CURRENT CONCEPTS OF THROMBOSIS
INTERNATIONAL CONSENSUS RECOMMENDATIONS Summary Statement and Additional Suggested Guidelines Rodger L. Bick, MD, PhD, FACE', and Sylvia K. Haas, MD
The primary purpose of a consensus conference is to provide optimal guidance to the clinician or practitioner in the setting of clinical practice. If the practice guidelines generated by consensus conferences are successful in this regard, they assist clinicians in decision making for their patients, and they provide protection against unjustified malpractice actions.14 Since the Consensus Conference of the National Institutes of Health in 1986, the developments in the field of prevention of deep venous thrombosis (DVT) and pulmonary embolism (PE) were mainly characterized by a more specific and extended use of new prophylactic agents such as low-molecular-weight heparins (LMWHs) as well as the perception that outpatients may also be at risk for thromboembolic complications. Therefore, in November 1991, a European Consensus Conference on the prevention of venous thromboembolism was held in Windsor, United Kingdom, to analyze the risk constellation(s) of various patient populations and to provide recommendations for primary prophylaxis in general surgery, urology, neurosurgery, orthopaedic and trauma surgery, obstetrics and gynecology, and medical patients. In addition, the costeffectiveness of the various methods of prophylaxis was highlighted, and questions of secondary prevention were addressed. The conference
From the Departments of Medicine and Pathology, University of Texas southwestern Medical Center (RLB), Dallas, Texas; and the Department of Medicine, Institute for Experimental Surgery, Technical University of Munich (SKH), Munich, Germany
MEDICAL CLINICS OF NORTH AMERICA VOLUME 82 NUMBER 3 MAY 1998
BICK & HAAS
was organized according to acknowledged guidelines of consensus conference organization, including a follow-up program and p~blication.~ To avoid conflict of interest and methodologic bias, the organizer was independent, and the participating experts were required to formulate their provisional standpoints before the conference. During the first part of the meeting, the delegates presented their views in plenary sessions. Controversial points were discussed in working groups, and the results were discussed in plenary sessions. Consensus, or lack thereof, was documented, and the remaining open questions were formulated to be answered by future studies. The published statements and recommendations are based on different levels of evidence. In the meantime, the recommendations have been carefully reviewed and updated by experts who represent the extensive range of opinions on the subject. This latest 1997 International Consensus Statement, which also critically reviewed the meanwhile published American College of Chest Physicians (ACCP) Consensus statement of 1995,'. has been p~blished.'~ Thus, the statements and recommendations of the International Consensus Statement and the ACCP Consensus provide a solid basis for the practical use of various methods of primary and secondary prevention of thromboembolism. It is anticipated that these recommendations will constantly improve with the data generated from ongoing and future clinical trials.
PROBLEM AND THE NEED FOR PREVENTION
Numerous studies have provided evidence that patients who undergo surgery are at risk for developing venous thromboembolic complications and that general medical patients may also be at risk. Thus, an important task for the medical profession is to prevent DVT and its complications. In this task, knowledge in epidemiology and pathophysiology is important to be able to define the risk groups in which prophylaxis must be considered. The attitudes and beliefs toward prophylaxis show great regional variations. This is true for the definition of risk groups, the proportion of patients receiving prophylaxis, and the prophylactic modality chosen (Fig. 1).
Surgical Patients Based on firm data from the literature, it can be summarized that the highest risk for DVT is in patients undergoing major trauma or orthopaedic surgery, followed by abdominal surgery, in which the incidence has been found to be almost 30%. The degree of risk is increased, however, by predisposing risk factors (i.e., age, morbidity, malignancy, obesity, prior history of thromboembolism, varicose veins, recent operative procedures, and thrombophilia). These factors are further modified
INTERNATIONAL CONSENSUS RECOMMENDATIONS
EXPOSING RISK(S): HIP I KNEE ARTHROPLASTY MAJOR SURGERY GENERAL SURGERY D 30 MIN. FEMORAL FRACTURE IMMOBILIZING PLASTER CAST
GENERAL SURGERY c 30 MIN. TRAUMA / TlBlAL FRACTURES AMBULATORY PLASTER CAST ARTHROSCOPY
H/GH = 3
MODERATE = 2
LOW = I
PREDISPOSING RISK(S): HISTORY OF DVT / PE AGE D 70 YEARS T H R O M B O P H IL I A' MALIGNANCY AGE D 60 YEARS OBESITY MAJOR VARICOSITIES ESTROGENS
1.5 1.5 1.5 1.5 1.o 0.5 0.5 0.5
HEREDITARY OR ACQUIRED THROMBOPHILIA
Figure 1. Individual risk assessment.
by general care, including duration and type of anesthesia, preoperative and postoperative immobilization, level of hydration, and the presence of sepsis. Thus, the individual risk is determined by the type of surgery and an accumulation of predisposing factors (i.e., patients undergoing minor surgery but bearing several additional risk factors may also be at high risk for thromboembolic complications). As recommended by the European Consensus, the risk classification of patients according to exposing and predisposing risk factors into (1) low, (2) medium, and (3) high risk of developing thromboembolism is summarized in Table 1.3 Medical Patients
Fewer data are available on medical than on surgical patients. An increased risk of venous thromboembolism, however, has been shown by prospective studies in patients with acute myocardial infarction, patients with stroke, and immobilized general medical patients.* RECOMMENDATIONS OF PROPHYLACTIC MODALITIES IN VARIOUS SURGICAL AND MEDICAL PATIENTS Low-Risk General Surgical Patients (Minor Surgery Without Risk Factors)
The data are insufficient to make any recommendations for lowrisk general surgical patients. On the basis of risk-to-benefit ratio and
BICK & HAAS
Table 1. RISK CATEGORIES IN SURGICAL PATIENTS
High Risk General and urological surgery in patients over 40 years with recent history of DVT or PE Extensive pelvic or abdominal surgery for malignant disease Major orthopaedic surgery of lower limbs Moderate Risk General surgery in patients over 40 years lasting 30 min or more and in patients below 40 years on oral contraceptives Low Risk Uncomplicated surgery in patients under 40 years without additional risk factors Minor surgery (i.e., less than 30 min) in patients over 40 years without additional risk factors
Calf Vein Thrombosis
Proximal Vein Thrombosis
Fatal Pulmonary Embolism
DVT = Deep venous thrombosis; PE = pulmonary embolism. Data from European Consensus Statement on the prevention of venous thromboembolism. Int Angiol 11:151, 1992.
