticoagulant therapy or received dextran or any type of anticoagulant or fibrinolytic therapy from admission to the time of first administration of the study drug or surgery were also excluded. Study Design Within 24 hours after admission and before surgery, patients were randomly assigned to treatment groups in blocks of four, with stratification according to center, with the use of a computer-generated randomization list. Patients were assigned to receive once-daily subcutaneous injections of either 2.5 mg of fondaparinux (Arixtra, Sanofi–Synthelabo, Paris, and NV Organon, Oss, the Netherlands) and a placebo or 40 mg of enoxaparin (Clexane/Lovenox, Aventis Pharmaceuticals, Bridgewater, N.J.) and a placebo. In the enoxaparin group, the first active dose was given 12±2 hours preoperatively and the second 12 to 24 hours postoperatively, according to the recommendation of the manufacturer. Since fondaparinux is a new compound, which differs from enoxaparin in its mechanism of action and pharmacokinetic properties, the starting time after surgery and the dose were determined during the early development of the drug 26 <http://content.nejm.org/cgi/content/full/345/18/#R26> ; the first dose of fondaparinux was administered 6±2 hours postoperatively and the second 12 hours or more after the first. However, if surgery was delayed until 24 to 48 hours after admission, administration of fondaparinux was initiated 12±2 hours before surgery. In both groups, omission of preoperative injections was recommended if spinal or epidural anesthesia or catheterization was planned, and any indwelling intrathecal or epidural catheter was to be removed at least two hours before the first postoperative injection. Day 1 was defined as the day of surgery. Treatment was scheduled to continue until day 5 to day 9, and the primary efficacy outcome was assessed between day 5 and day 11. Patients were then followed up in person, by mail, or by telephone between day 35 and day 49. During follow-up, patients were instructed to report any symptoms or signs of venous thromboembolism or bleeding and any other clinical event occurring since the completion of treatment. Investigators could extend prophylaxis during follow-up with any currently available therapy, but only after venography had been performed. If venous thromboembolism occurred during the study, treatment was left to the discretion of the investigators. The study was conducted according to the ethical principles stated in the Declaration of Helsinki and local regulations. The protocol was approved by independent ethics committees, and written informed consent was obtained from all patients before randomization. Medications Study medications were packaged in boxes of identical appearance, each containing 10 prefilled, single-dose syringes of active treatment and 10 prefilled, single-dose syringes of matching placebos. Each syringe contained either 2.5 mg of fondaparinux sodium in 0.25 ml of water for injectable preparations (a concentration of 10 mg per milliliter), 40 mg of enoxaparin sodium in 0.4 ml of water for injectable preparations (a concentration of 100 mg per milliliter), or placebo (0.25 or 0.4 ml of isotonic saline). Throughout the treatment period, the use of intermittent pneumatic compression, dextran, and thrombolytic, anticoagulant, or antiplatelet agents was prohibited. Centers were advised to avoid giving patients aspirin or nonsteroidal antiinflammatory drugs whenever possible. The use of graduated compression stockings and physiotherapy was recommended. Outcome Measures The primary efficacy outcome was assessed by the rate of venous thromboembolism (defined as deep-vein thrombosis, pulmonary embolism, or both) up to day 11. Secondary efficacy outcomes were total, proximal, or distal deep-vein thrombosis or symptomatic venous thromboembolism up to day 11 and symptomatic venous thromboembolism up to day 49. Patients were examined for deep-vein thrombosis by systematic bilateral ascending venography of the legs 27 <http://content.nejm.org/cgi/content/full/345/18/#R27> between day 5 and day 11, but no more than two days after the last dose of study drug, or earlier if thrombosis was clinically suspected. Symptomatic pulmonary embolism was confirmed by a lung scan indicating a high probability of pulmonary embolism, by pulmonary angiography, 28 <http://content.nejm.org/cgi/content/full/345/18/#R28> by helical computed tomography, or at autopsy. The primary safety outcome was the incidence of major bleeding, which included fatal bleeding; bleeding that was retroperitoneal, intracranial, or intraspinal or that involved any other critical organ; bleeding leading to reoperation; and overt bleeding with a bleeding index of 2 or more. The bleeding index was calculated as the number of units of packed red cells or whole blood transfused plus the hemoglobin values before the bleeding episode minus the hemoglobin values after the episode (in grams per deciliter). Secondary safety outcomes were death, minor bleeding, a need for transfusion, thrombocytopenia, and any other adverse event. Minor bleeding was defined as overt bleeding that did not meet the criteria for major bleeding. Efficacy and safety outcomes were adjudicated by a central independent committee whose members were unaware of the treatment assignments and included review of all venograms and reports of bleeding and death. Statistical Analysis Assuming an incidence of venous thromboembolism by day 11 of 22 percent in the enoxaparin group 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> and a risk reduction of about 30 percent (i.e., an incidence of 15 percent in the fondaparinux group), 600 patients were needed per group to provide the study with a power of 85 percent. The target number of recruited patients was 1700, a number that allowed for failure to obtain primary efficacy data in approximately 30 percent of patients. The analysis of the primary efficacy outcome included data on all patients who had received at least one dose of study medication, had undergone the appropriate surgery, and had had an adequate assessment for venous thromboembolism by day 11. The analysis of safety included data on patients who had received at least one dose of study medication. A two-tailed P value of less than 0.05 was considered to indicate statistical significance. The analysis of the primary efficacy outcome was performed with the use of a two-sided Fisher's exact test. Exact 95 percent confidence intervals for the absolute difference between fondaparinux and enoxaparin and the risk ratio were calculated. The treatment effect was also analyzed according to predefined categorical covariates with use of a logistic-regression model. The study was supervised by a steering committee of 10 people, which included 6 representatives of the sponsor (Sanofi–Synthelabo and NV Organon). The committee designed the study, interpreted the data, and wrote the article. The final statistical analysis was performed by the sponsor. The central adjudication committee and the data-monitoring committee operated independently of the sponsor. One planned interim analysis was conducted when half the projected patient population had been enrolled, for reestimation of the sample size, since the rate of venous thromboembolism in patients undergoing hip-fracture surgery was uncertain. Simulations demonstrated that the predefined procedure did not inflate the type I error. No change in the sample size was found to be necessary, and the study continued as planned. Results Study Population Between November 1998 and October 1999, 1711 patients were enrolled in 99 centers in 21 countries (listed in the Appendix). Thirty-eight patients did not receive either study drug ( Table 1 <http://content.nejm.org/cgi/content/full/345/18/#T1> ). Two patients did not undergo the appropriate surgery, and primary efficacy had not been assessed by day 11 in 421 patients. Thus, 1250 patients (73.1 percent) were included in the primary efficacy analysis, a percentage in line with other large multicenter studies that used venography after orthopedic surgery. 