Talal Dahhan, MD, Christian Castillo, MD and Victor F. Tapson, MD

1. Who first described pulmonary embolism?

    Pulmonary embolism (PE) was probably first reported in the early 1800s clinically, but Rudolf Virchow elucidated the mechanism by describing the connection between venous thrombosis and PE in the late 1800s. He also coined the term embolism.

2. What is Virchow’s triad?

    Virchow’s triad comprises the three broad categories of risk factors that contribute to thrombosis, including:

3. What percentage of patients with acute PE have clinical evidence of lower extremity deep vein thrombosis (DVT)?

    In the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) II study of 192 patients with angiographically documented PE, 47% had evidence of DVT by physical examination. The majority of patients with proven PE do have residual DVT present proven by imaging. More than 90% of emboli originate in the deep veins of the legs and thighs.

4. What percentage of patients with proximal DVT will develop PE?

    Approximately 50% of patients with proximal DVT will develop PE.

5. What is the usual cause of death in subjects with PE?

    It is well established that right ventricular (RV) failure is the cause of cardiovascular collapse and death in acute massive PE. Registry data have also suggested that patients with RV dysfunction on echocardiogram are at increased risk of all-cause mortality at 3 months.

6. What are the major risk factors for venous thromboembolism (VTE)?

    Risk factors are crucial in raising suspicion for acute VTE, although the disease may be idiopathic. Previous thromboembolism, immobility, cancer, advanced age, major surgery, trauma, acute medical illness, and certain thrombophilias impart significant risk. A list of risk factors for VTE is shown in Box 57-1.

7. How commonly are recommended VTE prevention measures used?

    Recent data suggest that worldwide, prophylaxis is underused. A recent study including more than 67,000 patients, the Epidemiologic International Day for the Evaluation of Patients at Risk for Venous Thromboembolism in the Acute Hospital Care Setting (ENDORSE) registry, indicated that in many countries less than half of patients deemed appropriate candidates for prophylaxis measures actually received them.

8. How should a patient with suspected DVT be initially evaluated?

    The symptoms and signs of DVT are nonspecific, and although clinical prediction rules can be useful, there should be a low threshold to proceed to compression ultrasonography. D-dimer testing may be useful; the diagnosis of DVT can be excluded without the need for ultrasound if the patient has a combination of a low or moderate clinical probability estimate and a negative D-dimer result. If the patient falls into the moderate or high clinical pretest probability or has a positive D-dimer test, further evaluation with ultrasound is advised. The enzyme-linked immunosorbent assay (ELISA)–based D-dimer tests have superior sensitivity (96% to 98%). DVT is discussed further in Chapter 50.

9. What are the symptoms and signs of acute PE?

    The symptoms and signs of PE are generally nonspecific but most commonly include dyspnea and chest pain, and may include the following:

If PE is associated with pulmonary hypertension, then elevated neck veins, a loud P2, RV heave, or a right-sided gallop may be noted.

10. What are four clinical syndromes seen with acute PE?

11. What is the Wells score for suspected acute PE?

    First described in 1998, the Wells score is a clinical prediction score based on simple, noninvasive clinical parameters. It has evolved over the years, has been validated, and is useful in determining pretest probability for suspected acute PE. Pretest probability can be defined based on the calculated score from the modified Wells score (Table 57-1). PE has ultimately been classified as “unlikely” if the clinical decision score was 4 or less, and “likely” with a score of more than 4 points. This cutoff was chosen because it has been shown to give an acceptable VTE diagnostic failure rate of 1.7% to 2.2% when in combination with a normal D-dimer test result. Moreover, with a score of 4 or less and a negative D-dimer test, no further testing appears to be necessary. A score greater than 4 requires further evaluation, and computed tomographic angiography (CTA) is usually performed. High clinical suspicion, however, should always take precedence, even if the Wells score is low.

12. What are the most common findings on electrocardiography (ECG) in acute PE?

    Sinus tachycardia is commonly present in acute PE. The “classic” S₁Q₃T₃ pattern (Fig. 57-1) is seen in a minority of patients. Other ECG findings include the following:

ST-T changes, when present, are often most marked in the right precordial leads. Although the ECG may be suggestive of PE, it cannot diagnose or rule out acute PE.

