eMedicine Specialties > Orthopedic Surgery > Systemic Diseases

Deep Venous Thrombosis Prophylaxis in Orthopedic Surgery

Author: Robert S Ennis, MD, FACS, Associate Professor, Department of Orthopedic Surgery, University of Miami School of Medicine; President, OrthoMed Consulting Services, Inc
Contributor Information and Disclosures

Updated: Jun 12, 2009

Introduction

An estimated 300,000 individuals are hospitalized annually in the United States for deep venous thrombosis, also known as deep vein thrombosis (DVT). This is especially significant, as up to three quarters of cases of DVT disease remain silent and do not come to medical attention.

Thrombosis is a naturally occurring physiologic process. Under normal circumstances, a physiologic balance is present between factors that promote and retard coagulation. A disturbance in this equilibrium may result in the coagulation process occurring at an inopportune time or location or in an excessive manor. Alternatively, failure of the normal coagulation mechanisms may lead to hemorrhage.

Virchow triad

More than 100 years ago, Virchow described a triad of factors of venous stasis, endothelial damage, and a hypercoagulable state that are associated with the equilibrium process. Venous stasis can occur as a result of anything that slows or obstructs the flow of venous blood. This results in an increase in viscosity and the formation of microthrombi, which are not washed away by fluid movement; the thrombus that forms may then grow and propagate.

Endothelial (intimal) damage in the blood vessel may be intrinsic or secondary to external trauma. It may result from accidental injury or surgical insult. A hypercoagulable state can occur due to a biochemical imbalance between circulating factors. This may result from an increase in circulating tissue activation factor, combined with a decrease in circulating plasma antithrombin and fibrinolysins. The Virchow triad can be summarized as follows:

• Venous stasis
  • More time for clotting
  • Small clots not washed away
  • Increased blood viscosity
• Vessel wall damage
  • Accidental trauma
  • Surgical trauma
• Blood coagulability increase
  • Increase in tissue factor
  • Presence of activated factors
  • Decrease in coagulation inhibitors (antithrombin III [ATIII])

Venous thrombi generally form in regions of stasis composed of red blood cells embedded in a mesh of fibrin strands and platelets, usually in response to a hypercoagulable state (see Image 1). In contrast, arterial thrombi usually occur in vessels that have a higher pressure gradient and flow rate. Arterial thrombi are composed mainly of platelet aggregates with relatively few fibrin strands, usually as a result of platelet reaction to intimal vessel damage. Thrombi, which occur in the proximal veins of the lower extremities, may break free and travel to the pulmonary vasculature where, if they are large enough, they can cause a fatal pulmonary embolism (PE) (see Image 2).

The coagulation cascade

For the most part, the coagulation mechanism consists of a series of self-regulating steps that result in the production of a fibrin clot. These steps are controlled by a number of relatively inactive cofactors or zymogens, which, when activated, promote or accelerate the clotting process. These reactions usually occur at the phospholipid surface of platelets, endothelial cells, or macrophages. Generally, the initiation of the coagulation process can be divided into 2 distinct pathways, an intrinsic system and an extrinsic system (see Image 3).

The extrinsic system operates as the result of activation by tissue lipoprotein, usually released as the result of some mechanical injury or trauma. The intrinsic system usually involves circulating plasma factors. Both of these pathways come together at the level of factor X, which is activated to form factor Xa. This in turn promotes the conversion of prothrombin to thrombin (factor II). This is the key step in clot formation, for active thrombin is necessary for the transformation of fibrinogen to a fibrin clot.

Once a fibrin clot is formed and has performed its function of hemostasis, mechanisms exist in the body to restore the normal blood flow by lysing the fibrin deposit. Circulating fibrinolysins perform this function. Plasmin digests fibrin and also inactivates clotting factors V and VIII and fibrinogen.

Three naturally occurring anticoagulant mechanisms exist to prevent inadvertent activation of the clotting process. These include the heparin-ATIII, protein C and thrombomodulin protein S, and the tissue factor inhibition pathways. When trauma occurs, or when surgery is performed, circulating ATIII is decreased. This has the effect of potentiating the coagulation process. Studies have demonstrated that levels of circulating ATIII is decreased more, and stay reduced longer, after total hip replacement (THR) than after general surgical cases (see Image 4). Furthermore, it has been demonstrated that patients who have positive venograms postoperatively tend to be those in whom circulating levels of ATIII are diminished (see Image 5).

Scope of the Problem

The nature of orthopedic illnesses and diseases, trauma, and surgical repair or replacement of hip and knee joints predisposes patients to the occurrence of venous thromboembolic (VTE) disease. These complications are predictable and are the result of alterations of the natural equilibrium mechanisms in various disease states.¹

As previously mentioned, an estimated 300,000 individuals are hospitalized annually in the United States for deep vein thrombosis (DVT) disease. This is especially significant, as up to three quarters of cases of DVT disease remain silent and do not come to medical attention. The overall incidence of DVT in the United States is estimated to be 84–150/100,000 annually. Pulmonary embolism is estimated to be responsible for about 150,000 deaths per year representing 5% of all perioperative mortality. DVT is thought to be the source of 90% of acute pulmonary emboli.

An analysis of deaths from VTE in 11,600 patients undergoing hip and knee replacement between 1976 and 1985 showed a 17-fold increase within 3 months of the surgery compared with the incidence for the rest of the year. In a study by Warwick and colleagues, the death rate from pulmonary embolism (PE) was 0.34% in 1162 patients after THR with no prophylaxis.² It is estimated that 2 to 3% of patients undergoing THR and 4-7% of patients undergoing surgery for hip fracture suffer nonfatal PE.

In a Scandinavian study by Bergqvist, PE was found in 23.6% of 1274 patients.^3,4 PE was believed to be the major causative factor in death in 6.4%. In the United States, this rate would translate to approximately 50,000-100,000 deaths annually due to PE. Similar to DVT, most cases of PE remain silent and clinically go undetected or undiagnosed. Thirty percent of patients presenting with acute PE had no prior symptoms. There is a 10-12% mortality rate reported for patients hospitalized with PE.