29 <http://content.nejm.org/cgi/content/full/345/18/#R29> , 30 <http://content.nejm.org/cgi/content/full/345/18/#R30> , 31 <http://content.nejm.org/cgi/content/full/345/18/#R31> The characteristics of patients excluded from the primary efficacy analysis did not differ from those of patients included in the analysis (data not shown). View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1298/T1> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1298/T1> Table 1. Patients Included in the Analyses and Reasons for Exclusion. Base-line characteristics did not differ significantly between the two groups of patients included in the analysis of safety ( Table 2 <http://content.nejm.org/cgi/content/full/345/18/#T2> ) or primary efficacy (data not shown). A total of 551 and 569 patients underwent surgery within 24 hours after admission in the fondaparinux and enoxaparin groups, respectively. Among the 626 patients in the fondaparinux group who were included in the primary efficacy analysis, fondaparinux was given preoperatively to 68 (10.9 percent) because surgery was delayed until 24 to 48 hours after admission; enoxaparin was given postoperatively, rather than preoperatively, to 464 of 624 patients assigned to that drug (74.4 percent) because of very early surgery after admission or planned regional anesthesia. The median time between surgery and the assessment of primary efficacy was eight days in both groups; most patients were assessed between day 5 and day 11 as planned. The two groups did not differ significantly with regard to the last day of active treatment or the use of concomitant treatments up to day 11 ( Table 3 <http://content.nejm.org/cgi/content/full/345/18/#T3> ). View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1298/T2> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1298/T2> Table 2. Base-Line Characteristics of the Patients. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1298/T3> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1298/T3> Table 3. Treatments Received during the Study Period by Patients Assessed for the Primary Efficacy Outcome. Overall, 829 patients treated with fondaparinux and 840 patients treated with enoxaparin returned for the follow-up visit on day 49. The duration of follow-up was similar between the two groups. During follow-up of patients who did not receive treatment for an acute thromboembolic event, 58.5 percent of patients treated with fondaparinux and 55.8 percent of patients treated with enoxaparin received prolonged thromboprophylaxis, primarily with a preparation of heparin or a vitamin K antagonist, after the study treatment. Incidence of Venous Thromboembolism The incidence of venous thromboembolism by day 11 was 8.3 percent in the fondaparinux group (52 of 626 patients) and 19.1 percent in the enoxaparin group (119 of 624 patients). This was a decrease of 10.8 percentage points, or a relative reduction in risk of 56.4 percent (95 percent confidence interval, 39.0 to 70.3 percent; P<0.001) ( Table 4 <http://content.nejm.org/cgi/content/full/345/18/#T4> ). A similar result was found in sensitivity analyses when patients who had had no primary efficacy assessment by day 11 were included in the primary efficacy analysis (data not shown). The incidence of total, proximal, and distal-only deep-vein thrombosis was significantly lower in the fondaparinux group (P<0.001 for all three comparisons). The incidence of symptomatic venous thromboembolism was low (6.5 percent), with no difference between the two groups. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1298/T4> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1298/T4> Table 4. Incidence of Venous Thromboembolic Events by Day 11. The superior efficacy of fondaparinux over enoxaparin was found when patients were grouped according to age, sex, body-mass index (the weight in kilograms divided by the square of the height in meters [<30 vs. >=30]), type of anesthesia (general, regional, or both), type of hip fracture (cervical, trochanteric, or subtrochanteric), type of surgery (implantation of half prosthesis, implantation of total prosthesis, or osteosynthesis), the use or nonuse of cement, or whether or not the patient had had previous venous thromboembolism (data not shown). The number of patients treated by participating physicians for a venous thromboembolic event by day 11 was significantly lower in the fondaparinux group (6.1 percent [43 of 702]) than in the enoxaparin group (11.7 percent [84 of 716], P<0.001). By day 49, the incidence of symptomatic venous thromboembolism was similar in the fondaparinux group (2.0 percent [17 of 831 patients]) and the enoxaparin group (1.5 percent [13 of 840 patients]). Fatal pulmonary embolism occurred in 8 of 831 patients in the fondaparinux group and 7 of 840 patients in the enoxaparin group; nonfatal pulmonary embolism occurred in 3 of 831 patients and 4 of 840 patients, respectively. Safety Outcomes Major bleeding occurred by day 11 in 18 of 831 patients treated with fondaparinux and 19 of 842 patients treated with enoxaparin (P=1.00) ( Table 5 <http://content.nejm.org/cgi/content/full/345/18/#T5> ). Most of these episodes occurred at the surgical site (14 of 18 patients in the fondaparinux group and 14 of 19 patients in the enoxaparin group). Minor bleeding occurred more often in the fondaparinux group (P=0.02). By day 49, three patients in the fondaparinux group and six patients in the enoxaparin group underwent reoperation because of bleeding. Transfusion requirements and the incidence of other adverse events during treatment or follow-up did not differ significantly between groups. The platelet count was lower than 100,000 per cubic millimeter in 40 of 822 patients in the fondaparinux group (4.9 percent) and 44 of 831 patients in the enoxaparin group (5.3 percent). No episode of decreased platelet count was reported as a serious adverse event in either group. The incidence of wound infection was low and was the same in both groups (0.7 percent [6 of 831 in the fondaparinux group and 6 of 842 in the enoxaparin group]). By day 49, 38 patients in the fondaparinux group (4.6 percent) and 42 in the enoxaparin group (5.0 percent) had died. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1298/T5> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1298/T5> Table 5. Safety Outcomes. Discussion This large study demonstrates that fondaparinux is significantly more effective than enoxaparin in preventing postoperative venous thromboembolism after surgery for hip fracture. The 19.1 percent incidence of venous thromboembolism in the enoxaparin group by day 11 is consistent with the results of previous studies of enoxaparin after hip-fracture surgery. 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> , 8 <http://content.nejm.org/cgi/content/full/345/18/#R8> , 9 <http://content.nejm.org/cgi/content/full/345/18/#R9> , 10 <http://content.nejm.org/cgi/content/full/345/18/#R10> By contrast, 8.3 percent of patients given fondaparinux had postoperative venous thromboembolism. Moreover, proximal deep-vein thrombosis, which is prone to embolize, occurred in 6 of 650 patients in the fondaparinux group and 28 of 646 patients in the enoxaparin group (P<0.001). 32 <http://content.nejm.org/cgi/content/full/345/18/#R32> , 33 <http://content.nejm.org/cgi/content/full/345/18/#R33> , 34 <http://content.nejm.org/cgi/content/full/345/18/#R34> Three other large studies in patients undergoing elective knee 35 <http://content.nejm.org/cgi/content/full/345/18/#R35> or hip-replacement 36 <http://content.nejm.org/cgi/content/full/345/18/#R36> , 37 <http://content.nejm.org/cgi/content/full/345/18/#R37> surgery also showed the superiority of fondaparinux over enoxaparin in preventing venous thromboembolism. The efficacy of fondaparinux may be attributed to its ability to inhibit factor Xa rapidly and selectively, its predictable linear pharmacokinetics, and its relatively long half-life, which permits the drug to achieve an antithrombotic effect for 24 hours. Physicians have been uncertain about effective and safe thromboprophylaxis after hip-fracture surgery. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 2 <http://content.nejm.org/cgi/content/full/345/18/#R2> Warfarin is moderately effective, 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 15 <http://content.nejm.org/cgi/content/full/345/18/#R15> , 16 <http://content.nejm.org/cgi/content/full/345/18/#R16> , 17 <http://content.nejm.org/cgi/content/full/345/18/#R17> , 18 <http://content.nejm.org/cgi/content/full/345/18/#R18> , 19 <http://content.nejm.org/cgi/content/full/345/18/#R19> and aspirin is not recommended in patients undergoing such surgery. