Issue: June 2012

June 05, 2012

15 min read

Save

This article is more than 5 years old. Information may no longer be current.

Two VTE prophylaxis guidelines are now in agreement

Issue: June 2012

In the act of balancing venous thromboembolism prophylaxis after total joint arthroplasty, guidelines by the American Academy of Orthopaedic Surgeons and American College of Chest Physicians have historically conflicted. However, this discord was seemingly resolved by the latest edition of the American College of Chest Physicians guidelines, which are based on data that evaluate symptomatic – rather than asymptomatic – events and recognize aspirin as beneficial in the prevention of venous thromboembolism.

"The American College of Chest Physicians (ACCP) guidelines were simply not embraced by the orthopedic community," Javad Parvizi, MD, FRCS, professor of orthopedic surgery at the Rothman Institute and Thomas Jefferson University in Philadelphia told Orthopedics Today. "The new guidelines are much closer to the American Academy of Orthopaedic Surgeons’ (AAOS) stance, so it is likely that a lot of orthopedic surgeons would be willing to consider following the ACCP guidelines."

In 2004, the ACCP released its seventh edition guidelines on the prevention of deep venous thrombosis (DVT) during hip and knee replacement. These initial guidelines included a 1A recommendation for treating postoperative patients with a pharmacologic anticoagulant other than aspirin, including low-molecular weight heparin (LMWH), warfarin or fondaparinux.

That year, the Surgical Care Improvement Project was implemented and adopted the ACCP guidelines as the gold standard — which was of concern for orthopedic surgeons, according to Norman A. Johanson, MD, professor and chairman of the department of orthopedic surgery at Drexel University College of Medicine and chief of orthopedics at Hahnemann University Hospital in Philadelphia.

Now that there are no serious conflicts between the guidelines, Norman A. Johanson, MD, said surgeons can focus on other aspects such as developing standard definitions of bleeds.

Image: Johanson NA

"[The 2004 ACCP guidelines] caused quite a bit of concern among the orthopedic community that they were getting railroaded into these highly potent anticoagulants without looking at some of the other developments that orthopedic surgeons around the country were finding good results with," Johanson told Orthopedics Today. "There were surgeons who were reporting on essentially thousands of cases with good control of symptomatic events."

The ACCP analyzed studies that used DVT as an endpoint in the creation of their guidelines, Geoffrey Westrich, MD, associate professor of orthopedic surgery at the Hospital for Special Surgery, Weill Medical College and Cornell University in New York, said.

"They dismissed almost all of the mechanical compression studies," he told Orthopedics Today.

First AAOS guidelines

By 2006, some orthopedic surgeons were reporting increased bleeding rates of up to 6% to 10% in hips and knees when the 1A recommendations from the ACCP were used, Johanson said. In 2007, Burnett and colleagues implemented a 10-day course of enoxaparan sodium (Lovenox, sanofi-aventis, Bridgewater, NJ) following the ACCP guidelines for 290 consecutive patients who underwent total hip and total knee arthroplasty. The investigators reported a 9% complication rate that included 4.7% re-admissions, 3.4% return to the operating room for wound incision and drainage, 5.1% prolonged hospitalization and 3.4% injection site complications.

"This study basically showed that when you follow the [ACCP] recommendations and put all these fresh surgery patients on potent anticoagulation to try to prevent a DVT or [pulmonary embolism] PE, you are causing more bleeding, which could be more problematic than the DVT," Westrich, who is also chairman of the thromboembolic disease committee at Hospital for Special Surgery in New York, told Orthopedics Today. "There was a huge backlash from these guidelines, and a lot of people in the AAOS basically said, ‘We need our own guidelines.’"

Geoffrey Westrich

The AAOS released its guidelines for prevention of symptomatic PE – a departure from the ACCP guidelines, which were based on symptomatic and asymptomatic DVT as detected by venography.

"The AAOS’s first purpose was to go back and re-examine the literature from the standpoint of symptomatic events – most importantly, PE," Johanson, who served as the chairman of the workgroup that created the first edition of the AAOS guidelines, said.

Johanson and the workgroup employed a methodology to evaluate event rates among several studies that pooled data.

"The conclusion was that there was no definite difference statistically that we could find among any of the agents, including aspirin," he said. "But we were seeing trends for increased bleeding with the heavy-duty pharmacologic agents, so we came to the conclusion that what you ought to aim for is risk stratification among patients."

This risk stratification involved giving more aggressive anticoagulants to the patients at higher risk for venous thromboembolism (VTE) and giving mechanical devices and aspirin to routine patients. In addition, if patients were at greater risk for bleeding, then the recommendations suggested not using potent anticoagulation.

