Section 6 Chapter 6 Venous Thromboembolism

Reviews suggest that pulmonary embolism (PE) occurs in approximately 700,000 people in the United States each year, of whom about 200,000 will die as a result.^3,4 In the absence of prophylaxis, fatal PE occurs in 4% to 7% of hip surgery patients and in 0.1% to 0.8% of general surgery patients.^3,⁵ In 40% to 60% of patients who die of PE, the diagnosis is not made clinically. PE may be responsible for as many as 5% of postoperative deaths, and it may occur in as many as 25% of patients admitted to the hospital.^6,7 Pulmonary infarction occurs in about 10% of PE patients.⁸

Significant PE is believed, as a rule, to arise from thrombosis of the deep veins of the thigh and the pelvis. Most studies of thromboembolism use DVT as a surrogate end point for PE. The only major study of thromboembolism prophylaxis to date that successfully used death from PE as an end point required more than 5,000 patients to reach statistical significance.⁹ Accordingly, the National Institutes of Health has concluded that using DVT as a surrogate for PE is a valid approach.¹⁰

Predictors of postoperative venous thromboembolism

The optimal treatment of thromboembolism is prevention, particularly in persons at high risk [see Table 1].^11,12 Risk factors for venous thromboembolism include increased age (� 30 years in some studies),¹³ major trauma (Injury Severity Score of 15 or greater or the presence of a pelvic or lower-extremity long bone fracture), morbid obesity, major operation, prolonged immobility, thrombophilia, and previous thromboembolism.

Pneumatic devices that compress the venous plexuses of the lower extremities are popular because they do not require anticoagulation and thus are not associated with increased bleeding risk. Intermittent pneumatic compression is capable of intermittently increasing venous flow velocities in the femoral and pelvic veins.¹⁶ It has been argued that some of the benefit might derive from the known tourniquet effect of enhancing fibrinolytic activity, attributed both to an increase in tissue plasminogen activator (t-PA) and to a decrease in plasminogen activator inhibitor (PAI). This argument seems to be valid for up to 24 hours of continuous use,¹⁷ but after that point, the effect is exhausted.¹⁶ Intermittent pneumatic compression is a safe, albeit somewhat uncomfortable, method of preventing clots in patients immobilized for prolonged periods. It is particularly useful in critical care units, where other forms of prophylaxis are inapplicable or contraindicated.¹¹

Several different pneumatic compression devices have been developed. The first to gain widespread acceptance was the full-length (calf and thigh) sequential compression stocking, which proved effective in a number of settings, including trauma.¹⁸ Because some injured patients were unable to wear the device, various modified compression devices were developed, including a calf-only device and a foot pump designed to compress the plantar venous plexus. Several studies have demonstrated that the various compression devices are not all equally effective in preventing thromboembolism. This is an important point because many of the devices are marketed solely on the basis of compliance data rather than efficacy data; those that are marketed without published evidence of efficacy often use cost advantages to gain market share.

The effectiveness of pneumatic compression devices is based on their ability to increase peak venous flow velocity in the large vessels of the thigh and the pelvis. Because some units create higher peak flow velocities, it would be logical to assume that these units would be more effective in reducing DVT. Support was lent to this assumption by a 2000 study in which more than 300 patients undergoing elective knee surgery were treated with different compression devices.¹⁹ Devices applying asymmetrical compression (which results in higher peak venous flow velocity) were compared with traditional compression devices, and calf-only devices were compared with calf-and-thigh devices. The two calf-and-thigh units studied proved superior to the calf-only unit, and the calf-and-thigh asymmetrical sequential compression device that generated higher peak flow velocities was superior to the traditional calf-and-thigh device. Similar results were obtained in a 2004 study that compared knee-high compression devices.²⁰

It is therefore important to recognize that although compliance is a critical component of compression devices, differences in efficacy must be taken into account. Given the absence of efficacy data for many of the compression devices now in use, great caution should be exercised in considering whether these products should be adopted, regardless of anticipated cost savings.

Because of these pitfalls, an adjusted-dose technique was devised in the mid-1980s.²² In this method, heparin is given (either subcutaneously or I.V.) in sufficient doses to elevate the activated partial thromboplastin time (aPTT) by 2 to 5 seconds, thereby compensating for depleted antithrombin levels in high-risk patients. This technique is superior to the low-dose method for preventing venous thromboembolism,²² and for practical purposes, it should replace the latter as the standard prophylactic dosing technique for unfractionated heparin.^22,23

LMWH possesses the same antithrombin-potentiating pentasaccharide chain that unfractionated heparin does. Consequently, it is similarly ineffective if antithrombin levels are depleted. The main advantage of LMWH over unfractionated heparin is that it has a more dependable half-life and bioavailability. Thus, it can be given without monitoring of drug effect or plasma heparin level.

Most of the clinical trials documenting the efficacy of LMWH evaluated patients undergoing elective hip^24,25 or knee operations. A few, however, addressed other patient populations (e.g., trauma patients).¹⁵ In these studies, the incidence of DVT in patients receiving unmonitored LMWH therapy was generally lower than that in patients receiving placebo²⁴ or low-dose heparin.^15,²⁵ LMWH therapy and adjusted-dose heparin therapy were of roughly equal efficacy.

In the light of these data, it appears that LMWH can be recommended over low-dose unfractionated heparin in elective, emergency, and trauma patients. Whether it is superior to adjusted-dose unfractionated heparin therapy remains uncertain, though it is clearly simpler to manage. Where compliance with monitoring and dose-adjusting protocols is an issue, unmonitored LMWH therapy may well be preferable.

Fondaparinux, a synthetic form of the specific pentasaccharide that interacts with antithrombin to potentiate its effect, has been approved by the Food and Drug Administration. This agent is smaller than the LMWHs and seems to have the same advantages over these substances that they have over unfractionated heparin�namely, increased bioavailability, longer half-life, and a more consistent effect. As the molecular weight of the antithrombin potentiator decreases, the antithrombotic effect focuses more sharply on inhibiting factor X (as opposed to factors II, IX, XII, etc.).

Before fondaparinux is uniformly adopted for thromboembolism prophylaxis, the supporting evidence should be carefully considered. The initial findings are quite promising. In a prospective, randomized trial that included approximately 1,700 hip fracture patients, fondaparinux was superior to enoxaparin with respect to the incidence of venographically identified DVT²⁶; however, the enoxaparin dosage (40 mg once daily) was lower than the dosage shown to be effective in trauma patients by previous investigators (30 mg twice daily).¹⁵ No significant difference in bleeding rates was observed. In another trial, which included approximately 700 patients undergoing elective knee surgery, fondaparinux, 2.5 mg once daily, proved superior to enoxaparin, 30 mg twice daily.²⁷ In this study, the bleeding rate was significantly higher with fondaparinux.

Ultimately, with regard to prophylactic anticoagulation, simple logic applies. The more potent the anticoagulant used, the lower the risk of PE and the higher the risk of serious bleeding. It is therefore appropriate to base one's selection of an anticoagulant regimen on a careful weighing of the expected benefits against the anticipated bleeding risk.

In some patients (e.g., those undergoing major gynecologic procedures), warfarin may be used instead of heparin.^12,²⁸ A low-dose regimen (as little as 1 mg/day) may offer some prophylactic benefit.²⁸ Warfarin anticoagulation must be started 3 or 4 days before the surgical procedure. The international normalized ratio (INR) [see General Principles of Anticoagulation and Lytic Therapy, below] should be kept below 3.0 to prevent excessive bleeding. Warfarin is not as easy to regulate as heparin is. In addition, the therapeutic effect takes several days to be realized and several more days to wear off. Frequently, postoperative patients are unable to resume a normal stable diet. Because warfarin interferes with the clotting factors in the vitamin K pathway, dosage management in the immediate postoperative period is challenging. Because the risk of intracranial bleeding is greater with warfarin than with heparin, great care should be taken in using warfarin for immediate perioperative prophylaxis.

Characteristic clinical manifestations of superficial thrombophlebitis [see Figure 1] include pain and slight swelling of the extremity, with most of the edema over the course of the involved vein. Unless the patient is obese, a palpable, tender subcutaneous cord is usually found (a pathognomonic finding). Erythema may be present in the overlying skin. The differential diagnosis includes cellulitis and streptococcal lymphangitis. For both conditions, there should be a proximal source (e.g., an open wound). If there is overt limb swelling in a patient who appears to have superficial phlebitis, DVT should be assumed and appropriate treatment [see Deep Vein Thrombosis, below] initiated.

Superficial thrombophlebitis is largely benign but is often overtreated out of fear that infection may be contributing to phlebitis. It is therefore important to differentiate between sterile and septic superficial thrombophlebitis.

When DVT develops in an outpatient, every effort should be made to determine the cause [see Figure 2]. Apparent spontaneous onset is often the manifestation of an underlying malignancy or even a congenital clotting tendency that will necessitate lifelong treatment [see Hypercoagulability States, below]. Conversely, when risk factors for DVT are identified in a hospitalized patient, it can be assumed that the cause is acquired and that the clotting tendency will be reversed upon recovery.

Before therapy is begun, the diagnosis should be verified. The differential diagnosis includes muscle contusion, plantar or gastrocnemius muscle rupture, ruptured Baker's cyst, popliteal artery aneurysm, arthritis of the knee or the ankle, cellulitis, and myositis. The gold standard for diagnosis of DVT is ascending venography. However, study of the entire lower-extremity venous system often involves injection of dye not only at the foot or ankle level [see Figures 3a and 3b] but also at the groin level for visualization of the iliac and femoral veins. This approach is uncomfortable and is associated with morbidity; in critically ill ICU patients, it may not be feasible. Accordingly, noninvasive evaluation techniques are favored.

The presence of intravascular clot can be confirmed by detecting D-dimer, a product formed when cross-linked fibrin is broken down by the fibrinolytic system. Both qualitative and quantitative assays are in current use. The various quantitative assays available have differing negative predictive values. The gold standard is the enzyme-linked immunosorbent assay (ELISA) method. Generally, 500 �g/L (in fibrin-equivalent units) or 250 �g/L (in standard units) is an acceptable threshold for a positive result. The latex agglutination test, though inexpensive, has an unacceptably low sensitivity and is the one quantitative method that should not be used. Because some amount of physiologic intravascular clot (e.g., at a wound site) is to be expected in many, if not most, patients at risk for DVT, a positive D-dimer assay is of little diagnostic value. D-dimer assay is therefore unsuitable for screening. In patients suspected of having DVT or PE, however, a negative D-dimer assay can, for the most part, rule out DVT and, by extension, PE.^32�34

Various forms of plethysmography (e.g., impedance plethysmography) have been used to identify nonocclusive DVT.^35,36 These techniques are accurate only when there is at least 50% obstruction of the lumen of a deep vein. The presence of large collateral vessels may result in a false-negative test result as well. Doppler ultrasonography can be performed quickly and easily, though interpretation of the results requires considerable experience. It has essentially the same drawbacks as plethysmography.³⁵

Real-time B-mode (duplex) ultrasonography can be valuable in this setting.^35,^37,38 It can actually visualize thrombus within a vessel. Inability to obliterate the vein with probe compression is additional evidence of thrombus. Often, experienced users can even differentiate between new and old thrombi on the basis of echogenicity. Duplex ultrasonography is quite sensitive and specific in patients with suspected DVT, and its diagnostic qualities can be further enhanced by the addition of color flow imaging. It has in fact become the noninvasive procedure of choice for assessment of clot in the neck and the extremity vessels. Unfortunately, it is less specific proximal to the axilla and the inguinal ligament, where compression of the vessels is difficult or impossible. Duplex ultrasonography is particularly valuable for detecting associated conditions that may confuse the diagnosis (e.g., muscle hematomas or a Baker's cyst).³⁸

Ultimately, ascending venography is the most accurate method of diagnosing DVT.^39,40 If a good contrast study fails to demonstrate the presence of clot, DVT is conclusively ruled out.