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 20 <http://content.nejm.org/cgi/content/full/345/18/#R20> , 38 <http://content.nejm.org/cgi/content/full/345/18/#R38> , 39 <http://content.nejm.org/cgi/content/full/345/18/#R39> Promising results have been reported in small studies of 40 mg of enoxaparin administered once daily, with treatment initiated preoperatively. 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> , 8 <http://content.nejm.org/cgi/content/full/345/18/#R8> , 9 <http://content.nejm.org/cgi/content/full/345/18/#R9> , 10 <http://content.nejm.org/cgi/content/full/345/18/#R10> In our study, because of planned regional anesthesia, early surgery after admission, or both, only 25.6 percent of patients received the preoperative injection of enoxaparin. This indicates the difficulty of administering low-molecular-weight heparin preoperatively in emergency situations. In our study, symptomatic events were rare during the treatment period, with a 0.2 percent incidence of fatal pulmonary embolism — similar to that reported in the large Pulmonary Embolism Prevention trial. 20 <http://content.nejm.org/cgi/content/full/345/18/#R20> However, the incidence of symptomatic events in our study should be interpreted with caution. Early detection by venographic screening and prolonged prophylaxis in nearly 60 percent of our patients probably prevented symptomatic venous thromboembolism. The incidence of fatal pulmonary embolism by day 49 was nevertheless nearly 1.0 percent in both groups. The duration of treatment may have been too short for some patients who were still at risk for venous thromboembolism when treatment was discontinued. Our study demonstrates that prophylactic fondaparinux is more effective than enoxaparin in preventing venous thromboembolism in patients undergoing hip-fracture surgery and does not increase the risk of clinically relevant bleeding. <http://weeklybriefings.org/feature.asp?strXmlDoc=3451802> Supported by NV Organon and Sanofi–Synthelabo. All authors have served as consultants to NV Organon and Sanofi–Synthelabo. Presented in abstract form at the 42nd Annual Meeting of the American Society of Hematology, San Francisco, December 1–5, 2000 (Blood 2000;9:490A, A2110). * Participants in the study are listed in the Appendix. <http://content.nejm.org/cgi/content/full/345/18/#RFN1> Source Information From the Department of Orthopedics, Sahlgrenska University Hospital–Östra, Göteborg, Sweden (B.I.E.); the Department of Medicine, Veterans Affairs Boston Healthcare System and Beth Israel Deaconess Medical Center, Boston (K.A.B.); the Department of Orthopedics, Hillerød University, Hillerød, Denmark (M.R.L.); and the Department of Medicine, Hamilton Health Sciences Corporation–General Division, Hamilton, Ont., Canada (A.G.G.T.). Address reprint requests to Dr. Eriksson at the Orthopedics Department, Sahlgrenska University Hospital–Östra, S-41685 Göteborg, Sweden, or at [log in to unmask] <mailto:[log in to unmask]> . References 1. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest 2001;119:Suppl:132S-175S. [Full Text] <http://content.nejm.org/cgi/ijlink?linkType=FULL&journalCode=chest&resid=11 9/1_suppl/132S> 2. Nicolaides AN. Prevention of venous thromboembolism: international consensus statement: guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1-37. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=11342993&link_type=MED> 3. Fisher CG, Blachut PA, Salvian AJ, Meek RN, O'Brien PJ. Effectiveness of pneumatic leg compression devices for the prevention of thromboembolic disease in orthopaedic trauma patients: a prospective, randomized study of compression alone versus no prophylaxis. J Orthop Trauma 1995;9:1-7. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=7714648&link_type=MED> 4. Hartman JT, Pugh JL, Smith RD, Robertson WW Jr, Yost RP, Janssen HF. Cyclic sequential compression of the lower limbs in prevention of deep venous thrombosis. J Bone Joint Surg Am 1982;64:1059-1062. [Abstract] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=jobojos&resid= 64/7/1059> 5. Moskovitz PA, Ellenberg SS, Feffer HL, et al. Low-dose heparin for prevention of venous thromboembolism in total hip arthroplasty and surgical repair of hip fractures. J Bone Joint Surg Am 1978;60:1065-1070. [Abstract] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=jobojos&resid= 60/8/1065> 6. Monreal M, Lafoz E, Navarro A, et al. 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. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2544742&link_type=MED> 7. Borris LC, Lassen MR, Poulsen KA, Jensen HP. Thromboembolic complications after hip fracture -- prophylaxis with low molecular weight- vs. unfractionated heparin. Thromb Haemost 1995;73:1104-1104.abstract 8. Barsotti J, Gruel Y, Rosset P, et al. Comparative double-blind study of two dosage regimens of low-molecular weight heparin in elderly patients with a fracture of the neck of the femur. J Orthop Trauma 1990;4:371-375. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2176235&link_type=MED> 9. The TIFDED Study Group. Thromboprophylaxis in hip fracture surgery: a pilot study comparing danaparoid, enoxaparin and dalteparin. Haemostasis 1999;29:310-317. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=10844404&link_type=MED> 10. Jorgensen PS, Strandberg C, Wille-Jorgensen P, et al. Early preoperative thromboprophylaxis with Klexane in hip fracture surgery: a placebo-controlled study. Clin Appl Thromb Hemost 1998;4:140-142. 11. Agnelli G, Cosmi B, Di Filippo P, et al. 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. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=1621239&link_type=MED> 12. Bergqvist D, Kettunen K, Fredin H, et al. Thromboprophylaxis in patients with hip fractures: a prospective, randomized, comparative study between ORG 10172 and dextran 70. Surgery 1991;109:617-622. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=1708528&link_type=MED> 13. Gent M, Hirsh J, Ginsberg JS, et al. Low-molecular-weight heparinoid Orgaran is more effective than aspirin in the prevention of venous thromboembolism after surgery for hip fracture. Circulation 1996;93:80-84. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=circulationaha &resid=93/1/80> 14. Gerhart TN, Yett HS, Robertson LK, Lee MA, Smith M, Salzman EW. Low-molecular-weight heparinoid compared with warfarin for prophylaxis of deep-vein thrombosis in patients who are operated on for fracture of the hip: a prospective, randomized trial. J Bone Joint Surg Am 1991;73:494-502. [Abstract] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=jobojos&resid= 73/4/494> 15. Borgström S, Greitz T, van der Linden W, et al. Anticoagulant prophylaxis of venous thrombosis in patients with fractured neck of the femur: a controlled clinical trial using venous phlebography. Acta Chir Scand 1965;129:500-508. 16. Hamilton JW, Crawford JS, Gardiner JH, Wiley AM. Venous thrombosis in patients with fracture of the upper end of the femur: a phlebographic study of the effect of prophylactic anticoagulation. J Bone Joint Surg Br 1970;52:268-289. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=5445407&link_type=MED> 17. Powers PJ, Gent M, Jay RM, et al. A randomized trial of less intense postoperative warfarin or aspirin therapy in the prevention of venous thromboembolism after surgery for fractured hip. Arch Intern Med 1989;149:771-774. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2650646&link_type=MED> 18. Bronge A, Dahlgren S, Lindquist B. Prophylaxis against thrombosis in femoral neck fractures -- a comparison between dextran 70 and dicumarol. Acta Chir Scand 1971;137:29-35. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=4112897&link_type=MED> 19. Bergquist E, Bergqvist D, Bronge A, Dahlgren S, Lindquist B. An evaluation of early thrombosis prophylaxis following fracture of the femoral neck: a comparison between dextran and dicoumarol. Acta Chir Scand 1972;138:689-693. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=4118601&link_type=MED> 20. Pulmonary Embolism Prevention (PEP) Trial Collaborative Group. