RISK FACTORS	SPECIFIC NOTES
Previous history of PE or DVT	Inquire about the setting and circumstances of the previous VTE
Recent trauma or surgery	In general, trauma requiring admission or surgery requiring general anesthesia within the previous month. Recent long-bone, vascular, or trauma surgery may especially increase the risk
Cancer	In general, patients with currently treated cancer or palliative care
Central or long-term vascular catheters
Age	Risk significantly increases above the age of 50 to 60 years
Oral contraceptives	Especially third-generation formulations
Hormone replacement therapy	Currently less common than in the past
Pregnancy	Risk increases along with the duration of pregnancy; it peaks at term and then decreases over a period of 4 to 6 weeks postpartum
Immobility	Includes casts or splints, as well as permanent limb or generalized body immobility, including that from general hospitalization
Factor V Leiden mutation	Most common in northern European populations. The heterozygous carrier state exists in 3% to 7% of many samples. A homozygous mutation is less common and confers three times greater risk for VTE relative to the normal genotype.
Antiphospholipid antibody syndrome	Very potent risk factor. Associated with large and recurrent PE. May be associated with anticardiolipin antibodies, stroke, myocardial infarction, and frequent first-trimester miscarriages
Prothrombin mutation
Hyperhomocysteinemia	Can occur as a result of inadequate folate and B vitamin intake, as well as a genetic mutation in methyltetrahydrofolate reductase. The degree of elevation in risk is controversial
Deficient levels of clotting factors	Protein C, protein S, antithrombin III
Congestive heart failure	May result from generalized immobility or vascular stasis
Chronic obstructive pulmonary disease	May result from generalized immobility
Air travel	Primary risk with travel in excess of 5000 km (3100 miles) and concurrent other risk factors. The degree of elevation in risk is controversial
Obesity	The degree of elevation in risk is controversial

The timing of development of DVT is difficult to know, but it typically starts with small asymptomatic abnormalities in flow at the level of the endothelium. It may be precipitated by venous valve dysfunction or immobility, both of which decrease the normal function of the venous-muscle pump system for return of blood from the legs.

DVT can also develop in the pelvic veins (typically associated with comorbid pelvic or gynecologic conditions) or upper extremity (typically with venous catheters), but by far most cases of DVT diagnosed in the ED are in the legs.

Understanding the venous anatomy of the leg is crucial for interpreting risk and diagnostic tests. The most obvious and concerning cases of DVT occur in the proximal deep system. In order from distal to proximal, the proximal deep veins are the popliteal, femoral, deep femoral, common femoral, and finally the external iliac. Occasionally, the femoral vein is referred to as the superficial femoral vein, which results in confusion because it is a deep vein and clots there should be treated as DVT. Thrombosis may develop in the distal deep system (the calf), and these veins are the anterior tibial, the posterior tibial, and the peroneal. These are deep veins and thrombosis here is DVT, but management of isolated distal DVT is somewhat less certain than that for proximal DVT. Finally, the superficial venous system of the leg can be subject to thrombosis and superficial thrombophlebitis, but this is managed differently from thrombosis in the deep system. These veins are the lesser and greater saphenous and the perforating veins. This is not DVT, and isolated thrombosis here is not typically treated with systemic anticoagulation.

The morbidity associated with venous thrombosis is not just due to subsequent embolization. Even with anticoagulation there is the possibility of injury to the endothelium and venous valve system, both of which can result in chronic venous stasis and recurrent clots.

Presenting Signs and Symptoms

Patients can have a variety of symptoms that may include fullness, cramping, achiness, or vague pain in the calf or posterior part of the leg. They may also report swelling. It is important to distinguish between unilateral symptoms, which may be more likely to be due to DVT, and bilateral symptoms, which may be more likely to be a consequence of some other disease process. The exception would be the rare case of obstruction of the inferior vena cava or simultaneous bilateral DVT. Signs on examination may include edema, redness, and tenderness, particularly in the posterior aspect of the calf or the upper part of the leg.

Swelling may be minimal or extensive and involve the entire extremity. It can be challenging to distinguish chronic symptoms of venous stasis from acute DVT based on a single examination. Therefore, a careful history to establish the time course of the findings in the leg is critical, and examination of the past medical record can be highly valuable.

Use of examination alone to determine the presence or absence of DVT is problematic. Classic teaching is that the accuracy of physical examination alone for the diagnosis of DVT is 50%. This is a bit of a generalization, but it is clear that findings may be subtle, patients may not have detectable tenderness or edema early in the course or spectrum of disease severity, and some patients may be nearly asymptomatic. The often mentioned, but rarely understood or performed Homan sign—calf pain elicited by passive dorsiflexion of the ankle—is insensitive, nonspecific, and useless.