Once nonocclusive DVT is diagnosed, the treatment of choice is initial therapeutic-dose heparin therapy followed by warfarin therapy. If the patient is responsible and reasonably well educated, initial heparin anticoagulation can be done on an outpatient basis with subcutaneous LMWH.^41,42 If this approach is not appropriate, inpatient therapeutic-dose I.V. heparin anticoagulation is employed. After 3 or 4 days, depending on the response, heparin is replaced with warfarin.

Lower-extremity occlusive DVT [see Figure 4] is usually associated with swelling; however, if good collateral circulation or duplicate veins are present, especially in the thigh, only local inflammation may be apparent. Typical findings include pain and tenderness over the involved veins as well as swelling in the distal limb (which may be minimal with the patient supine). The differential diagnosis is essentially the same as for nonocclusive DVT. In addition, lower-extremity DVT can be associated with PE: free-floating clot may occur in conjunction with occlusive clot.

If both legs are swollen, the proximal extent of the thrombus is likely in the vena cava. If one entire leg is swollen, the proximal extent must be in the iliac veins. If the swelling is limited to the lower leg below the knee, the thrombus is probably in the superficial femoral vein. If the only manifestations are minimal swelling and calf tenderness, the thrombus is probably limited to the sural vein, the gastrocnemius vein, or both.

If the patient has a history of DVT, is hospitalized, and is at risk for recurrence, heparin therapy may be started before test results are available (provided that there is no contraindication to anticoagulation). If the patient is an outpatient, is hospitalized but lacks risk factors for DVT, or has a contraindication to anticoagulation, a D-dimer assay should be performed. If the assay is negative, an alternative diagnosis should be sought. If it is positive, the diagnostic workup of DVT should proceed. If diagnostic studies yield equivocal results and venography is difficult or impossible, treatment should proceed as if the diagnosis had been confirmed.

The treatment of choice is immobilization in bed, elevation of the limb (with or without elastic compression), and therapeutic-dose heparin (unfractionated heparin or LMWH), followed by 3 to 6 months of warfarin therapy. If the episode is mild, recovery is usually prompt. If pain and swelling do not respond promptly to anticoagulation, either the diagnosis is wrong or anticoagulation is inadequate. Lytic therapy combined with heparin anticoagulation may be superior to heparin anticoagulation alone, leading to better clearance of clot from the valves with improved function and less risk of postphlebitic syndrome.⁴³

For all practical purposes, upper-extremity DVT [see Figure 5] involves only the axillary, the subclavian, or the innominate vein (or a combination thereof). Involvement of the superior vena cava is rare, mainly occurring in chronic conditions (e.g., long-term venous catheterization). Arm thrombophlebitis is characterized by pain and swelling with tenderness over the involved vein. Often, it is relatively asymptomatic: because of the excellent collateral circulation in the arm, thrombosis must be extensive to produce marked swelling.

Spontaneous onset of axillary or subclavian vein thrombosis can occur in association with thoracic compression syndromes (effort thrombosis) or as a complication of so-called Saturday night palsy, in which an alcoholic sleeps with the axilla compressed by the arm of a chair. If a potential mechanical cause is not apparent, other possible causes must be explored. The onset of swelling, tenderness, or fever in a patient with a central venous catheter is an indication for removal of the catheter. If there is no bacteremia or fever, if there has been a catheter in the vein, and if the problem developed spontaneously, sterile thrombophlebitis may be assumed. In these cases, once the catheter is removed, anticoagulation is unnecessary.⁴⁴

Subclinical nonocclusive clot is probably of little significance because documented PE from the upper extremity is quite rare. Noninvasive tests [see Nonocclusive, above] typically yield positive results when the upper extremity is involved.³⁹ Moreover, distal vein catheterization is easy, and phlebography is relatively uncomplicated. These techniques should be used whenever the diagnosis is in doubt.

The morbidity of occlusive upper-extremity DVT can be substantial. Thus, if the patient presents with massive swelling of the upper limb, therapeutic-dose heparin anticoagulation should be initiated immediately, and consideration should be given to lytic therapy.⁴³ If phlebography shows compression of arm veins at the thoracic inlet after lytic therapy or spontaneous recovery from the thrombotic process, resection of rib 1 may be indicated, particularly if positional morbidity is present.^45,46

Septic DVT is more common in the upper extremity than in the lower, primarily because upper-extremity veins are more frequently catheterized and more often used for injection of illicit drugs. If phlebitis occurs in a catheterized vein with fever and sepsis, the catheter should be removed immediately, the tip cultured, and Gram staining done on any clot present. Broad-spectrum antibiotics should be administered until more specific antibiotic therapy can be instituted. Heparin anticoagulation is the primary treatment unless contraindicated.

Ligation and drainage are not as practical for deep veins as for superficial veins, but either may be indicated on rare occasions if the process does not respond to conventional therapy within 3 or 4 days and marked swelling and fever persist. Drainage is done on the most accessible portion of the phlebitic process. For ligation, the proximal end of the process should be identified via surgery or phlebography and the vein then ligated proximally.

Phlegmasia cerulea dolens [see Figure 6] is most apt to occur in dehydrated, cachectic patients and is usually superimposed on another critical illness. It can involve either the upper or the lower extremity but more commonly affects the lower. In the lower extremity, there is usually simultaneous thrombosis of the iliac, femoral, common femoral, and superficial femoral veins. The limb is massively swollen, bluish, and mottled. Eventually, it becomes nonviable as arterial flow stops because of arterial spasm associated with venous outflow obstruction. The problem is compounded by acute massive fluid loss into the limb, which can result in hypovolemic shock.

Treatment involves rapid and aggressive fluid replacement, elevation of the limb, and aggressive heparin anticoagulation or catheter-directed lytic therapy.^47,48 If the patient does not respond, thrombectomy may be considered, provided that the associated disease does not carry a fatal prognosis.⁴⁹ The procedure is best done transfemorally with a limited incision so that anticoagulation can be continued postoperatively. If anticoagulation cannot be continued, thrombophlebitis will recur immediately.

It is widely agreed that PE is grossly underdiagnosed.^3,4,^50,51 Most episodes (up to 90%) are unsuspected,^52,53 and only a minority (10% to 25%) of fatal episodes are diagnosed before death. Clinical manifestations include dyspnea, hemoptysis, pleurisy, heart failure, and cardiovascular collapse; however, each of these is also associated with other conditions [see Table 2]. Risk factors for PE are similar to those for DVT [see Table 1].

PE should be distinguished from pulmonary infarction. Of the approximately 10% of all pulmonary emboli that are recognized clinically, only 10% are associated with pulmonary infarction.⁸ Because the lung has excellent collateral circulation, obstruction of the larger pulmonary arteries rarely leads to death of lung tissue. When pulmonary infarction does occur, the diagnosis is usually obvious; hemoptysis, pleuritic chest pain, and a wedge-shaped density on chest x-ray are the classic manifestations. In most PE patients (i.e., those without pulmonary infarction), these findings are absent, and the chest x-ray may even be normal.

For the purposes of clinical diagnosis and treatment, PE is best classified as minor (or suspected), moderate, or catastrophic [see Figures 7a, 7b, and 7c].

Manifestations of minor PE [see Figure 7a] may include transient tachypnea (with perhaps a slight change in blood gas values) and cardiac irritability (with frequent premature beats or tachyarrhythmias).^54�56 These changes may resolve in moments, and the patient may then appear perfectly normal. In these circumstances, the embolus probably either is small or is composed of relatively fresh clot that produces only transient obstruction when it enters the pulmonary vascular tree.

It was long believed that after operation, hospitalization, or injury, the earliest PE might occur was 4 to 7 days after the insult. A 1997 study of previously healthy trauma patients, however, found that approximately 25% of the PE episodes occurred in the first 4 days after injury.⁵⁷ Accordingly, the presence of clinical signs and symptoms consistent with PE in a patient with risk factors calls for appropriate workup, regardless of how soon after the insult they appear.

The differential diagnosis includes acute respiratory distress syndrome (ARDS), aspiration, atelectasis, heart failure, pneumonia, and systemic infection. If the diagnosis is not obvious but the risk of PE is substantial and there is no contraindication to anticoagulation, heparin therapy (therapeutic-dose unfractionated heparin or LMWH) may be instituted while diagnostic tests are being selected and performed.

If PE is unlikely, the risk from anticoagulation is high, or other serious diagnostic possibilities cannot be ruled out, specific studies (e.g., intravascular coagulation, spiral computed tomography, or pulmonary angiography) should be ordered before anticoagulation.

In a stable patient with suspected PE, a blood D-dimer level should be obtained. If the result is negative, PE can be excluded and further diagnostic studies canceled. As with DVT, a positive D-dimer assay does not confirm DVT or PE. The negative predictive value of the assay for DVT and PE is between 90% and 100% when an appropriate assay with an appropriate cutoff value is used.^31�33 If a patient is experiencing a life-threatening respiratory event consistent with PE, the D-dimer assay should be skipped, therapy instituted, and formal diagnostic studies performed.

Noninvasive evaluation of the legs may establish the presence of DVT necessitating anticoagulation. This circuitous way of establishing the diagnosis of PE has severe limitations. When noninvasive assessment aimed at detecting clot in the major leg veins is done before documented PE, it yields positive results in only 33% to 45% of cases.⁵⁸ Venography is more sensitive than duplex ultrasonography. When it is used to diagnose venous thrombosis, as many as 30% to 40% of patients with PE are found not to have clot in the major veins of the leg or the abdomen. If the duplex scan or venogram is positive for DVT and there are no contraindications, anticoagulation may be begun. If, on the other hand, a patient is suspected of having PE but is sent for a duplex scan in place of a pulmonary arteriogram, and the duplex scan is negative, workup must continue. It is unacceptable in such cases to assume that the negative result excludes PE.