Postthrombotic syndrome (PTS) is a complication that occurs frequently in patients with untreated DVT. Symptoms, which may not become apparent for months or years, consist of pain, chronic swelling, venous insufficiency, induration, or skin breakdown. Kahn reported that in treated individuals with DVT, the incidence of PTS at 2 years is approximately 25% to 50% and up to 75% after 5-7 years.⁵

Untreated general surgical patients have a postoperative risk of DVT of 19-25% depending on the method used for diagnosis (see Table 1).

Open table in new window

In comparison, the DVT rate among high-risk orthopedic patients is substantially greater. Untreated patients following THR have a DVT rate of 50-60%, with a 20-30% proximal DVT rate. The overall incidence is even greater in patients after total knee replacement (TKR), with a 60-85% DVT rate, although, the proximal DVT rate is less, 9-20%. Patients with hip fractures have a DVT rate of 30-60%, with a proximal DVT rate of up to 36%. In these same series, the risk of fatal PE ranges 0.4-12.9%. Therefore, because VTE disease is often silent, with significant consequences in morbidity and mortality in untreated DVT and PE, offering those high-risk patients protection in the form of DVT prophylaxis is obligatory.⁶

A survey of the American Association of Hip and Knee Surgeons indicated that 100% of their members were providing some method of DVT prophylaxis in high risk cases.⁷ In 2003, the Hip and Knee Registry reported that 1 or more types of thromboprophylaxis were used in 99% of patients. About 89% of total hip arthroplasty (THA) patients and 91% of total knee arthroplasty (TKA) patients received therapy matching the prophylaxis recommendations of the American College of Chest Physicians (ACCP).⁸

Risk Analysis and Stratification in Orthopedic Surgery

Certain conditions appear to predispose individuals to thromboembolic disease.The ACCP Seventh Consensus Conference on Antithrombotic and Thrombolytic Therapy, published in September, 2004, listed the major risk factors. Joseph Caprini, MD, and others, attempted to quantify these factors in a reproducible manner to assist the clinician in performing preoperative risk-factor assessment.⁹ Risk factors are grouped according to severity and are added to produce an overall risk factor score, which corresponds to low through very high potential for deep vein thrombosis (DVT) development. (See Table 2)

• Risk factor assessment – 1 Point
  • Minor surgery
  • Age 41-60
  • History of major surgery within 1 month
  • Pregnancy or postpartum within 1 month
  • Varicose veins
  • Inflammatory bowel disease
  • Swelling of legs
  • Obesity (BMI >25)
  • Oral contraceptives, patch or hormone replacement therapy
• Risk factor assessment – 2 Points
  • Age over 60 years
  • Malignancy or current chemotherapy or radiation therapy
  • Major surgery (>45 minutes)
  • Laparoscopic surgery (>45 minutes)
  • Confined to bed more than 72 hours
  • Immobilizing cast less than 1 month
  • Central venous access less than 1 month
  • Tourniquet time over 45 minutes
• Risk factor assessment – 3 Points
  • History of DVT or pulmonary embolism (PE)
  • Family history of thrombosis
  • Age over 75 years
  • Factor V Leiden /activated protein C resistance
  • Medical patient with risk factors of myocardial infarction, congestive heart failure, or chronic obstructive pulmonary disease
  • Congenital or acquired thrombophilia
• Risk factor assessment – 5 Points
  • Major, elective lower extremity arthroplasty TKR, THR
  • Hip, pelvis or leg fracture within 1 month
  • Stroke within 1 month
  • Multiple trauma within 1 month
  • Acute spinal cord injury with paralysis within 1 month

Open table in new window

These factors include those that diminish venous flow or return, increase viscosity, or alter mobility. Age is one of the most easily definable factors.¹⁰ The risk of DVT increases in exponential fashion with increasing age (see Image 6).

In addition, recognition of certain hypercoagulable states caused by congenital hematologic abnormalities is growing. They include the following:

• Activated protein C resistance (factor V Leiden)
• Prothrombin variant 20210A
• Antiphospholipid antibodies (lupus anticoagulant and anticardiolipin antibody)
• ATIII, protein C, protein S, heparin cofactor II deficiency
• Dysfibrinogenemia
• Plasminogen deficit
• Hyperhomocysteinemia
• Polycythemia vera
• Myeloproliferative disorders

Patients with a prior history of DVT or PE have demonstrated an incidence of a congenital thrombophilic abnormality as high as 25%. Congenital abnormalities are as follows:

• Protein C resistance gene - Mutation factor V, present in 3-5% of population (white, northern European)
• Prothrombin gene mutation - 1-3% of population; increases risk of DVT 2-3 times
• Past history of DVT or PE - 25% have congenital thrombophilia mutation.

Diagnosis of DVT and PE

Screening methods in use for the diagnosis of deep vein thrombosis (DVT) range in their specificity and selectivity. Studies that report on the incidence of DVT vary considerably, depending on the endpoint of diagnosis used.¹¹

Studies indicate that a reliable and inexpensive test, the D-dimer assay test can be used as a rapid screening measure in cases where leg swelling exists in the face of equivocal or negative clinical or radiologic findings. D-dimer is formed when cross-linked fibrin contained within a thrombus is broken down by plasmin. The assay test is highly sensitive test, with a negative predictive value (it rules out DVT if negative). Forty percent of patients with a negative clinical examination and negative D-dimer test require no further clinical evaluation. Similarly, subjects with an elevated D-dimer test at 1 month following anticoagulant cessation have a significantly higher risk of recurrent VTE.¹²

In the hands of an experienced technician, compression ultrasonography and duplex Doppler ultrasonography can be used to reliably image the venous vasculature from calf to groin.¹³  Compressibility of the vessel to be imaged results in diminished blood flow beneath the transducer. Thrombosed vessels are noncompressible. The iliac and pelvic veins and vessels below the popliteal vein are difficult to access and therefore are less reliable in the detection of thrombi. Nevertheless, ultrasonography remains the first choice for diagnosis in most institutions, because it is noninvasive and relatively easy to use. Proximal thrombus diagnosis is 95-99% accurate, but this falls to under 50% for calf thrombi.