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) trial. Lancet 2000;355:1295-1302. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=10776741&link_type=MED> 21. 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. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=bmj&resid=308/ 6923/235> 22. Petitou M, Lormeau J-C, Choay J. Chemical synthesis of glycosaminoglycans: new approaches to antithrombotic drugs. Nature 1991;350:Suppl:30-33. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2017262&link_type=MED> 23. van Boeckel CAA, Petitou M. The unique antithrombin III binding domain of heparin: a lead to new synthetic antithrombotics. Angew Chem Int Ed Engl 1993;32:1671-1690. 24. Lormeau JC, Herault JP. The effect of the synthetic pentasaccharide SR 90107/ORG31540 on thrombin generation ex vivo is uniquely due to ATIII-mediated neutralization of factor Xa. Thromb Haemost 1995;74:1474-1477. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=8772223&link_type=MED> 25. Walenga JM, Bara L, Petitou M, Samama MM, Fareed J, Choay J. The inhibition of the generation of thrombin and the antithrombotic effect of a pentasaccharide with sole anti-factor Xa activity. Thromb Res 1988;51:23-33. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=3166242&link_type=MED> 26. Turpie AGG, Gallus AS, Hoek JA. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 2001;344:619-625. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=344 /9/619> 27. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972;104:134-144. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=5008903&link_type=MED> 28. The PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA 1990;263:2753-2759. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2332918&link_type=MED> 29. Eriksson BI, Ekman S, Kälebo P, Zachrisson B, Bach D, Close P. Prevention of deep-vein thrombosis after total hip replacement: direct thrombin inhibition with recombinant hirudin, CGP 39393. Lancet 1996;347:635-639. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=8596376&link_type=MED> 30. Eriksson BI, Wille-Jørgensen P, Kälebo P, et al. A comparison of recombinant hirudin with a low-molecular-weight heparin to prevent thromboembolic complications after total hip replacement. N Engl J Med 1997;337:1329-1335. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=337 /19/1329> 31. Hull RD, Pineo GF, Francis C, et al. Low-molecular-weight heparin prophylaxis using dalteparin in close proximity to surgery vs warfarin in hip arthroplasty patients: a double-blind randomized comparison. Arch Intern Med 2000;160:2199-2207. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=10904464&link_type=MED> 32. Havig Ö. Deep vein thrombosis and pulmonary embolism: an autopsy study with multiple regression analysis of possible risk factors. Acta Chir Scand Suppl 1977;478:1-20. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=146995&link_type=MED> 33. Moser KM, LeMoine JR. Is embolic risk conditioned by location of deep venous thrombosis? Ann Intern Med 1981;94:439-444. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=7212500&link_type=MED> 34. Eriksson BI, Kälebo P, Anthmyr BA, Wadenvik H, Tengborn L, Risberg B. Prevention of deep-vein thrombosis and pulmonary embolism after total hip replacement: comparison of low-molecular-weight heparin and unfractionated heparin. J Bone Joint Surg Am 1991;73:484-493. [Abstract] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=jobojos&resid= 73/4/484> 35. Bauer K. The PENTAMAKS Study: comparison of the first synthetic factor Xa inhibitor with low molecular weight heparin in the prevention of venous thromboembolism (VTE) after elective major knee surgery. Blood 2000;96:491a-491a.abstract 36. Lassen MR. The EPHESUS Study: comparison of the first synthetic factor Xa inhibitor with low molecular weight heparin (LMWH) in the prevention of venous thromboembolism (VTE) after elective hip replacement surgery. Blood 2000;96:490a-490a.abstract 37. Turpie G. The PENTATHLON 2000 Study: comparison of the first synthetic factor Xa inhibitor with low molecular weight heparin in the prevention of venous thromboembolism (VTE) after elective hip replacement surgery. Blood 2000;96:491a-491a.abstract 38. Sors H, Meyer G. Place of aspirin in prophylaxis of venous thromboembolism. Lancet 2000;355:1288-1289. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=10776735&link_type=MED> 39. Cohen A, Quinlan D. PEP trial. Lancet 2000;356:247-247. Fondaparinux Compared with Enoxaparin for the Prevention of Venous Thromboembolism after Elective Major Knee Surgery Kenneth A. Bauer, M.D., Bengt I. Eriksson, M.D., Michael R. Lassen, M.D., Alexander G.G. Turpie, F.R.C.P., for the Steering Committee of the Pentasaccharide in Major Knee Surgery Study ABSTRACT Background Despite thromboprophylaxis, major knee surgery carries a high risk of venous thromboembolism. Fondaparinux, the first of a new class of synthetic antithrombotic agents, may reduce this risk. Methods In a double-blind study, we randomly assigned 1049 consecutive patients undergoing elective major knee surgery to receive subcutaneous doses of either 2.5 mg of fondaparinux once daily or 30 mg of enoxaparin twice daily, with both treatments initiated postoperatively. The primary efficacy outcome was venous thromboembolism up to postoperative day 11, defined as deep-vein thrombosis detected by mandatory bilateral venography, documented symptomatic deep-vein thrombosis, or documented symptomatic pulmonary embolism. The primary safety outcome was major bleeding. Results The primary efficacy outcome was assessed in 724 patients. The fondaparinux group had a significantly lower incidence of venous thromboembolism by day 11 (12.5 percent [45 of 361 patients]) than the enoxaparin group (27.8 percent [101 of 363 patients]; reduction in risk, 55.2 percent; 95 percent confidence interval, 36.2 to 70.2; P<0.001). Major bleeding (including overt bleeding with a bleeding index of 2 or more) occurred more frequently in the fondaparinux group (P=0.006), but there were no significant differences between the two groups in the incidence of bleeding leading to death or reoperation or occurring in a critical organ. Conclusions In patients undergoing elective major knee surgery, postoperative treatment with 2.5 mg of fondaparinux once daily was significantly more effective in preventing deep-vein thrombosis than 30 mg of enoxaparin twice daily. _____ Venous thromboembolism is a frequent, life-threatening postoperative complication of total-knee-replacement surgery. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> , 2 <http://content.nejm.org/cgi/content/full/345/18/#R2> Without thromboprophylaxis, the prevalence rate is 40 to 84 percent for venographically verified postoperative deep-vein thrombosis and 2 to 7 percent for pulmonary embolism. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> Thromboprophylaxis that is effective in hip-replacement surgery, such as low-dose heparin, low-molecular-weight heparin, or warfarin, is less successful in knee-replacement surgery and reduces the prevalence of deep-vein thrombosis only to 31 to 47 percent. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> For this reason, more effective antithrombotic prophylaxis is needed in knee-replacement surgery. Fondaparinux is one of a new class of antithrombotic agents, the selective inhibitors of activated factor X (factor Xa). 3 <http://content.nejm.org/cgi/content/full/345/18/#R3> , 4 <http://content.nejm.org/cgi/content/full/345/18/#R4> , 5 <http://content.nejm.org/cgi/content/full/345/18/#R5> , 6 <http://content.nejm.org/cgi/content/full/345/18/#R6> Fondaparinux is an entirely synthetic pentasaccharide that is structurally related to the antithrombin-binding site of heparin. In contrast to heparin, which interacts with many plasma components, the pentasaccharide selectively binds to antithrombin, causing it to rapidly inhibit factor Xa, a key enzyme in the coagulation pathway. Recent dose-ranging studies suggested that a once-daily subcutaneous injection of 2.5 mg of fondaparinux can prevent venous thromboembolism after hip-replacement 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> or knee-replacement surgery (unpublished data). This multicenter, randomized, double-blind trial was part of a program that also evaluated fondaparinux as prophylaxis against venous thromboembolism in patients undergoing surgery for hip fracture 8 <http://content.nejm.org/cgi/content/full/345/18/#R8> and elective hip replacement. 