- Email: [email protected]
Warfarin for Thromboprophylaxis Following Total Joint Arthroplasty
The Journal of Arthroplasty 28 (2013) 1251–1253
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Warfarin for Thromboprophylaxis Following Total Joint Arthroplasty Are Patients Safely Anti-Coagulated? Michael Aynardi, MD, P. Benedict Brown, BS, Zachary Post, MD, Fabio Orozco, MD, Alvin Ong, MD The Rothman Institute at Thomas Jefferson University Hospital, Philadelphia, Pennsylvania
a r t i c l e
i n f o
Article history: Received 11 September 2012 Accepted 18 December 2012 Keywords: joint replacement warfarin International Normalized Ratio (INR) thromboprophylaxis complications
a b s t r a c t The purpose of this study was to investigate whether any correlation exists between INR level at discharge and postoperative complications or readmission rates. From 2010–2011, INR levels on discharge, complications and readmissions within 30 days were recorded on 441 patients undergoing joint arthroplasty. Eighty percent (352 of 441) patients had a subtherapeutic INR level at discharge. The overall complication rate was 15% with an 8.6% readmission rate. A supratherapeutic INR level at discharge was associated with both higher readmission rate as well as increased number of complications (P b 0.048). Most patients taking warfarin are nontherapeutic at the time of discharge; notably, a supratherapeutic INR places patients at risk for increased complications and readmissions rates following surgery. © 2013 Elsevier Inc. All rights reserved.
Venous thromboembolism (VTE) is one of the most common complications following total joint arthroplasty [1]. While multimodal thromboprophylaxis has lowered the incidence of VTE, no single entity has emerged as the ideal method of prophylaxis [2,3]. Additionally, anticoagulation has its own set of complications including increased risk of bleeding, difficulty with wound healing, and even infection [4–6]. Discrepancy between currently accepted clinical guidelines has caused resurgence in the debate for the ideal method of thromboprophylaxis following total joint arthroplasty [2,3,7]. Traditionally, warfarin with an International Normalized Ratio (INR) goal of 2.0–2.5 has been widely accepted as an appropriate method of VTE prophylaxis; however, accelerated rehabilitation programs are leading to decreased lengths of hospital stay which have increased the probability of nontherapeutic INR levels at the time of discharge [8,9]. As a result, it is crucial that both the safety and efficacy of warfarin be reevaluated to ensure patients are being properly anticoagulated after major joint arthroplasty. The purpose of this study was to investigate whether any correlation exists between INR level at the time of discharge and postoperative complications or readmission rates following total knee and hip arthroplasty in patients taken warfarin.
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2012.12.007. Reprint requests: Michael Aynardi, MD, The Rothman Institute at Thomas Jefferson University, 1015 Walnut Street, Room 801 Curtis Building, Philadelphia, PA 19107. 0883-5403/2808-0002$36.00/0 – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2012.12.007
Materials and Methods Under approval from the institutional review board, 441 consecutive patients undergoing primary total joint arthroplasty were retrospectively reviewed from 2010–2011. All patients were enrolled in our institution's total joint arthroplasty protocol preoperatively. Each patient underwent mandatory screening and risk stratification by both their primary care physician, anesthesiologist, and in most instances a cardiologist 3 to 6 weeks prior to surgery. All patients underwent hypotensive regional spinal anesthesia during the procedure. Operations were performed by one of the two senior authors, FO and AO, in rooms with laminar flow with all members of the surgical team wearing body exhaust suits. Hip arthroplasty was performed in the supine position through a modified anterolateral approach. Knee arthroplasty was done using a medial parapatellar approach. Postoperatively, all patients received identical multimodal VTE prophylaxis; mobilization on the afternoon