Evaluation of the safety and efficacy of enoxaparin and warfarin for prevention of deep vein thrombosis after total knee arthroplasty

The Journal of Arthroplasty Vol. 15 No. 2 2000

Evaluation of the Safety and Efficacy of Enoxaparin and Warfarin for Prevention of D e e p Vein Thrombosis After Total K n e e Arthroplasty Steven

H. Stern, MD, Richard

L. W i x s o n ,

MD, and Daryl O'Connor,

MD

Abstract: Of 263 patients who underwent total knee arthroplasty, 122 received adjusted low-dose warfarin and 141 received enoxaparin as deep vein thrombosis (DVT) prophylaxis. Three patients in the warfarin group and 3 in the enoxaparin group developed ultrasound-detectable DVT (P > .05). Although the amount of perioperative blood transfused was equivalent in both groups, the overall hemoglobin drop was greater (P < .005) in the enoxaparin group (2.9 g/dL) as compared with the warfarin group (2.3 g/dL). Five patients (4.6%) in the warfarin group and 16 (11.3%) in the enoxaparin group had bleeding complications (P < .05). Our data support earlier published reports suggesting that reductions, if any, in the incidence of DVT associated with enoxaparin are offset by a significant increase in bleeding complications as compared with adjusted-dose warfarin. We continue to use adjusted-dose warfarin as primary thromboembolic prophylaxis after total knee arthroplasty. Key words: enoxaparin, knee arthroplasty, deep vein thrombosis prophylaxis, warfarin.

A n e w form of deep v e n o u s thrombosis (DVT) prophylaxis became available in the United States w h e n the U.S. Food and Drug Administration a p p r o v e d the use of enoxaparin, a l o w - m o l e c u l a r weight heparin, for use after total joint arthroplasty [1-3]. L o w - m o l e c u l a r weight heparin offered several conceptual advantages o v e r o t h e r pharmacologic DVT prophylaxis methods. Theoretically, l o w - m o l e c u l a r weight heparin does not require routine m o n i t o r ing, can be administered in identical doses for all patients, and requires a shorter period of anticoagulation t h e r a p y [3]. N u m e r o u s reports h a v e been published on the results of l o w - m o l e c u l a r weight heparin w h e n used as a DVT prophylaxis agent after

total knee arthroplasty (TKA) [4-10]. W h e n enoxaparin was used in routine clinical practice at our institution, the surgeons noted increased w o u n d drainage, bleeding, and leg swelling. These increased bleeding complications were believed to be significant and led to a return to use of adjusteddose warfarin for pharmacologic DVT prophylaxis after TKA. We hypothesized that although e n o x a p a rin was as efficacious as adjusted-dose warfarin in preventing thrombosis formation, its use also resulted in an increased incidence of bleeding.

Materials and Methods Patient Population

From the Department of Orthopaedic Surgery, Northwestern University Medical School, Chicago, Illinois. Submitted August 26, 1998; accepted May 13, 1999. No benefits or funds were received in support of this study. Reprint requests: Steven H. Stern, MD, Suite 450, 676 North St. Clair, Chicago, IL 60611. Copyright © 2000 by Churchill Livingstone® 0883-5403/00/1502-0003510.00/0

F r o m 1990 to 1996, 271 patients u n d e r w e n t consecutive index TKAs u n d e r the care of 2 senior surgeons (R.L.W, and S.H.S.). All procedures were done at a tertiary care teaching institution. For the purpose of this retrospective study, inclusion criteria w e r e p r i m a r y TKA use of either e n o x a p a r i n or

154

The Journal of Arthroplasty Vol. 15 No. 2 February 2000

adjusted-dose warfarin as thromboembolic prophylaxis. Eight patients were excluded because of use of another method of prophylaxis. Of 263 patients included, 141 received enoxaparin, and 122 received adjusted-dose warfarin. These 2 groups were not randomized. The preferred prophylaxis protocol at our institution was adjusted-dose warfarin during 1990 to 1992 and enoxaparin from 1993 to 1996. There was no statistical significant difference in the demographics between the 2 groups (Table 1). More patients received regional anesthesia in the enoxaparin group than in the adjusted-dose warfarin group (26% vs 8%). This difference reflected the changing practice of our anesthesia department.

treated with antiembolism stockings and compression foot pumps.