"It was the first guideline to look at the patient rather than just whether the patient has a clot on a venogram," Westrich said.

The AAOS guidelines also recommended regional anesthesia rather than general anesthesia and early mobilization, according to Westrich.

Additionally, the AAOS guidelines recommended against routine ultrasound of patients who underwent total joint arthroplasty.

"The problem with [routine screening] is that if it was postoperative day 3 and [the ultrasound] was negative, [then] we would send patients home and most surgeons would take them off of prophylaxis," Westrich said. "But the problem with that is we know clots can occur after 3 days."

Both sets of guidelines now recommend against routine postoperative duplex ultrasonography screening for asymptomatic patients who undergo total hip or knee replacement.

In 2008, the ACCP released another set of guidelines that, according to Johanson, gave more credence to mechanical devices because more data existed to support their use. However, the ACCP "still basically stated that they recommended against aspirin as the sole agent for anticoagulation, which was not technically contrary to what the AAOS said because we were recommending mechanical devices to everyone and so aspirin would have never been the sole agent per se," Johanson said. "But that resulted in a sharp conflict between the orthopedic community and the ACCP."

In 2011, the AAOS released an updated set of guidelines – this time using a methodology of network meta-analysis, which allowed for making comparisons between agents or devices that were not necessarily compared by specific literature articles.

"They basically came to the same conclusion," Johanson said. "If you look at the results, there was no significant difference among any of the agents that were out there for prevention of symptomatic events."

For the latest guidelines released in February, the ACCP employed a methodology known as the GRADE profile, which allowed for more a quantitative comparison of the differences in asymptomatic, symptomatic and bleeding events.

"That is a major breakthrough," Johanson, who served on the panel for the ninth edition of the ACCP guidelines, said. "The previous articles had not mentioned what is important to the patient, and as it turns out, a symptomatic nonfatal PE is about equivalent to a major bleeding event in the eyes of the patient."

In the development of the 2012 ACCP guidelines, no individual receiving funds from pharmaceutical companies could vote. This was in contrast to the 2008 guidelines, which were created by a panel of seven authors, six of whom had financial relationships with a mean of six pharmaceutical companies per author, according to Johanson.

"What you saw in 2012 was that perhaps a more objective view of the literature is achieved when you eliminate that kind of conflict," he said.

Types of anticoagulants

In the latest sets of guidelines, the ACCP recommends the use of newer agents including rivaroxaban (Xarelto, Janssen Pharmaceuticals Inc., Titusville, NJ), a direct factor Xa inhibitor, and dabigatran, an oral antithrombin agent, in addition to LMWH, fondaparinux, apixaban, aspirin, low-dose unfractionated heparin and adjusted-dose vitamin K antagonist — although the ACCP expressed a preference for LMWH over the other anticoagulants. The AAOS does not recommend for or against specific anticoagulants because they note that the current evidence is unclear. For duration of prophylaxis, the AAOS suggested that patients and physicians discuss the decision and does not give an exact duration because of the absence of reliable evidence. The ACCP recommends a minimum duration of 10 days to 14 days.

Surgeons are seeing similar rates of DVT and PE with these newer agents, which can be taken orally and do not require checking international normalized ratio, making them more attractive for compliance purposes.

"Studies have shown that the compliance rate [with LMWH] is somewhere between 60% and 95%," John J. Murnaghan, MD, MSc, MA, FRCSC, associate professor of surgery at Holland Orthopedic Arthritic Center, Sunnybrook Health Sciences Center and University of Toronto in Canada, said. In contrast, at Murnaghan’s institution, compliance with rivaroxaban is between 94% and 97%, according to data from a poster presented at this year’s World Congress on Osteoarthritis in Barcelona.

However, "there are two major concerns with these new agents: One is the lack of a specific antidote and two is the risk for bleeding both locally into a hip or knee wound and also at nonsurgical sites when these new agents are used in a wide variety of patients rather than just carefully selected patients," Paul F. Lachiewicz, MD, attending surgeon at Chapel Hill Orthopedics Surgery & Sports Medicine and Consulting Professor, Department of Orthopaedic Surgery, Duke University, said.

In January 2011, Jensen and colleagues reported a 9% rate of hematoma with rivaroxaban compared with no hematoma in patients who received LMWH. In addition, they reported a 2.5% infection rate with rivaroxaban compared with 1% for LMWH, and a rate of 3.94% return to operating room for patients who received rivaroxaban compared with 1.8% among those who received LMWH.

Data from Murnaghan’s institution to be presented at the Canadian Orthopaedic Association meeting in June shows rates of less than 1% for DVT, PE and major bleeding.