9 <http://content.nejm.org/cgi/content/full/345/18/#R9> , 10 <http://content.nejm.org/cgi/content/full/345/18/#R10> The aim of the study was to compare the efficacy and safety of a once-daily subcutaneous injection of 2.5 mg of pentasaccharide with twice-daily subcutaneous injections of 30 mg of enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. Methods Patients Patients were considered for inclusion if they were at least 18 years of age and were undergoing elective major knee surgery — that is, surgery requiring resection of the distal end of the femur or proximal end of the tibia or revision of at least one component of a previously implanted total-knee prosthesis. Patients were excluded if surgery in the contralateral knee was performed at the same time or within two weeks after enrollment. Women were excluded if they were pregnant or not using effective contraception. Other main reasons for exclusion were active bleeding; a documented congenital or acquired bleeding disorder; current ulcerative or angiodysplastic gastrointestinal disease; hemorrhagic stroke or brain, spinal, or ophthalmologic surgery within the previous three months; insertion of an indwelling intrathecal or epidural catheter during the treatment period; unusual difficulty in administering epidural or spinal anesthesia (e.g., more than two attempts); hypersensitivity to heparin, low-molecular-weight heparins, porcine products, or iodinated contrast medium; a contraindication to anticoagulant therapy; a current addictive disorder; a serum creatinine concentration above 2 mg per deciliter (177 µmol per liter) in a well-hydrated patient; and a platelet count below 100,000 per cubic millimeter. Finally, patients who required anticoagulant therapy were also excluded. The use within one week before randomization of dextran or any type of anticoagulant, fibrinolytic, or antiplatelet agent was discouraged. Study Design Immediately after surgery, patients were randomly assigned (in a ratio of 1:1 in blocks of four, stratified according to center), through a central computer-derived randomization scheme to receive subcutaneous doses of either 2.5 mg of fondaparinux (Arixtra, NV Organon, Oss, the Netherlands, and Sanofi–Synthelabo, Paris) once daily and a placebo once daily or 30 mg of enoxaparin (Clexane/Lovenox, Aventis Pharmaceuticals, Bridgewater, N.J.) twice daily. In the enoxaparin group, the first dose was given between 12 and 24 hours after surgery, according to the recommendation of the manufacturer. Since fondaparinux is a new compound, which differs from enoxaparin in its mechanism of action and pharmacokinetic properties, the starting time after surgery and the dose were determined during the early development of the drug 7 <http://content.nejm.org/cgi/content/full/345/18/#R7> ; the first postoperative injection was administered 6±2 hours after surgery and the second injection 12 hours or more after the first. The day of surgery was defined as day 1. Treatment was scheduled to continue until day 5 to day 9, and the primary efficacy outcome was assessed between day 5 and day 11. Patients were then followed up in person, by mail, or by telephone between day 35 and day 49. During follow-up, patients were instructed to report any symptoms or signs of venous thromboembolism or bleeding and any other clinical event occurring since the completion of treatment. Investigators could extend prophylaxis during follow-up with any currently available therapy, but only after venography had been performed. If venous thromboembolism occurred during the study, treatment was left to the discretion of the investigator. The study was conducted according to the ethical principles stated in the Declaration of Helsinki and local regulations. The protocol was approved by independent local institutional review boards, and written informed consent was obtained from all patients before randomization. Medications Study medications were packaged in boxes of identical appearance, each containing 19 prefilled, single-dose syringes: 10 syringes of fondaparinux and 9 syringes of placebo for each patient assigned to fondaparinux, or 18 syringes of enoxaparin and 1 syringe of placebo for each patient assigned to enoxaparin. Each syringe contained 2.5 mg of fondaparinux sodium in 0.25 ml of water for injectable preparations (a concentration of 10 mg per milliliter), 30 mg of enoxaparin sodium in 0.3 ml of water for injectable preparations (a concentration of 100 mg per milliliter), or placebo (0.25 or 0.3 ml of isotonic saline). Each syringe was loaded inside an opaque autoinjector (Autoject, Owen Mumford, Woodstock, United Kingdom) to maintain blinding. Throughout the treatment period, the use of intermittent pneumatic compression, dextran, and any other anticoagulant, thrombolytic, or antiplatelet agent was prohibited. Centers were advised to avoid giving patients aspirin or nonsteroidal antiinflammatory drugs whenever possible. The use of graduated-compression stockings and physiotherapy was recommended. Outcome Measures The primary efficacy outcome was assessed by the rate of venous thromboembolism (defined as deep-vein thrombosis, pulmonary embolism, or both) up to day 11. Secondary efficacy outcomes were total, proximal, or distal deep-vein thrombosis or symptomatic venous thromboembolism up to day 11 and symptomatic venous thromboembolism up to day 49. Patients were examined for deep-vein thrombosis by systematic bilateral ascending venography of the legs 11 <http://content.nejm.org/cgi/content/full/345/18/#R11> between day 5 and day 11, but no more than two days after the last injection of study drug, or earlier if thrombosis was clinically suspected. Symptomatic pulmonary embolism was confirmed by a lung scan indicating a high probability of pulmonary embolism, pulmonary angiography, 12 <http://content.nejm.org/cgi/content/full/345/18/#R12> or helical computed tomography or at autopsy. The primary safety outcome was the incidence of major bleeding, which included fatal bleeding; bleeding that was retroperitoneal, intracranial, or intraspinal or that involved any other critical organ; bleeding leading to reoperation; and overt bleeding with a bleeding index of 2 or more. The bleeding index was calculated as the number of units of packed red cells or whole blood transfused plus the hemoglobin values before the bleeding episode minus the hemoglobin values after the episode (in grams per deciliter). Secondary safety outcomes were death, other bleeding, a need for transfusion, thrombocytopenia, and any other adverse event. Efficacy and safety outcomes were adjudicated by a central independent committee whose members were unaware of the treatment assignments and included reviews of all venograms and reports of bleeding and death. Statistical Analysis Assuming an incidence of venous thromboembolism by day 11 of 34 percent in the enoxaparin group and a risk reduction of about 30 percent (i.e., an incidence of 24 percent in the fondaparinux group), 319 patients were needed in each group (for a total of 638 patients) to provide the study with a power of 85 percent. The target number of recruited patients was 912, a number that allowed for failure to obtain primary efficacy data in up to 30 percent of patients. The analysis of the primary efficacy outcome included data on all patients who had received at least one dose of study medication, had undergone the appropriate surgery, and had had an adequate assessment for venous thromboembolism by day 11. The analysis of safety included data on patients who had received at least one dose of study medication. A two-tailed P value of less than 0.05 was considered to indicate statistical significance. The analysis of the primary efficacy outcome was performed with the use of a two-sided Fisher's exact test. Exact 95 percent confidence intervals for the absolute difference between fondaparinux and enoxaparin and the risk ratio were calculated. The treatment effect was also analyzed according to predefined categorical covariates with the use of a logistic-regression