Surgical Techniques and Postoperative Care All components in all knees were implanted with the use of bone-cement. The institution's routine postoperative protocol for that time period included the use of suction drains. Rehabilitation included use of a continuous motion machine supplemented with standard physical therapy. Ice packs were routinely used to control swelling, although no formal cryotherapy machines were employed.

Thromboembolic Prophylaxis

Outcome Measures

Adjusted-dose warfarin prophylaxis consisted of an initial 5-mg dose, followed by an adjustedZdose regimen with the goal being to maintain the prothrombin time at approximately 1.5 to 2.0 times control. Adjusted-dose warfarin anticoagulation prophylaxis was continued for 4 to 6 weeks postoperatively, with biweekly blood checks and dose adjustment to maintain the appropriate prothrombin time level. Patients receiving fixed-dose enoxaparin received 30 mg subcutaneously twice a day. The average time interval between the end of surgery and first enoxaparin dose was 16 hours (standard deviation [SDI, 9 hours). Enoxaparin prophylaxis was routinely continued fol" 2 weeks. No routine laboratory monitoring was used in the patients receiving enoxaparin therapy. No further pharmacologic methods were used after the enoxaparin or the adjusted-dose warfarin course was completed. All patients successfully completed the recommended length of prophylaxis. In addition to the pharmacologic anticoagulation listed, patients were routinely

Clinical efficacy was determined by the frequency of documented DVT or symptomatic pulmonary embolism. At the time under study in this report, routine duplex Doppler ultrasound studies were performed as part of our institution's protocol on the 4th or 5th postoperative day. All studies were performed by 1 of several specially trained Doppler technicians at our institution. Of the patients analyzed in this report, 95% (249 of 263 patients) successfully completed the postoperative venous Doppler ultrasound study. Patients were considered to have developed DVT if the Doppler ultrasound examination revealed significant noncompressible vein segments [ 11 ]. Clinical safety was determined by the amount of blood loss during the hospitalization. Clinically significant bleeding episodes were defined as any postoperative bleeding event documented in the patient's medical record. Major bleeding was defined as i) any documented overt bleeding event that decreased hemoglobin levels by 2 g/dL in 1 24-hour period, excluding the first 24 hours postoperatively, or ii) hemarthrosis requiring discontinuation of prophylaxis for at least 24 hours. Minor bleeds were events documented in the medical record (ie, wound drainage or bleeding at other sites) that did not reach the criteria for major bleeds. A mean blood loss index was calculated for all patients to obtain a more objective measure of bleeding. The blood loss index by convention is defined as the sum of the difference between the patient's preoperative and discharge hemoglobin levels added to the number of units of blood transfused [10].

Table 1. Patient Demographics

Warfarin No. patients Age (y) Sex Female Male Weight (Ib) Height (in.) Diagnosis Osteoarthritis (%) Anesthesia General Epidural Spinal

Lovenox*

Total

122 67 _+ l0

141 70 +_ 9

263 69 +_ 9

73 47 177 _+ 43 66 - 4

83 58 174 +_ 40 66 - 4

158 105 176 _+ 41 66 +- 4

94

93

93

I 13 4 5

105 33 3

218 37 8

*Trade name for low-molecular weight heparin.

Statistical Analysis The rates of DVT and wound or bleeding complications in the 2 groups were compared using the

Enoxaparin and Warfarin After TKA

chi-square test with Fisher's exact test. Blood loss, time intervals, and other baseline patient demographic data were compared using Student's t-test for independent samples.