Iskander and colleagues reported that even with adequate anticoagulant prophylaxis, 36% of trauma patients with calf DVT had thrombus propagation while in hospital.¹⁴ Serial scans for calf DVT may therefore be indicated if an initial DVT is identified.

The use of duplex Doppler imaging as a screening test for asymptomatic patients on discharge is unreliable in predicting future DVT (see Image 7).

Venography has been used as the criterion standard for DVT diagnosis, and it is accurate in demonstrating distal and proximal thrombi (see Image 8). The disadvantages of venography are that it is painful and invasive. In addition, it requires the use of iodinated contrast agent, which may provoke allergic reactions in susceptible individuals. Renal insufficiency may be a relative contraindication to the use of these agents. In patients with severe leg edema or obesity, it can still provide accurate results. Disagreement also exists about the clinical significance of the thrombi detected. However, despite these shortcomings, it is still the method of comparison used in many of the clinical studies of the effectiveness of prophylactic regimens. Spiral computed tomography (CT) venography has been utilized in patients with leg swelling and equivocal or negative Doppler ultrasonography.

Diagnosing intrapelvic thrombi has been difficult until recently, with the advent of magnetic resonance (MR) venography. This technique has provided physicians with a sensitive and specific method to diagnose intrapelvic and proximal DVT in patients in whom venography is not possible or is unlikely to be diagnostic. Magnetic resonance imaging (MRI) cannot be used in patients with magnetic metallic implants or implanted devices.

Pulmonary embolism (PE) is commonly diagnosed on the basis of symptoms of chest pain and dyspnea (shortness of breath). Electrocardiography (ECG) may demonstrate ST-segment changes. The arterial oxygen saturation (PaO₂) level may be lowered. All or none of these findings may be present, and the embolization may remain subclinical or silent.

DVT Prophylaxis

Mechanical methods of DVT prophylaxis

Mechanical methods have been shown to be a useful adjunct to anticoagulation therapy in reducing the incidence of deep vein thrombosis (DVT). In a study of the efficacy of IPC in multiple postoperative patient groups versus no use of prophylaxis, Urbankova reported in 2005 that the incidence of DVT was reduced by 60%.¹⁵  However, the use of mechanical means of prophylaxis alone is not effective in cases of moderate or high risk. Modalities include passive devices, such as graduated compression (elastic) knee or thigh-high stockings (GCS); active (external pneumatic compress or IPC) devices;¹⁶ or venous foot pumps (VFP).¹⁷

IPC devices are designed to decrease venous stasis, improve blood flow velocity, and increase the level of circulating fibrinolysins. IPC devices have the advantage of requiring no monitoring, with no increase in bleeding. Generally, they are well tolerated. These devices come in a wide variety and can be applied to the foot, calf, or thigh. A study comparing asymmetrical versus circumferential intermittent compression devices following TKR seemed to support the asymmetrical device.¹⁸

Patient compliance is an issue with IPC devices, and the efficacy is dependent on the time of use. Evidence from clinical trials has also shown that, while the rate of distal thrombi is reduced significantly, proximal thrombi are not. This finding may lead to a false sense of security because, although the total number of deep venous thrombi may be similar to the numbers observed with pharmacologic prophylaxis, the proportion of the relatively more dangerous proximal clots is increased (see Table 3).

Shorter lengths of hospital stays make the use of mechanical methods alone ineffective in preventing DVT in the critical weeks after joint replacement.

Open table in new window

Although all 3 types of mechanical compression reduce the incidence of DVT below that found when prophylaxis is absent, these modalities are generally less effective at producing such reductions than are pharmacologic methods. Shorter lengths of hospital stays make the use of mechanical methods alone ineffective in preventing DVT in the critical weeks after joint replacement. No mechanical prophylaxis method has been shown to reduce the risk of pulmonary embolism or death. The use of IPC devices is therefore recommended primarily as an adjunct to anticoagulant-based prophylaxis or in patients who are at high risk of bleeding.

Pharmacologic methods

Many pharmacologic agents are currently available to prevent thrombosis. Agents that retard or inhibit the process belong under the general heading of anticoagulants. Agents that prevent the growth or formation of thrombi are properly termed antithrombotics and include anticoagulants and antiplatelet drugs, whereas thrombolytic drugs lyse existing thrombi. For the importance of prevention, see Hull's 1998 study.^19,20,21

Platelet-active drugs

Platelet-active drugs such as aspirin or cyclooxygenase-1 (COX-1) inhibitors have been used to prevent thrombosis.²² Aspirin is effective as a platelet inhibitor at very low doses (50-100 mg/d). This dose is significantly less than that necessary to produce an anti-inflammatory effect. A meta-analysis of the effect of aspirin following THR completed in 1994 had equivocal results.^23,24

A large study performed in Europe, the Pulmonary Embolism Prevention (PEP) study, included 13,356 patients with hip fractures and 4088 patients with THR. The patients were given a 160-mg/d dose compared with placebo and evaluated at day 35. Approximately 40% of the patients also were given LDH or LMWH. With this regimen, the overall DVT rate was decreased 30% compared with placebo, and the overall pulmonary thrombosis (PE) rate was decreased by 40%.