extrapolation from studies in moderate-risk patients, it is the practice in some countries to use graduated compression stockings in addition to early ambulation and adequate hydrati~n.’~ Moderate-Risk General Surgical Patients (Major Surgery, Age Over 40 Years, Without Any Additional Risk Factors) The use of low-dose heparin, LMWH, dextran, or aspirin is recommended for all moderate-risk patients. An alternative recommendation is intermittent pneumatic compression used continuously until the patient is ambulant, graduated elastic compression stockings, or a combination of both. These are grade A recommendations based on level I or level I1 data. Further studies are needed to assess the effect of using graduated compression stockings or intermittent pneumatic compression (IPC) in addition to pharmacologic methods and to assess the combined effects of different pharmacologic methods, such as heparin plus aspirin versus heparin alone.
INTERNATIONAL CONSENSUS RECOMMENDATIONS
High-Risk General Surgical Patients (Major Surgery, Age Over 60 Years, or Presence of Additional Risk Factors) All high-risk patients should receive prophylaxis as for moderaterisk patients (grade A recommendation). In addition to single modalities, such as low-dose heparin or LMWH, combined modalities of pharmacologic and mechanical methods should be considered because they may be more effective (grade B recommendation). In moderate-risk and high-risk patients, dextran and aspirin are not the methods of choice because of their limited efficacy on DVT prevention, anaphylactoid reactions and danger of cardiac overload associated with the former, high dose of aspirin (1000 to 1500 mg/d) required, and the fact that oral medications are not possible for several days in patients having abdominal surgery.
Neurosurgery Neurosurgery patients should be considered for mechanical methods of prophylaxis. In three randomized, controlled studies involving a total of 422 patients, the incidence of DVT was reduced from 21.3% in controls to 6.0% in the prophylactic groups using pneumatic compression (relative risk, 0.28; 95% confidence interval [CI], 0.16 to 0.51) (grade A recommendation).
Orthopaedic Surgery and Trauma Elective Hip Replacement Fixed low-dose unfractionated heparin (5000 IU every 8 or 12 hours) is effective for reducing DVT (level I data) and PE (level I1 data) in patients having elective hip replacement. Increasing the dose leads to a greater risk of bleeding. Adjusting the dose against a coagulation assay may enhance efficacy but is more difficult to manage. The addition of dihydroergotamine may increase efficacy but carries a risk of vasospasm. LMWH may be superior to unfractionated heparin in reducing both DVT and PE for hip replacement surgery, but more studies are needed. Fixed minidose oral anticoagulant therapy is not effective. Adjusteddose oral anticoagulants to a desired international normalized ratio [INRI (2.0 to 3.0) improves efficacy but is difficult to manage (level I evidence). A meta-analysis has demonstrated that antiplatelet therapy (mainly aspirin) in elective hip surgery is only moderately effective for protection against DVT (relative risk, 0.70; 95% CI, 0.61 to 0.82) (level I1 evidence), but the observed reduction in the risk of PE is substantial (relative risk,
BICK & HAAS
0.49; 95% CI, 0.26 to 0.92) (level I1 evidence). The effect of aspirin on PE needs confirmation. Dextran is only moderately effective in preventing DVT (level I evidence) and has inherent risks, such as fluid overload and anaphylactoid reactions. The latter can be minimized by hapten inhibition (level I evidence). There are few data as to the efficacy of graduated elastic compression by itself after hip replacement, but its use in orthopaedic surgery would be supported by data extrapolated from general surgery (grade C recommendation). IPC is effective (level I evidence). Data demonstrate that combined foot impulse technology with graduated elastic compression is effective in reducing the incidence of proximal DVT in patients having hip and knee surgery. In contrast with pharmacologic agents, mechanical methods are not associated with hemorrhagic complications. Regional anesthesia in patients having hip surgery (spinal or epidural) reduces the frequency of DVT, although its benefit is less than that of other methods and may not be additive in the presence of other techniques for prophylaxis. From extensive experience, there is no evidence that preoperative low-dose unfractionated heparin or LMWH increases the risk of spinal hematoma after regional anesthesia. Elective Knee Replacement
There are fewer available studies on prophylaxis after knee replacement. Data from hip replacement should not be extrapolated to knee replacement. IPC is effective (level I evidence). There is evidence that LMWH is more effective than warfarin and more effective than unfractionated heparin (level I evidence). Some data support the combined use of regional anesthesia with graduated elastic compression, or foot impulse technology. In most randomized, controlled trials, after knee replacement, the absolute risk of DVT remains high despite prophylaxis. The safety, efficacy, and cost-effectiveness of each method need to be documented further, and more effective methods or combinations need to be sought. Duration of Prophylaxis in Elective Orthopaedic Surgery
The optimal duration of prophylaxis in elective orthopaedic surgery has not been established. Because intraoperative risk factors are probably important, prophylaxis should ideally be started before surgery. A randomized comparison of preoperative and postoperative commencement of pharmacologic prophylaxis is necessary. Most trials have studied prophylaxis for 7 to 10 days or until the patients are ambulatory. Three randomized, controlled studies in patients having hip arthroplasty indicate, however, that prolonged thromboprophylaxis with LMWH decreases the frequency of venographically detected DVT. Further studies are needed before general recommendations can be made.