Results Incidence of Deep Vein Thrombosis Objective testing (venous ultrasound) was available in 93% of patients (113 of 122) in the adjusteddose warfarin group and 96% of patients (136 of 141) in the enoxaparin group. Venous ultrasound confirmed the presence of DVT in 3 of 113 patients (2.7%) who received adjusted-dose warfarin and 3 of 136 patients (2.2%) who received enoxaparin (Table 2). Proximal DVT occurred in 1 patient in the adjusted-dose warfarin group and 2 patients in the enoxaparin group. Although bilateral venous ultrasonography was performed on all patients, all of the detected occurrences of DVT were isolated to the operative extremity.

Incidence of Bleeding Complications Clinically significant major bleeds were seen in 2 patients in the enoxaparin group as compared with 1 patient in the adjusted-dose warfarin group. Minor bleeding complications were noted in 14 patients in the enoxaparin group as compared with 4 in the adjusted-dose warfarin group. The overall bleeding rate was statistically higher in the patients receiving enoxaparin prophylaxis (11.3% vs 4.1%) compared with patients receiving adjusted-dose war-

Table 2. Results Warfarin No. patients No. Doppler No. DVT (%)

122 113 3 (2.7%)

Lovenox* 141 136 3 (2.2%)

Total 263 249 6 (2.4%)

Bleeding Minor Major Total Blood loss index Preoperative Hgb Discharge Hgb Hgb drop Units transfused (no.) Blood loss index

4 1 5 (4.1%)

14 2 16 (11.3%)

18 3 21 (8.0%)-I-

12.6 -+ 1.4 10.4 _+ 1.2 2.3 - 1.4

13.0 _+ 1.3 10.2 -+ 1.2 2.9 _+ 1.6

12.8 +- 1.4 10.3 _ 1.2 2.6 - 1.6:1:

2.1 - . 9 4.1 -+ 1.7

2.1 +-.8 4.7 +- 1.7

2.1 -+ .9 4.4 _+ 1.7§

*Trade n a m e for low-molecular weight heparin. Manufactured by Aventis Pharmaceuticals Inc., Collegeville, PA. t P = .039. {P = .002. § P = .015. DVT, deep venous thrombosis; Hgb, hemoglobin.

•

Stern et al.

155

farin anticoagulation (P < .05) (Table 2). Of the 64 patients who received enoxaparin within the first 12 hours of surgery, 6 manifested episodes of overt bleeding (9%) as compared with 10 overt bleeds in the 77 patients (13 %) whose initial dose of enoxaparin was > 12 hours after surgery (not significant). The average reduction in hemoglobin levels from presurgical levels to discharge studies was 2.9 g/dL (SD, 1.6) in the enoxaparin group and 2.3 g/dL (SD, 1.4) in the adjusted-dose warfarin group. The amount of perioperative blood transfused was equivalent in both treatment groups (average 2.1 units in both groups). The mean blood loss index was 4.7 (SD, 1.7) in the enoxaparin group as compared with 4.1 (SD, 1.7) in the adjusted-dose warfarin group (P < .02) (Table 2). The small number of patients with major bleeding episodes required slightly longer hospital stays, but the long-term clinical outcome in these patients was unaffected. Although the minor bleeding episodes were concerning, they did not result in further surgery, longer hospital stays, or any obvious alteration in clinical outcomes. No patient required hospital readmission for bleeding or thromboembolic complications.

Discussion Multiple studies have examined the efficacy of low-molecular weight heparins by comparing them with placebo [6,7] or intravenous heparin [6,12,13]. Fewer studies have directly compared low-molecular weight heparins with adjusted-dose warfarin for use after TKA. In general, these studies have reported that low-molecular weight heparins were efficacious in reducing the incidence of DVP, but this may be at the expense of increased bleeding or wound complications [5,10,14,15]. Leclerc et al. [8] compared the effectiveness and safety of fixed-dose enoxaparin and adjusted-dose warfarin after TKA. These authors found 51.7% (109 of 211) of the adjusted-dose warfarin recipients had DVT as compared with 36.9% (76 of 206) of the enoxaparin patients (P--.003). The incidence of the more significant proximal clots (approximately 10%) was almost identical in the 2 groups. The incidence of major and minor bleeding was 26.6% in the adjusted-dose warfarin group as compared with 30.1% in the enoxaparin group. Hull et al. [5] reported on another large series of patients comparing the efficacy of adjusted-dose warfarin and fixed-dose, low-molecular weight heparin (Logiparin, Novo Nordisk, Bagsvaerd, Denmark). In this double-blind, randomized report, the incidence of DVP diagnosed by venography was