In a concomitant study of 4088 patients with THRs, a 25% reduction of DVT was observed in comparison with the placebo control. No decrease was noted in the rate of PE. This trial did not show a clear benefit to using aspirin as the primary method of venous prophylaxis in patients undergoing either total hip or total knee surgery. The Seventh ACCP Conference did not recommend the use of aspirin alone as a prophylactic agent for any patient group, because aspirin is less effective than other options. However, reports by Lotke and Lonner and by Berend and Lombardi have suggested that the use of aspirin combined with optimally used IPC devices may be effective in some circumstances in preventing fatal PE.^25,26

Coumarins

Coumarins are a class of oral anticoagulant drugs, which act as antagonists to vitamin K. The mechanism of action is to interfere with the interaction between vitamin K and coagulation factors II, VII, IX, and X. Vitamin K acts as a cofactor at these levels. Coumarins produce their anticoagulant effect by inhibiting the carboxylation necessary for biologic activity. Warfarin is a mixture of 2 isomers; the R and S forms in roughly equal proportions. Warfarin is absorbed rapidly from the GI tract and bound to plasma proteins. Although it has high bioavailability, warfarin requires 36-72 hours to reach a stable loading dose. The dose response in patients taking warfarin is variable, and it is influenced by various genetic and environmental factors.

Numerous drug interactions and disease states may affect its pharmacokinetics. Coumadin, therefore, requires continuous laboratory monitoring.

The effectiveness of Coumadin anticoagulation is measured by determining the prothrombin or protime against a standard control. The use of international normalized ratio (INR) has supplanted the protime for hospital use. INR uses a standardized protime, which allows for comparisons between hospitals and laboratories.

For DVT prophylaxis, the optimal INR level is between 2.0 and 3.0, with a target of 2.5. When used for DVT prophylaxis after THR, warfarin reduces total DVT by 60% and proximal DVT by 70%. Disadvantages of warfarin use include its long onset of action, the necessity to monitor INR values frequently to obtain stable dosage, the long half-life that may require vitamin K reversal in incidents of hemorrhage, frequent drug and dietary interaction, and variable patient response. Hemorrhagic complications are reported in up to 3-5% of patients on warfarin prophylaxis.

If adjusted-dose warfarin is to be used, it is started the night prior to surgery and continued postoperatively during the discharge period. INR target levels usually are not reached until the third postoperative day.

Heparins

Standard unfractionated heparin (UFH) is recognized as an acceptable anticoagulant modality. UFH has been used for this purpose in various forms since its discovery by McLean in 1916. UFH acts in conjunction with a circulating plasma cofactor, ATIII and, in its presence, catalyzes the inactivation of factors IIa, Xa, IXa, and XIIa.

By inactivating thrombin, heparin not only prevents fibrin formation but also inhibits thrombin-induced activation of factor V and factor VIII. Of these, factors IIa and Xa are most sensitive. Therefore, heparin has anticoagulant and antithrombotic properties.

Heparin is a heterogeneous mixture of molecules that contain a range of molecular weights of 3,000-30,000, with an average of approximately 15,000. Only one third of the heparin molecules have an active binding site for ATIII, and this fraction is responsible for most of the anticoagulant activity. Heparin is effective when given by intravenous (IV) or subcutaneous (SC) administration but is inactivated in the GI tract. Heparin has a rapid onset of action, its half-life is brief in comparison to warfarin, and it binds to platelets, endothelial cells, and macrophages in vivo. Therapeutic levels of heparin are measured by the activated partial thromboplastin time (aPTT). Because of the rapid clearance of heparin from the bloodstream, therapeutic levels (aPTT of 1.2-1.5 times control) are more likely achieved with continuous IV infusion.

Postoperative DVT prophylaxis with UFH usually is achieved by administering a bolus of 5000 U every 8 hours. This LDH regimen results in a 60-70% reduction of DVT and PE in low- or moderate-risk patients. However, this method is not as effective in patients who are at high risk for development of DVT or PE. In these patients, adjusted-dose heparin with aPTT monitoring is preferred to maintain the desired anticoagulant level. Studies have demonstrated a high hemorrhagic complication rate of 8-15% when this method is used for postoperative DVT prophylaxis.

Heparin overdosage is reversible with protamine sulfate, which itself is an anticoagulant. Each milligram of protamine sulfate can neutralize approximately 100 U of heparin activity. It must be administered very slowly by IV infusion over a 10-minute period in doses not to exceed 50 mg. Because heparin is cleared rapidly from the circulation, the amount of protamine required decreases rapidly as the time from initial heparin administration increases. The final dosage required is titrated according to coagulation studies.

Disadvantages of UFH therapy include variable pharmacokinetics, the requirement for aPTT monitoring for adjusted-dose regimens, short half-life and low bioavailability, and lack of oral dosage form (although an oral form is currently in clinical trials). In addition, a small percentage of patients (2-4%) are susceptible to the development of heparin-induced thrombocytopenia (HIT), which is an antibody-mediated adverse reaction that can cause venous and arterial thrombosis. HIT is heralded by an otherwise unexpected fall in platelet count of greater than 50% from prior levels. HIT can result in disseminated intravascular coagulation and gangrene in severe cases. Treatment with danaparoid sodium or recombinant hirudin, such as lepirudin, may be effective in life-threatening cases.

A 2006 comparison study by McGarry and colleagues of outcomes of thromboprophylaxis between an LMWH (enoxaparin) and UFH revealed a 74% lower incidence of VTE in the LMWH group.²⁷ There was no significant difference in side effects, deaths in the hospital, or economics.

Low-molecular-weight heparins

LMWHs are manufactured when standard heparin is treated by a variety of enzymatic or chemical methods to select those lower molecular weight moieties that contain the active ATIII binding site. The average molecular weight of fractionated heparin is 4500 in comparison to the usual 15,000. The molecular weight threshold under which anti–factor Xa activity is maximized is 5400 Da.

The polysaccharide side chain of the heparin molecule is decreased from 18 U to approximately 13 U. As the length of the side chain is decreased, the ability of the molecule to prolong the aPTT is lost, but the ability to complex with ATIII is retained. LMWHs do not require monitoring of either aPTT or INR (see Image 13).