of postoperative day 0 by a physical therapist, intermittent pneumatic compression devices, and pharmacologic thromboprophylaxis with warfarin. All 441 patients received 5 mg of warfarin on the evening of surgery and were dosed daily according to our institutional dosing nomogram (Appendix 1) with an International Normalized Ratio goal of 2.0–2.5 in adherence with national guidelines [2,3]. Daily warfarin dosing and INR levels were followed and managed by a warfarin clinic with patients continuing the drug for 6 weeks postoperatively unless otherwise indicated. Demographic information, daily international normalized ratio levels, patient readmissions, and any postoperative complications within 30 days of the index surgery were recorded. Patients were seen in the office by the orthopaedic surgeon at two weeks, six weeks, and six months postoperatively. Patients were
1252
M. Aynardi et al. / The Journal of Arthroplasty 28 (2013) 1251–1253
subdivided and analyzed according to INR level at the time of discharge: subtherapeutic b2.0, therapeutic 2.0–2.5, and supratherapeutic N2.5. Statistical analysis was performed utilizing the chisquare test, student's t-test, and logistical regression to compare and analyze the INR level at the time of discharge, postoperative complications, and readmission rates among the three group calculating 95% confidence intervals. Post hoc power analysis was conducted. At beta = 0.8, a difference in overall complication rates, INR level at discharge, infection, readmission rates, and hematoma formation could be detected. The sample size was too small to detect a statistical difference in rates of mortality and venous thromboembolic events between groups (Minitab v13.2). Results In this series, 80% (352/441) of patients had a subtherapeutic (b2.0) INR level at discharge, 13% (56/441) were therapeutic (2.0– 2.5), and 7% (33/441) were supra-therapeutic (N 2.5). The mean INR level at the time of discharge was 1.61. The average hospital stay was 2.79 days. The overall 30 day complication rate was 15% (67/441), while the hospital readmission rate was 8.6% (38/441) (Fig. 1). There were no mortalities or diagnosed pulmonary emboli. DVT occurred in 3 of 441 patients, or 0.68%. All three had subtherapeutic INR at the time of complication. Hematoma formation or major bleeding events occurred in 1.5% of patients (7/441). In this study, hematoma formation was defined by the need for surgical evacuation and irrigation and debridement within 30 days of index procedure with preoperative aspiration negative for deep infection. The rate of deep infection requiring operative intervention was 1.1% (5/441) while 7 patients developed a superficial cellulitis requiring oral antibiotics (1.6%). Seven patients developed post operative hematoma, 1 (14%) of these had a supratherapeutic INR. A complete list of complications can be seen in (Table 1). Statistical analysis demonstrated an association between supratherapeutic INR at discharge with an increased number of complications postoperatively (P b 0.048). For each point of INR level increase, there was an odds ratio of 1.435 for increasing likelihood of complication (95% CI 1.002–2.054). Furthermore, supratherapeutic INR at discharge lead to a higher readmission rate, odds ratio 1.361 (95% CI 0.956–1.940). Importantly, elevated BMI (P = 0.001), younger age (P b 0.001), and male sex (P = 0.01) were all identified as patient variables which were associated with a supratherapeutic INR at the time of discharge.
30 25
Complications Readmissions
20 % 15 10 5 0 <2.0
2.0-2.5 INR Level
>2.5
Fig. 1. Complications and Readmissions by INR level.
Table 1 Complications. Major Complications Pulmonary Embolism Deep Vein Thrombosis Myocardial Infarction Infection Deep Superficial Major Bleeding⁎ Manipulation Under Anesthesia Lysis of Adhesions Foot drop Fracture Other⁎⁎ Overall Complications
Number 0 3 1 12 5 7 7 13 3 1 1 26 67
% 0.00 0.68 0.23 2.72 1.13 1.59 1.59 2.95 0.68 0.23 0.23 5.90 15.19
⁎ Major bleeding defined as bleeding resulting in complication or readmission. One hematoma was also infected. ⁎⁎ Syncope, pneumonia, negative chest pain work up, leg pain, extremity swelling, atrial fibrillation.