"This is not one of those studies where we tested everybody," Murnaghan told Orthopedics Today. "We only tested those patients who had symptoms."

"The timing and administration of these drugs is going to be critical, and we will need to learn about the benefits and the complications as they are used more frequently in real-life orthopedic situations vs. randomized trials," Orthopedics Today Editorial Board member, Jay R. Lieberman, MD, director of the New England Musculoskeletal Institute and professor and chairman of the department of orthopedic surgery at the University of Connecticut Health Center, said. "What surgeons want to see with these new agents is how they are working in real-life situations with respect to preventing symptomatic PE and symptomatic DVT, and the risk associated with using these types of drugs with respect to increased bleeding."

Pneumatic compression devices

Both guidelines recommend use of pneumatic compression devices in conjunction with aspirin or another anticoagulant. The AAOS also recommends that patients with a bleeding disorder or active liver disease use pneumatic compression devices.

Newer portable pneumatic compression devices have been shown to increase compliance. A 2009 study by Froimson and colleagues at the Cleveland Clinic revealed better compliance (83% vs. 49%), lower DVT rates (1.3% vs. 3.6%) and a reduction in clinically important PE (0% vs. 66%) with a miniaturized, portable, sequential, pneumatic compression device compared with a non-mobile, nonsequential device in combination with LMWH in high-risk patients after total joint arthroplasty.

"We learned from a technology perspective that we can actually pump more blood with a smaller device that has different hemodynamic parameters that are high-flow impulse compression devices," Westrich said.

There are concerns with the newer devices, however.

"Our concern with the new mechanical devices is that we have no biomechanical data on how these devices affect venous velocity and venous volume," Lachiewicz told Orthopedics Today. "There are issues with cost reimbursement, and the use and compliance of this device among patients who are not in a closely monitored study."

Paul F. Lachiewicz

Lachiewicz said some patients at his institution have reported malfunction of the devices at home.

Highest-risk patients

According to both sets of guidelines, patients at highest risk for VTE include those with a history of pre-existing documented VTE either with or without surgery — although the AAOS guidelines grade this recommendation as weak. The AAOS workgroup also recommended that surgeons assess a patient’s risk for known bleeding disorders or active liver disease. Because data are inconclusive about other factors, the AAOS said it is unable to recommend for or against assessing them.

"I think that could be modified with better studies using registries and inclusive databases because if you look at the literature for VTE and randomized studies, head-to-head drug trials, they exclude all of the people that they feel are at high risk, yet it has never been proven that they are — like obesity, diabetes, history of varicose veins," Johanson said.

Westrich’s institution, the Hospital for Special Surgery, is currently conducting a study with Quest Diagnostics to develop tests for patients with certain genetic factors, such as factor V Leiden deficiency and protein C and protein S deficiencies, which put them at greater risk for VTE. In the study, which has enrolled about a third of the patients necessary, the researchers are evaluating patients’ genetic profiles to determine the prevalence of different types of genetic abnormalities affecting their risk for DVT or bleeding.

"I believe what is going to happen is that someday as the price of this test continues to go down, where once you identify which genes you want to test, it can all go on one little chip, and with one drop of blood, we will be able to get a print-out of patients’ genetic hematologic profiles," Westrich said. "If we had that prior to surgery, we would be able to stratify our prophylaxis much more efficiently."

Other hypercoaguable states that place patients at a higher risk are cancer malignancy and immobility. Age also has been shown to be a relative risk factor.

Parvizi and colleagues have submitted data for publication that involves 22,000 patients with a 5% incidence of VTE.

"The risk factors we found include old age, male gender, obesity, long operative time, previous history of DVT and PE, among others," Parvizi said.

Inferior vena cava filters

The ACCP guidelines recommend against the use of inferior vena cava (IVC) filters, but the AAOS guidelines do not take a stance on this issue.

"Traditionally, we have put in a filter and it has been left in place … but there are lots of downsides to filters," Westrich said. "Patients could clot off at the filter and their whole lower extremities could become dilated. Filters can move, they could migrate, they could become problematic when trying to put them in, so there is a real complication rate."

Parvizi and colleagues presented data at the AAOS 2012 Annual Meeting that showed a 3.9% rate of PE with the use of IVC filters.

"[These data] point to the fact that PE can arise itself in the lungs," Parvizi, who is an Orthopedics Today Editorial Board member, said. "There is less and less enthusiasm for the use of IVC filters. … At this point, the exact indications and the efficacy of IVC filters remain unknown."

Moving forward

With the discrepancies between the guidelines resolved, surgeons and researchers now feel they can move on to other areas of concern in VTE prophylaxis.