There are several commercially available LMWH medications, including the following:

• Enoxaparin (Lovenox) – Dosage 30 mg sc every 12 hours starting 12-24 hours postoperative
• Dalteparin (Fragmin) – 5000 IU sc daily starting 12-24 hours postoperative³⁰
• Danaparoid (Orgaran) – 750 U sc every 12 hours starting 12-24 hours postoperative
• Nadroparin (Fraxiparine) – 38 U/kg sc daily starting 12-24 hours postoperative
• Tinzaparin (Innohep) – 75 U/kg/d sc starting 12-24 hours postoperative
• Ardeparin (Normiflo) – Knee surgery, 50 IU/kg sc every 12 hours starting 12-24 hours postoperative
• Synthetic factor Xa Inhibitors³¹
  • Fondaparinux sodium (Arixtra) - A synthetic pentasaccharide, this agent selectively binds to antithrombin III and potentiates neutralization of factor Xa, inhibiting thrombin formation and thrombus development. Fondaparinux acts rapidly but has a long half life (18 hours). A dose of 2.5mg sc daily can be started 6-8 hours postoperative. Renal clearance requires minimal kidney function of CL_cr of greater than 30 ml/min or a weight of over 110 lb. In a controlled study by Bauer and colleagues, fondaparinux was more effective than enoxaparin in preventing DVT after TKR, but episodes of major bleeding were more frequent.³² For a comparison between fondaparinux and enoxaparin, see Turpie et al.³³

Combination Therapies

Keeney and colleagues reported on the use of early mobilization with a combination of IPC and adjusted-dose, short-duration warfarin in a group of 700 primary and revision total hip arthroplasties.³⁸ They recorded a low incidence of clinical DVT (as measured by ultrasonography) on postoperative day 3 or 4 and of clinical DVT and PE within 90 days postoperative. Further clinical investigations with larger numbers of patients are necessary to determine the optimal levels and duration of anticoagulation with the appropriate risk/benefit ratio.

DVT Prophylaxis Based on Risk Stratification Levels

In separate studies, Rosendaal, Kearon, and Bulger analyzed the relative contribution of each of these risk factors to the development of deep vein thrombosis (DVT).^39,40,41 When more than 1 risk factor is present, the risk is cumulative; however, there is as yet no good model of how the individual risk factors interact.⁴²

Nonetheless, several attempts have been made to quantify the risk factors associated with VTE disease.^9,43 The use of a checklist to stratify patients and assign them to categories of relative propensity for DVT development is helpful in deciding on an appropriate treatment regimen (see Table 2). A list can be constructed using the ACCP risk categories (see the ACCP risk factor assessment guidelines). These figures include a list of the pertinent factors, which are arbitrarily assigned a risk level of between 1 and 5. An individual aged 61-75 years is assigned 2 units; a person older than 75 years is assigned a score of 3, as is an individual with a prior history of thrombosis, inherited thrombophilia, antiphospholipid antibodies, or lupus anticoagulant. The total score is then added.

By using the risk criteria listed above, orthopedic patients can be categorized into 4 risk groups, ranging from low to very high. Appropriate methods of prophylaxis may be applied to each level, as follows:

• Low-risk patients - These patients have a score of 1 or less. They are patients below age 40 years who are undergoing a minor surgical procedure and have no additional risk factors. The risk of calf DVT in this group is estimated to be 2-5% without prophylaxis, and the risk of clinical pulmonary thrombosis (PE) is 0.2% No specific prophylaxis is required in this group other than early and aggressive mobilization.
• Moderate-risk patients - These individuals have a score of 2 or less. They are patients in the above group who have additional risk factors or are persons aged 40-60 years who are undergoing nonmajor surgery and have no additional risk factors. Other risk factors are surgery requiring a tourniquet (eg, arthroscopy), lower-extremity fractures, cast immobilization, or spinal surgery. Major surgery in patients younger than 40 years poses a moderate risk of DVT, which is estimated at 10-20%. The risk of clinical PE in this group is 1-2%. Successful prevention strategies in this group consist of low-dose unfractionated heparin (LDUH; q12h), LMWH (<3,400 U daily), and GCS or IPC.
• High-risk patients - Patients in this category have a score of 3 or 4. They include persons who are older than 60 years, as well as patients aged 40-60 years who have additional risk factors, such as prior VTE, malignancy, or hypercoagulability. The risk of calf DVT is estimated at 20-40% in this group, with clinical PE occurring in 2-4%. Successful prevention strategies in this group consist of LDUH (q8h), LMWH (>3,400 U daily), with or without IPC.
• Highest-risk patients - These patients have a score of 5 or greater. They are persons older than 40 years who have additional risk factors, who are undergoing hip or knee replacement surgery, or who have suffered hip fracture, open lower-leg fracture, multiple trauma, or spinal cord injury (SCI). Hip fracture patients have the highest risk of dying from a fatal PE. Additional risk factors may include a history of VTE, malignancy, or hypercoagulable state. These risk factors carry an estimated risk of calf DVT of 40-80% without prophylaxis, with clinical PE occurring in 4-10% and fatal PE in 0.2-5%. Successful prevention strategies include LMWH (>3,400 U daily), fondaparinux, and coumarins (INR 2-3). Dose-adjusted LDUH or LMWH may be used with or without IPC/GCS.

The ACCP provides the following recommendations for the treatment of patients following elective total hip replacement:

• LMWH preoperatively 12 hours before surgery and/or 12-24 hours after surgery or 4-6 hours after surgery at one half the dose initially, followed by a full dose on the next day. Alternatively, fondaparinux (2.5 mg) started 6-8 hours postoperative or Coumadin started preoperatively or after surgery (INR target 2.5, range 2–3). Data from 5 trials reported at the Seventh ACCP Conference showed that LMWH rates of DVT were 33% lower than with Coumadin but that bleeding rates were somewhat higher. Overall rates were equivalent at 3 months. In the large European pentasaccharide trial, fondaparinux (2.5 mg postoperative) was compared with enoxaparin (40 mg preoperative and postoperative), with reduction of overall VTE rates by more than 50% and equivalent bleeding risk.⁴⁴
• LDH, aspirin, dextran, or GCS or IPC alone are not recommended.