The initial enthusiasm for the use of lung scans to diagnose or screen for PE has diminished.^59,60 Current thinking about the use of scans for this purpose may be summarized as follows. If a scan is read as high probability, there is roughly an 85% chance that the diagnosis is correct. If a scan is read as normal, there is roughly a 5% chance that the patient had PE. If a scan is read as low or intermediate probability (the most likely scenario), the likelihood that the diagnosis is correct is little better than random chance.

As a result of the dramatic improvements in CT imaging, many proposed replacing pulmonary angiography with CT. There were several good arguments for this proposal. First, CT scanning is less invasive than pulmonary angiography. Second, it does not require the immediate presence of a radiologist. Third, it is less costly at most institutions. Finally, CT scanning is usually more easily obtained than pulmonary angiography. The initial results from CT scanning for PE were quite promising.⁶¹

In the past few years, CT has undergone rapid increases in sophistication. Specifically, as CT technology has moved from single-detector to multiple-detector (light-speed) scanners, the sensitivity achievable with this modality has skyrocketed. As a result, the negative predictive value of CT in this setting�that is, the degree to which a negative CT pulmonary angiogram can be relied on to rule out PE�has risen dramatically.⁶² For most institutions, the result of this technologic improvement is that CT angiography is now the primary radiologic diagnostic study for excluding PE. It has been argued that CT angiography cannot yet replace pulmonary angiography altogether, because thrombolytic therapy (if required) may be provided during pulmonary angiography, whereas it typically cannot be provided during CT angiography. This argument probably applies only to the sickest PE patients, who are the ones most likely to benefit from thrombolytic therapy and least likely to be able to tolerate a second load of dye (as when pulmonary angiography is done for treatment purposes after CT angiography).

For critically ill patients, who tolerate diagnostic testing poorly, pulmonary angiography is a more appropriate initial study [see Figure 8].^3,^50,⁶³ If the angiogram is obtained immediately after the clinical episode, particularly if the patient is still symptomatic, a negative result rules out PE. However, if the patient improves or recovers before angiography, the angiogram may be falsely negative, implying that the clot was minimal or was disposed of by natural lytic processes. Thus, a negative angiogram in such a patient does not unequivocally rule out PE.⁶⁴ The pulmonary angiogram does, however, establish the degree of patency of the pulmonary vasculature, which affects prognosis.

PE can occur even immediately after injury in previously healthy persons.⁵⁷ These early emboli are generated from fresh clot and thus are more easily fragmented and broken down. They are much more likely to be found in the periphery.⁶⁵ For these reasons, CT is less sensitive in detecting them.⁶⁵ Accordingly, patients with suspected PE shortly after injury or operation should undergo pulmonary arteriography.

If all diagnostic tests for PE yield negative results, therapeutic anticoagulation is not indicated; however, if risk factors are present, prophylaxis is indicated. If test results are suggestive or indicative of PE, therapeutic heparin anticoagulation should be continued.

Manifestations of moderate PE [see Figure 7b] include transient hypotension, tachycardia or other cardiac dysrhythmias, tachypnea with a significant fall in arterial oxygen and carbon dioxide tension, apprehension, and symptoms or signs of pulmonary infarction^54,56; there may be signs of right heart failure as well. Electrocardiography is rarely helpful in the differential diagnosis. Acute right axis deviation, new incomplete right bundle branch block, and changes in S₁, Q₃, or T₃ are thought to characterize this disorder but are found in only a small percentage of patients with proven PE.

If the diagnosis is probable and there are no likely alternatives, heparin therapy should be initiated. Lytic therapy is of debatable utility in these patients: compared with standard heparin therapy, it appears not to reduce mortality or pulmonary dysfunction significantly, yet it carries a higher risk of bleeding.^3,⁶⁶ Moreover, lytic therapy is often contraindicated because of recent surgery, injury, or vascular punctures.

If there is a relative contraindication to anticoagulation (e.g., an acute surgical wound, a previous bleeding episode, or an allergic reaction to heparin) or alternative diagnoses are likely, specific diagnosis is required, ideally via pulmonary angiography. Peripheral noninvasive venous studies may be helpful because if they show significant venous obstruction, the likelihood of PE increases and the need for therapy is documented. Ventilation-perfusion scanning, again, is valuable only if strongly positive.

If there is no contraindication to heparin therapy and the diagnosis is strongly suspected but not confirmed, therapeutic-dose heparin anticoagulation should be started.^67,68 Treatment is continued if the diagnosis is verified and stopped if the diagnosis is excluded.^69�72

Vena caval interruption has often been recommended for patients with documented PE despite apparently adequate systemic anticoagulation, but many authorities now advocate vena caval filter placement even in patients who do not have documented thromboembolism but are at high risk and in whom anticoagulation is contraindicated.^73�75 Supporting data come largely from studies with historical controls. In the one prospective, randomized, controlled trial involving patients with thromboembolic disease, there was no reduction in mortality at any time, nor was there even a reduction in PE at 2 years; however, there was an increased incidence in DVT over that period.²⁹ These findings suggest that vena cava filters should be reserved for patients with documented DVT, a contraindication to anticoagulation, and a high risk of subsequent PE.

In 2005, an 8-year follow-up analysis of the patients enrolled in the aforementioned prospective, randomized trial²⁹ was published.⁷⁶ Over the 8-year period, the incidence of PE was lower in the group that received inferior vena cava filters than in the group that did not; however, there was an increased rate of recurrent DVT in the filter group. Overall, there was no significant difference in mortality between the two groups. The authors reiterated their recommendation that vena cava filters be used with restraint.

If vena caval interruption is considered necessary, it should be done percutaneously via either the jugular or the femoral route. If it is done by the latter route, phlebography (from the insertion site through the vena cava) should be performed first to document the absence of clot along the planned route. A number of different vena caval filters are currently on the market. Each is slightly different from the others, but none has demonstrated clear superiority in preventing PE or reducing caval thrombosis. Surgeons should be aware of the advantages and disadvantages of the types available at their institutions.

Removable filters are now widely available. Several have been approved by the FDA and are in clinical use. Because these devices were approved comparatively recently, data on long-term efficacy and complication rates are not yet available.

One of the most critical questions that remains unanswered is, what will be the long-term consequences when the damage caused by the venotomy required for filter removal is superimposed on the endothelial damage caused by the filter? This question should be kept in mind, as should the reservations expressed by the authors of the only prospective, randomized trial of inferior vena caval filters published to date.^29,⁷⁶ Still, it appears that removable vena caval filters are likely to have a role to play in this setting. They should be considered in patients who are at extremely high risk for PE, who have an absolute contraindication to anticoagulation for a finite period, and in whom anticoagulation can be instituted upon removal of the filter.

Catastrophic PE [see Figure 7c] is most apt to be superimposed on a critical illness or a major operation. The peak incidence is 7 to 10 days after the procedure or the onset of clinical illness, though emboli may occur at any time.⁵⁴ The reason for this apparent delay is that for the clot to remain intact after embolization to the pulmonary vasculature, it must mature in the vascular system, a process that takes several days. Fresh clot breaks up readily and dissipates promptly, whereas older clot is resistant to lysis. The manifestation of early embolization of fresh clot to the pulmonary vasculature is ARDS. Embolization of older clot can produce acute pulmonary obstruction and acute right heart failure, making radiologic diagnosis of PE relatively easy.⁵⁴ Occlusion of large portions of the vasculature is associated with hemodynamic catastrophe.

Typically, the clinical onset of catastrophic PE comes when a patient, having just been mobilized, performs a vigorous Valsalva maneuver in the course of his or her first postoperative bowel movement. The great abdominal veins distend, and any clot present tends to be stripped loose. If a large clot embolizes, immediate collapse and cardiac arrest may result; in some cases, bradyarrhythmia or severe hypotension precedes the actual arrest. Immediate emergency treatment comprises intubation and administration of 100% oxygen, heparin anticoagulation, and, if cardiac arrest occurs, cardiopulmonary resuscitation. A Swan-Ganz catheter should be inserted as soon as possible so that the effects of therapy can be monitored. Cardiotonic agents (e.g., dopamine, 2.0 to 5.0 �g/min, or dobutamine, 2.5 to 10.0 �g/kg/min) should be administered to strengthen myocardial function. If sudden arrest occurs in circumstances that permit emergency thoracotomy, Trendelenburg's procedure can be performed; however, it is rarely indicated and even more rarely successful. If the patient survives initial emergency treatment and improves, high-dose heparin therapy should be continued and lytic therapy considered.^71,⁷⁷

Unfractionated heparin therapy is also frequently referred to as conventional anticoagulation.^78�80 Before therapy is begun, a clotting battery should be performed, consisting of the aPTT, the INR, the platelet count, and levels of fibrinogen, antithrombin, and D-dimer. High fibrinogen levels and platelet counts are seen in patients with chronic clotting syndromes,⁸¹ probably representing overcompensation for increased utilization. Elevated D-dimer levels suggest intravascular clotting with activation of the fibrinolytic system.

In the average patient, therapeutic-dose heparin anticoagulation begins with administration of 5,000 to 10,000 units, followed by continuous I.V. infusion at a rate sufficient to double or triple the aPTT�typically, 1,000 to 2,000 units/hr. When dosages higher than 2,000 units/hr are required, antithrombin depletion is highly probable.

Tight control of the aPTT change as a result of heparin therapy is not as important as monitoring for evidence of bleeding and platelet depletion. Clinical evidence of bleeding is not necessarily a contraindication to anticoagulation. Minimal amounts of blood may be lost in the urine or through the GI tract; if the patient has a clearly identifiable need for anticoagulation, such minor blood loss should be accepted. Only when transfusion is indicated to maintain the hematocrit should discontinuance of heparin be considered. At that point, if the risks of bleeding seem to outweigh the benefits of anticoagulation, heparin infusion can be stopped or reduced to prophylactic levels. It is important to watch for falls in the hematocrit indicative of significant bleeding. The most common sites for hemorrhagic complications are surgical wounds and the retroperitoneum. Retroperitoneal bleeding is generally asymptomatic until the patient progresses to hemorrhagic hypovolemic shock.

As a rule, the therapeutic dose of LMWH is twice the prophylactic dose. The various LMWHs currently on the market all have slightly different activities and half-lives. Enoxaparin may be taken as prototypical. The accepted prophylactic dose for enoxaparin is 30 mg twice daily, and the therapeutic dose is 60 mg (or 1 mg/kg) twice daily.

A major benefit of using LMWHs to treat DVT and PE is that therapeutic doses can be given subcutaneously.^41,42 As a result, patients may be treated as outpatients both in the acute phase of the disease and in the subacute phase, during the transition to oral anticoagulants. This approach requires that patients be clinically stable and able to follow dosing instructions. Because there are no validated methods of monitoring LMWH therapy, an initial assessment of antithrombin activity is appropriate. If this is low, unfractionated heparin therapy in conjunction with aPTT monitoring is probably preferable.