156

The Journal of Arthroplasty Vol. 15 No. 2 February 2000

54.9% (12.3% for proximal clots) in the adjusteddose warfarin group as compared with 45.0% (7.8% for proximal clots) in the low-molecular weight heparin group. Patients receiving the low-molecular weight heparin showed a slight increased propensity for bleeds with 2.4% of the patients manifesting bleeds on adjusted-dose warfarin prophylaxis (0.9% major bleeds) as compared with 4.4% (2.8% major bleeds) of the low-molecular weight heparin recipients. Wound hematomas were more common in the low-molecular weight heparin group (8.8%) compared with the adjusted-dose warfarin group (5.9%). Our data showed no significant difference in the rate of DVT between the enoxaparin (2.2%) and adjusted-dose warfarin (2.7%) groups. The overall rates were considerably lower than in previous studies, however, suggesting either superior anticoagulation or inferior detection methods at our center. The relatively low rate of DVT may be related to the use of Doppler ultrasound as the detection test mechanism in our series. Many other reports have used venography, which has been shown to be more sensitive than venous ultrasound, as the method to detect thrombosis [11,16-20]. The incidence of proximal DVT was not significantly different between the 2 groups, similar to previous studies that showed no significant difference [5,8,10,14]. The results in the present study showed a significant increased incidence of bleeding and wound complications in the patientsreceiving enoxaparin. Although many of the differences in the 2 groups were in the increased incidence of minor bleeds, the results did confirm the surgeon's clinical perception of increased hemorrhage with enoxaparin. The blood loss index was significantly greater in patients receiving enoxaparin. The index has the advantage of attempting to quantify the blood loss associated with surgery. The index does not depend on the subjective views or bias of an observer, which can be especially problematic in examining wounds for drainage or hematoma or in differentiating a minor from a major bleed. Statistically significant differences in the blood loss index (4.7 vs 4.1) between the enoxaparin and adjusted-dose warfarin groups represent a clear quantified measure of increased hemorrhage in the low-molecular weight heparin recipients. Our finding was similar to what was reported in another study comparing the blood loss index in patients receiving adjusted-dose warfarin or ardeparin (RD Heparin) [10]. The present report is retrospective with the known limitations of this method. Prospective reports, however, theoretically have an increased risk of a sentinel effect, in which the study itself causes increased

vigilance. In addition, one of the hypothetical benefits of enoxaparin is its ease of use (ie, no monitoring needed, identical doses for all patients) as compared with adjusted-dose warfarin. One would expect the advantages with enoxaparin to be even more pronounced w hen used in the routine clinical setting. Conversely, adjusted-dose warfarin, which requires monitoring, should benefit from the increased vigilance present in a controlled study. Despite this theoretical advantage, the patients receiving enoxaparin had a similar incidence of DVT and an increased incidence of hemorrhage in this study. Limitations of the present report include the use of Doppler venous ultrasonography for the detection of DVT. Venography is associated with complications such as pain, allergic reactions, and contrastrelated inflammatory responses and is not considered cost-effective for routine screening [21-23]. Doppler ultrasound was routinely used at our institution during the period under review. Ultrasound imaging, however, has been shown to have a lower sensitivity than venography in detecting asymptomatic postoperative DVT [11,16-20]. This previously demonstrated lower sensitivity paired with the low incidence of DVT in our study suggests the possibility of missed asymptomatic DVT. Alternatively, the low incidence of DVT may also be related to the effectiveness of the mechanical prophylaxis measures (foot pumps and thromboembolic disease stockings) employed. The absolute prevalence of DVT in the 2 treatment groups should not be regarded as accurate estimates of the true DVT incidence. Additionally, even though the incidence of thrombosis was similar in the 2 treatment groups, this does not necessarily indicate any difference in efficacy between the 2 drugs. The sample size in our report is not large enough to provide sufficient statistical power to say definitively that there is no difference in efficacy between enoxaparin and adjusted-dose warfarin in terms of DVT prophylaxis [2]. Another limitation of our report is its hospital endpoint with regards to monitoring for postoperative complications. Studies have shown that DVTs continue to occur 3 weeks after surgery [24-29], and bleeding and wound complications can occur throughout the postoperative period. The incidence of DVTs as well as wound and bleeding complications is most likely underestimated. Similarly, the use of a retrospective chart review would also tend to underestimate the true extent of bleeding complications. Our study supports the recommendation of the fifth American College of Chest Physicians (ACCP). Consensus Conference on Antithrombotic Therapy