The following are ACCP recommendations for specific treatment of patients following TKR:

• LWMH 12-24 hours postoperative, fondaparinux 2.5 mg started 6 to 8 hours postoperatively, or adjusted-dose warfarin administered preoperative and/or postoperatively and with an INR range of 2-3 (target INR of 2.5)
• The optional use of IPC devices is an alternative to anticoagulant prophylaxis when they are used intraoperatively or are immediately employed postoperatively; the devices are worn continuously until the patient is fully ambulatory. However, this is not possible under most circumstances due to shortened hospital stays, poor patient compliance, and an inability to use the devices continuously due to the necessity for early mobilization. IPC is most often used, therefore, as an adjunct to anticoagulant prophylaxis.
• LDUH and VFPs are not recommended for sole use as prophylaxis

ACCP recommendations for patients with hip fractures are as follows:

• LMWH or fondaparinux or adjusted-dose warfarin immediately administered postoperatively, with a target INR of 2.5 (range of 2-3) if bleeding is controlled; LDUH may be alternative (limited data)
• Aspirin alone is not recommended

If surgery is delayed, then prophylaxis with LDUH or LMWH should be initiated at the time of admission and discontinued prior to surgery. The immediate action, reversibility, and short half-life of these agents make them ideal for interim prophylaxis during the period between admission and surgery.^45,46

ACCP recommendations for knee arthroscopy are as follows:

• Clinicians should not use routine thrombosis prophylaxis to treat patients undergoing arthroscopic knee surgery.
• Early mobilization alone is recommended.
• Patients with additional preexisting risk factors for VTE or prolonged tourniquet time should be given LMWH for prophylaxis.

For an excellent monograph on the use of prophylaxis in knee arthroscopy, see the article by Muntz and Friedman.⁴⁷

ACCP recommendations for elective spine surgery are as follows:

• For patients who have no additional risk factors, antithrombotic prophylaxis following elective spine surgery is not recommended.
• Patients at high risk for developing postoperative VTE may be treated with LDUH, LMWH, or perioperative IPC.
• Multiple risk factors may require the combined use of mechanical and pharmacologic measures.

Epstein reported a 2.8% incidence of DVT and a 0.7% incidence of PE in 139 patients, following multilevel lumbar spine surgery treated with IPC and early mobilization.⁴⁸ For prevention of thromboembolism in spinal cord injuries, see the recommendations of the Consortium for Spinal Cord Medicine.⁴⁹

Anesthesia

Anticoagulant prophylaxis should be used with caution in patients receiving spinal or indwelling catheter epidural anesthesia. Although the risk of spinal hematoma is very small (0.0025% with spinal and 0.03% with epidural), care should be taken to delay the initiation of thromboprophylaxis for at least 2 hours after catheter removal. Patients with known bleeding disorders should not receive preoperative prophylaxis if they are to receive spinal anesthesia. In cases of traumatic spinal tap with bloody spinal fluid, postoperative administration of thromboprophylaxis should be done with caution.

Initiation of Prophylaxis Therapy

In Europe, it is common practice to begin anticoagulant prophylaxis 10-12 hours prior to surgery. In North America, the practice is to begin treatment 12-24 hours following surgery. The Seventh ACCP conference suggested that for most major, elective orthopedic surgery, the first dose of LMWH may be administered either before or after surgery, although meta-analyses suggest little advantage for the preoperative initiation. A study by Bergqvist and Hull seems to suggest that starting half dose anticoagulation 6 hours after surgery may deliver more effective prophylaxis without a significant increase in bleeding risk. Patients with a high risk of bleeding should have the first postoperative dose delayed 12-24 hours after surgery. In a meta-analysis of 33 trials, Leonardi and colleagues reported an approximately 3% rate of bleeding complications from DVT prophylaxis in which the bleeding was severe enough to require a change of care.⁵⁰

Extended Anticoagulant Treatment

Studies have demonstrated that patients undergoing THR remain at increased risk for the development of deep vein thrombosis (DVT) for up to 3 months postoperatively. Two definite peaks have been demonstrated in DVT initiation, POD 4 and POD 13-14 (see Image 14).

In patients undergoing THR or hip fracture surgery, as well as in other high-risk patients, prophylaxis should last 28-35 days or longer, postoperative. This therapy with LMWH, fondaparinux, or Coumadin can reduce total and proximal DVT by at lease 50% without increasing major bleeding events. Patients with a history of DVT or PE and those who have inherited thrombophilia may require even longer treatment. Patients with acute SCI have a significantly increased risk of dying from a PE during the first 3 months after injury, and extended prophylaxis during this period is recommended. The risk decreases to 19 times the normal risk for years 2-5, and to 8.9 times the normal risk after year 5.

A study by Barrett and colleagues of bilateral TKR suggested that the adjusted risk of pulmonary embolism in the 3 months following a simultaneous procedure (1.44%) is 80% higher than in the 3 months after a single procedure (0.81%).⁵¹

Open table in new window

*From 3 studies, 876 subjects.

Long-term anticoagulant prophylaxis or treatment of established DVT usually requires continued treatment after hospitalization. At the time of this writing, there are no FDA-approved pharmacologic agents that can be given orally without laboratory monitoring.

Outpatient therapy with LMWHs is safe and effective. However, a mechanism should be available for either self-administration of SC medication or to teach caregivers to administer the medication. Visiting nursing services may also provide this service. Unfortunately, many insurance companies do not cover out-of-hospital expenses. For this reason, a common practice is to overlap the initiation of oral warfarin therapy with LMWH treatment while the patient is still hospitalized. When the INR has been at a therapeutic level for 48 hours, LMWH can be discontinued.