INTERNATIONAL CONSENSUS RECOMMENDATIONS
Emergency Orthopaedic Surgery The risks of DVT and PE, including fatal PE, are high in patients with hip fracture. Prophylaxis should be started as soon as possible after diagnosis and should be the same as recommended for elective hip surgery, with the exception of dihydroergotamine and heparin, which is contraindicated, and adjusted low-dose unfractionated heparin, for which no studies exist. Low-dose unfractionated heparin is effective in reduction of DVT, and although an overview of trials has not demonstrated a significant reduction in total PE, the observed effect was not different from that observed in general surgery, and there was a significant reduction in fatal PE. The best results so far have been obtained from studies using LMWH, adjusted-dose oral anticoagulants, dextran, one heparinoid, and IPC or foot impulse technology (level I data). More studies are needed for all methods. A meta-analysis has demonstrated that antiplatelet therapy in traumatic orthopaedic surgery is only slightly effective for protection against DVT (relative risk, 0.86; 95% CI, 0.73 to 1.0) (level I1 data), but the observed reduction in the risk of PE is substantial (relative risk, 0.40; 95% CI, 0.22 to 0.71) (level I1 data). The effect of aspirin on PE needs confirmation; the multicenter study on PE prevention in patients with femoral neck fracture is currently in progress and results are eagerly awaited. Recommendations for unfractionated heparin, LMWH, heparinoid, oral anticoagulants, dextran, and antiplatelet therapy are grade A. Because of the limited effect of dextran and aspirin on the incidence of DVT, these are not regarded as the methods of choice. Gynecologic Surgery Low-Risk Patients
Low-risk patients may receive prophylaxis. Turner et a P a reported a level I study that demonstrated a lower DVT rate with the use of graduated elastic compression (0% versus 4%; P<.05). On the basis of this study, the risk-to-benefit ratio, and extrapolation from moderaterisk patients, graduated compression stockings may be used in addition to early ambulation and adequate hydration. Moderate-Risk Patients Low-dose unfractionated heparin (5000 units every 12 hours) and LMWH have been demonstrated to be an effective prophylaxis in medium-risk gynecologic surgery patients (grade A recommendation based on level I data). By extrapolation from other types of surgery, antiplatelet therapy may be considered (grade C recommendation). IPC should also be considered because it is effective in higher-risk patients (grade A
BICK & HAAS
recommendation). Perioperative dextran 70 may also be used in selected patients (grade A recommendation based on level I data). Adjusted-dose warfarin is not recommended for routine prophylaxis but may have a role when low-dose heparin is contraindicated (e.g., where there is a history of heparin-induced thrombocytopenia) (grade A recommendation). Data relating to graduated elastic compression in moderate-risk gynecologic surgery are not sufficient at present to make a recommendation. Any use of elastic compression is based on extrapolation from studies in general surgery. High-Risk Patients Low-dose heparin (5000 units every 8 hours) or IPC used continuously for at least 5 days provides effective prophylaxis (grade A recommendation). When these two modalities were compared in a randomized trial, their efficacy appeared to be equal, but there were more bleeding complications associated with the use of low-dose heparin. Prophylaxis with combined methods, such as the addition of aspirin or compression stockings to heparin, and for extended periods needs to be studied further. Data evaluating LMWH and graduated compression stockings in high-risk gynecologic surgery patients are currently insufficient. Combination Estrogen-Containing Oral Contraceptives Combination estrogen-containingoral contraceptives may be associated with increased risk of DVT in patients undergoing gynecologic surgery. Discontinuation of oral contraceptives 4 to 6 weeks before surgery should be considered, but this should be balanced against the risk of unwanted pregnancy. In the absence of other risk factors, there is insufficient evidence to support a policy of routinely stopping the combined oral contraceptive pill before major surgery. Heparin prophylaxis is advisable when oral contraceptives have not been discontinued and additional risk factors are present. For emergency surgery, thromboprophylaxis should be provided in women taking the combined oral contraceptive pill because the risk of DVT is greater (grade C recommendation). Hormone-Replacement Therapy In the absence of other risk factors, there is currently insufficient evidence to support a policy of routinely stopping hormone replacement therapy before surgery. Although there are no data, heparin prophylaxis is advisable when hormone replacement therapy has not been discontinued before surgery. In practice, most patients receiving hormone replacement therapy have additional risk factors, particularly age, which
INTERNATIONAL CONSENSUS RECOMMENDATIONS
in themselves would be indications for thromboprophylaxis (grade C recommendation). Pregnancy Low-dose heparin prophylaxis is commonly used in pregnant patients at high risk for DVT and PE, such as previous DVT or PE and certain thrombophilias (see later), although data regarding efficacy from controlled trials are lacking (grade C recommendation). There are insufficient data on both the optimal timing and dosing schedule of low-dose heparin prophylaxis. There is substantial evidence for safety with lowdose aspirin use in pregnancy. Although there is no direct evidence regarding aspirin in preventing DVT or PE in pregnancy, its efficacy in other settings suggests that it may be worth considering (grade C recommendation). Oral anticoagulants are contraindicated for prophylaxis of venous thromboembolism in the first trimester because of increased risk of embryopathies, and available data indicate that they are associated with increased fetal and maternal-fetal bleeding in the second and particularly the third trimester. In the presence of contraindications to heparin, oral anticoagulants can be considered, when justified, for prophylaxis in the second trimester because bleeding is uncommon at this stage (grade C recommendation). The benefits of prophylaxis have not been demonstrated in patients undergoing cesarean section who do not have additional risk factors. Perioperative and postpartum prophylaxis should be considered if there are risk