13. What are the common chest radiographic findings in patients with acute PE?

    The chest radiograph is commonly abnormal in acute PE, although a significant minority of patients have a normal film. When it is abnormal, however, the findings are nonspecific and include an elevated hemidiaphragm, focal or multifocal infiltrates, pleural effusion, platelike atelectasis, enlarged pulmonary arteries, focal oligemia (the Westermark sign), and RV enlargement.

    Hampton and Castleman described in detail the radiographic findings in PE and pulmonary infarction in 1940, having assembled 370 cases of PE and infarction. In pulmonary infarction, they suggested that the cardiac margin of the opacity of a pulmonary infarction on a chest radiograph is rounded or hump-shaped (i.e., the Hampton hump).

14. What are typical arterial blood gas (ABG) findings in patients with PE?

    Low Pao₂, low Paco₂, high alveolar-arterial oxygen difference. Although nonspecific, one of these findings is likely to be present in up to 97% of cases. A normal ABG does not absolutely exclude PE.

15. When should a ventilation-perfusion () scan be performed?

    The scan is most useful when the chest radiograph is normal and there is no cardiopulmonary disease. In this setting, it has the highest likelihood of being either normal or high probability. Diagnostic tests (especially those for PE) must be interpreted in light of the patient’s pretest probability score. It is crucial to remember that commonly the scan is low or intermediate probability, even when PE is present.

     scanning may also be useful in patients with abnormal renal function, in which a CTA using iodinated contrast agent may be contraindicated.

16. Does a negative CTA indicate that PE is not present with certainty?

    No, but a good quality CTA is extremely sensitive. We learned from the PIOPED II study, published in 2006, that clinical probability is extremely important when considering CTA results. Approximately 60% (9/15) of patients who had high clinical pretest probability but a negative CTA were ultimately diagnosed with PE. Similarly, 42% (16/38) of patients with a low clinical pretest probability and a positive CTA did not have PE. A recent study of more than 3000 patients with suspected acute PE by the Christopher Investigators suggested that if CTA is negative, outcome at 3 months is excellent without therapy. Nonetheless, it is prudent to consider additional imaging when a negative CTA is accompanied by high clinical suspicion; furthermore, imaging quality is not uniformly high in all clinical settings. Finally, computed tomographic (CT) scanning has evolved such that modern day multislice scanners are more sensitive than their predecessors. A large PE, documented by CTA, is shown in Figure 57-2. A diagnostic algorithm, which can be used as a guide, is offered in Figure 57-3.

17. How does echocardiography aid in the diagnosis of PE?

    Transthoracic echocardiography can detect RV strain and dilation. Risk stratification should be strongly considered in patients diagnosed with PE. The McConnell sign, a normally contracting right ventricular apex associated with severe hypokinesis of the mid–free wall, is an echocardiographic pattern that suggests, but is not diagnostic of, acute PE.

18. What is the most appropriate initial therapy for patients with documented acute PE?

    In patients with acute PE, a therapeutic level of anticoagulation should ideally be achieved within 24 hours, because this reduces the risk of recurrence. The 9th American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (ACCP-9) from February 2012, recommend initiation of treatment while awaiting diagnostic tests if clinical suspicion is deemed high. Initial treatment with low-molecular-weight heparin (LMWH), unfractionated heparin (UFH), or fondaparinux for at least 5 days, together with initiation of oral anticoagulant therapy until the international normalized ratio (INR) is 2.0 or more for at least 24 hours, is recommended. Warfarin should be started on the first treatment day. ACCP-9 also recommends that in patients with acute nonmassive PE, initial treatment with LMWH rather than intravenous UFH be used, if feasible, based on advantages of LMWH, including subcutaneous rather than intravenous delivery, much less need for monitoring, and a lower rate of heparin-induced thrombocytopenia. The LMWH preparations are also more bioavailable and thus, more predictable than standard UFH in terms of degree of anticoagulation. Anticoagulation clearly improves survival in patients with acute symptomatic PE.