Differential Diagnosis and Medical Decision Making

Many other diseases may be accompanied by pain, swelling, and tenderness of the leg. Box 71.1 presents a differential diagnosis of conditions that should be considered.

Clinicians often have a difficult time distinguishing between cellulitis and DVT. The pain, swelling, erythema, and unilateral nature of cellulitis may be clinically similar to findings in patients with DVT, and the two conditions can even coexist. A break in the skin, fever, or warmth may more strongly suggest cellulitis. However, distinguishing between the two may be impossible without objective testing for DVT.

Compartment syndrome can be manifested as pain and swelling in the leg and is an emergency, time-dependent diagnosis. Extreme pain, exquisite tenderness in the compartment, coolness of the extremity, and in later stages, diminished pulses and decreased sensation may all indicate this diagnosis. It is most common in the context of trauma to the leg from either penetrating, blunt, or fracture mechanisms. If being considered, it must be assessed by measuring compartment pressures and, if appropriate, emergency orthopedic or traumatic surgery consultation. Trauma of this significance can be associated with concomitant DVT and compartment syndrome, so if compartment pressures are normal, testing for DVT can and should proceed.

Many of the other conditions in the list of differential diagnoses are easier to differentiate from DVT by close attention to the time course and other symptoms as revealed from a careful patient history. Overall edematous conditions are listed (congestive heart failure, renal failure, liver failure), but they are commonly associated with gradual or recurrent episodes of bilateral edema. The possible important exception is a patient with congestive heart failure who has undergone coronary artery bypass grafting with removal of the saphenous vein. These patients may, if asked, report that they always have some asymmetry in their swelling and leg size.

Baker cyst rupture or inflammation is a common cause of pain behind the knee and even in the calf. It may be very painful and can result in redness and swelling. An acute onset associated with bending over or being on flexed knees suggests this diagnosis, but confirmation is often achieved with an ultrasound study that is otherwise negative for DVT.

Diagnostic Testing

Pretest Probability

Just like any possible diagnosis in the ED for which further testing is considered, the first step should be an assessment of what the pretest probability of disease is.^2,³ This can be done easily by using a clinical decision model such as that derived and validated by Wells et al.⁴ to stratify patients as low risk or high risk for having DVT (Table 71.2). It has been also advocated by some authors that experienced clinicians may empirically use clinical gestalt to estimate the probability of DVT as low or not low. In either case, documentation on the medical record should clearly indicate the physician’s pretest probability estimate for DVT and what the supporting evidence for this may be from the history and examination.

Table 71.2 Wells Model for Estimating the Pretest Probability for Deep Vein Thrombosis

CLINICAL VARIABLE	SCORE
Active cancer (treatment ongoing, within the previous 6 months, or palliative)	1
Paralysis, paresis, or recent plaster immobilization of the lower extremities	1
Bedridden for 3 days or more or surgery requiring general anesthesia in the last 3 mo	1
Localized tenderness along the deep venous system	1
Swelling of the entire leg	1
Calf swelling measured to be 3 cm greater than other leg (at 10 cm below the tibial tuberosity)	1
Pitting edema in the symptomatic leg	1
Collateral superficial veins (nonvaricose)	1
Previous deep vein thrombosis	1
Alternative diagnosis at least as likely as deep vein thrombosis	−2
A score of 1 or less indicates that deep vein thrombosis is not likely.

Adapted from Wells PS, Anderson DR, Rodger M, et al. Evaluation of D-dimer in the diagnosis of suspected deep-vein thrombosis. N Engl J Med 2003;349:1227-35.

Ultrasonography

Venous duplex ultrasonography is noninvasive, not associated with radiation exposure, and accurate in detecting proximal DVT. It has become the first-line imaging test for DVT. When performed by a certified sonographer and interpreted by a radiologist or other credentialed expert, it has a sensitivity and specificity of approximately 95% for proximal thrombosis. Ultrasonography does present some problems, however, including limited availability at night in many institutions, operator variability, occasional difficulty in visualization because of body habitus, and at times interobserver disagreement regarding distal DVT. Nonetheless, it is the imaging test of choice for most ED clinicians, and absence of DVT on ultrasonography should be reassuring that no DVT requiring emergency treatment is present. The question of whether to include visualization of the infrapopliteal veins and the clinical significance of isolated distal DVT are controversial.^5,⁶