Enoxaparin and Warfarin AfterTKA

that postoperative adjusted-dose warfarin is n o w a r e c o m m e n d e d prophylaxis regimen in patients undergoing TKA surgery [30]. This current r e c o m m e n dation is in contrast to the fourth ACCP Consensus Conference on Antithrombotic Therapy, w h i c h did not include adjusted-dose warfarin as a recomm e n d e d prophylaxis regimen [1]. Our data as well as multiple prior reports have revealed a trend toward more bleeding complications with lowm o l e c u l a r - w e i g h t heparin. Because optimizing w o u n d healing and minimizing w o u n d bleeding and drainage are of utmost importance in TKA, any medication that compromises the w o u l d envelope by subjecting it to increased h e m o r r h a g e after TKA must be used with caution. We recognize that m a n y prophylaxis regimens are effective and safe in preventing t h r o m b o e m b o l i s m after TKA. Based on data from this study and others, we continue to use adjusted-dose warfarin as their primary prophylaxis in patients u n d e r g o i n g TKA.

Summary Our data showed that enoxaparin, administered twice daily after TKA, appears to offer comparable efficacy to adjusted-dose warfarin in terms of ultrasound-detectable DVT prophylaxis. Enoxaparin appeared to be associated with significantly more bleeding complications w h e n compared with adjusted-dose warfarin, however. For this reason, we continue to advocate the use of adjusted-dose warfarin for routine DVT prophylaxis after TKA.

References 1. Clagett GP, Anderson FA, Heir J, et al: Prevention of venous thromboembolism: fourth ACCP consensus conference on antithrombotic therapy. Chest 108: 312S, 1995 2. Green D, Hirsh J, Heir J, et al: Low molecular weight heparin: a critical analysis of clinical trials. Pharmacol Rev 46:89, 1994 3. Wolf H: Low-molecular-weight heparin. Med Clin North Am 78:33, 1994 4. Hamulyak K, Lensing AWA, van der Meet J, et al: Subcutaneous low-molecular weight heparin or oral anticoagulants for the prevention of deep-vein thrombosis in elective hip and knee replacement. Thromb Haemost 74:1428, 1994 5. Hull R, Raskob G, Pineo G, et ah A comparison of subcutaneous low-molecular weight heparin with warfarin sodium for prophylaxis against deep vein thrombosis after hip or knee implantation. N Engl J Med 329:1370, 1993 6. Jorgensen LN, Wille-Jorgensen P, Hauch O: Prophylaxis of postoperative thromboembolism with low molecular weight heparins. Br J Surg 80:689, 1993

•

Stern et al.