Discussion Current practice guidelines support the use of several different pharmacological methods of thromboprophylaxis following joint arthroplasty [2,3]. While therapeutic levels of warfarin are effective at preventing VTE following joint reconstruction [10], we hypothesized that a high percentage of patients treated with warfarin for VTE prophylaxis were being discharged at subtherapeutic INR levels due to rapid recovery and subsequent early discharge. Our concern was that nontherapeutic INR levels could lead to an increased incidence of VTE or other complications. While we did find a staggering number of TJA patients are being discharged at subtherapeutic INR levels, our findings also demonstrated an association between higher INR levels and readmission as well as an increased complication rate. Additionally, the study identified several risk factors which correlate with elevated INR level at discharge. Together, these issues have raised concerns in our practice regarding warfarin and its effectiveness and safety. The number of patients discharged with a therapeutic INR was surprisingly low, 13% (56/441), with the vast majority of patients being discharged with a subtherapeutic INR, 80% (352/441). These results are contrary to prior studies which have demonstrated a greater proportion of patients being discharged with therapeutic INR [8,9,11]. However, in earlier reports, patients often became therapeutic by POD 4 or 5 and tended to have a longer average length of stay [8,9]. Unfortunately, current practices in orthopaedics are geared towards accelerated patient discharge and early mobilization with shorter lengths of stay [12], as evidenced by our average length of stay of 2.79 days. As a result, patients are not achieving therapeutic INR levels until a few days after their day of discharge, which must now be accomplished through outpatient monitoring. Furthermore, response to warfarin is quite variable in the first days of dosing and is often unreliable until a steady-state has been reached [7]. While studies employing dosing nomograms have improved the effectiveness of warfarin, many were modeled for treatment of patients in the hospital [8,9]. Consequently, patients discharged at subtherapeutic levels require close monitoring until a steady-state can be achieved, which becomes a very difficult task in an outpatient setting. Warfarin dosed and managed outside of the hospital also requires a very diligent and compliant patient. These barriers combine to make warfarin therapy for VTE prophylaxis potentially unsafe in a modern TJA practice. Importantly, this study demonstrated an association between both complications and readmissions for patients with supratherapeutic INR levels at discharge. The overall complication rate of 15% (67/441) is higher than what is reported in the literature [4,5,13], but likely reflects increased surveillance for minor complications such as cellulitis and medical complications which may not have a direct
M. Aynardi et al. / The Journal of Arthroplasty 28 (2013) 1251–1253
clinical association with warfarin use. Hematoma formation and major bleeding events occurred in 1.5% of patients (7/441), which is similar to previous reports [5,8,14]. The relationship between hematoma formation and subsequent infection in association with a high INR has been described [4,5]. Furthermore, we identified elevated BMI, younger age, and male sex as risk factors for a supratherapeutic INR at discharge. Perhaps these risk factors are related to the variable response to warfarin as well as patient compliance. BMI, age, and other genetic factors have been shown to impact INR levels [15]. Despite the number of patients subtherapeutic at discharge, there were no mortalities or pulmonary emboli. There was one myocardial infarction which occurred on postoperative day 0. Likewise, the rate of DVT was low at 0.68%, occurring in only 3 of 441 patients. There are several potential explanations for these findings. First, patients underwent aggressive rehabilitation protocols and were mobilized the evening of surgery while receiving intermittent pneumatic compression devices both in bed and in the chair; thus, limiting venous stasis. Secondly, it is possible that some cases of VTE were missed due to early discharge. While most VTE occur within 7 to 14 days of surgery, it is possible for clots to occur as long as 6 weeks after surgery [3], which may have gone undetected in our study. Third, our institutional protocols are designed to investigate only clinically symptomatic DVT and PE. The institution's diagnostic algorithm for PE has been previously reported [16]. To complicate matters, the rate of symptomatic and asymptomatic VTE often varies in the literature [3,17] and INR level has been shown to be a poor predictor of which patients will develop pulmonary embolism [16]. These findings warrant a larger, prospective investigation to establish if subtherapeutic INR at discharge correlates with the rate of VTE. There are several limitations of this study. While the power was large enough to detect differences between the groups of INR levels and complications, the cohort was not large enough to comment with certainty on rarer events such as mortality and pulmonary embolism. We also lacked the power to comment of the association of VTE and subtherapeutic INR at discharge. Finally, our study was conducted in a retrospective, non-randomized fashion which limits our ability to record and detect events; thus, the rate of VTE may have been underreported. With these limitations in mind, we believe the data raise valid concerns and demonstrate important associations between INR levels and complications after joint arthroplasty. While warfarin dosed with a therapeutic INR has been used for decades as effective VTE prophylaxis [10], several concerns have recently arisen [5,8,9,13,16,18]. Parvizi et al demonstrated that excessive anticoagulation with warfarin was a significant risk factor for postoperative wound complications, hematoma formation, and periprosthetic infection after TJA [5] and the findings of this investigation support these reports. Furthermore, Hansen et al were unable to find any correlation between pulmonary embolism and INR level; notably, a therapeutic INR did not seem to be preventative for pulmonary embolism [16]. The authors suggest that DVT and PE are separate entities and should be treated as such, and that an elevated INR carries associated risks of bleeding and wound complications with no afforded benefit of pulmonary embolism prevention [16]. Additionally, a common misconception is that warfarin is inexpensive. Botteman et al using economic models showed that anticoagulation with low molecular weight heparin is more cost-effective than warfarin demonstrating a savings of $3733 per quality-adjusted lifeyears overall [18]. Lastly, changes to our health system have resulted in a push for decreased lengths of hospital stay and accelerated rehabilitation programs; in turn, this has resulted in many patients being discharged at nontherapeutic INR levels [8,9,12]. Frequent INR monitoring, early patient discharge at subtherapeutic levels, a high rate of postoperative complications and increased readmission rates were shown in this study to all be concerns with using warfarin as VTE prophylaxis in joint arthroplasty. Given these