Contraindications to anticoagulant therapy

Complications of anticoagulant treatment include major and minor bleeding, hematoma formation, compartment syndrome, and HIT. Major bleeding is defined as hemorrhage that alters the clinical course of the patient's treatment or changes the clinical outcome. Major bleeding may prolong the hospital stay, necessitate a return to the operating room, or result in unexpected transfusion. DVT prophylaxis should be delayed or terminated in these cases. Rehabilitation or mobilization may also be delayed.

Absolute contraindications to anticoagulant therapy include active hemorrhage or an unstable condition of a patient with multiple traumas. Patients with HIT should not be given standard anticoagulants; instead, they should be treated with one of the newer methods specifically approved for such use. Warfarin should not be used in patients who are pregnant. Patients who have sustained severe head trauma or acute SCI should not undergo anticoagulation. Indwelling spinal catheters should be withdrawn, with a 2-hour delay before initiating anticoagulation.

Relative contraindication to anticoagulation includes patients with previous history of cerebral or GI hemorrhage. Patients with history of thrombocytopenia or coagulopathy may have circulating heparin antibodies. Patients with active intracranial lesions or neoplasm may be at increased risk for bleeding on anticoagulant therapy. Proliferative retinopathy also may put patients at increased risk for intraocular bleeding.

While it is beyond the scope of this discussion, the employment of inferior vena caval filters is useful in patients who have a contraindication to the use of anticoagulants, who have major bleeding, or in whom treatment has not been successful. An IVC filter provides a mechanical barrier to clot propagation or migration from the lower extremities. The ability to use temporary or removable filters has broadened the indications for these devices in situations where short-term anticoagulation is called for, such as in major emergency or elective surgery.
 
A 2006 study by Mivahara of 33 removable filters indicated the mean duration of filter placement to be 10+/-7 days, and there was no case of PE during filter protection.⁵² Filter-related complications, however, occurred in 27% of patients. Permanent filter protection from PE was significant in short-term studies, but it became less effective after 1-2 years.⁵³

Bates and Ginsberg have reviewed the current treatment of established DVT.⁵⁴

Emerging Treatments in DVT Prophylaxis and Conclusion

An ideal anticoagulant should be easy to administer (preferably oral), should be effective and safe with minimum possible complications or adverse effects, have rapid onset, have a therapeutic half-life, and require minimal or no monitoring. The action of the anticoagulant should be predictable with few drug or dietary interactions, and it should be reversible. The drug should also be inexpensive.

These criteria are often difficult to achieve. Several anticoagulant agents exist today, and each of them incorporates some of these characteristics, but no one combines all these attributes.

Current research in anticoagulants involves investigations into drugs that act on various phases of the coagulation cascade. For convenience, the authors can arbitrarily divide the process into 3 phases: the initiation phase, the propagation phase, and the thrombin activity phase.

Drugs under investigation that act in the initiation phase include tissue factor pathway inhibitors (TFPIs) and nematode anticoagulant peptide (NAPc2). These drugs act through inhibition of the factor VIIa/tissue factor complex.

A number of new synthetic, direct or indirect antithrombin-dependent inhibitors of FXa are being tested. These have similarities to the currently approved fondaparinux. Phase II dose-finding trials in which the metapentasaccharide idraparinux was administered subcutaneously once each week to prevent the development of secondary VTE have been completed. A second class of orally active direct FXa inhibitors, which includes razaxaban, is also undergoing clinical phase II trials.

Drugs that act on the third stage of the coagulation cascade, the thrombin activity phase, include the direct thrombin inhibitors. These drugs specifically inactivate thrombin and are independent of antithrombin ATIII. Included in this group are the hirudins and its derivatives made by recombinant DNA techniques. Originally, hirudin was isolated from leech salivary gland tissue. The new drugs include bivalirudin (Angiomax) and lepirudin. A randomized, multicenter, double-blind study of hirudin versus heparin in patients with THRs demonstrated that deep vein thrombosis (DVT) and proximal DVT rates were decreased substantially in the hirudin group. Another class of drugs acting at the third level of the coagulation cascade includes the noncovalent inhibitor argatroban, which is a carboxylic acid derivative that has been approved for use in the treatment of HIT.

A promising new direct thrombin inhibitor consisting of an oral prodrug of melagatran named ximelagatran (Exanta; AstraZeneca) was reported in 2003. It is rapidly absorbed through the GI tract, where it is converted to its active form, melagatran. It does not require monitoring; it has a rapid onset of action, a predictable dose-response and a therapeutic half-life. Also, like the other direct thrombin inhibitors, it does not affect the aPTT or PT.

The results reported by Francis and colleagues in the New England Journal of Medicine s howed ximelagatran and warfarin did not differ significantly with respect to the incidence of major or minor bleeding.⁵⁵ The report also determined that ximelagatran was significantly more effective in preventing DVT after TKR than was warfarin.

In the US, 4 studies have shown that postoperatively initiated ximelagatran (24 mg twice daily) had efficacy similar to that of enoxaparin or warfarin in the prevention of VTE in patients undergoing hip or knee replacement. Overall, the incidence of bleeding events and transfusion rates were not markedly different from those of comparator anticoagulants. Some patients experienced transient elevations of liver enzyme levels, which returned to normal after cessation of treatment.⁵⁶

Large phase III clinical trials have been published describing the investigational oral anticoagulant, rivaroxaban, for prevention of thromboembolism following total knee or total hip arthroplasty. Rivaroxaban is administered once daily and has shown significant superiority in preventing deep-vein thrombosis, nonfatal pulmonary embolism, or death compared with subcutaneous enoxaparin following arthroplastic surgery. If approved in the United States, it will be the first orally active direct inhibitor of coagulation factor Xa.^34,35,36,37