factors, particularly age over 35 years, obesity, previous DVT or PE, or thrombophilia (grade C recommendation). Dextran should be avoided in pregnancy because an anaphylactoid reaction may precipitate acute fetal distress. Dextran should be withheld during cesarean section until after delivery of the infant. Insufficient data are available on the use of LMWHs or mechanical methods in pregnancy. Multicenter trials comparing standard heparin with LMWH in high-risk pregnant patients are urgently needed to assess efficacy, safety, and possible side effects, such as osteoporosis. Women who develop thromboembolism during pregnancy should be treated with therapeutic levels of adjusted-dose subcutaneous heparin, which should be continued throughout pregnancy, labor, and delivery. Higher doses are required in late pregnancy but the subcutaneous dose should be reduced in labor or before cesarean section to reduce the risk of hemorrhage at delivery. Anticoagulation is usually continued for at least 6 weeks postpartum, but the optimal duration of antithrombotic therapy has not been established in this setting (grade C recommendation). Patients who develop thromboembolism during pregnancy or the puerperium should be referred for hematologic consultation. The management of thrombophilic conditions throughout pregnancy usually re-
BICK & HAAS
quires adjusted-dose heparin with monitoring of the heparin effect (grade C recommendation). Medical Patients
Compared with surgical patients, there are fewer randomized trials of DVT prophylaxis using objective diagnostic measures in hospitalized medical patients. Available data suggest, however, that prophylaxis with anticoagulants (usually low-dose unfractionated heparin or LMWH) or with antiplatelet therapy (usually aspirin) shows a similar relative risk reduction in the incidence of DVT diagnosed by objective screening methods in medical compared with surgical patients, that is, prevention of about two thirds of cases of DVT. There are no reported trials of mechanical methods of prophylaxis, such as graduated elastic compression stockings or IPC, in medical patients. Although there is no reason to believe that such methods would be less effective than in surgical patients, further studies are needed before clear recommendations can be made. All medical patients admitted to the hospital should be assessed for risk of venous thromboembolism, and prophylaxis should be considered for those at moderate or high risk. Acute Myocardial Infarction
Historically, in the absence of antithrombotic therapy (aspirin or anticoagulants) or thrombolytic therapy, patients with acute myocardial infarction had a moderate risk of DVT (24%) and a moderate risk of clinical PE (2.6%to 6.1%). The risk of DVT increases with age and in the presence of cardiac failure. Most studies assessing the incidence of DVT were performed in the 1970s before the introduction of routine aspirin and selective thrombolytic therapy. There is no evidence that earlier mobilization reduces the risk of DVT? lo but thrombolytic therapy and aspirin have both been shown to reduce mortality. In addition, aspirin also reduces the rate of reinfarction and stroke. In high-risk medical patients, antiplatelet therapy has also been shown to prevent DVT and PE. Aspirin should therefore be used routinely in acute myocardial infarction and unstable angina, unless there is some particular definite contraindication (grade A recommendation). Some patients with acute myocardial infarction receive full-dose anticoagulants (heparin or warfarin or both), as prophylaxis either to prevent rethrombosis in the coronary artery(s) after thrombolytic therapy or to prevent systemic embolism,from a left ventricular mural thrombus (such as patients with anterior Q-wave infarction, severe left ventricular dysfunction, or mural thrombus at echocardiography) or from atrial
INTERNATIONAL CONSENSUS RECOMMENDATIONS
fibrillation. In the absence of aspirin, full-dose anticoagulants appear to be effective in preventing venous thromboembolism, but there is not clear evidence that anticoagulants produce the additional benefit of PE reduction when given with aspirin, which is now used routinely because of proven benefits. In patients with acute myocardial infarction who are at high risk for DVT but in whom anticoagulants are contraindicated because of overt or high risk of bleeding, the addition of graduated elastic compression stockings or IPC to aspirin may be considered (grade C recommendation based on extrapolation of data from trials in surgical patients). RECOMMENDATIONS FOR PREVENTION OF VENOUS THROMBOEMBOLISM GIVEN BY THEACCPCONSENSUS
1. In low-risk general surgery patients who are undergoing minor operations, are younger than 40 years of age, and have no clinical risk factors, no specific prophylaxis other than early ambulation is recommended. This is a grade C recommendation based primarily on the low risk in such patients.’ 2. It is recommended that elastic stockings, low-dose unfractionated heparin given 2 hours before and every 12 hours after operation, or IPC be used in moderate-risk general surgery patients who are older than 40 years of age and are undergoing major operations but who have no additional clinical risk factors for venous thromboembolism. IPC and elastic stockings should be applied during the operation, if possible, and throughout the postoperative period. These are grade A recommendations based on level I data. 3. In higher-risk general surgery patients who are prone to wound complications, such as hematomas and infection, IPC is a good alternative choice for prophylaxis. This is a grade A recommendation based on level I data. 4. In high-risk general surgery patients with multiple risk factors it is recommended that effective pharmacologic methods (lowdose unfractionated heparin, LMWH, or dextran) be combined with IPC. This is a grade B recommendation based on level I1 data and on extrapolation of data from other patient groups. Low-dose unfractionated heparin and LMWH therapy should be started preoperatively and dextran given intraoperatively. IPC should also be applied intraoperatively, if possible. 5. In selected high-risk general surgery patients, perioperative warfarin (INR, 2.0 to 3.0) therapy may be used. This is a grade A recommendation based on level I data. 6. It is recommended that aspirin not be used for prophylaxis in general surgery patients because other measures (as recommended previously) are more efficacious. This is a grade A recommendation based on level I data.