High-dose heparin therapy is reserved for patients who are dying of PE or are at risk for immediate limb loss from phlegmasia cerulea dolens. Such therapy consists of administering a large enough dose of heparin to elevate the aPTT off the scale. The maximum aPTT that can be measured by our laboratory is 150 seconds; high-dose heparin treatment should therefore yield an aPTT higher than this value. Theoretically, given that a fully anticoagulated patient should not form clot at all, the aPTT should be infinite. This method of treatment may be used in patients with immediately life-threatening PE or phlegmasia cerulea dolens when the more conventional technique, catheter-directed thrombolytic therapy, is unavailable.

In most patients, high-dose therapy begins with a 20,000 unit I.V. bolus, followed by infusion of 5,000 units/hr I.V.^82�86 The end point of therapy is clinical evidence of improvement. In patients being treated for PE, pulmonary function should improve.

Because complete anticoagulation is the essential principle of high-dose heparin therapy, there is no need to be concerned about an upper limit for the dosage: if the patient cannot clot, doubling or even tripling the dosage should not increase the risk of bleeding. Moreover, because the incidence of bleeding is very low in the first 2 or 3 days of therapy, regardless of the dosage,^85,86 high initial dosages do not carry an unacceptable bleeding risk. After heparin has been observed to have an effect and a prolonged aPTT documented, the high dosage should be continued for at least 24 hours, then decreased by 500 to 1,000 units/hr over the next 24 hours. If the clinical effect is maintained and improvement continues, the dosage can be decreased by another 500 to 1,000 units/hr over the following 24 hours. In theory, once all clotting stops, natural antithrombin levels should recover, allowing lower dosages of heparin to be effective. After 3 or 4 days of therapy, the dosage may be reduced to more conventional levels [see Therapeutic Dose, above].

If the initial improvement is lost, the dosage should be restored to its previous high level and maintained there for several days before any attempt is made to reduce it again. The platelet count and the hematocrit should be carefully monitored, the latter at least four times a day. Heparin should be discontinued or the dosage reduced only when the risks of bleeding and transfusion exceed the benefits of anticoagulation. In a monitored environment, patients very rarely die of hemorrhage; rather, they die of the consequences of clotting.

The most devastating hemorrhagic risk of heparin therapy�fortunately, a rare one�is intracranial bleeding. The risk of major hemorrhage ranges from 4% to 9% and is directly affected by how tightly the INR is controlled.⁸⁷ The risk is greatest in elderly patients, particularly women,⁷⁹ but it is still small in comparison with the obvious risks posed by the clotting episode. Nevertheless, the existence of this risk makes it appropriate to use high-dose heparin primarily in life-threatening conditions.

A more common complication of full heparin anticoagulation is retroperitoneal bleeding. This problem is accentuated in elderly patients. Because aging is associated with loss of connective tissue elasticity, bleeding into retroperitoneal connective tissue that would normally be insignificant can become life-threatening. Usually, this is not a serious problem if the hematocrit is followed, heparin dosing adjusted, and lost blood replaced. When the perceived risk of bleeding outweighs the thrombotic risk, heparin should be discontinued.

Two forms of acute heparin-induced thrombocytopenia (HIT) have been reported.^88�91 Mild HIT occurs in 2% to 5% of patients 2 to 15 days after the initiation of therapy. The platelet count usually remains at about 100,000/mm³, and treatment can be continued without undue risk of bleeding or thrombosis. Severe HIT is much less frequent. It usually occurs about 7 to 14 days after the initiation of heparin therapy and is reversible once the drug is discontinued. It is not dependent on the heparin dose given. Clinical manifestations include a substantial (at least 50%) drop in the platelet count followed by a thrombotic episode (frequently both arterial and venous). An ELISA directed at the platelet factor 4-heparin complex is generally accepted for laboratory confirmation of the diagnosis. Treatment consists of discontinuance of heparin. If the patient still requires anticoagulation, a different anticoagulant must be used. The most widely accepted agent for this purpose is lepirudin (a direct thrombin inhibitor).⁹² Other direct thrombin inhibitors (e.g., argatroban) and heparinoids (e.g., danaproid) have also been successfully used to provide anticoagulation in patients with HIT. None of the LMWHs are acceptable in this setting, and at present, the FDA does not allow use of the pentasaccharide in severe HIT.

Unlike warfarin, heparin does not cross the placenta and has not been associated with fetal malformations; thus, it is preferred for thrombotic complications of pregnancy. Heparin can be administered subcutaneously in an outpatient setting for 3 to 6 months. Long-term administration can lead to osteoporosis and spontaneous vertebral fractures.⁹³

Very rarely, heparin therapy can lead to adrenal hemorrhage and consequent adrenal insufficiency.⁷⁹ If acute adrenal insufficiency is suspected, anticoagulant therapy should be discontinued and high-dose steroid therapy (preferably with hydrocortisone) initiated. Treatment should not await laboratory confirmation. CT scanning may be useful. Heparin may suppress aldosterone synthesis, especially with prolonged use.⁷⁹

The anticoagulant effect of heparin disappears within hours after discontinuance. If the effect must be reversed quickly, the patient should receive protamine sulfate I.V. This agent binds the heparin and prevents it from activating antithrombin. Protamine sulfate should be given in the smallest dosages that still evoke the desired result�typically, about 1 mg for every 100 units of heparin remaining in the patient. It should be administered slowly over 5 to 10 minutes; rapid infusion can cause shortness of breath, flushing, bradycardia, hypotension, or anaphylaxis. On rare occasions, patients previously sensitized to protamine may experience massive platelet aggregation, as manifested by catastrophic arterial thrombosis. Patients particularly likely to manifest this adverse reaction include diabetics and persons with fish allergies. Protamine has little or no capacity for reversing either LMWH or fondaparinux. Research aimed at developing specific protamines to inactivate the LMWHs is now being carried out, but at present, no such agents are available in the United States.

Historically, warfarin dosage has been regulated by monitoring the prothrombin time (PT), with a PT 1.5 to 2.5 times normal (11 or 12 seconds) generally considered to represent the optimal level. In response to the wide variations in PT reported by different laboratories, the World Health Organization (WHO) has recommended substituting the INR for the PT ratio so that all laboratory assessments will be comparable.⁹⁴ An INR of 2.0 to 3.0 corresponds to a PT that is 1.3 to 1.5 times normal (moderate dose); an INR of 3.0 to 4.5 corresponds to a PT that is 1.5 to 2.0 times normal (high dose). Lower INRs are recommended for all but extremely high-risk patients (e.g., those with mechanical heart valves) [see Table 3].⁹⁴

Initially, the daily dose of warfarin required to increase the INR to between 2.0 and 3.0 is estimated and administered. The INR is then checked every morning. If it suddenly overshoots the target range, the warfarin dosage is reduced. If the INR has not reached or surpassed 1.5 after the third dose, the dosage is increased. The maintenance dosage averages about 5 mg/day but may range from 1 to 10 mg/day.

While the maintenance dosage is being determined, the INR should be checked daily. Once the patient stabilizes, the INR can be checked less often: twice weekly for the first few weeks, once weekly for the next several months, and once monthly thereafter if the patient is stable. The patient should be cautioned about drug interactions. If the dosages of other medications are changed, the impact on the INR should be investigated and the warfarin dosage adjusted as appropriate.

There is no general agreement on how long oral anticoagulant therapy should be continued after a thromboembolic event. Current data suggest that the duration of therapy should be based on the level of underlying risk rather than on the severity of the event. For patients with a limited risk period (e.g., a young patient with a femur fracture and no other risk factors), an 8- to 12-week course is as efficacious as a longer course.⁹⁵ For patients with a lifelong risk (e.g., a patient with a congenital hypercoagulability syndrome or cancer), a therapeutic dosage for 3 to 6 months, followed by a low dosage for the remainder of the patient's life, is indicated.⁸⁷ Lengthening the duration of full anticoagulant therapy in patients with long-term risk factors appears only to delay the recurrence of thromboembolism, not to prevent it.

Response to warfarin is affected not only by various bodily factors but also by drug interactions [see Table 4]. Such interactions are most dangerous when drugs administered in parallel are taken intermittently.^79,80 Increased metabolic clearance of the drug can result from administration of barbiturates, rifampin, or phenytoin; long-term use of alcohol; ingestion of large amounts of vitamin K; and rich foods. Elevated levels of coagulation factors during pregnancy also decrease warfarin's effectiveness.

Decreased metabolism or displacement from protein-binding sites caused by phenylbutazone, sulfinpyrazone, metronidazole, disulfiram, allopurinol, cimetidine, amiodarone, or acute intake of ethanol can elevate the INR and increase the risk of hemorrhage. Relative vitamin K deficiency, resulting from inadequate diet or the elimination of the intestinal flora by antimicrobial agents, may have similar effects. For these reasons, warfarin should be used with great caution in patients who are receiving antibiotics or who cannot tolerate a regular diet.

There are some serious interactions that increase the risk of bleeding without altering the INR. These include inhibition of platelet function by drugs such as aspirin and gastritis or gastric ulceration induced by anti-inflammatory drugs. Obviously, when placing a patient on more than one anticoagulant simultaneously, great care must be taken.

Bleeding is the major complication of oral anticoagulation. Tight control of warfarin therapy is essential for minimizing this complication.⁸⁷ Bleeding is rare when the INR is kept below 3.0. When bleeding does occur, a preexisting lesion is likely. If the bleeding is minor, the warfarin dosage should be adjusted; if it is major, the drug may have to be discontinued. The risk of intracerebral or subdural hematoma is greater with warfarin than with heparin, particularly in patients older than 50 years. If there is any sign of hemorrhage, the next anticoagulant dose should be withheld and the INR measured. For continued or serious bleeding, 5 to 10 mg of vitamin K1 oxide (phytonadione) I.V. is effective. Several hours may pass before hemostasis improves significantly, and 24 hours or longer may be needed for maximal effect. If immediate restoration of hemostatic competence is necessary, levels of vitamin K-dependent coagulation factors can be raised by giving fresh frozen plasma, 10 to 20 ml/kg body weight, or prothrombin complex concentrate.⁹⁶

Administration of warfarin during pregnancy can cause birth defects and abortion and therefore is contraindicated. Warfarin-induced skin necrosis is a rare complication of oral anticoagulant therapy.⁹⁰ This syndrome, characterized by the appearance of skin lesions shortly after initiation of treatment, may be the result of a transient hypercoagulable state caused by depletion of the natural anticoagulants (proteins C and S) before the onset of warfarin's effect. To mitigate the initial hypercoagulable state, some advocate starting warfarin therapy only after initial heparinization.

A newer oral anticoagulant, the direct thrombin inhibitor ximelagatran, has been approved for use in Europe and has been widely studied; however, it has not been approved by the FDA for use in the United States. Initial studies of this agent yielded promising results, raising hopes that there might be an effective oral anticoagulant that is safer and easier to administer than warfarin.