In 2004, the FDA Cardiovascular and Renal Drug Advisory Committee (CRAC) reviewed the ximelagatran clinical program. The 3 indications that were proposed included the prevention of VTE in patients undergoing TKR, the prevention of stroke and other thromboembolic complications associated with atrial fibrillation, and the secondary prevention of VTE after an episode of acute DVT. The committee reviewed data from 30,698 subjects and included 5 phase III studies. Ximelagatran hepatic toxicity was a key feature that led the CRAC to conclude that the benefit/risk ratio of ximelagatran was unfavorable for the 3 proposed indications. A 2005 report by Colwell and Mouret, however, indicated that melagatran/ximelagatran has been approved in the European Union for the prevention of VTE in patients undergoing elective hip or knee replacement surgery.⁵⁷ In 2006, Boudes reported on the challenges and risk analysis to be learned from the ximelagatran FDA CRAC.⁵⁸

Conclusion

Major surgical and high-risk orthopedic procedures place patients at risk for DVT and VTE, including pulmonary embolism (PE). Complications of DVT include postphlebitic syndrome or death from PE. Therefore, prophylaxis with anticoagulant medications, as well as the adjunct use of mechanical devices, is essential. The most effective treatment protocol for a patient must be determined on a case-by-case basis and account for the risk-benefit ratio in each situation. A risk stratification protocol, such as that developed by the ACCP, is recommended to determine the appropriate level and method of treatment.

Preventing VTE is always a tradeoff between the potential life-saving benefit of prophylaxis and the risk of hemorrhage. It is important to emphasize, therefore, that even with adequate anticoagulant prophylaxis, DVT can and does develop. A 2005 study by Schiff revealed a 14% incidence of VTE following major orthopedic procedures, particularly TKR, where standard prophylactic measures had been applied.⁵⁹ Continued vigilance and a high index of suspicion on the part of the medical staff is called for in this group of patients.

For excellent patient education resources, visit eMedicine's Circulatory Problems Center and Lung and Airway Center. Also, see eMedicine's patient education articles Deep Vein Thrombosis (Blood Clot in the Leg, DVT) and Pulmonary Embolism.

Multimedia

(Enlarge Image)

Media file 1: Venous thrombus.

(Enlarge Image)

Media file 2: Pulmonary embolus.

(Enlarge Image)

Media file 3: Coagulation pathway.

(Enlarge Image)

Media file 4: Postoperative antithrombin III levels.

(Enlarge Image)

Media file 5: Antithrombin III levels and deep venous thrombosis formation.

(Enlarge Image)

Media file 6: Age-adjusted venous thromboembolic risk.

(Enlarge Image)

Media file 7: Doppler sonogram shows deep venous thrombosis.

(Enlarge Image)

Media file 8: Venogram shows proximal deep venous thrombosis.

(Enlarge Image)

Media file 9: Lung scan.

(Enlarge Image)

Media file 10: Spiral CT scan shows a pulmonary thrombus.

(Enlarge Image)

Media file 11: Normal pulmonary angiogram.

(Enlarge Image)

Media file 12: Positive pulmonary angiogram.

(Enlarge Image)

Media file 13: Comparison of binding sites for standard heparin and low-molecular-weight heparin.

(Enlarge Image)

Media file 14: Time course of deep venous thrombosis risk.

Keywords

deep venous thrombosis prophylaxis, deep venous thrombosis, deep vein thrombosis, DVT prophylaxis, pulmonary embolus, pulmonary embolism, PE, venous thromboembolism, VTE, thromboprophylaxis, DVT prevention, PE prevention, VTE prevention, venous stasis, Virchow triad, Virchow's triad, heparin, low-dose heparin, LDH, low-molecular weight heparin, LMWH, unfractionated heparin, UFH, intermittent pneumatic compression, IPC, elastic stockings, ES, post-thrombotic syndrome, postthrombotic syndrome, PTS, heparin-induced thrombocytopenia, HIT

Related eMedicine topics

Bedside Ultrasonography, Deep Vein Thrombosis

Deep Venous Thrombosis and Thrombophlebitis

Deep Venous Thrombosis, Lower Extremity

Deep Venous Thrombosis, Upper Extremity

Clinical guidelines

Suspected upper extremity deep vein thrombosis (DVT). American College of Radiology.  1995 (revised 2005).  5 pages.  NGC:004795
 
Guidelines for the management of severe traumatic brain injury. Deep vein thrombosis prophylaxis. Brain Trauma Foundation - Disease Specific Society.  2000 (revised 2007).  5 pages.  NGC:005773
 
Venous thromboembolism.
Institute for Clinical Systems Improvement - Private Nonprofit Organization.  1998 Jun (revised 2007 Jun).  91 pages.  NGC:005885

Prevention of deep vein thrombosis.
University of Iowa Gerontological Nursing Interventions Research Center, Research Translation and Dissemination Core - Academic Institution.  1999 Jun 10 (revised 2006 Feb).  40 pages.  NGC:004960

Management of venous thromboembolism: a clinical practice guideline from the American College of Physicians and the American Academy of Family Physicians.
American College of Physicians - Medical Specialty Society.  2007 Feb 6.  8 pages.  NGC:005524

Clinical trials

Deep Vein Thrombosis Treatment With the Oral Direct Factor Xa Inhibitor Rivaroxaban in Patients Using a Strong CYP 3A4 Inducer

Selective D-Dimer Testing Compared With Uniform D-Dimer Testing in the Diagnosis of Deep Vein Thrombosis (SELECT)

Residual Vein Thrombosis and the Optimal Duration of Low Molecular Weight Heparin in Cancer Patients With Deep Vein Thrombosis

Prophylaxis Against DVTs After Hip and Knee Surgery

Oral Direct Factor Xa Inhibitor Rivaroxaban In Patients With Acute Symptomatic Deep-Vein Thrombosis Without Symptomatic Pulmonary Embolism: Einstein-DVT Evaluation

• Search for CME/CE on This Topic »

• Lung and Airway Center
• Pulmonary Embolism Overview
• Circulatory Problems Center
• Leg Blood Clot Overview
• Leg Blood Clot Symptoms
• Leg Blood Clot Treatment
• Leg Blood Clot Causes