BICK & HAAS
7. In patients undergoing total hip replacement surgery, postoperative subcutaneous twice-daily fixed-dose unmonitored LMWH, low-intensity (INR = 2.0 to 3.0) oral anticoagulation (started preoperatively or immediately after operation), and adjusteddose unfractionated heparin (started preoperatively) are the most effective anticoagulant-based prophylaxis regimens and are recommended for routine use. These are grade A recommendations based on level I data. Adjuvant prophylaxis with elastic stockings or IPC may provide additional efficacy. Although other agents such as low-dose unfractionated heparin, aspirin, dextran, and IPC reduce the overall incidence of venous thromboembolism, they are less effective and should not be used routinely. 8. In patients undergoing total knee replacement surgery, postoperative subcutaneous twice-daily fixed-dose unmonitored LMWH is the most effective anticoagulant-based prophylaxis regimen. IPC is the most effective nonpharmacologic prophylaxis regimen and provides a reduction in relative risk comparable to LMWH. Either LMWH or IPC is recommended for routine prophylaxis. This is a grade A recommendation based on level I data. 9. In patients undergoing hip fracture surgery, either preoperative subcutaneous fixed-dose unmonitored LMWH or oral anticoagulation (INR, 2.0 to 3.0) is the most effective prophylactic agent and is recommended for routine use. This is a grade A recommendation based on level I data. IPC combined with either of the recommended anticoagulant-based prophylaxis regimens may provide additional benefit. 10. Data are insufficient to recommend prophylactic inferior vena cava filter placement for orthopaedic surgery patients. Prophylactic inferior vena cava filter placement should be limited to high-risk patients in whom other forms of anticoagulant-based prophylaxis are not feasible because of active bleeding. This is a grade C recommendation based on level IV and V data. 11. It is recommended that IPC with or without elastic stockings be used in patients undergoing intracranial neurosurgery. Lowdose heparin therapy may be an acceptable alternative. These are grade A recommendations based on level I data. IPC and low-dose unfractionated heparin may be more effective in combination than individually and should be considered in highrisk patients. 12. In patients with acute spinal cord injury with paralysis, treatment with adjusted-dose heparin or LMWH is recommended for prophylaxis. These are grade B recommendations based on level I1 data. Warfarin prophylaxis may also be effective. This a grade C recommendation based on extrapolation of results from other patient groups. Low-dose unfractionated heparin, elastic stockings, and IPC when used alone appear ineffective and are
INTERNATIONAL CONSENSUS RECOMMENDATIONS
not recommended. These methods may have benefit when used together. This is a grade B recommendation based on level I1 data. 13. In multiple trauma patients, it is recommended that IPC, warfarin, or LMWH be used when feasible. These are grade C recommendations based on extrapolation of data from other patient groups. Because of the high risk of venous thromboembolism, the inability to apply standard methods of prophylaxis in many trauma patients, and the uncertainty about the efficacy of preventive measures, serial surveillance with duplex ultrasonography may be a successful strategy. In selected high-risk patients, prophylactic caval filter placement may be used. These are grade C recommendations based on level I11 and IV data. 14. It is recommended that low-dose unfractionated heparin be used in patients with myocardial infarction. Full-dose anticoagulation is also effective. These are grade A recommendations based on level I data. IPC and possibly elastic stockings may be useful when heparin therapy is contraindicated. These are grade C recommendations based on extrapolation of data from other patient groups. 15. In patients with ischemic stroke and lower extremity paralysis, low-dose unfractionated heparin and LMWH are effective. These are grade A recommendations based on level I data. IPC and elastic stockings are also probably effective. These are grade C recommendations based on extrapolation of data from other patient groups. 16. In general medical patients with clinical risk factors for venous thromboembolism, particularly those with congestive heart failure or chest infections, low-dose unfractionated heparin and LMWH are effective. These are grade A recommendations based on level I data. 17. It is recommended that warfarin 1 mg daily be used in patients with long-term indwelling central vein catheters to prevent axillary-subclavian venous thrombosis. This is a grade A recommendation based on level I data. 18. It is recommended that low-dose unfractionated heparin (every 8 hours) or LMWH be used in higher-risk general surgery patients who are older than 40 years, undergoing major operations, and have additional risk factors. These are grade A recommendations based on level I data. RECOMMENDATIONS FOR TREATMENT OF VENOUS THROMBOEMBOLISM GIVEN BY THE ACCP CONSENSUS
1. Patients with DVT or PE should be treated with intravenous heparin or adjusted-dose subcutaneous heparin sufficient or pro-
BICK & HAAS
long the activated partial thromboplastin time to a range that corresponds to a plasma heparin level of 0.2 to 0.4 U/mL by protamine titration method or an anti-Xa level of 0.3 to 0.7 U/ mL. This grade A recommendation is based on level I studies in patients with PE and DVT and level I1 studies on the relationship between the activated partial thromboplastin time and effectiveness6 2. It is recommended that treatment with heparin should be continued for 5 to 10 days and that oral anticoagulation should be overlapped with heparin therapy for 4 to 5 days. For many patients, heparin and warfarin therapy can be started together and heparin therapy discontinued on day 5 or 6 if the prothrombin time is therapeutic. For massive PE or ileofemoral thrombosis, a longer period of heparin therapy may be considered. This grade A recommendation is based on five level I studies.6 3. In many countries, LMWH is used in place of unfractionated heparin. Dosing requirements are individualized for each individual product. LMWH should be administered for 5 to 10 days and therapy overlapped with oral anticoagulation as outlined in no. 2 above. 4. Long-term anticoagulant therapy should be continued for at least 3 months using oral anticoagulants to prolong the prothrombin time to an INR of 2.0 to 3.0. When oral anticoagulants are either contraindicated or inconvenient, adjusted-dose heparin therapy should be given to prolong the activated partial thromboplastin time to a time that corresponds to a plasma heparin level greater than 0.2 U/mL by protamine titration method or greater than 0.3 U/mL by anti-Xa assay, or beyond the middose interval. This grade A recommendation is based on two level I studies. 5. It is recommended that patients with recurrent venous thrombosis or a continuing risk factor, such as antithrombin I11 deficiency, protein C or S deficiency, lupus anticoagulant, other hereditary or acquired thrombophilic state, or malignancy, should be treated indefinitely. This grade C recommendation is not based on published data. 6. Accumulated level I, 11, and IV evidence indicates that symptomatic isolated deep calf vein thrombosis should be treated with anticoagulation for 3 months. If for any reason anticoagulation cannot be given, serial noninvasive studies of the lower extremity should be performed to assess for proximal extension of thrombus. 7. The use of thrombolytic agents in the treatment of venous thromboembolism continues to be highly individualized. Further clinical investigation is needed before more definite recommendations can be made. 8. Inferior vena cava filter placement is recommended when there is a contraindication or complication of anticoagulant therapy in an individual with or at high risk for proximal vein thrombosis