model. The study was supervised by a steering committee of 11 people, which included 7 representatives of the sponsors (NV Organon and Sanofi–Synthelabo). The committee designed the study, interpreted the data, and wrote the article. The final statistical analysis was performed by the sponsor. The central adjudication committee and the data monitoring committee operated independently of the sponsor. One planned interim analysis was conducted by an independent statistical center when half the projected patient population had been enrolled, for reestimation of the sample size, since the rate of venous thromboembolism in patients undergoing knee surgery was uncertain. Simulations demonstrated that the predefined procedure did not inflate the type I error. No change in the sample size was found to be necessary, and the study continued as planned. Results Study Populations Between December 1998 and January 2000, 1049 patients were enrolled in 64 centers in North America. Fifteen patients did not receive any study drug, leaving 1034 available for the safety analysis ( Table 1 <http://content.nejm.org/cgi/content/full/345/18/#T1> ); primary efficacy had not been assessed by day 11 in 310 patients. Thus, 724 patients (69.0 percent) were included in the primary efficacy analysis, a percentage in line with other large multicenter trials that used venography after orthopedic surgery. 13 <http://content.nejm.org/cgi/content/full/345/18/#R13> , 14 <http://content.nejm.org/cgi/content/full/345/18/#R14> , 15 <http://content.nejm.org/cgi/content/full/345/18/#R15> , 16 <http://content.nejm.org/cgi/content/full/345/18/#R16> The characteristics of patients excluded from the primary efficacy analysis did not differ from those of patients included in the analysis (data not shown). View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1305/T1> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1305/T1> Table 1. Patients Included in the Analyses and Reasons for Exclusion. Base-line characteristics did not differ significantly between the two groups of patients included in the analysis of safety ( Table 2 <http://content.nejm.org/cgi/content/full/345/18/#T2> ) or primary efficacy (data not shown). Among patients analyzed for primary efficacy, the median time between surgery and the qualifying examination for venous thromboembolism was seven days in both groups; most patients underwent this examination between day 5 and day 11, as planned. The two groups did not differ with regard to the last day of active treatment or the use of concomitant treatments up to day 11 ( Table 3 <http://content.nejm.org/cgi/content/full/345/18/#T3> ). View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1305/T2> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1305/T2> Table 2. Base-Line Characteristics of the Patients. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1305/T3> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1305/T3> Table 3. Treatments Received during the Study Period by Patients Assessed for the Primary Efficacy Outcome. Overall, 514 patients treated with fondaparinux and 511 patients treated with enoxaparin returned for the follow-up visit on day 49. The duration of follow-up was similar in the two groups. During follow-up of patients who were not treated for an acute thromboembolic event, 19.1 percent of patients assigned to fondaparinux (86 of 450) and 20.2 percent of patients assigned to enoxaparin (82 of 406) received prolonged thromboprophylaxis, primarily with a preparation of heparin or a vitamin K antagonist. Incidence of Venous Thromboembolism The incidence of venous thromboembolism by day 11 was 27.8 percent (101 of 363 patients) in the enoxaparin group and 12.5 percent (45 of 361 patients) in the fondaparinux group (reduction in risk, 55.2 percent; 95 percent confidence interval, 36.2 to 70.2 percent; P<0.001) ( Table 4 <http://content.nejm.org/cgi/content/full/345/18/#T4> ). A similar result was found in sensitivity analyses when patients who had had no primary efficacy assessment by day 11 were included in the primary efficacy analysis (data not shown). As compared with enoxaparin, fondaparinux reduced the incidence of proximal deep-vein thrombosis by 54.5 percent (P=0.06) and distal deep-vein thrombosis by 55.9 percent (P<0.001). The incidence of symptomatic venous thromboembolism was low and did not differ significantly between the two groups ( Table 4 <http://content.nejm.org/cgi/content/full/345/18/#T4> ). Overall, the superiority of fondaparinux over enoxaparin with respect to primary efficacy was consistent according to age, sex, body-mass index (the weight in kilograms divided by the square of the height in meters [<30 vs. >=30]), type of anesthesia (general, regional, or both), type of surgery (primary or revision), use or nonuse of cement, and whether or not patients had had previous venous thromboembolism (data not shown). The number of patients treated by participating physicians for a venous thromboembolic event by day 11 was significantly lower in the fondaparinux group (15.1 percent [67 of 443]) than in the enoxaparin group (25.1 percent [111 of 442], P<0.001). View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1305/T4> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1305/T4> Table 4. Incidence of Venous Thromboembolic Events by Day 11. By day 49, the incidence of symptomatic venous thromboembolism did not differ significantly between the fondaparinux group (1.0 percent [5 of 517 patients]) and the enoxaparin group (1.9 percent [10 of 517 patients]). Fatal pulmonary embolism occurred in one patient in each group; nonfatal pulmonary embolism occurred in two patients in the fondaparinux group and four in the enoxaparin group. Safety Outcomes There were no instances of fatal bleeding or bleeding in a critical organ in either treatment group; bleeding requiring reoperation occurred in two patients in the fondaparinux group and one in the enoxaparin group. In all three patients, drainage of a knee effusion was performed. In the fondaparinux group, nine episodes of overt bleeding were associated with a bleeding index of 2 or more; seven occurred at the surgical site, and only three led to discontinuation of study treatment. The total for the primary safety outcome was therefore 11 major bleeding episodes in the fondaparinux group and 1 in the enoxaparin group (P=0.006) ( Table 5 <http://content.nejm.org/cgi/content/full/345/18/#T5> ). The incidence of minor bleeding, a need for transfusion, and other adverse events during treatment or follow-up did not differ significantly between the two groups. Platelet counts lower than 100,000 per cubic millimeter were measured in 14 patients (2.7 percent) in the fondaparinux group as compared with 19 (3.7 percent) in the enoxaparin group (P=0.27). No episode of a decreased platelet count was reported as a serious adverse event in either group. By day 49, two patients in the fondaparinux group (0.4 percent) and three in the enoxaparin group (0.6 percent) had died from causes unrelated to the treatment. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1305/T5> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1305/T5> Table 5. Safety Outcomes. Discussion This study demonstrates that fondaparinux is significantly more effective than enoxaparin in preventing venous thromboembolism after elective major knee surgery. Deep-vein thrombosis has been more difficult to prevent with anticoagulation therapy after knee surgery than after total hip replacement, even with low-molecular-weight heparin, the most effective thromboprophylactic therapy to date. 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> The 27.8 percent incidence of venous thromboembolic events in the enoxaparin group by day 11 is consistent with the incidence of 19.0 to 25.0 percent in other trials of enoxaparin after knee surgery. 16 <http://content.nejm.org/cgi/content/full/345/18/#R16> , 17 <http://content.nejm.org/cgi/content/full/345/18/#R17> , 18 <http://content.nejm.org/cgi/content/full/345/18/#R18> , 19 <http://content.nejm.org/cgi/content/full/345/18/#R19> In our study, the reduction to 12.5 percent in the fondaparinux group is consistent with the results of three other large studies in patients undergoing surgery for hip fracture 8 <http://content.nejm.org/cgi/content/full/345/18/#R8> or elective hip replacement. 