"Now that there is no serious conflict between the two guidelines, we can start working on things that need to be done, such as standardization of definitions of bleeds, which would include surgical site bleeding and drainage as important events in the eyes of the surgeon, where many studies that have been done with head-to-head drug trials do not mention the surgical site," Johanson said.

Some surgeons are beginning to question the propagation theory — that the higher in the leg a DVT occurs, the greater a patient’s risk for PE.

"The major question we have to answer is, does PE arise from limb DVT or not?" Parvizi said. "If it does not, it will put into question the dogma that we have believed in for many decades practicing prophylaxis and thinking that prevention of distal DVT will lead to a lower incidence of PE. There is more evidence emerging that DVT in the limbs and PE could all be independent … Further studies need to be done to untangle this association that has been taken at face value for so many decades, but never been questioned."

"People are questioning how much do you need to treat and for how long?" Murnaghan said. "There are a few practitioners saying that the risks of using blood thinners exceed the risk for dying from a PE. Are we treating 99.9% of people to prevent a 0.1% event? I think the newest question is, do we understand enough about this whole condition to say everybody needs to be treated? I think that remains to be seen." – by Tina DiMarcantonio

References:

American Academy of Orthopaedic Surgeons. Guidelines on preventing venous thromboembolic disease in patients undergoing elective hip and knee arthroplasty. Retrieved from www.aaos.org/research/guidelines/VTE/VTE_guideline.asp.
Burnett RS, Clohisy JC, Wright RW, et al. Failure of the American College of Chest Physicians: A protocol for lovenox in clinical outcomes for thromboembolic prophylaxis. J Arthroplasty. 2007;
22(3):317-324.
Falck-Ytter Y Francis CW, Johanson NA, et al. Prevention of VTE in orthopedic surgery patients: Antithrombotic therapy and prevention of thrombosis, 9th Ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2012;141;e278S-e325S.
Froimson MI, Murray TG, Fazekas AF, et al. Venous thromboembolic disease reduction with a portable pneumatic compression device. J Arthroplasty. 2009;24(2):310-316.
Jensen CD, Steval A, Partington PF, et al. Return to theatre following total hip and knee replacement, before and after the introduction of rivaroxaban: A retrospective cohort study. J Bone Joint Surg Br. 2011;93(1):91-95.
Rodriguez-Elizalde S, Murnaghan J, Murnaghan D, Razmjou H, et al. Patient-reported compliance with thromboprophylaxis using an oral factor Xa inhibitor (Rivaroxaban) following hip and knee arthroplasty. Poster #577. Presented at the World Congress on Osteoarthritis 2012. Barcelona. April 22-39.
Zmistowski B, Restrepo C, Casper D, et al. Is inferior vena cava filter effective at preventing pulmonary embolism? Paper #31. Presented at the American Academy of Orthopaedic Surgeons 2012 Annual Meeting. Feb. 7-11. San Francisco.

For more information:

Norman A. Johanson, MD, can be reached at University Orthopaedic Institute at Hahnemann, 216-220 North Broad St., Feinstein Building, 2nd Fl., Philadelphia, PA 19102; 215-762-2663; email: norman.johanson@tenethealth.com.
Paul F. Lachiewicz, MD, can be reached at Chapel Hill Orthopedics Surgery & Sports Medicine, 101 Conner Drive, Suite 200, Chapel Hill, NC 27514; 919-986-6008; e-mail: paul.lachiewicz@gmail.com.
Jay R. Lieberman, MD, can be reached at New England Musculoskeletal Institute, Medical Arts & Research Building, UConn Health Center, 263 Farmington Ave., Farmington, CT 06030; 310-825-7687; fax: 310-206-0063; email: jlieberman@uchc.edu.
John J. Murnaghan, MD, FRCSC, can be reached at Sunnybrook Health Sciences Centre, Holland Orthopaedic & Arthritic Centre, 43 Wellesley St. E., Suite 319, Toronto, ON M4Y 1H1; 416-967-8778; fax: 416-967-8780; email: j.murnaghan@utoronto.ca.
Javad Parvizi, MD, FRCS, can be reached at the Rothman Institute, 925 Chestnut St., 5th Floor, Philadelphia, PA 19107; 267-339-3617; email: parvj@aol.com.
Geoffrey Westrich, MD, can be reached at Hospital for Special Surgery, 535 East 70th St., New York, NY; 212-606-1000; email: westrichg@hss.edu.
Disclosures: Johanson, Lieberman and Parvizi have no relevant financial disclosures. Lachiewicz is on the speaker’s bureau for Cadence Pharmaceuticals and his institution receives research support from Zimmer. Murnaghan receives research funding from Bayer Healthcare. Westrich is a consultant for DonJoy Orthopedics, Exactech and Stryker, who also provide research support for his institution.