1253
findings we call into question the safety of continued use of warfarin for routine VTE prophylaxis in modern TJA patients where there is a trend toward shorter length of stay. The findings of this study have caused our institution to re-evaluate the routine use of warfarin for VTE prophylaxis. Currently, warfarin is only prescribed for patients who are at increased risk of VTE or have a contraindication to using some other form of prophylaxis. The increasing popularity of other methods for VTE prophylaxis which require no monitoring or have limited risk of bleeding may soon lead to a decrease in the use of warfarin. The results of this study further demonstrate the need for a large, prospective, randomized investigation involving warfarin in direct comparison with other methods of VTE prophylaxis paying particular attention to the non-thrombotic complications associated with VTE prophylaxis after TJA. Appendix 1 Daily Warfarin Dosing Nomogram. INR b1.24 1.25-1.5 1.51-1.75 1.76-1.99 N2.0 *5 mg given POD 0
Dose (mg) 5 4 2 1 Hold
References 1. Harris WH, Sledge CB. Total hip and knee replacement. N Engl J Med 1990;323:725. 2. Johanson NA, Lachiewicz PF, Lieberman JR, et al. American Academy of Orthopaedic Surgeons. Clinical Practice Guideline on prevention of symptomatic pulmonary embolism in patients undergoing total hip or knee arthroplasty. J Am Acad Orthop Surg 2009;17:183. 3. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(Suppl):381S. 4. Pulido L, Ghanem E, Joshi A, et al. Periprosthetic Joint Infection: The Incidence, Timing, and Predisposing Factors. Clin Orthop Relat Res 2008;466(7):1710. 5. Parvizi J, Ghanem E, Joshi A, et al. Dose “Excessive” Anticoagulation Predispose to Periprosthetic Infection? J Arthroplasty 2007;22(6):24. 6. Amstutz HC, Friscia DA, Dorey F, et al. Warfarin prophylaxis to prevent mortality form pulmonary embolism after total hip replacement. J Bone Joint Surg Am 1989;71:321. 7. Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the vitamin K antagonists: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(Suppl):160S. 8. Anderson DR, Wilson J, Blundell J, et al. Comparison of a nomogram and physicianadjusted dosage of warfarin for prophylaxis against deep vein thrombosis after arthroplasty. J Bone Joint Am 2002;84:1992. 9. Asnis P, Gardner M, Ranawat A, et al. The Effectiveness of Warfarin Dosing from a Nomogram Compared with House Staff Dosing. J Arthroplasty 2007;8:213. 10. Xing K, et al. Has the Incidence of Deep Vein Thrombosis in Patients Undergoing Total Hip/Knee Arthroplasty Changed Over Time? A Systematic Review of Randomized Controlled Trials. Thromb Res 2008;123:24. 11. Freedman KB, Brookenthal KR, Fitzgerald RH, et al. A meta-analysis of thromboembolic prophylaxis following elective total hip arthroplasty. J Bone Joint Surg Am 2000;82:929. 12. Husted H, Holm G. Predictors of Length of Stay and Patients Satisfaction after Hip and Knee Replacement Surgery: Fast Track Experience in 712 Patients. Acta Orthop 2008;79(20):168. 13. Pulido L, Parvizi J, Macgibeny M, et al. In Hospital Complications After Total Joint Arthroplasty. J Arthroplasty 2008;23(6):139. 14. Colwell Jr CW, Berkowitz SD, Lieberman JR, et al. Oral direct thrombin inhibitor ximelagatran compared with warfarin for the prevention of venous thromboembolism after total knee arthroplasty. J Bone Joint Surg Am 2005;87:2169. 15. Hirsh J, Fuster V, Ansell J, et al. American Heart Association/American College of Cardiology Foundation Guide to warfarin therapy. Circulation 2003;107:1692. 16. Hansen P, Zmistowski B, Restrepo C, et al. Does International Normalized Ratio Level Predict Pulmonary Embolism? Clin Orthop Relat Res 2012;470:547. 17. Dager W. Warfarin for Venous Thromboembolism Prophylaxis After Elective Hip or Knee Arthroplasty: Exploring the Evidence, Guidelines, and Challenges Remaining. Ann Pharmacother 2012;46:79. 18. Botteman M, et al. Results of an Economic Model to Assess the Cost-Effectiveness of Enoxaparin, a Low-Molecular-Weight Heparin, Versus Warfarin for the Prophylaxis of Deep Vein Thrombosis and Associated Long-Term Complications in Total Hip Replacement Surgery in the United States. Clin Ther 2002;24:1960.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Warfarin for Thromboprophylaxis Following Total Joint Arthroplasty Are Patients Safely Anti-Coagulated? Michael Aynardi, MD, P. Benedict Brown, BS, Zachary Post, MD, Fabio Orozco, MD, Alvin Ong, MD The Rothman Institute at Thomas Jefferson University Hospital, Philadelphia, Pennsylvania
a r t i c l e
i n f o
Article history: Received 11 September 2012 Accepted 18 December 2012 Keywords: joint replacement warfarin International Normalized Ratio (INR) thromboprophylaxis complications
a b s t r a c t The purpose of this study was to investigate whether any correlation exists between INR level at discharge and postoperative complications or readmission rates. From 2010–2011, INR levels on discharge, complications and readmissions within 30 days were recorded on 441 patients undergoing joint arthroplasty. Eighty percent (352 of 441) patients had a subtherapeutic INR level at discharge. The overall complication rate was 15% with an 8.6% readmission rate. A supratherapeutic INR level at discharge was associated with both higher readmission rate as well as increased number of complications (P b 0.048). Most patients taking warfarin are nontherapeutic at the time of discharge; notably, a supratherapeutic INR places patients at risk for increased complications and readmissions rates following surgery. © 2013 Elsevier Inc. All rights reserved.