9 <http://content.nejm.org/cgi/content/full/345/18/#R9> , 10 <http://content.nejm.org/cgi/content/full/345/18/#R10> The superior efficacy of fondaparinux may be related to its ability to initiate selective inhibition of factor Xa, its predictable linear pharmacokinetics, the choice of dose, and the starting time after surgery. The low incidence of symptomatic events in our study should be interpreted with caution, as it is likely to be lower than would be observed in typical clinical situations. As in other trials of thromboprophylaxis, most of our asymptomatic patients with positive venograms had been receiving an anticoagulant at a therapeutic dose. Moreover, about 20 percent of our patients were receiving prophylaxis after the study treatment period ended. Both of these factors may have prevented symptomatic venous thrombosis. Major bleeding was significantly more frequent in the fondaparinux group (11 patients, including 9 with a bleeding index of 2 or more) than in the enoxaparin group (1 patient). Administration of fondaparinux was continued in six of the nine patients with a bleeding index of 2 or more. Nevertheless, the two groups did not differ significantly with respect to fatal bleeding, bleeding in critical organs, or bleeding leading to reoperation. <http://weeklybriefings.org/feature.asp?strXmlDoc=3451802> Supported by NV Organon and Sanofi–Synthelabo. All authors have served as consultants to NV Organon and Sanofi–Synthelabo. Presented in abstract form at the 42nd Annual Meeting of the American Society of Hematology, San Francisco, December 1–5, 2000 (Blood 2000;9:491A, A2111). * Participants in the study are listed in the Appendix. <http://content.nejm.org/cgi/content/full/345/18/#RFN1> Source Information From the Department of Medicine, Veterans Affairs Boston Healthcare System and Beth Israel Deaconess Medical Center, Boston (K.A.B.); the Department of Orthopedics, Sahlgrenska University Hospital–Östra, Göteborg, Sweden (B.I.E.); the Department of Orthopedics, Hillerød University, Hillerød, Denmark (M.R.L.); and the Department of Medicine, Hamilton Health Sciences Corporation–General Division, Hamilton, Ont., Canada (A.G.G.T). Address reprint requests to Dr. Bauer at the Beth Israel Deaconess Medical Center, 330 Brookline Ave., Boston, MA 02215, or at [log in to unmask] <mailto:[log in to unmask]> . References 1. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest 2001;119:Suppl:132S-175S. [Full Text] <http://content.nejm.org/cgi/ijlink?linkType=FULL&journalCode=chest&resid=11 9/1_suppl/132S> 2. Nicolaides AN. Prevention of venous thromboembolism: international consensus statement: guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1-37. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=11342993&link_type=MED> 3. Lormeau JC, Herault JP. The effect of the synthetic pentasaccharide SR 90107/ORG31540 on thrombin generation ex vivo is uniquely due to ATIII-mediated neutralization of factor Xa. Thromb Haemost 1995;74:1474-1477. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=8772223&link_type=MED> 4. Walenga JM, Bara L, Petitou M, Samama MM, Fareed J, Choay J. The inhibition of the generation of thrombin and the antithrombotic effect of a pentasaccharide with sole anti-factor Xa activity. Thromb Res 1988;51:23-33. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=3166242&link_type=MED> 5. Petitou M, Lormeau J-C, Choay J. Chemical synthesis of glycosaminoglycans: new approaches to antithrombotic drugs. Nature 1991;350:Suppl:30-33. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2017262&link_type=MED> 6. van Boeckel CAA, Petitou M. The unique antithrombin III binding domain of heparin: a lead to new synthetic antithrombotics. Angew Chem Int Ed Engl 1993;32:1671-1690. 7. Turpie AGG, Gallus AS, Hoek JA. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 2001;344:619-625. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=344 /9/619> 8. Eriksson BI, Bauer KA, Lassen MR, Turpie AGG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. N Engl J Med 2001;345:1298-1304. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=345 /18/1298> 9. Lassen MR. The EPHESUS Study: comparison of the first synthetic factor Xa inhibitor with low molecular weight heparin (LMWH) in the prevention of venous thromboembolism (VTE) after elective hip replacement surgery. Blood 2000;96:490a-490a.abstract 10. Turpie G. The PENTATHLON 2000 Study: comparison of the first synthetic factor Xa inhibitor with low molecular weight heparin in the prevention of venous thromboembolism (VTE) after elective hip replacement surgery. Blood 2000;96:491a-491a.abstract 11. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972;104:134-144. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=5008903&link_type=MED> 12. The PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA 1990;263:2753-2759. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=2332918&link_type=MED> 13. Eriksson BI, Ekman S, Kälebo P, Zachrisson B, Bach D, Close P. Prevention of deep-vein thrombosis after total hip replacement: direct thrombin inhibition with recombinant hirudin, CGP 39393. Lancet 1996;347:635-639. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=8596376&link_type=MED> 14. Hull RD, Pineo GF, Francis C, et al. Low-molecular-weight heparin prophylaxis using dalteparin in close proximity to surgery vs warfarin in hip arthroplasty patients: a double-blind, randomized comparison. Arch Intern Med 2000;160:2199-2207. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=10904464&link_type=MED> 15. Heit JA, Berkowitz SD, Bona R, et al. Efficacy and safety of low molecular weight heparin (ardeparin sodium) compared to warfarin for the prevention of venous thromboembolism after total knee replacement surgery: a double-blind, dose-ranging study. Thromb Haemost 1997;77:32-38. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=9031445&link_type=MED> 16. Leclerc JR, Geerts WH, Desjardins L, et al. Prevention of venous thromboembolism after knee arthroplasty: a randomized, double-blind trial comparing enoxaparin with warfarin. Ann Intern Med 1996;124:619-626. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=8607589&link_type=MED> 17. Leclerc JR, Geerts WH, Desjardins L, et al. Prevention of deep vein thrombosis after major knee surgery -- a randomized, double-blind trial comparing a low molecular weight heparin fragment (enoxaparin) to placebo. Thromb Haemost 1992;67:417-423. [Medline] <http://content.nejm.org/cgi/external_ref?access_num=1321509&link_type=MED> 18. Spiro TE, Fitzgerald RH, Trowbridge AA, et al. Enoxaparin, a low molecular weight heparin and warfarin for the prevention of venous thromboembolic disease after elective knee replacement surgery. Blood 1994;84:Suppl:246a-246a.abstract 19. Spiro TE, Colwell CW, Bona RD, et al. A clinical trial comparing the efficacy and safety of enoxaparin, a low molecular weight heparin and unfractionated heparin for the prevention of deep venous thrombosis after elective knee replacement surgery. Blood 1993;:410a-410a.abstract Choosing a Parenteral Anticoagulant Agent For decades, if a patient required a parenteral anticoagulant agent, the choice was simple: unfractionated heparin was the only such agent available. Unfractionated heparin has a long track record of effectiveness in both the treatment of and prophylaxis against arterial and venous thromboembolic disease. However, because of differences among batches of heparin and problems related to the bioavailability of the drug, monitoring of the anticoagulant effect of heparin has been not only necessary but also problematic. In addition, heparin-induced bleeding and thrombocytopenia can threaten life and limb. Moreover, there have always been problems with the use of unfractionated heparin to prevent thrombosis in high-risk patients. In that group, it is clear that fixed-dose prophylaxis is inadequate, yet treatment with adjusted-dose heparin or adjusted-dose warfarin poses the additional problem of monitoring therapy. Since 1987, when the first low-molecular-weight heparin was approved for use in the United States, there has been an explosion in the number of available parenteral anticoagulant drugs. There are currently four low-molecular-weight heparins, one heparinoid, two hirudin derivatives, and one direct thrombin inhibitor approved for use, all of which have defined roles in patients requiring anticoagulation. To this plethora of agents is now added fondaparinux, a synthetic sulfated pentasaccharide that was derived from the activated factor X (factor Xa)–binding moiety of unfractionated heparin ( Table 1 <http://content.nejm.org/cgi/content/full/345/18/#T1> ). 