Ultrasound Performed by Emergency Department Clinicians

A limited study compared ultrasound of the common femoral and popliteal veins performed by emergency physicians versus formal duplex ultrasound performed by certified ultrasonographers in radiology or vascular departments.⁷^–¹⁰ A sensitivity of 89% to 100% and a specificity of 75% to 91.9% were reported. However, the extent of operator training and the specific protocols used have varied widely among published studies.¹¹ One study prospectively investigated the performance of ED-based ultrasound by a heterogeneous mix of 56 emergency physicians with varied levels of ultrasound experience and found a summary sensitivity of 70%, which is low for the evaluation of a disease with significant potential morbidity.¹¹ Not surprisingly, sonographers with the most experience had the most accuracy in this study. It remains to be seen what the best method is to train emergency medicine ultrasonographers for evaluation of DVT and what the overall diagnostic performance would be when done outside academic centers or institutions without ultrasound fellowships and dedicated training. In the future this may be a cost-effective and safe option as this modality evolves. Presently, a clearly negative result of a limited scan performed by a properly trained emergency physician suggests the absence of DVT in the proximal leg veins. If the results are positive or equivocal or involve patients with a high probability, confirmatory studies by the radiology or vascular service should be performed.

Additional or Follow-up Ultrasonography

There can be significant confusion and disagreement between the emergency medicine physician consultant and the literature over the need for additional ultrasonographic studies in patients with negative ultrasound findings. Some suggest that if the initial ultrasound visualized only the proximal system, a second ultrasound as an outpatient is needed. However, studies have found this practice to yield a very low rate of disease. A more common and feasible strategy would be to refer patients with equivocal findings in the distal system or those with a high pretest probability of DVT for additional studies done as an outpatient. Integration of the ultrasound results with pretest probability and D-dimer testing as indicated later can limit confusion on this issue.

Venography

Venography has been the historical “gold standard” for the diagnosis of DVT, but as with pulmonary angiography, this invasive and painful test is now rarely performed. Complications include contrast reaction, contrast nephropathy, vascular injury, and the possibility of subsequent DVT in previously nonthrombosed vessels.

Runoff Venography After Chest Computed Tomography for Pulmonary Embolism

When computed tomography (CT) of the chest is performed as part of imaging for possible PE, some institutions also use a delayed imaging protocol that visualizes the pelvis and legs to assess for the presence of DVT. This may add a marginal value of about 2% to the overall sensitivity of this testing strategy. However, many sites do not use this protocol because of radiation exposure, cost, time needed to read the images, and less than optimal interobserver agreement on the venography aspect of the study. In general, this is a hospital- and radiology-specific decision with no conclusive evidence in the literature on its value. CT venography done independent of a PE protocol study is not typically performed because of the alternative nonradiating and noninvasive nature of duplex ultrasonography.

D-Dimer Test

The D-dimer test is a blood-based assay that uses antibodies to the D-dimer protein to measure the presence or level of circulating D-dimer. D-dimer is a breakdown product of fibrin, and elevated levels are associated with elevated fibrin or clot somewhere in the body. Unfortunately, elevated D-dimer can occur in a variety of conditions other than acute PE and DVT, most notably malignancy, pregnancy (particularly later trimesters), chronic inflammatory conditions, recent surgery, infection, stroke, myocardial infarction, and even advanced age. For many versions of this test, levels below 500 ng/mL are considered negative, but this may vary by laboratory and assay used.

Combining of Pretest Probability, D-Dimer Testing, and Duplex Ultrasonography

This integrated strategy has been advocated as a way to optimize the speed of evaluation, cost, need for imaging, and safety. Similar to the risk stratification and D-dimer approach to PE testing, it starts with physician estimation of the probability of disease in a symptomatic ED patient. If DVT is determined to be “unlikely” by the Wells DVT score (1 or less), a D-dimer test that is negative is reassuring to stop ED testing for DVT and consider other diagnoses.¹² If D-dimer is positive, ultrasound is the next step, with results guiding further decision making. One approach for those who are not “DVT unlikely” still uses D-dimer but mandates that all patients, regardless of the D-dimer result, undergo ultrasonography, and D-dimer is then used to guide who needs follow-up in 1 week for an additional ultrasound. This integrated strategy is shown in Figure 71.1.

Fig. 71.1 Pretest probability–based diagnostic algorithm for the evaluation of potential deep vein thrombosis (DVT) in emergency department patients.

LMWH, Low-molecular-weight heparin.

Treatment

Subcutaneous injection of low-molecular-weight heparin (LMWH, including enoxaparin, dalteparin, and others) is the standard of care used to treat acute DVT and provide rapid therapeutic anticoagulation until the effects of oral warfarin are sufficient to result in therapeutic anticoagulation (usually 4 to 7 days). The goal of long-term warfarin treatment is to prevent recurrence of DVT or PE, which is achieved by targeting the international normalized ratio (INR) to a level between 2.0 and 3.0 during the treatment period. Therapeutic anticoagulation may also lower the risk for postphlebitic syndrome,¹³ in addition to the risk for recurrent VTE.