INTERNATIONAL CONSENSUS RECOMMENDATIONS
or PE. Filter placement is also recommended for recurrent thromboembolism that occurs despite adequate anticoagulation, in the presence of a large free-floating vena caval thrombus, in patients with chronic recurrent embolism with pulmonary hypertension, and during the concurrent performance of surgical pulmonary embolectomy or pulmonary endarterectomy. Whenever possible, anticoagulation should be given after placement of a vena caval filter in a patient with proximal DVT of the lower extremity.6 SECONDARY PREVENTION OF VENOUS THROMBOEMBOLISM: METHODS AND RECOMMENDATIONS OF THE INTERNATIONAL CONSENSUS STATEMENT
The recommendations for treatment and secondary prevention of venous thromboembolism given by the International Consensus statementI5are similar to those provided by the ACCP Consensus. In particular, for patients treated with anticoagulants, it has been emphasized that initial heparin therapy and long-term oral anticoagulation therapy are both necessary (grade A recommendation) with DVT; initial therapy with oral anticoagulants alone is associated with an unacceptable rate of recurrence. Concerning other strategies of treatment, there is full agreement of the two consensus documents. IMPACT OF THE CONSENSUS STATEMENTS ON THE CLINICAL ACCEPTANCE OF LOW-MOLECULARWEIGHT HEPARINS
The favorable assessment of LMWHs and the recommendation of their prophylactic use in high-risk patients has encouraged numerous clinicians in Europe to abandon low-dose unfractionated heparin and to administer fixed doses of LMWH instead. The mode of administration is according to the statement of the European Consensus: “Prophylaxis should preferably be started before operation and continued for 7 to 10 days after operation or until fully ambulatory.” Higher doses of LMWH are given 12 hours preoperatively (i.e., in the evening of the day before operation). This common practice will not be abandoned unless and until the key question about the necessity of a randomized comparison of the preoperative and postoperative commencement of pharmacologic modalities is answered. The use of LMWH in patients at high risk has become standard in many European countries; LMWH-controlled studies are required by several health authorities for the registration of new compounds, and fixed-dose unfractionated heparin-controlled trials are no longer accepted in Germany and France and perhaps in other countries. New compounds and formulations will have to compete with fixed-dose
BICK & HAAS
LMWH or activated partial thromboplastin time-adjusted unfractionated heparin in future clinical studies. Also, various meta-analyses have provided evidence that the bleeding risk of LMWH prophylaxis is not higher than that of unfractionated heparin pro phyla xi^.^,^ It may be even less under particular circumstances as shown by Kakkar.’ Thus, LMWH has also been used for prevention of venous thromboembolism in patients at moderate and even at low risk, and its use is limited only by cost. The European Consensus Statement has initiated various attempts to define the minimal risk of DVT not requiring routine prophylaxis and a risk-adapted prophylaxis in various patient populations. According to the Consensus Statement, the cost-effectiveness of pharmacologic prophylaxis seems to be clear for moderate-risk and high-risk patients: “In medium-risk and high-risk patients, the costs of screening, diagnosis, and treatment of thromboembolism are so high that the currently recommended methods of primary prophylaxis are cost effective”; however, “In low-risk patients no data is available at present concerning the costeffectiveness of the recommended prophylactic methods.” Not only costeffectiveness is discussed when considering pharmacologic prophylaxis in low-risk patients, but also the benefit-to-risk ratio plays an important role. Heparin prophylaxis may be associated with the potential risk of heparin-induced thrombocytopenia; thus this complication has to be weighed out against the risk of venous thromboembolism. The period since the European Consensus Statement 1991 has been mainly characterized by preclinical and clinical trials to elucidate the pathophysiologic mechanism(s) of heparin-induced thrombocytopenia and to determine its frequency in various patient populations. Warkentin et all7 analyzed heparin-dependent IgG antibodies in patients undergoing hip surgery and receiving either unfractionated heparin or LMWH as prophylaxis. The investigators were able to show that heparin-induced thrombocytopenia, associated thrombotic events, and heparin-dependent IgG antibodies are more common in patients treated with unfractionated heparin than in those treated with LMWH. Thus, the acceptance of LMWHs as prophylaxis has become greater than that of unfractionated heparin, and this is also true when the continuing risk of DVT after discharge from the hospital is considered: ”The risk for venous thromboembolism may continue beyond the period of hospitalization. This risk needs to be assessed in prospective trials. Although the evidence is weak, consideration should be given to extending prophylaxis when the hospital stay is prolonged or the risk continues.” This key question of the European Consensus Statement has initiated a series of clinical trials to prevent DVT and its sequelae in outpatients, and the same is true for patients undergoing ambulatory surgery. The beneficial safety and efficacy profile of LMWHs and the necessity of only single daily injections have facilitated home treatment and self-administration of the drug. The European Consensus Statement has also contributed to the understanding of the mode of action of LMWHs. There is a consensus that these compounds should be considered as separate drugs and that
INTERNATIONAL CONSENSUS RECOMMENDATIONS
dosage equivalents between these products should not be assumed. According to this view, individual dose-ranging studies are therefore necessary for each product to achieve an appropriate prophylactic dose. The above clearly has regulatory implications, and each manufacturer is required to demonstrate that the various characteristics of their product have been well defined. For example, a manufacturer would be expected to provide full details on consistency of production on a batch-to-batch basis, detailed pharmacokinetics and pharmacodynamics in volunteers, and clinical evidence that the product is both safe and effective in various risk categories. This does not mean, however, that there are not broad similarities between the various products. It is remarkable how effective and safe the various preparations have been over a relatively narrow dose range. Although investigators agree that the anti-Xa and anti-IIa activities of LMWHs are only partial expressions of their biologic and antithrombotic potential, this is no reason to ignore completely the value of these indices. The fact that most of the available LMWHs are used at a dose range of 2000 to 4000 anti-Xa IU/d by once-daily subcutaneous injection for the prophylaxis of DVT and that they all give broadly comparable results strongly indicates that LMWHs at least belong to the same family of drugs. The International and the ACCP Consensus Statement have provided the first evidence that unfractionated heparin may also be replaced by LMWHs in the treatment of DVT Low-molecular-weight heparins given subcutaneously have been shown to be as safe and effective as standard heparin therapy for the initial treatment of DVT in terms of reduction of the thrombus size as assessed by repeat venography. Preliminary results demonstrate that fixed-dose, body weight-adjusted unmonitored low-molecular-weight heparins are as effective as adjusted-dose intravenous standard heparin in the prevention of symptomatic recurrent venous thromboembolism during long-term follow-up.