What strategy or agents do you employ to manage bleeding risk or prevent bleeding during total joint arthroplasty?

Timing of antithrombotic administration essential to safety

Louis M. Kwong

On the issue of venous thromboembolism (VTE) prophylaxis following total joint arthroplasty, for many surgeons, the matters of potential bleeding risk and complications associated with systemic anticoagulation are a greater concern than the risks and potential sequelae of the VTE events themselves. This leads some surgeons to use a non-evidence-based approach to thromboprophylaxis either involving the use of unproven agents or to use otherwise proven agents off-label in an ineffective manner in an attempt to increase safety.

In optimizing the balance between efficacy of an antithrombotic and safety in terms of bleeding, the surgeon should follow a few guiding principles. The surgeon needs to always specify a specific start time for administration of the first dose of antithrombotic agent. This is to help safeguard against an agent being administered too early. Most importantly, it is essential that the surgeon make a thoughtful decision as to when that first dose is best given. All modern agents used for VTE prophylaxis specify a time range to start drug administration (e.g., 6 hours to 10 hours) either referenced from the time of wound closure or from the time that primary hemostasis of the wound has occurred. Although in some cases primary hemostasis is achieved at the time of wound closure; in many cases, the achievement of primary wound hemostasis may not be achieved until 8 hours to 12 hours after wound closure.

By considering the issues of complexity of the procedure, size of the patient, magnitude of deformity, need for soft tissue releases and comorbidities, the surgeon stratifies bleeding risk and uses medical judgment as to when primary wound hemostasis may likely be achieved. For example, if it is estimated that primary wound hemostasis will likely be achieved 12 hours after wound closure and the surgeon desires initiation of the first dose of the antithrombotic agent 10 hours after that, then the specific start time written in the chart will be 22 hours after wound closure. This is important, as any antithrombotic administered too close to the time of surgery (or when primary wound hemostasis has not yet occurred) will increase the risk for bleeding.

In terms of risk stratification, careful consideration of what we as surgeons do at the time of surgery that may increase bleeding is essential in reducing the potential for largely preventable postoperative hemorrhagic complications.

For more information:

Louis M. Kwong, MD, FACS, is professor and vice chairman of the department of orthopedic surgery, medical director of the orthopedic clinic, chief of the Orthopaedic Arthritis Service and program director of residency training at Harbor–UCLA Medical Center in Torrance, Calif. He is also co-director of the Thrombosis Research Institute at LA BioMed (TRIAL Group).
Disclosure: He receives research funding from Astellas, Bayer, Pfizer, Purdue and Takeda; serves as a consultant for Zimmer, Center of Innovation for Biomaterials in Orthopaedic Research and Portola; receives royalties from Zimmer; and is a speaker for ConvaTec and Janssen.

Preoperative evaluation and risk stratification are important

Matthew S. Austin

The management of bleeding risk during total joint arthroplasty begins with the preoperative evaluation. The patient is specifically queried about a prior history of coagulation disorders and medications that may increase the risk for bleeding. Patients with a prior bleeding history are referred to a hematologist for evaluation and optimization. Medications such as nonsteroidal anti-inflammatory medications, antiplatelet agents and anticoagulants are discontinued prior to surgery. A medical consultant is invaluable in timing discontinuation and restarting of antiplatelet and anticoagulation agents for patients with medical conditions requiring these drugs.

We prefer to perform the operation under spinal, hypotensive anesthesia. Appropriately selected patients are given tranexamic acid (TA), which is an antifibrinolytic agent that blocks the lysine binding site on plasminogen. At our institution, TA is given 15 minutes prior to incision for total hip arthroplasty and 15 minutes prior to tourniquet deflation for total knee arthroplasty. The literature has been supportive of the role of TA in reducing the need for transfusion without significantly increasing the incidence of venous thromboembolic disease. Of course, these modalities are no substitute for appropriate hemostasis, which must be achieved during the surgery.

Furthermore, postoperative venous thromboembolic prophylaxis is stratified according to the American Academy of Orthopaedic Surgeons’ clinical practice guideline taking into account patients’ risks for venous thromboembolic disease balanced with their risks for bleeding.

For more information:

Matthew S. Austin, MD, is the director of joint replacement services at the Rothman Institute in Philadelphia.
Disclosure: He receives research support from DePuy, is a consultant for Zimmer and Biomet and receives royalties from Zimmer.