157

7. 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 67:417, 1992 8. Leclerc JR, Geerts WH, Desjardins L, et al: Prevention of venous thromboembolism after knee arthroplasty. Ann Intern Med 124:619, 1996 9. Lieberman JR, Geerts WH: Prevention of venous thromboembolism after total hip and knee arthroplasty. J Bone Joint Surg Am 76:1239, 1994 10. RD Heparin Arthroplasty Group: RD Heparin compared with warfarinJor prevention of venous thromboembolic disease following total hip or knee arthroplasty. J Bone Joint Surg Am 76:11745, 1994 11. Kraay MJ, Goldberg VM, Herbener TE: Vascular ultrasonography for deep venous thrombosis after total knee arthroplasty. Clin Orthop 286:18, 1993 12. Anderson DR, O'Brien BJ, Levine MN, et al: Efficacy and cost of low-molecular weight heparin compared with standard heparin for the prevention of deep vein thrombosis after total hip arthroplasty. Ann Intern Med 119:1105, 1993 13. Levine MN, Hirsh J, Gent M, et al: Prevention of deep vein thrombosis after elective hip surgery: a randomized trial comparing low molecular weight heparin with standard unfractionated heparin. Ann Intern Med 114:545, 1991 14. Heir J, Berkowitz S, Bona R, et al: Efficacy and safety of normiflo (a LMWH) for prevention of venous thromboembolism following total knee replacement: a double-blind, dose-ranging study (abstract). Thromb Haemost 73:978A, 1995 15. 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 (abstract). Blood 84:246A, 1994 16. Borris LC, Christiansen HM, Lassen MR, et al, the Venous Thrombosis Group: Comparison of real-time B-mode ultrasonography and bilateral ascending phlebography for detection of postoperative deep vein thrombosis following elective hip surgery. Thromb Haemost 61:363, 1989 17. Davidson BL, Elliott CG, Lensing AW, RD Heparin Arthroplasty Group: Low accuracy of color Doppler ultrasound in the detection of proximal leg vein thrombosis in asymptomatic high-risk patients. Ann Intern Med 117:735, 1992 18. Ginsberg JS, Caco CC, BrilI-Edwards PA, et al: Venous thrombosis in patients who have undergone major hip or knee surgery: Detection with compression US and impedance plethysmography. Radiology 181: 651, 1991 19. Heijboer H, ten Cate JW, Bullet HR: Diagnosis of venous thrombosis. Semin Thromb Hemost 17:259, 1991 20. Heijboer H, Bullet HR, Lensing AW, et al: A comparison of real-time compression ultrasonography with impedence plethysmography for the diagnosis of

158

21. 22.

23.

24.

25.

The Journal of Arthroplasty Vol. 15 No. 2 February 2000 deep-vein thrombosis in symptomatic outpatients. N EnglJ Med 329:1365, 1993 Albrechtsson U, Olsson CG: Thrombotic side effects of lower-limb phlebography. Lancet 1:723, 1976 Bettmann MA, Paulin S: Leg phlebography: the incidence, nature, and modification of undesirable side effects. Radiology 122:101, 1977 Hoek JA, Lensing AW, ten Care JW, et al: The clinical utility of objective diagnostic tests for diagnosing deep vein thrombosis of the leg. Br J Clin Pract 65(suppl): 26, 1989 Arceles JI, Caprini JA, Traverso CI: Venous thromboembolism after hospital discharge. Semin Thromb Hemost 19(suppl 1):142, 1993 Fitzgerald RH: Post-discharge prevention of deep vein thrombosis following total joint replacement. Orthopedics (suppl)19:15, 1996

26. Johnson R, Green JR, Charnley J: Pulmonary embolism and its prophylaxis following Charnley total hip replacement. Clin Orthop 127:123, 1977 27. Kakkar VV, Howe CT0 Flanc C, et al: Natural history of postoperative deep-vein fllrombosis. Lancet 2:230, 1969 28. Trowbridge A, Boese CK, Woodruff B, et al: Incidence of posthospitalization proximal deep venous thrombosis after total hip arthroplasty: a pilot study. Clin Orthop 299:203, 1994 29. Turpie AGG: Deep vein thrombosis prophylaxis in the outpatient setting: preventing complications following hospital discharge. Orthopedics (suppl) 18:15, 1995 30. Clagett GP, Anderson FA, Geerts W, et al: Prevention of venous thromboembolism: Fifth ACCP consensus conference on amithrombotic therapy. Chest 114: 531S, 1998