In the meantime, numerous clinical trials have provided evidence that the therapeutic use of LMWHs in secondary prevention of DVT is lo,l1 efficacious, safe, and cost-effe~tive.~, Finally, the great impact of the European Consensus on the clinical acceptance of LMWHs has initiated a variety of clinical trials to answer the open key questions formulated in the Statement of 1997. These questions and demands are as follows: The risk of venous thromboembolism may continue beyond hospitalization. This risk needs to be assessed in prospective studies. The efficacy of continued prophylaxis beyond hospitalization deserves investigation. A proper randomized comparison of preoperative and postoperative commencement of pharmacologic prophylactic modalities is necessary. An urgently needed study in moderate-risk patients is a direct comparison of fixed-low-dose heparin with LMWH with respect to total mortality and confirmed fatal PE.
BICK & HAAS
Further studies are needed to determine whether graduated compression stockings or IPC enhances the efficacy of pharmacologic methods. A prospective register (postmarketing surveillance) on the prevalence of spinal or epidural anesthesia-induced hemorrhage in patients pretreated with prophylactic anticoagulant agents is needed. The risk of DVT in the new minimally invasive abdominal surgical procedures needs to be established. A multicenter trial comparing standard heparin with LMWH in high-risk pregnant patients assessing efficacy, safety, and side effects such as osteoporosis is also needed.
A Recent Unsolved Problem
Recent studies have opened the door to another key question, perhaps the primary question of the next decade: What is the role of outpatient management for DVT? Particularly in North America, where cost-containment, often driven by managed care, is of primary concern and clinicians are under increasing pressure to decrease costs through decreased hospital admissions, publications presenting results of LMWHs in randomized trials demonstrating the clear safety and efficacy of home treatment or early discharge management of DVT have become of key i n t e r e ~ t . ~ , ~These , ’ ~ ,studies ~ ~ , ~ ~are too recent to have been as yet considered in consensus conferences, and firm guidelines have not yet been established. One of the authors (RLB) has adopted the following general principles of outpatient or early discharge management of DVT pending more results of randomized trials or, preferably, consensusdriven recommendations that provide additional information.
Suggested Guidelines for the Outpatient or €ar/y Discharge Management of Deep Venous Thrombosis
1. Admit for 24 hours if no comorbid condition, or treat as outpatient if no comorbid condition and all of the following can be accomplished. 2. Complete blood count and platelet count on admission. 3. Prothrombin time and activated partial thromboplastin time on admission. 4. Teach patient applicable antiembolic exercises (see later). 5. Start subcutaneous LMWH as dalteparin (Fragmin) at 200 U/ kg every 24 hours (available as 2500 U/0.2 mL or 5000 U/0.2 mL)I3or enoxaparin (Lovenox) at 1 mg/kg every 12 hours (100 U/kg every 12 hours) (30 mg ampules).12 6. Instruct patient in self-injection of subcutaneous LMWH in ante-
INTERNATIONAL CONSENSUS RECOMMENDATIONS
rior thighs or anterior abdominal wall (anterior thighs preferred-use rotating injection sites). 7. Measure for medium compression panty hose or upper extremity hose for use during waking hours only. 8. Start warfarin at 5 mg/d if less than 70 kg total body weight or 10 mg/d if greater than 70 kg total body weight (see exceptions at no. 17). 9. Discharge at 24 hours if no comorbid conditions, or discharge as soon as comorbid condition h o t DVT) allows. 10. Arrange home health if patient or family cannot self-inject LMWH. 11. Arrange outpatient prothrombin time and INR and complete blood count and platelet count at home on day 3. 12. Evaluate patient clinically (in office or clinic) on day 7; obtain prothrombin time and complete blood count and platelet count day 5 and 7; stop LMWH when INR is approximately 2.0 then adjust warfarin dose accordingly. 13. See patient weekly until stable on long-term antithrombotic therapy. 14. If young age patient with unexplained DVT, consider evaluation for common blood coagulation protein(s) or platelet defects leading to thrombosis: antiphospholipid syndrome (lupus anticoagulant or anticardiolipin antibodies), antithrombin, protein C, protein S, activated protein C resistance (factor V Leiden), sticky platelet syndrome, plasminogen activator (PA) defects. 15. If clinically warranted and all of the above are negative, consider evaluating patient for uncommon blood defects associated with unexplained thrombosis: plasminogen deficiency, plasminogen activator inhibitor Type 1 (PAI-1) elevation, dysfibrinogenemia, heparin cofactor-I1defects, lipoprotein(a), and homocysteinemia. Always consider occult malignancy if clinically warranted. 16. While hospitalized, when patient is in bed, raise foot of bed straight, with feet elevated 7 to 10 degrees above hips; never put pillow(s) under popliteal fossae. 17. Depending on clinical parameters (thrombophilia, platelet defects), alternatives to oral anticoagulants may be indicated. Antiembolic Leg Exercises
The patient should be instructed in antiembolic leg exercises consisting of dorsiplantar flexion of each foot at a time with the legs supported at the feet, legs straight and elevated above the hips approximately 7 degrees to 10 degrees. These are to continue for 3 to 5 minutes, or until the calf muscle group is fatigued, with initiation of muscle pain; then the opposite leg is exercised in a likewise manner. The patient should be told to do these exercises four to six times a day. The patient also should be instructed not to remain in a sitting position or indulge in other activities, such as car or airplane trips, with the thighs and
BICK & HAAS
knees bent, for more than 20 minutes at a time without straightening the legs through brief ambulation or leg stretching for a few minutes. Antiembolic Arm Exercises The patient should be instructed in antiembolic arm exercises consisting of palmar squeezing of each hand at a time with the arms elevated and straightened above the head. Holding a tennis ball in the hand, while squeezing is excellent for this exercise. These should continue for 3 to 5 minutes, or until the arm muscles are fatigued, with initiation of muscle pain, then the opposite arm should be exercised in a likewise manner. The patient should be told to do these exercises four to six times a day. The patient also should be instructed not to remain in a position or indulge in other activities with the arms tightly bent at the elbow or shoulder for more than 20 minutes at a time, without stretching the arms straight for a few minutes.