In a prospective, double-blind study from 2003, more than 1,800 patients undergoing knee surgery were randomly assigned to receive either ximelagatran or warfarin.⁹⁷ Venography was performed on day 7. Warfarin was given in standard dosages, with a target INR of 2.0 to 3.0. Ximelagatran was given in two different doses (as a dosing trial), without monitoring. The authors observed a significantly lower incidence of DVT in the ximelagatran group, with no significant difference in bleeding. After the publication of this study, several reports of adverse reactions (specifically, liver failure) were reported to the FDA.

In a subsequent study, almost 4,000 patients with atrial fibrillation were randomly assigned to receive either warfarin or ximelagatran for stroke prophylaxis.⁹⁸ The authors found no significant differences in the rates of stroke, embolism, or death but did note a significantly higher incidence of bleeding in the warfarin group. They also found that a significant percentage of the patients in the ximelagatran group had increases in their hepatic enzymes that were correlated with administration of the drug. In addition, there were cases of fatal hepatic failure that could have been associated with the drug.

Thus, ximelagatran appears to hold great promise, but it is unclear whether it will be approved for the American market.

A number of different lytic agents have been studied [see Table 5]. At present, however, lytic therapy is generally understood to refer to administration of streptokinase, urokinase, or t-PA,^79,80,^99�102 all of which act on the endogenous fibrinolytic system to convert plasminogen to plasmin. Streptokinase combines with plasminogen to form streptokinase-plasminogen complexes that are converted to streptokinase-plasmin complexes, which then convert residual plasminogen to plasmin.^79,¹⁰² Urokinase directly cleaves a peptide bond in the plasminogen molecule to form plasmin. t-PA binds to fibrin via lysine binding sites at the N-terminal.

Lytic therapy is most effective when it can be initiated within hours. It is worth attempting when the clot has been present for less than 1 week, particularly if it has been present for less than 3 days. When lytic therapy is begun, heparin therapy usually is temporarily discontinued because of the theoretical possibility of increased bleeding risk; it may be resumed immediately upon completion of lytic therapy. If, however, the problem is immediately life-threatening (e.g., myocardial infarction or massive PE), anticoagulation should probably be done in parallel with lytic therapy to prevent rethrombosis.

The indications for lytic therapy are being extended.⁹⁶ Urokinase, t-PA, and, to a much lesser degree, streptokinase are being used for venous thrombotic conditions, such as symptomatic obstruction of major upper-extremity veins. The morbidity of axillary vein thrombosis can be considerable; clearance of clot may not only help restore patency but also help identify the underlying cause. In the lower extremities, more thorough clearance of clot should, in theory, help restore valve function and prevent so-called postphlebitic syndrome.^101,¹⁰³

Contraindications to lytic therapy include surgery in the previous 10 days, serious GI bleeding in the previous 3 months, a history of hypertension, an active bleeding or hemorrhagic disorder, a previous cerebrovascular accident, and an active intracranial process. As with heparin, the risk of intracranial bleeding is increased in older patients; the risk appears to be higher with t-PA than with streptokinase or urokinase.

Streptokinase Streptokinase is a 47 kd protein produced by b-hemolytic streptococci. Because it is not endogenous, circulating antibodies to it (from previous streptococcal infections) often are already present in plasma. When streptokinase is infused, a loading dose must be given to overcome these antibodies. Once the antibodies are depleted, the half-life of streptokinase is about 80 minutes. Achievement of the desired therapeutic effects is confirmed by documenting a rise in the thrombin time (TT), a fall in the fibrinogen level, or an abrupt rise in the D-dimer level. Because streptokinase may deplete circulating plasminogen after a few hours, the optimal approach may be to administer it for 6 hours by continuous infusion every 24 hours for 2 or 3 days, then to administer heparin in the intervals between infusions.

Urokinase Urokinase is a 34 kd globulin originally found in human urine and now isolated from cultured human cells. It has a half-life of 15 minutes and is metabolized by the liver. It was removed from the U.S. market for several years as a result of concerns expressed by the FDA, but it subsequently was reintroduced after these concerns were satisfactorily addressed. For catheter clearance, a solution containing 5,000 units/ml should be infused into the obstructed tubing. Urokinase is not fibrin-specific and therefore produces a systemic lytic state. Its primary disadvantage is that it costs far more than streptokinase.

t-PA Tissue plasminogen activator is an enzymatic glycoprotein composed of 527 amino acids that is produced from a human melanoma cell line by means of recombinant DNA technology. Its half-life is approximately 4 minutes; it is metabolized by the liver, and approximately 80% of the dose is excreted in the urine within 18 hours. It is not antigenic and does not promote antibody formation. Theoretically, t-PA should be somewhat more specific for fibrin clot than urokinase or streptokinase. In practice, however, its effects are clinically indistinguishable from those of urokinase. Like urokinase, t-PA is extremely expensive.

The traditional method of administering lytic therapy, venous infusion, is widely used to treat coronary artery thrombosis104; however, it is only modestly effective against peripheral arterial occlusion and is associated with a high rate of hemorrhagic complications.¹⁰⁵ Current methods focus more closely on the site of occlusion, particularly with the development of intra-arterial infusion techniques.

Lytic therapy for acute PE has attractive theoretical benefits; however, the FDA currently approves this approach only for patients with so-called massive PE (i.e., PE resulting in both shock and heart failure). Clinical trials demonstrated that in patients who were hemodynamically unstable as a result of PE, thrombolysis achieved greater improvements in intermediate end points (e.g., right ventricular function) than heparin alone did, though survival was not improved.¹⁰⁶ Subsequent studies evaluated recombinant t-PA in patients with so-called submassive PE, with similar results.⁶⁶

Although monitoring of lytic therapy is less standardized than monitoring of anticoagulant therapy, several principles should be followed. The effects must be monitored from both a clinical and a laboratory perspective. Clinical monitoring involves following improvements on the angiograms. Laboratory monitoring has three components. First, D-dimer levels should be measured; a marked increase signals that cross-linked fibrin is undergoing breakdown. Second, adequate stores of plasminogen should be documented; without plasminogen, none of the thrombolytic drugs are effective. Third, fibrinogen levels should be followed to prevent exhaustion of native clotting; most authorities recommend that lytic therapy be discontinued once fibrinogen levels fall below 50 mg/dl. Previously, the TT was used for monitoring thrombolysis; today, however, given the recommendation for concurrent use of heparin,¹⁰⁴ the TT is considered to be of little value in this setting.

The major toxicity of all three major lytic agents is hemorrhage, resulting from (1) lysis of physiologic thrombi occurring at vascular injury sites and (2) a systemic lytic state caused by the systemic formation of plasmin. The incidence of bleeding is many times higher after lytic therapy than after anticoagulant therapy and is dependent on both the dosage and the duration of lytic therapy. Careful administration of lytic agents can keep the incidence of major hemorrhage below 5% and the incidence of intracranial hemorrhage below 1%.¹⁰²

A potential major complication is distal embolism of partially lysed clot. In theory, this possibility should rule out lytic therapy for treating thrombus in the heart or in the cerebrovascular system.¹⁰² Surprisingly, however, dislodgment of intracardiac clot as a result of lytic therapy is rare.

Streptokinase causes several adverse reactions that urokinase and t-PA do not. When first produced, streptokinase was associated with a very high incidence of antigenicity and severe pyrogenic reactions. The current purified formulation is relatively free of pyrogens and has a reduced incidence of allergic side effects, but it is still antigenic and may still cause allergic reactions or, in rare instances, anaphylaxis. Streptokinase may also induce the formation of additional antibodies that make re-treatment impossible. In contrast, retreatment with urokinase may be carried out as often as necessary with minimal risk of allergic reactions.

Screening for acquired clotting conditions [see Table 6] is based on the history, physical examination, and laboratory assessment. The history should include medications, diseases, and surgical procedures or other injuries.^108�110 Examination may disclose causes of hypercoagulability.¹¹¹ Soft tissue injury, for example, is a potent activator of the coagulation system. If the injury is severe enough, it may be capable of causing a severe acquired coagulopathy. The problem is usually obvious, but on occasion, detailed study may be necessary to identify tissue damage or ischemic injury to bowel or extremities. Hypovolemia�especially hypovolemic shock�markedly reduces clotting time: blood from a patient in profound shock may clot instantaneously in the syringe as it is being drawn. The breakdown of red cells in a hemolytic transfusion reaction can cause clotting. Severe infection, especially from gram-negative organisms, is a potent activator of coagulation.¹¹²

Of the acquired hypercoagulability syndromes, Trousseau syndrome is a particularly important condition for surgeons to recognize because it occurs in the surgical population (cancer patients) and must be treated with heparin (it is unresponsive to warfarin). It occurs when an adenocarcinoma secretes a protein recognized by the body as tissue factor, resulting in multiple episodes of venous thromboembolism over time (migratory thrombophlebitis). Simple depletion of vitamin K-dependent factors is ineffective. Patients should receive therapeutic-dose heparin indefinitely or until the cancer is brought into remission.¹¹³

Laboratory screening may facilitate diagnosis. A complete blood count may document the presence of polycythemia or leukemia. Thrombocythemia may be a manifestation of a hypercoagulable disorder, and thrombocytopenia after the administration of heparin raises the possibility of intravascular platelet aggregation. A prolonged aPTT is suggestive of lupuslike anticoagulant. Increased levels of D-dimers, fibrin degradation products (FDPs), or fibrin monomers in the plasma may reflect low-grade intravascular coagulation.

Congenital clotting tendencies can result from deficiencies in inhibitors of thrombosis (antithrombin, proteins C and S, and possibly heparin cofactor II), dysfibrinogenemias, or dysfibrinolysis [see Table 7]. Most congenital clotting defects are transmitted as an autosomal dominant trait. A negative family history does not preclude inherited thrombophilia, because the defects have a low penetrance, and fresh mutations may have occurred.

The most common congenital causes of accelerated clotting are mutations of prothrombin (prothrombin G20210A mutation) and factor V (Leiden mutation, or activated protein C resistance).^114�116 The prevalence of each of these ranges from 1% to 5% in the general population and may be much higher in specific ethnic subpopulations.¹¹⁷ Each mutation may be identified conclusively by means of polymerase chain reaction (PCR) techniques. Detection of these mutations, unlike assays for antithrombin and proteins C and S, is not dependent on the patient's current inflammatory state. It must be remembered that the presence of one of these mutations, especially in the heterozygous form, does not imply that it is the sole cause of thrombosis. In many patients, a second precipitating factor must be present for the pathologic genetic thrombotic potential to be manifested.

The prothrombin G20210A mutation is known to involve a single amino acid substitution in the prothrombin gene, but precisely how this increases the risk of venous thromboembolism is unclear. The one apparent manifestation of the mutation is a 15% to 40% increase in circulating prothrombin. Regardless of the mechanism at work, patients who are at least heterozygous for the trait are at two- to sixfold greater risk for venous thromboembolism than those without the mutation.¹¹⁸

Resistance of human clotting factors to inactivation by activated protein C is believed to be the most common inherited procoagulant disorder.¹¹⁴ Normally, activated factor V is degraded by activated protein C in the presence of membrane surface as part of normal regulation of thrombosis. Activated protein C resistance is caused by a single substitution mutation in the factor V gene, which is passed in an autosomal dominant fashion. The mutant factor V that results, termed factor V Leiden, is resistant to inactivation by activated protein C and thus has a greater ability to activate thrombin and accelerate clotting.