Venous thromboembolism (VTE) is one of the most common complications following total joint arthroplasty [1]. While multimodal thromboprophylaxis has lowered the incidence of VTE, no single entity has emerged as the ideal method of prophylaxis [2,3]. Additionally, anticoagulation has its own set of complications including increased risk of bleeding, difficulty with wound healing, and even infection [4–6]. Discrepancy between currently accepted clinical guidelines has caused resurgence in the debate for the ideal method of thromboprophylaxis following total joint arthroplasty [2,3,7]. Traditionally, warfarin with an International Normalized Ratio (INR) goal of 2.0–2.5 has been widely accepted as an appropriate method of VTE prophylaxis; however, accelerated rehabilitation programs are leading to decreased lengths of hospital stay which have increased the probability of nontherapeutic INR levels at the time of discharge [8,9]. As a result, it is crucial that both the safety and efficacy of warfarin be reevaluated to ensure patients are being properly anticoagulated after major joint arthroplasty. The purpose of this study was to investigate whether any correlation exists between INR level at the time of discharge and postoperative complications or readmission rates following total knee and hip arthroplasty in patients taken warfarin.
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2012.12.007. Reprint requests: Michael Aynardi, MD, The Rothman Institute at Thomas Jefferson University, 1015 Walnut Street, Room 801 Curtis Building, Philadelphia, PA 19107. 0883-5403/2808-0002$36.00/0 – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2012.12.007
Materials and Methods Under approval from the institutional review board, 441 consecutive patients undergoing primary total joint arthroplasty were retrospectively reviewed from 2010–2011. All patients were enrolled in our institution's total joint arthroplasty protocol preoperatively. Each patient underwent mandatory screening and risk stratification by both their primary care physician, anesthesiologist, and in most instances a cardiologist 3 to 6 weeks prior to surgery. All patients underwent hypotensive regional spinal anesthesia during the procedure. Operations were performed by one of the two senior authors, FO and AO, in rooms with laminar flow with all members of the surgical team wearing body exhaust suits. Hip arthroplasty was performed in the supine position through a modified anterolateral approach. Knee arthroplasty was done using a medial parapatellar approach. Postoperatively, all patients received identical multimodal VTE prophylaxis; mobilization on the afternoon of postoperative day 0 by a physical therapist, intermittent pneumatic compression devices, and pharmacologic thromboprophylaxis with warfarin. All 441 patients received 5 mg of warfarin on the evening of surgery and were dosed daily according to our institutional dosing nomogram (Appendix 1) with an International Normalized Ratio goal of 2.0–2.5 in adherence with national guidelines [2,3]. Daily warfarin dosing and INR levels were followed and managed by a warfarin clinic with patients continuing the drug for 6 weeks postoperatively unless otherwise indicated. Demographic information, daily international normalized ratio levels, patient readmissions, and any postoperative complications within 30 days of the index surgery were recorded. Patients were seen in the office by the orthopaedic surgeon at two weeks, six weeks, and six months postoperatively. Patients were
1252
M. Aynardi et al. / The Journal of Arthroplasty 28 (2013) 1251–1253
subdivided and analyzed according to INR level at the time of discharge: subtherapeutic b2.0, therapeutic 2.0–2.5, and supratherapeutic N2.5. Statistical analysis was performed utilizing the chisquare test, student's t-test, and logistical regression to compare and analyze the INR level at the time of discharge, postoperative complications, and readmission rates among the three group calculating 95% confidence intervals. Post hoc power analysis was conducted. At beta = 0.8, a difference in overall complication rates, INR level at discharge, infection, readmission rates, and hematoma formation could be detected. The sample size was too small to detect a statistical difference in rates of mortality and venous thromboembolic events between groups (Minitab v13.2). Results In this series, 80% (352/441) of patients had a subtherapeutic (b2.0) INR level at discharge, 13% (56/441) were therapeutic (2.0– 2.5), and 7% (33/441) were supra-therapeutic (N 2.5). The mean INR level at the time of