CONCLUSIONS The European Consensus Statement, the ACCP Consensus 1995, and the International Consensus Statement 1997 have had a great impact on the clinical acceptance of LMWHs by clinicians, physicians, and health authorities and in particular have helped To initiate various clinical trials to answer open key questions. To provide further evidence on the safety and efficacy profile of LMWHs. To consider LMWHs as individual compounds. To facilitate home treatment. To consider cost-effectiveness. To set a new standard required by health authorities to use LMWH as a reference in future prophylaxis trials in high-risk patients.
References 1. Clagett GP, Anderson FA, Heit J, et al: Prevention of venous thromboembolism. Fourth ACCP Consensus Conference on Antithrombotic Therapy. Chest 108 (suppl):312, 1995 2. Clement DI, Gheeraert P, De Buysere M, et al: Medical patients. In Bergqvist D, Comerota AJ, Nicolaides AN, et a1 (eds): Prevention of Venous Thromboembolism. London, Med-Orion, 1994, p 319 3. European Consensus Statement on the prevention of venous thromboembolism. Int Angiol 11:151, 1992 4. Hirsh J, Siragusa S, Cosmi B, et al: Low molecular weight heparins (LMWH) in the treatment of patients with acyte venous thromboembolism. Thromb Haemost 74:360, 1995 5. Hull RD, Raskob GE, Pineo GF, et al: The treatment of proximal vein thrombosis with subcutaneous low molecular weight heparin compared with continuous intravenous
INTERNATIONAL CONSENSUS RECOMMENDATIONS
heparin. The Canadian-American Thrombosis Study Group. Clin Appl Thromb Hemost 1:151, 1995 6. Hyers TM, Hull RD, Weg JG: Antithrombotic therapy for venous thromboembolic disease. Fourth ACCP Consensus Conference on Antithrombotic Therapy. Chest 108 (suppl):335,1995 7. Kakkar V V Effectiveness and safety of low molecular weight heparins (LMWH) in the prevention of venous thromboembolism. Thromb Haemost 74:364, 1995 8. Koopman MMW, Prandoni P, Piovella F, et al: Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous low-molecular-weight heparin administered at home. N Engl J Med 334:682, 1996 9. Leizorovicz A, Naugh MC, Chapuis FR, et al: Low molecular weight heparins in the prevention of perioperative thrombosis. BMJ 305:913, 1992 10. Leizorovicz A, Simonneau G, Decousus H, et al: Comparison of efficacy of lowmolecular weight heparins and unfractionated heparin in initial treatment of deep vein thrombosis: A meta-analysis. BMJ 309:299, 1994 11. Lensing AWA, Prins MH, Davidson BL, et al: Treatment of deep-vein thrombosis with low molecular weight heparin: A meta-analysis. Arch Intern Med 155:601, 1995 12. Levine M, Gent M, Hirsh J, et al: A comparison of low-molecular-weight heparin administered primarily at home with unfractionated heparin administered in the hospital for proximal deep-vein thrombosis. N Engl J Med 334:677, 1996 13 Lindmarker P, Holmstrom KM, Granquist S, et al: Comparison of once-daily subcutaneous Fragmin with continuous intravenous unfractionated heparin in the treatment of deep vein thrombosis. Thromb Haemost 72:186, 1994 14 McIntyre K Medicolegal implications of consensus statements. Fourth ACCP Consensus Conference on Antithrombotic Therapy. Chest 108 (suppl): 502, 1995 15. Prevention of venous thromboembolism. International Consensus Statement (guideline according to scientific evidence). Int Angiol 16:3, 1997 16. Shafer AI: Low-molecular-weight heparin-an opportunity for home treatment of venous thrombosis . N Engl J Med 334:724,1996 16a. Turner GM, Cole SF, Brooks JH: The efficacy of graduated compression stockings in the prevention of deep vein thrombosis after major gynecological surgery. Brit J Obstet Gynecol91:588-591, 1984 17. Warkentin TE, Levine MN, Hirsh J, et al: Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 332:1330. 1995
Address reprint requests to Rodger L. Bick, MD, PhD, FACP 10455 North CGntral Expressway, Suite 109 Dallas, TX 75231