Two techniques are commonly used to diagnose this disorder. The first is a functional assay that compares a standard aPTT to one performed in the presence of exogenous activated protein C. If the latter aPTT does not exhibit significant prolongation, the patient is probably resistant to activated protein C. The results of this assay must be interpreted with caution if the patient is still in the acute phase of the illness. The second technique, which is more reliable, involves direct detection of the mutation via PCR analysis of DNA.

Antithrombin (once termed antithrombin III) is a 65 kd protein that decelerates the coagulation system by inactivating activated factors�primarily factor Xa and thrombin but also factors XII, XI, and IX.^119,120 Antithrombin therefore acts as a scavenger of activated clotting factors. Its activity is enhanced 100-fold by the presence of heparans on the endothelial surface and 1,000-fold by administration of exogenous heparin.

Congenital antithrombin deficiency occurs in approximately 0.01% to 0.05% of the general population and 2% to 4% of patients with venous thrombosis.¹¹⁹ The trait is passed on as an autosomal dominant trait, with the heterozygous genotype being incompatible with life. Antithrombin-deficient patients are at increased risk for thromboembolism when their antithrombin activity falls below 70% of normal.¹²¹

Patients with congenital antithrombin deficiency frequently present after a stressful event. They usually have DVT but sometimes have PE. If anticoagulation is not contraindicated, the treatment of choice is heparin at a dosage sufficient to raise the aPTT to the desired level, followed by warfarin. If anticoagulation is contraindicated (as it is during the peripartum period), antithrombin concentrate should be given to raise the antithrombin activity to 80% to 120% of normal during the period when anticoagulants cannot be given.

Acquired antithrombin deficiency is a well-recognized entity. In most patients undergoing severe systemic stress, antithrombin levels fall below normal.¹²² Patients with classic risk factors for venous thromboembolism tend to have the lowest levels.

Protein C is a 62 kd glycoprotein with a half-life of 6 hours. Because it is vitamin K dependent, a deficiency will develop in the absence of vitamin K. Acquired protein C deficiency is seen in liver disease, malignancy, infection, the postoperative state, and disseminated intravascular coagulation.⁴⁹ Protein C deficiency occurs in approximately 4% to 5% of patients younger than 40 to 45 years who present with unexplained venous thrombosis.¹²³ It is transmitted as an autosomal dominant trait, and the family history is usually positive for a clotting tendency. Protein C levels range from 70% to 164% of normal in patients without a clotting tendency; levels below 70% of normal are associated with a thrombotic tendency. The most appropriate tests for screening are functional assays; there are cases of dysfunctional protein C deficiency in which protein C antigen levels are normal but protein C activity is low, and these would not be detected by the usual immunoassays.

Protein S is a vitamin K-dependent protein that acts as a cofactor for activated protein C by enhancing protein C-induced inactivation of activated factor V. The incidence of protein S deficiency is similar to that of protein C deficiency.¹²³ It is transmitted as a dominant trait, and the family history is often positive for a thrombotic tendency.

Although hyperhomocysteinemia is more commonly associated with cardiac disease and arterial thrombosis, it may also be associated with an increased incidence of venous thromboembolism.¹²⁴ This association is not as strong as those already discussed (see above). Accordingly, anticoagulation of asymptomatic patients with elevated homocysteine levels is not currently recommended.

More than 100 qualitative abnormalities of fibrinogen (dysfibrinogenemias) have been reported.¹²⁵ Dysfibrinogenemias are inherited in an autosomal dominant manner, with most patients being heterozygous. Most patients with dysfibrinogenemia have either no clinical symptoms or symptoms of a bleeding disorder; a minority (about 11%) have clinical features of a recurrent thromboembolic disorder.^126,127 Congenital dysfibrinogenemias associated with thrombosis account for about 1% of cases of unexplained venous thrombosis occurring in young people. The most commonly observed functional defect in such dysfibrinogenemias is abnormal fibrin monomer polymerization combined with resistance to fibrinolysis. Decreased binding of plasminogen and increased resistance to lysis by plasmin have been noted.

In addition to a prolonged TT, patients who have dysfibrinogenemia associated with thromboembolism may have a prolonged INR. The diagnosis is confirmed if the reptilase time is also prolonged. Measured with clotting techniques, fibrinogen levels may be slightly or moderately low; measured immunologically, levels may be normal or even increased.

Fibrinolysis can be impaired by inherited deficiencies of plasminogen, defective release of t-PA from the vascular endothelium, and high plasma levels of regulatory proteins (e.g., t-PA inhibitors).^127,128 In addition, factor XII (contact factor) deficiency may induce failure of fibrinolysis activation.

Inherited plasminogen deficiency is probably only rarely responsible for unexplained DVT in young patients. It is transmitted as an autosomal dominant trait. In heterozygous persons with a thrombotic tendency, plasminogen activity is about one half normal (3.9 to 8.4 �mol/ml). The euglobulin clot lysis time is prolonged. Functional assays should be carried out, and there should be full transformation of plasminogen into plasmin activators.

The important role of t-PA inhibitors I and II in the regulation of fibrinolysis is well defined.^128,129 In normal plasma, t-PA inhibitor I is the primary inhibitor for both t-PA and urokinase. Release of t-PA inhibitor I by platelets results in locally increased concentrations where platelets accumulate. The ensuing local inhibition of fibrinolysis may help stabilize the hemostatic plug. t-PA inhibitor II is present in and secreted by monocytes and macrophages.

Factor XII deficiency is a rare cause of impaired fibrinolysis. Initial contact activation of factor XII not only results in activation of the clotting cascade and of the inflammatory response but also leads to plasmin generation. This intrinsic activation of fibrinolysis requires factor XII, prekallikrein, and high-molecular-weight kininogen. Patients with factor XII deficiencies can be identified by a prolonged aPTT in the absence of clinical bleeding.^107,¹³⁰

Patients with activated protein C resistance who present with venous thrombosis should be treated with heparin in the standard fashion. They also should receive genetic counseling and refrain from using oral contraceptives.

Treatment of antithrombin deficiency associated with active clotting involves initiating heparin anticoagulation at a dosage sufficient to ensure a significant rise in the aPTT. Warfarin is the drug of choice for long-term prophylaxis and should be given at a dosage sufficient to maintain an INR of 2.0 to 3.0. When anticoagulation is contraindicated, a purified form of antithrombin may be administered directly. Patients with acquired antithrombin deficiency should receive prophylaxis in the form of heparin at a dosage sufficient to raise the aPTT 5 seconds above the upper limit of the normal laboratory value.

Treatment of clotting states related to protein C or protein S deficiency involves administering fresh frozen plasma or factor IX concentrate. Therapeutic-dose heparin followed by warfarin may be appropriate for long-term treatment.

Treatment of thromboembolism associated with dysfibrinogenemia involves therapeutic-dose heparin followed by long-term warfarin. Treatment of thromboembolic disorders associated with dysfibrinolysis is essentially the same as that of dysfibrinogenemia. Some patients with these qualitative plasminogen defects and acute massive thrombotic events may not respond to fibrinolytic treatment with urokinase or streptokinase.

1. Dunmire SM: Pulmonary embolism. Emerg Med Clin North Am 7:339, 1989 [PMID 2653804]

2. Shackford , Moser KM: Deep venous thrombosis and pulmonary embolism in trauma patients. J Intensive Care Med 3:87, 1988

3. Hirsh J, Hull RD: Venous Thromboembolism: Natural History, Diagnosis, and Management. CRC Press, Boca Raton, Florida, 1988

4. Bergqvist D: Postoperative Thromboembolism: Frequency, Etiology, Prophylaxis. Springer-Verlag, New York, 1983

5. Kakkar VV, Stringer MD: Prophylaxis of venous thromboembolism. World J Surg 14:670, 1990 [PMID 2122604]

6. Bell WR, Simon TL: Current status of pulmonary thromboembolic disease: pathophysiology, diagnosis, prevention and treatment. Am Heart J 103:2398, 1982

7. Sabiston DC Jr: Pathophysiology, diagnosis and management of pulmonary embolism. Am J Surg 138:384, 1979 [PMID 382885]

8. Dalen JE, Haffajee CI, Alpert JS 3rd, et al: Pulmonary embolism, pulmonary hemorrhage and pulmonary infarction. N Engl J Med 296:1431, 1977 [PMID 865513]

9. Kakkar VV, Corrigan TP, Fossard DP, et al: Prevention of fatal postoperative pulmonary embolism by low doses of heparin: reappraisal of results of international multicentre trial. Lancet 1:567, 1977 [PMID 65660]

10. Prevention of venous thrombosis and pulmonary embolism. NIH Consensus Development. JAMA 256:744, 1986

11. Blaisdell FW: Preventing postoperative thromboembolism. West J Med 151:188, 1989

12. Reilly DT: Prophylactic methods against thromboembolism. Acta Chir Scand Suppl 550(suppl):115, 1989

13. Knudson , Lewis FR, Clinton A, et al: Prevention of venous thromboembolism in trauma patients. J Trauma 37:480, 1994 [PMID 8083913]

14. McLeod RS, Geerts WH, Sniderman KW, et al: Subcutaneous heparin versus low-molecular-weight heparin as thromboprophylaxis in patients undergoing colorectal surgery: results of the Canadian colorectal DVT prophylaxis trial: a randomized, double-blind trial. Ann Surg 233:438, 2001 [PMID 11224634]

15. Geerts WH, Jay RM, Code KI, et al: A comparison of low-dose heparin with low-molecular-weight heparin as prophylaxis against venous thromboembolism after major trauma. N Engl J Med 335:701, 1996 [PMID 8703169]

16. Christen Y, Wutschert R, Weimer D, et al: Effects of intermittent pneumatic compression on venous haemodynamics and fibrinolytic activity. Blood Coagul Fibrinol 8:185, 1997

17. Comerota AJ, Chouhan V, Harada RN, et al: The fibrinolytic effects of intermittent pneumatic compression: mechanism of enhanced fibrinolysis. Ann Surg 226:306, 1997 [PMID 9339937]

18. Elliott CG, Dudney TM, Egger M, et al: Calf-thigh sequential pneumatic compression compared with plantar venous pneumatic compression to prevent deep-vein thrombosis after non-lower extremity trauma. J Trauma 47:25, 1999 [PMID 10421182]

19. Howard A, Zaccagnini D, Ellis M, et al: Randomized clinical trial of low molecular weight heparin with thigh-length or knee-length antiembolism stockings for patients undergoing surgery. Br J Surg 91:842, 2004 [PMID 15227689]