discharge was 1.61. The average hospital stay was 2.79 days. The overall 30 day complication rate was 15% (67/441), while the hospital readmission rate was 8.6% (38/441) (Fig. 1). There were no mortalities or diagnosed pulmonary emboli. DVT occurred in 3 of 441 patients, or 0.68%. All three had subtherapeutic INR at the time of complication. Hematoma formation or major bleeding events occurred in 1.5% of patients (7/441). In this study, hematoma formation was defined by the need for surgical evacuation and irrigation and debridement within 30 days of index procedure with preoperative aspiration negative for deep infection. The rate of deep infection requiring operative intervention was 1.1% (5/441) while 7 patients developed a superficial cellulitis requiring oral antibiotics (1.6%). Seven patients developed post operative hematoma, 1 (14%) of these had a supratherapeutic INR. A complete list of complications can be seen in (Table 1). Statistical analysis demonstrated an association between supratherapeutic INR at discharge with an increased number of complications postoperatively (P b 0.048). For each point of INR level increase, there was an odds ratio of 1.435 for increasing likelihood of complication (95% CI 1.002–2.054). Furthermore, supratherapeutic INR at discharge lead to a higher readmission rate, odds ratio 1.361 (95% CI 0.956–1.940). Importantly, elevated BMI (P = 0.001), younger age (P b 0.001), and male sex (P = 0.01) were all identified as patient variables which were associated with a supratherapeutic INR at the time of discharge.
30 25
Complications Readmissions
20 % 15 10 5 0 <2.0
2.0-2.5 INR Level
>2.5
Fig. 1. Complications and Readmissions by INR level.
Table 1 Complications. Major Complications Pulmonary Embolism Deep Vein Thrombosis Myocardial Infarction Infection Deep Superficial Major Bleeding⁎ Manipulation Under Anesthesia Lysis of Adhesions Foot drop Fracture Other⁎⁎ Overall Complications
Number 0 3 1 12 5 7 7 13 3 1 1 26 67
% 0.00 0.68 0.23 2.72 1.13 1.59 1.59 2.95 0.68 0.23 0.23 5.90 15.19
⁎ Major bleeding defined as bleeding resulting in complication or readmission. One hematoma was also infected. ⁎⁎ Syncope, pneumonia, negative chest pain work up, leg pain, extremity swelling, atrial fibrillation.
Discussion Current practice guidelines support the use of several different pharmacological methods of thromboprophylaxis following joint arthroplasty [2,3]. While therapeutic levels of warfarin are effective at preventing VTE following joint reconstruction [10], we hypothesized that a high percentage of patients treated with warfarin for VTE prophylaxis were being discharged at subtherapeutic INR levels due to rapid recovery and subsequent early discharge. Our concern was that nontherapeutic INR levels could lead to an increased incidence of VTE or other complications. While we did find a staggering number of TJA patients are being discharged at subtherapeutic INR levels, our findings also demonstrated an association between higher INR levels and readmission as well as an increased complication rate. Additionally, the study identified several risk factors which correlate with elevated INR level at discharge. Together, these issues have raised concerns in our practice regarding warfarin and its effectiveness and safety. The number of patients discharged with a therapeutic INR was surprisingly low, 13% (56/441), with the vast majority of patients being discharged with a subtherapeutic INR, 80% (352/441). These results are contrary to prior studies which have demonstrated a greater proportion of patients being discharged with therapeutic INR [8,9,11]. However, in earlier reports, patients often became therapeutic by POD 4 or 5 and tended to have a longer average length of stay [8,9]. Unfortunately, current practices in orthopaedics are geared towards accelerated patient discharge and early mobilization with shorter lengths of stay [12], as evidenced by our average length of stay of 2.79 days. As a result, patients are not achieving therapeutic INR levels until a few days after their day of discharge, which must now be accomplished through outpatient monitoring. Furthermore, response to warfarin is quite variable in the first days of dosing and is often unreliable until a steady-state has been reached [7]. While studies employing dosing nomograms have improved the effectiveness of warfarin, many were modeled for treatment of patients in the hospital [8,9]. Consequently, patients discharged at subtherapeutic levels require close monitoring until a steady-state can be achieved, which becomes a very difficult task in an outpatient setting. Warfarin dosed and managed outside of the hospital also requires a very diligent and compliant patient. These barriers combine to make warfarin therapy for VTE prophylaxis potentially unsafe in a modern TJA practice. Importantly, this study demonstrated an association between both complications and readmissions for patients with supratherapeutic INR levels at discharge. The overall complication rate of 15% (67/441) is higher than what is reported in the literature [4,5,13], but likely reflects increased surveillance for minor complications such as cellulitis and medical complications which may not have a direct