Unfractionated heparin administered via continuous infusion is used less often now than in the past, but it can be an option if there is a potential need to rapidly stop the effects of anticoagulation at a later time in care. Other than this unique situation, LMWH has become the first choice for anticoagulation in patients with acute VTE, largely because of the ease of administration, its efficiency in achieving anticoagulation, and the potential reduced bleeding. In the initial treatment of DVT, once-daily dosing appears to be as effective and safe as twice-daily dosing, but the frequency may vary with the agent used.

The duration of warfarin treatment may vary by patients. Typically, it is taken for 3 to 6 months. Patients with provoked DVT (from a specific temporary event such as an injury, surgery, or pregnancy) are more likely to have warfarin discontinued at 3 to 6 months than are patients with active cancer or unprovoked idiopathic VTE. Some oncologists and hematologists recommend daily LMWH for oncology patients with VTE rather than transition to warfarin, and some patients with nontemporary risk factors for VTE, such as incurable malignancy, may be maintained on an anticoagulation regimen for life. Both LMWH and warfarin reduce the development of additional thrombosis but do not have a mechanism of action that dissolves or lyses the existing clot. The natural fibrinolytic system accomplishes this task to a varied degree in each individual patient, depending on many factors. Pharmacologically, lysis can be facilitated by use of thrombolytic agents, but at present such treatment is indicated only for extensive limb-threatening thrombosis not amenable to vascular surgery and other rare instances.

Other anticoagulants, such as direct thrombin inhibitors and factor Xa inhibitors, have been developed and are undergoing additional testing in large trials. These anticoagulants have the advantage of rapid time to therapeutic effect and ease of oral dosing and appear to be equally effective as LMHW plus warfarin, but they are not easily reversed and may incur a significant increase in cost relative to warfarin alone.

Disposition

Most patients with acute DVT who have the ability to self-administer subcutaneous injections of LMHW, continue warfarin anticoagulation as an outpatient, have a means of determining the INR, and have additional follow-up arranged can be discharged from the ED. Of course, if other conditions are thought to be present that mandate inpatient testing or treatment, patients are admitted for inpatient care. Patients who are thought to have concomitant PE based on clinically significant signs or symptoms may require testing for this condition and admission for monitoring of oxygen saturation and the degree of pulmonary symptoms, but pharmacologic treatment remains the same for PE and DVT. The counterintuitive practice pattern whereby patients are typically discharged from the ED with uncomplicated DVT but often admitted for uncomplicated PE is subject to ongoing debate, with some experts advocating for outpatient care of PE in patients who do not have an oxygen requirement and are otherwise at low risk for adverse events. In the absence of randomized trials or large-scale observational data, this debate lingers as an evolving area of potential health services research. In any case, clinicians are not mandated to order pulmonary angiography or chest CT in patients without pulmonary or chest symptoms and should not embark on a quest to find PE in every patient who has DVT. Such a strategy would be costly, expose the patient to radiation and other potential risks such as contrast reaction and nephropathy, and would not change management. Explaining this to patients requires deliberate, clear communication but may be helpful in reducing subsequent anxiety as they learn more about VTE in the outpatient setting. Finally, follow-up must be emphasized as being important. The clinical course of acute DVT is impossible to predict, even with appropriate anticoagulation treatment.

Special Cases Related to Treatment or Disposition Decisions

Superficial Thrombophlebitis

Painful inflammation of superficial veins because of thrombus is called superficial thrombophlebitis. Because it possible for DVT to be present with superficial thrombophlebitis, it is not unreasonable to consider ultrasound evaluation in these patients. If deep venous involvement is not present, a thrombus in the superficial vein is not generally thought to pose a threat for significant embolization, but appropriate follow-up and return instructions are needed because in a minority of these patients, DVT or worsening of their superficial disease may eventually occur.

Outpatient treatment of superficial thrombophlebitis usually consists of nonsteroidal antiinflammatory medications, heat, and compression. The pain may persist for weeks, and the condition may recur.

Isolated Calf Deep Vein Thrombosis

Treatment of isolated calf DVT with anticoagulants is controversial. Advocates of treatment cite the not-insignificant risk for proximal propagation and potential subsequent PE.⁶ Opponents argue that isolated calf DVT is less likely to be associated with clinically significant PE.⁵ A period of several days to a week of outpatient treatment with 325 mg of aspirin daily and ultrasound repeated within 1 week may be an option in these patients if discussed with the patient and primary physician.