20. Lachiewicz PF, Kelley SS, Haden LR: Two mechanical devices for prophylaxis of thromboembolism after total knee arthroplasty. A prospective, randomised study. J Bone Joint Surg Br 86:1137, 2004 [PMID 15568526]

21. Rosenberg RD: Action and interactions of antithrombin and heparin. N Engl J Med 292:145, 1975

22. Leyvraz PF, Richard J, Bachmann F, et al: Adjusted versus fixed-dose subcutaneous heparin in the prevention of deep-vein thrombosis after total hip replacement. N Engl J Med 309:954, 1983 [PMID 6621623]

23. Owings JT, Blaisdell FW: Low-dose heparin thromboembolism prophylaxis. Arch Surg 131:1069, 1996 [PMID 8857904]

24. Turpie AG, Levine MN, Hirsh J, et al: A randomized controlled trial of a low-molecular-weight heparin (enoxaparin) to prevent deep-vein thrombosis in patients undergoing elective hip surgery. N Engl J Med 315:925, 1986 [PMID 3531851]

25. Levine MN, Hirsh J, Gent M, et al: Prevention of deep vein thrombosis after elective hip surgery: a randomized trial comparing low molecular weight heparin with standard unfractionated heparin Ann Intern Med 114:545, 1991 [PMID 1848054]

26. Eriksson BI, Bauer KA, Lassen MR, et al: Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. Steering Committee of the Pentasaccharide in Hip-Fracture Surgery Study. N Engl J Med 345:1298, 2001 [PMID 11794148]

27. Bauer KA, Eriksson BI, Lassen MR, et al: Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after elective major knee surgery. Steering Committee of the Pentasaccharide in Major Knee Surgery Study. N Engl J Med 345:1305, 2001 [PMID 11794149]

28. MacCallum PK, Thomson JM, Poller L: Effects of fixed minidose warfarin on coagulation and fibrinolysis following major gynaecological surgery. Thromb Haemost 64:511, 1990 [PMID 2084936]

29. Decousus H, Leizorovicz A, Parent F, et al: A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. N Engl J Med 338:409, 1998 [PMID 9459643]

30. Belcaro G, Nicolaides AN, Errichi BM, et al: Superficial thrombophlebitis of the legs: a randomized, controlled, follow-up study. Angiology 50:523, 1999 [PMID 10431991]

31. Geerts WH, Code KI, Jay RM, et al: A prospective study of venous thromboembolism after major trauma. N Engl J Med 331:1601, 1994 [PMID 7969340]

32. Owings JT, Gosselin RC, Battistella FD, et al: Whole blood D-dimer assay: an effective non-invasive method to rule out pulmonary embolism. J Trauma 48:795, 2000 [PMID 10823521]

33. Owings JT, Gosselin RC, Anderson JT, et al: Practical utility of the whole blood D-dimer assay for excluding thromboembolism in severely injured trauma patients. J Trauma 51:425, 2001 [PMID 11535885]

34. Gosselin RC, Owings JT, Jacoby RC, et al: Evaluation of a new automated quantitative D-dimer, Advanced D-Dimer, in patients suspected of venous thromboembolism. Blood Coag Fibrinol 13:323, 2002

35. Bergqvist D, Bergentz SE: Diagnosis of deep vein thrombosis. World J Surg 14:679, 1990 [PMID 2238670]

36. Hirsh J: Reliability of non-invasive tests for the diagnosis of venous thrombosis (editorial). Thromb Haemost 65:221, 1991 [PMID 2048045]

37. Krupski WC, Bass A, Dilley RB, et al: Propagation of deep venous thrombosis identified by duplex ultrasonography. J Vasc Surg 12:467, 1990 [PMID 2214041]

38. Lensing AW, Prandoni P, Brandjes D: Detection of deep-vein thrombosis by real-time B-mode ultrasonography. N Engl J Med 320:342, 1989 [PMID 2643771]

39. Haire WD, Lynch TG, Lund GB: Limitations of magnetic resonance imaging and ultrasound directed (duplex) scanning in the diagnosis of subclavian vein thrombosis. J Vasc Surg 13:391, 1991 [PMID 1999858]

40. Barnes RW, Nix ML, Barnes CL: Perioperative asymptomatic venous thrombosis: role of duplex scanning versus venography. J Vasc Surg 9:25, 1989

41. Simonneau G, Sors H, Charbonnier B, et al: A comparison of low-molecular-weight heparin with unfractionated heparin for acute pulmonary embolism. The THE-SEE Study Group. Tinzaparine ou Heparine Standard: Evaluations dans l'Embolie Pulmonaire. N Engl J Med 337:663, 1997 [PMID 9278462]

42. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. The Columbus Investigators. N Engl J Med 337:657, 1997

43. Hirsh J, Turpie AG: Use of plasminogen activators in venous thrombosis. World J Surg 14:688, 1990 [PMID 2238671]

44. Sakakibara Y, Shigeta O, Ishikawa S, et al: Upper extremity vein thrombosis: etiologic categories, precipitating causes, and management. Angiology 50:547, 1999 [PMID 10431994]

45. DiFelice GS, Paletta GA Jr, Phillips BB, et al: Effort thrombosis in the elite throwing athlete. Am J Sports Med 30:708, 2002 [PMID 12239007]

46. Schmacht DC, Back MR, Novotney ML, et al: Primary axillary-subclavian venous thrombosis: is aggressive surgical intervention justified? Vasc Surg 35:353, 2001 [PMID 11565039]

47. Wells PS, Forster AJ: Thrombolysis in deep vein thrombosis: is there still an indication? Thromb Haemost 86:499, 2001 [PMID 11487040]

48. Patel NH, Plorde JJ, Meissner M: Catheter-directed thrombolysis in the treatment of phlegmasia cerulea dolens. Ann Vasc Surg 12:471, 1998 [PMID 9732427]

49. Lord RS, Chen FC, DeVine TJ, et al: Surgical treatment of acute deep venous thrombosis. World J Surg 14:694, 1990 [PMID 2238672]

50. LeClerk JR: Venous Thromboembolic Disorders. Lea & Febiger, Philadelphia, 1991 , p 54

51. Smith GT, Dammin GJ, Dexter L: Postmortem arteriographic studies of the human lung in pulmonary embolization. JAMA 188:143, 1964 [PMID 14107239]

52. Goldhaber SZ, Hennekens CH, Evans DA, et al: Factors associated with correct antemortem diagnosis of major pulmonary embolism. Am J Med 73:822, 1982 [PMID 7148876]

53. Karwinski B, Svendsen E: Comparison of clinical and postmortem diagnosis of pulmonary embolism. J Clin Pathol 42:135, 1989 [PMID 2921354]

54. Moser KM, Hull R, Saltzman HA, et al: Recent advances in diagnosis of pulmonary embolism and deep venous thrombosis. Am Rev Respir Dis 138:1046, 1988 [PMID 3202431]

55. Coon WW: Risk factors in pulmonary embolism. Surg Gynecol Obstet 143:385, 1976 [PMID 959958]

56. Boneu B, Bes G, Pelzer H, et al: D-dimers, thrombin antithrombin III complexes and prothrombin fragments 1 + 2 diagnostic value in clinically suspected deep vein thrombosis. Thromb Haemost 65:28, 1991

57. Owings JT, Kraut EJ, Battistella FD, et al: Timing of the occurrence of pulmonary embolism in trauma patients. Arch Surg 132:862, 1997 [PMID 9267270]

58. Killewich LA, Nunnelee JD, Auer AI: Value of lower extremity venous duplex examination in the diagnosis of pulmonary embolism. J Vasc Surg 17:934, 1993 [PMID 8487362]

59. PIOPED Investigators Value of ventilation/perfusion scan in acute pulmonary embolism: results of prospective investigation of pulmonary embolism diagnosis (PIOPED). JAMA 263:2753, 1990

60. Hull RD, Hirsh J, Carter CJ, et al: Diagnostic value of ventilation-perfusion lung scanning in patients with suspected pulmonary embolism. Chest 88:819, 1985 [PMID 4064769]

61. Remy-Jardin M, Remy J, Deschildre F, et al: Diagnosis of pulmonary embolism with spiral CT: comparison with pulmonary angiography and scintigraphy. Radiology 200:699, 1996 [PMID 8756918]

62. Prologo JD, Gilkeson RC, Diaz M, et al: The effect of single-detector CT versus MDCT on clinical outcomes in patients with suspected acute pulmonary embolism and negative results on CT pulmonary angiography. AJR Am J Roentgenol 184:1231, 2005 [PMID 15788601]

63. Hull RD, Raskob GE, Hirsh J: The diagnosis of clinically suspected pulmonary embolism: practical approaches. Chest 89:4175, 1986

64. Ferris EJ, Holder JC, Lim WN, et al: Angiography of pulmonary emboli: digital studies and balloon-occlusion cineangiography. Am J Roentgenol 142:369, 1984

65. Anderson JT, Jeng T, Bain M, et al: Diagnosis of post traumatic pulmonary embolism: is chest computed tomographic angiography acceptable? J Trauma 54:472, 2003 [PMID 12634525]

66. Konstantinides S, Geibel A, Heusel G, et al: Management Strategies and Prognosis of Pulmonary Embolism-3 Trial Investigators. Heparin plus alteplase compared with heparin alone in patients with submassive pulmonary embolism. N Engl J Med 347:1143, 2002 [PMID 12374874]

67. Atik M, Broghamer WL Jr: The impact of prophylactic measures on fatal pulmonary embolism. Arch Surg 114:366, 1979 [PMID 435051]

68. Collins R, Scrimgeour A, Yusuf S, et al: Reduction in fatal pulmonary embolism and venous thrombosis by perioperative administration of subcutaneous heparin: overview of results of randomized trials in general, orthopedic, and urologic surgery. N Engl J Med 318:1162, 1988 [PMID 3283548]

69. Geerts WH: Pulmonary embolism. Conn's Current Therapy 1992. Rakel RE, Ed. WB Saunders Co, Philadelphia, 1992 , p 179

70. Moser KM: State of the art: pulmonary embolism. Am Rev Respir Dis 115:829, 1977 [PMID 855959]

71. Thomas DP: Therapeutic role of heparin in acute pulmonary embolism. Curr Ther Res 18:21, 1975 [PMID 808387]

72. Silver D: Pulmonary embolism: prevention, detection and nonoperative management. Surg Clin North Am 54:1089, 1974 [PMID 4608794]

73. Rohrer MJ, Scheidler MG, Wheeler HB, et al: Extended indications for placement of an inferior vena cava filter. J Vasc Surg 10:44, 1989 [PMID 2501522]

74. Fink JA, Jones BT: The Greenfield filter as the primary means of therapy in venous thromboembolic disease. Surg Gynecol Obstet 172:253, 1991 [PMID 2006447]