M. Aynardi et al. / The Journal of Arthroplasty 28 (2013) 1251–1253
clinical association with warfarin use. Hematoma formation and major bleeding events occurred in 1.5% of patients (7/441), which is similar to previous reports [5,8,14]. The relationship between hematoma formation and subsequent infection in association with a high INR has been described [4,5]. Furthermore, we identified elevated BMI, younger age, and male sex as risk factors for a supratherapeutic INR at discharge. Perhaps these risk factors are related to the variable response to warfarin as well as patient compliance. BMI, age, and other genetic factors have been shown to impact INR levels [15]. Despite the number of patients subtherapeutic at discharge, there were no mortalities or pulmonary emboli. There was one myocardial infarction which occurred on postoperative day 0. Likewise, the rate of DVT was low at 0.68%, occurring in only 3 of 441 patients. There are several potential explanations for these findings. First, patients underwent aggressive rehabilitation protocols and were mobilized the evening of surgery while receiving intermittent pneumatic compression devices both in bed and in the chair; thus, limiting venous stasis. Secondly, it is possible that some cases of VTE were missed due to early discharge. While most VTE occur within 7 to 14 days of surgery, it is possible for clots to occur as long as 6 weeks after surgery [3], which may have gone undetected in our study. Third, our institutional protocols are designed to investigate only clinically symptomatic DVT and PE. The institution's diagnostic algorithm for PE has been previously reported [16]. To complicate matters, the rate of symptomatic and asymptomatic VTE often varies in the literature [3,17] and INR level has been shown to be a poor predictor of which patients will develop pulmonary embolism [16]. These findings warrant a larger, prospective investigation to establish if subtherapeutic INR at discharge correlates with the rate of VTE. There are several limitations of this study. While the power was large enough to detect differences between the groups of INR levels and complications, the cohort was not large enough to comment with certainty on rarer events such as mortality and pulmonary embolism. We also lacked the power to comment of the association of VTE and subtherapeutic INR at discharge. Finally, our study was conducted in a retrospective, non-randomized fashion which limits our ability to record and detect events; thus, the rate of VTE may have been underreported. With these limitations in mind, we believe the data raise valid concerns and demonstrate important associations between INR levels and complications after joint arthroplasty. While warfarin dosed with a therapeutic INR has been used for decades as effective VTE prophylaxis [10], several concerns have recently arisen [5,8,9,13,16,18]. Parvizi et al demonstrated that excessive anticoagulation with warfarin was a significant risk factor for postoperative wound complications, hematoma formation, and periprosthetic infection after TJA [5] and the findings of this investigation support these reports. Furthermore, Hansen et al were unable to find any correlation between pulmonary embolism and INR level; notably, a therapeutic INR did not seem to be preventative for pulmonary embolism [16]. The authors suggest that DVT and PE are separate entities and should be treated as such, and that an elevated INR carries associated risks of bleeding and wound complications with no afforded benefit of pulmonary embolism prevention [16]. Additionally, a common misconception is that warfarin is inexpensive. Botteman et al using economic models showed that anticoagulation with low molecular weight heparin is more cost-effective than warfarin demonstrating a savings of $3733 per quality-adjusted lifeyears overall [18]. Lastly, changes to our health system have resulted in a push for decreased lengths of hospital stay and accelerated rehabilitation programs; in turn, this has resulted in many patients being discharged at nontherapeutic INR levels [8,9,12]. Frequent INR monitoring, early patient discharge at subtherapeutic levels, a high rate of postoperative complications and increased readmission rates were shown in this study to all be concerns with using warfarin as VTE prophylaxis in joint arthroplasty. Given these