1 <http://content.nejm.org/cgi/content/full/345/18/#R1> In this issue of the Journal, two groups report that once-daily treatment with 2.5 mg of fondaparinux initiated in the early postoperative period is more effective than enoxaparin, a low-molecular-weight heparin, in preventing venous thromboembolism after hip 2 <http://content.nejm.org/cgi/content/full/345/18/#R2> or knee 3 <http://content.nejm.org/cgi/content/full/345/18/#R3> surgery and does not increase the risk of bleeding. If we assume that there will be no problems with the Food and Drug Administration approval process, fondaparinux will join the many other agents now available to prevent and treat thromboembolic disease. The question is which agent should be used for a particular indication. View this table: [in this window] <http://content.nejm.org/cgi/content/full/345/18/1340/T1> [in a new window] <http://content.nejm.org/cgi/content-nw/full/345/18/1340/T1> Table 1. Indications for and Contraindications to Parenteral Anticoagulant Agents. The low-molecular-weight heparins are chemical or physical fractions of unfractionated heparin. All are excreted by the kidney, and all have greater bioavailability than unfractionated heparin. For this reason, their anticoagulant activity is much more predictable than that of unfractionated heparin, making administration without monitoring a reality in the vast majority of patients. They are at least as effective as adjusted-dose warfarin in preventing perioperative venous thromboembolism in high-risk patients, especially when administered preoperatively. They are also as effective as fixed-dose unfractionated heparin for preventing venous thromboembolism in patients at moderate risk for deep venous thrombosis. Low-molecular-weight heparins are less likely to cause thrombocytopenia than unfractionated heparin, since they bind poorly to platelet surfaces. For these reasons, I recommend that the low-molecular-weight heparins replace unfractionated heparin altogether for prophylaxis in patients at moderate risk for venous thromboembolic disease (for example, after a myocardial infarction or abdominal surgery). The value of low-molecular-weight heparins for perioperative prophylaxis in high-risk patients is limited by the need to administer the drug preoperatively to obtain maximal benefit. The move from general to regional anesthesia for surgical procedures also limits the use of the low-molecular-weight heparins. All these drugs (and danaparoid, a heparinoid) carry a warning against their use in patients undergoing regional anesthesia, because of the risk of epidural hematomas. This potential complication has severely curtailed the perioperative use of low-molecular-weight heparins. Fondaparinux may find its initial niche in patients undergoing regional anesthesia, since it seems to be at least as effective as the best current treatment, even when it is given only postoperatively. At higher doses, the low-molecular-weight heparins enoxaparin, dalteparin, and tinzaparin are all effective for the treatment of venous thromboembolism. The relatively long half-life of these drugs and the fact that in most cases monitoring is unnecessary simplify inpatient care and make outpatient treatment of uncomplicated venous thromboembolic disease a reality. The true potential of low-molecular-weight heparins will be seen when we develop better systems for the outpatient management of thromboembolic disease. Enoxaparin and dalteparin have now been approved for the treatment of unstable angina. Here, the differences in efficacy and safety between unfractionated heparin and low-molecular-weight heparins are modest, and the long half-life of the newer drugs can be problematic. Most procedure-oriented cardiology departments have shied away from these agents, since unfractionated heparin is nearly as effective and its anticoagulant effects can be neutralized should an urgent procedure become necessary. Bivalirudin, a direct inhibitor of thrombin, is also approved for patients with unstable angina who are undergoing coronary angioplasty, but other uses of the drug are still being determined. The low-molecular-weight heparins are contraindicated in patients with type II heparin-induced thrombocytopenia because the drugs cross-react with antiheparin antibodies. Since we now recognize heparin-induced thrombocytopenia as a prothrombotic rather than an antithrombotic condition, the need for effective anticoagulant drugs to treat the disorder has become much clearer. The three agents currently approved for heparin-induced thrombocytopenia are danaparoid, a heparinoid compound with little cross-reactivity with antiheparin antibodies; lepirudin, a hirudin derivative; and argatroban, a small-molecule direct thrombin inhibitor. Danaparoid is the only agent approved for prophylaxis against venous thromboembolism in high-risk patients with heparin-induced thrombocytopenia. It is also often used for treating thrombosis in this situation. Danaparoid has a prolonged half-life, and its effects are only partially reversible — characteristics that make its use problematic in patients in need of procedures. Since danaparoid is derived from heparin, there is also the theoretical risk of cross-reactivity with antiheparin antibodies (though this has not proved to be a significant clinical problem). Lepirudin and argatroban are both direct thrombin inhibitors that exert their anticoagulant effect independently of antithrombin III. They are not derived from heparin, so there is no chance of cross-reactivity with antiheparin antibodies. Both are administered as intravenous infusions, and both have a very short half-life (30 minutes to 1 hour). Both are monitored with the use of the activated partial-thromboplastin time, in much the same way as heparin is traditionally monitored. Lepirudin is excreted by the kidney in patients with renal dysfunction; therefore, substantial dose reduction is necessary. Argatroban is metabolized in the liver, so substantial dose reduction is required in patients with liver disease. Neither agent has reversible effects, and both interact with warfarin. Lepirudin and argatroban seem to be equally effective in the management of thrombosis due to heparin-induced thrombocytopenia. Since many patients with heparin-induced thrombocytopenia have evidence of multiorgan dysfunction, the choice of an agent often depends on other conditions that the patient might have. Is there ever a situation in which the use of unfractionated heparin is appropriate? In the operating room, in the intensive care unit, and in patients with renal failure (and in the absence of heparin-induced thrombocytopenia), unfractionated heparin is still the agent of choice, given its short half-life, easy reversibility, and extrarenal metabolism. However, there are clear advantages to the newer agents, and we should use them appropriately, especially in the area of prophylaxis against venous thromboembolism. David L. Diuguid, M.D. College of Physicians and Surgeons of Columbia University New York, NY 10032 References 1. Turpie AGG, Gallus AS, Hoek JA. A synthetic pentasaccharide for the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 2001;344:619-625. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=344 /9/619> 2. Eriksson BI, Bauer KA, Lassen MR, Turpie AGG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. N Engl J Med 2001;345:1298-1304. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=345 /18/1298> 3. Bauer KA, Eriksson BI, Lassen MR, Turpie AGG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. N Engl J Med 2001;345:1305-1310. [Abstract/Full Text] <http://content.nejm.org/cgi/ijlink?linkType=ABST&journalCode=nejm&resid=345 /18/1305> Edward E. Rylander, M.D. Diplomat American Board of Family Practice. Diplomat American Board of Palliative Medicine.