75. Wells I: Inferior vena cava filters and when to use them. Clin Radiol 40:11, 1989 [PMID 2646059]

76. PREPIC Study Group: Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study. Circulation 112:416, 2005

77. Schmitz-Rode T, Janssens U, Duda SH, et al: Massive pulmonary embolism: percutaneous emergency treatment by pigtail rotation catheter. J Am Coll Cardiol 36:375, 2000 [PMID 10933345]

78. Rooke TW: Deep venous thrombosis of the extremities. Conn's Current Therapy 1992. Rakel RE, Ed. WB Saunders Co, Philadelphia, 1992 , p 289

79. Majerus PW, Broze GJ Jr, Miletich JP, et al: Anticoagulant, thrombolytic and antiplatelet drugs. Goodman & Gilman's The Pharmacological Basis of Therapeutics. Goodman AG, Rall TW, Nies AS, et al, Eds. Pergamon Press, New York, 1990 , p 1311

80. USP DI, Drug Information for the Health Care Professional vol IB. The United States Pharmacopeial Convention, Inc, Rockville, Maryland, 1992 , pp 1505, 2357, 2658

81. Owen CA Jr, Bowie EJ, et al: Chronic intravascular coagulation syndromes: a summary. Mayo Clin Proc 49:673, 1974 [PMID 4606336]

82. Blaisdell FW, Graziano CJ, Effeney DJ: In vivo assessment of anticoagulation. Surgery 82:827, 1977 [PMID 929373]

83. Blaisdell FW, Steele M, Allen RE: Management of acute lower extremity arterial ischemia due to embolism and thrombosis. Surgery 84:822, 1978 [PMID 715701]

84. Blaisdell FW, Graziano CJ: Assessment of clotting by the determination of fibrinogen catabolism. Am J Surg 135:436, 1978 [PMID 626326]

85. Conti S, Daschbach M, Blaisdell FW: A comparison of high-dose versus conventional-dose heparin therapy for deep vein thrombosis. Surgery 92:972, 1982 [PMID 7147193]

86. Kashtan J, Conti S, Blaisdell FW: Heparin therapy for deep venous thrombosis. Am J Surg 140:836, 1980 [PMID 7457709]

87. Ridker PM, Goldhaber SZ, Danielson E, et al: for the PREVENT Investigators: Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med 348:1425, 2003 [PMID 12601075]

88. Silver D, Kapsch DN, Tsoi EK: Heparin induced thrombocytopenia, thrombosis and hemorrhage. Ann Surg 198:301, 1983 [PMID 6615052]

89. Becker PS, Miller VT: Heparin-induced thrombocytopenia. Stroke 20:1449, 1989 [PMID 2683243]

90. Celoria GM, Steingart RH, Banson B, et al: Coumarin skin necrosis in a patient with heparin-induced thrombocytopenia: a case report. Angiology 39:915, 1988 [PMID 3177957]

91. Walker AM, Jick H: Predictors of bleeding during heparin therapy. JAMA 244:1209, 1980 [PMID 7411781]

92. Mudaliar JH, Liem TK, Nichols WK, et al: Lepirudin is a safe and effective anticoagulant for patients with heparin-associated antiplatelet antibodies. J Vasc Surg 34:17, 2001 [PMID 11436069]

93. Ginsberg JS, Kowalchuk G, Hirsh J, et al: Heparin effect on bone density. Thromb Haemost 64:286, 1990 [PMID 2270535]

94. Hirsh J, Poller L, Deykin D, et al: Optimal therapeutic range for oral anticoagulants. Chest 95(2 suppl):5s, 1989

95. Pinede L, Ninet J, Duhaut P, et al: Investigators of the 'Duree Optimale du Traitement AntiVitamines K' (DOTAVK) Study. Comparison of 3 and 6 months of oral anticoagulant therapy after a first episode of proximal deep vein thrombosis or pulmonary embolism and comparison of 6 and 12 weeks of therapy after isolated calf deep vein thrombosis. Circulation 103:2453, 2001

96. Yasaka M, Minematsu K, Naritomi H, et al: Predisposing factors for enlargement of intracerebral hemorrhage in patients treated with warfarin. Thromb Haemost 89:278, 2003 [PMID 12574807]

97. Francis CW, Berkowitz SD, Comp PC, et al: Comparison of ximelagatran with warfarin for the prevention of venous thromboembolism after total knee replacement. N Engl J Med 349:1703, 2003 [PMID 14585938]

98. Albers GW, Diener HC, Frison L, et al: Ximelagatran vs warfarin for stroke prevention in patients with nonvalvular atrial fibrillation: a randomized trial. JAMA 293:690, 2005 [PMID 15701910]

99. Blaisdell FW: Hemostasis and thrombosis. Vascular Surgery: A Comprehensive Review 2nd ed. Moore WS, Ed. Grune & Stratton, New York, 1986 , p 909

100. Meyerovitz MF, Goldhaber SZ, Reagan K, et al: Recombinant tissue-type plasminogen activator versus urokinase in peripheral arterial and graft occlusions: a randomized trial. Radiology 175:75, 1990 [PMID 2107563]

101. Turpie AG: Thrombolytic agents in venous thrombosis. J Vasc Surg 12:196, 1990

102. Marder VJ, Sherry S: Thrombolytic therapy: current status. N Engl J Med 318:1585, 1988 [PMID 3287158]

103. Goldhaber SZ, Buring JE, Lipnick RJ, et al: Pooled analyses of randomized trials of streptokinase and heparin in phlebographically documented acute deep venous thrombosis. Am J Med 76:393, 1984 [PMID 6367454]

104. The effects of tissue plasminogen activator, streptokinase, or both on coronary-artery patency, ventricular function, and survival after acute myocardial infarction. The GUSTO Angiographic Investigators. N Engl J Med 329:1615, 1993

105. Amery A, Deloof W, Vermylen J, et al: Outcome of recent thromboembolic occlusions of limb arteries treated with streptokinase. Br Med J 4:639, 1970 [PMID 5488380]

106. Meyer GJ, Sors H, Charbonnier B, et al: Effects of intravenous urokinase versus alteplase on total pulmonary resistance in acute massive pulmonary embolism: a European multicenter double-blind trial. The European Cooperative Study Group for Pulmonary Embolism. J Am Coll Cardiol 19:239, 1992 [PMID 1732347]

107. Schafer AI: The hypercoagulable states. Ann Intern Med 102:814, 1985 [PMID 3158262]

108. Baehner RL: Alterations in blood coagulation with trauma. Pediatr Clin North Am 22:289, 1975 [PMID 1124216]

109. Jansson IG, Hetland O, Rammer LM, et al: Effects of phospholipase C, a tissue thromboplastin inhibitor, on pulmonary microembolism after missile injury of the limb. J Trauma 28:S222, 1988 [PMID 3339692]

110. Effeney DJ, McIntyre KS, Blaisdell FW, et al: Fibrinogen kinetics in major human burns. Surg Forum 29:56, 1978 [PMID 401261]

111. Blaisdell FW: Acquired and congenital clotting syndromes. World J Surg 14:664, 1990 [PMID 2238669]

112. Hauptman JG, Hassouna HI, Bell TG, et al: Efficacy of antithrombin III in endotoxin induced disseminated intravascular coagulation. Circ Shock 25:111, 1988 [PMID 3390893]

113. Callander N, Rapaport SI: Trousseau's syndrome. West J Med 158:364, 1993 [PMID 8317122]

114. Dahlb�ck B, Carlsson M, Svensson PJ: Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc Natl Acad Sci USA 90:1004, 1993

115. Bertina RM, Reitsma PH, Rosendaal FR, et al: Resistance to activated protein C and factor V Leiden as risk factors for venous thrombosis. Thromb Haemost 74:449, 1995 [PMID 8578504]

116. De Stefano V, Martinelli I, Mannucci PM, et al: The risk of recurrent deep venous thrombosis among heterozygous carriers of both factor V Leiden and the G20210A prothrombin mutation. N Engl J Med 341:801, 1999 [PMID 10477778]

117. Hessner MJ, Luhm RA, Pearson SL, et al: Prevalence of prothrombin G20210A, factor V G1691A (Leiden), and methylenetetrahydrofolate reductase (MTHFR) C677T in seven different populations determined by multiplex allele-specific PCR. Thromb Haemost 81:733, 1999 [PMID 10365746]

118. Marder VJ, Matei DE: Hereditary and acquired thrombophilic syndromes. Hemostasis and Thrombosis. Colman, Hirsch, Marder, et al, Eds. Lippincott Williams & Wilkins, Philadelphia, 2001

119. Egeberg O: Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrhag 13:516, 1965

120. High KA: Antithrombin-III, protein-C, and protein-S: naturally occurring anticoagulant proteins. Arch Pathol Lab Med 112:28, 1988 [PMID 2962557]

121. Bauer KA, Goodman TL, Kass BL, et al: Elevated factor Xa activity in the blood of asymptomatic patients with congenital antithrombin deficiency. J Clin Invest 76:826, 1985 [PMID 3875633]

122. Owings JT, Bagley M, Gosselin R, et al: Effect of critical injury on plasma antithrombin activity: low antithrombin levels are associated with thromboembolic complications. J Trauma 41:396, 1996 [PMID 8810955]

123. Gladson CL, Scharrer I, Hach V, et al: The frequency of type I heterozygous protein-S and protein-C deficiency in 141 unrelated young patients with venous thrombosis. Thromb Haemost 59:18, 1988 [PMID 2966450]

124. den Heijer M, Koster T, Blom HJ, et al: Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med 334:759, 1996 [PMID 8592549]

125. Rocha E, Paramo JA, Aranda A, et al: Congenital dysfibrinogenemias: a review. Ric Clin Lab 15:205, 1985 [PMID 3909359]

126. Liu Y, Lyons RM, McDonagh J: Plasminogen San Antonio: an abnormal plasminogen with a more cathodic migration, decreased activation and associated thrombosis. Thromb Haemost 59:49, 1988 [PMID 3363534]

127. Nilsson IM, Ljungner H, Tengborn L: Two different mechanisms in patients with venous thrombosis and defective fibrinolysis: low concentration of plasminogen activator or increased concentration of plasminogen activator inhibitor. Br Med J 290:1453, 1985

128. Kruithof EK, Gudinchet A, Bachmann F: Plasminogen activator inhibitor 1 and plasminogen activator inhibitor 2 in various disease states. Thromb Haemost 59:7, 1988 [PMID 3129809]

129. Juhan-Vague I, Roul C, Alessi MC, et al: Increased plasminogen activator inhibitor activity in non insulin dependent diabetic patients�relationship with plasma insulin. Thromb Haemost 61:370, 1989

130. Rodgers GM, Shuman MA: Congenital thrombotic disorders. Am J Hematol 21:419, 1986 [PMID 2937290]

131. Blaisdell FW: What's new in clotting and anticoagulation. Progress in Vascular Surgery. Najarian JS, Delaney JP, Eds. Year Book Medical Publishers, Chicago, 1988 , p 75

Dec	JAN	Feb
	27
2011	2012	2013