1253
findings we call into question the safety of continued use of warfarin for routine VTE prophylaxis in modern TJA patients where there is a trend toward shorter length of stay. The findings of this study have caused our institution to re-evaluate the routine use of warfarin for VTE prophylaxis. Currently, warfarin is only prescribed for patients who are at increased risk of VTE or have a contraindication to using some other form of prophylaxis. The increasing popularity of other methods for VTE prophylaxis which require no monitoring or have limited risk of bleeding may soon lead to a decrease in the use of warfarin. The results of this study further demonstrate the need for a large, prospective, randomized investigation involving warfarin in direct comparison with other methods of VTE prophylaxis paying particular attention to the non-thrombotic complications associated with VTE prophylaxis after TJA. Appendix 1 Daily Warfarin Dosing Nomogram. INR b1.24 1.25-1.5 1.51-1.75 1.76-1.99 N2.0 *5 mg given POD 0
Dose (mg) 5 4 2 1 Hold
References 1. Harris WH, Sledge CB. Total hip and knee replacement. N Engl J Med 1990;323:725. 2. Johanson NA, Lachiewicz PF, Lieberman JR, et al. American Academy of Orthopaedic Surgeons. Clinical Practice Guideline on prevention of symptomatic pulmonary embolism in patients undergoing total hip or knee arthroplasty. J Am Acad Orthop Surg 2009;17:183. 3. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(Suppl):381S. 4. Pulido L, Ghanem E, Joshi A, et al. Periprosthetic Joint Infection: The Incidence, Timing, and Predisposing Factors. Clin Orthop Relat Res 2008;466(7):1710. 5. Parvizi J, Ghanem E, Joshi A, et al. Dose “Excessive” Anticoagulation Predispose to Periprosthetic Infection? J Arthroplasty 2007;22(6):24. 6. Amstutz HC, Friscia DA, Dorey F, et al. Warfarin prophylaxis to prevent mortality form pulmonary embolism after total hip replacement. J Bone Joint Surg Am 1989;71:321. 7. Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the vitamin K antagonists: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133(Suppl):160S. 8. Anderson DR, Wilson J, Blundell J, et al. Comparison of a nomogram and physicianadjusted dosage of warfarin for prophylaxis against deep vein thrombosis after arthroplasty. J Bone Joint Am 2002;84:1992. 9. Asnis P, Gardner M, Ranawat A, et al. The Effectiveness of Warfarin Dosing from a Nomogram Compared with House Staff Dosing. J Arthroplasty 2007;8:213. 10. Xing K, et al. Has the Incidence of Deep Vein Thrombosis in Patients Undergoing Total Hip/Knee Arthroplasty Changed Over Time? A Systematic Review of Randomized Controlled Trials. Thromb Res 2008;123:24. 11. Freedman KB, Brookenthal KR, Fitzgerald RH, et al. A meta-analysis of thromboembolic prophylaxis following elective total hip arthroplasty. J Bone Joint Surg Am 2000;82:929. 12. Husted H, Holm G. Predictors of Length of Stay and Patients Satisfaction after Hip and Knee Replacement Surgery: Fast Track Experience in 712 Patients. Acta Orthop 2008;79(20):168. 13. Pulido L, Parvizi J, Macgibeny M, et al. In Hospital Complications After Total Joint Arthroplasty. J Arthroplasty 2008;23(6):139. 14. Colwell Jr CW, Berkowitz SD, Lieberman JR, et al. Oral direct thrombin inhibitor ximelagatran compared with warfarin for the prevention of venous thromboembolism after total knee arthroplasty. J Bone Joint Surg Am 2005;87:2169. 15. Hirsh J, Fuster V, Ansell J, et al. American Heart Association/American College of Cardiology Foundation Guide to warfarin therapy. Circulation 2003;107:1692. 16. Hansen P, Zmistowski B, Restrepo C, et al. Does International Normalized Ratio Level Predict Pulmonary Embolism? Clin Orthop Relat Res 2012;470:547. 17. Dager W. Warfarin for Venous Thromboembolism Prophylaxis After Elective Hip or Knee Arthroplasty: Exploring the Evidence, Guidelines, and Challenges Remaining. Ann Pharmacother 2012;46:79. 18. Botteman M, et al. Results of an Economic Model to Assess the Cost-Effectiveness of Enoxaparin, a Low-Molecular-Weight Heparin, Versus Warfarin for the Prophylaxis of Deep Vein Thrombosis and Associated Long-Term Complications in Total Hip Replacement Surgery in the United States. Clin Ther 2002;24:1960.