    Rivaroxaban, a new oral anticoagulant approved for the treatment of DVT/PE, is discussed in Question 27 below.

19. How should one stratify patient risk with acute PE?

    Outcomes of PE depend on a number of factors. The Pulmonary Embolism Severity Index (PESI) allows stratification on a clinical basis. Several therapeutic implications exist for patients with PE. High-risk patients (who represent about 5% of all symptomatic patients, with about a 15% short-term mortality) should be considered for aggressive treatment with thrombolytic therapy or surgical or catheter embolectomy. Low-risk patients (most patients with PE, with a short-term mortality of about 1%) benefit from anticoagulation therapy, and can sometimes be monitored and followed as outpatients. Intermediate-risk patients (who represent about 30% of all symptomatic patients) should be admitted to the hospital, anticoagulated, and be considered for thrombolytic therapy if indicated. Low-risk and intermediate-risk categories can be referred to as nonmassive PE. The PESI has been simplified (Table 57-2) to ease clinical application. The 305 of 995 patients (30.7%) who were classified as low risk by the simplified PESI had a 30-day mortality of 1.0% (95% confidence interval [CI] of 0.0% to 2.1%) compared with 10.9% (95% CI of 8.5% to 13.2%) in the high-risk group.

20. Are there any laboratory investigations that may help in the prognosis of patients with PE?

    B-type natriuretic peptide (BNP) and N-terminal of B-type natriuretic peptide (NT-proBNP) levels and cardiac biomarkers such as troponin T or I may help to identify severity of PE. Elevated troponin has been correlated with increased mortality in acute PE. Neither of these tests, however, is sensitive or specific.

21. What is the primary indication for thrombolytic therapy?

    Proven PE with cardiogenic shock is the clearest indication. ACCP-9 also notes that in selected high-risk patients without hypotension who are judged to have a low risk of bleeding, thrombolytic therapy can be considered. This received a grade 2C recommendation—that is, a decision made based on limited data and expert opinion to support the suggestion. An example would be in submassive PE (RV dilation and hypokinesis without hypotension). The decision to use thrombolytic therapy depends on the clinician’s assessment of PE severity, prognosis, and risk of bleeding. Thus, it is often considered in patients with hypotension but without shock. Box 57-2 summarizes recommendations with regard to thrombolytic therapy in PE.

22. What are some complications and contraindications of thrombolytic therapy?

    Intracranial hemorrhage is the most devastating complication of thrombolytic therapy and has been reported in approximately 1% of patients in clinical trials, but in about 3% of patients from the International Cooperative Pulmonary Embolism Registry (ICOPER) representing a more “real world” patient population. Other complications include retroperitoneal and gastrointestinal bleeding and bleeding from surgical wounds or from sites of recent invasive procedures. Contraindications to thrombolytic therapy are divided into major and relative contraindications and are listed in Box 57-3.

23. Has thrombolytic therapy been shown to improve mortality from PE?

    No. Thrombolytic therapy has never been shown to improve mortality from PE in a clinical trial. It has been shown to improve hemodynamics and lung scans with a suggestion that younger (less than 50 years old) patients, and patients with new emboli (less than 48 hours old) and larger emboli respond better. However, based upon a review of the available medical literature, the standard of care in the absence of absolute contraindications is to administer thrombolytic therapy in the setting of massive PE.

24. What are the indications for inferior vena caval (IVC) filter placement?

    ACCP-9 states that the primary indications for placement of an IVC filter include contraindications to anticoagulation, major bleeding complications during anticoagulation, and recurrent PE despite adequate anticoagulation. Although there are no firm trial data, some experts suggest filter placement in the case of massive PE when it is believed that additional emboli might be lethal, particularly if thrombolytic therapy is contraindicated.

25. What are some complications of IVC filter placement?

    IVC filters increase the subsequent incidence of DVT (in about 20% of patients) and have not been shown to increase overall survival. Other complications of IVC filters include procedural-related insertion-site thrombosis (8%), pneumothorax, air embolism, or hematoma, and late complications of IVC thrombosis (2% to 10%), postthrombotic syndrome, IVC penetration, and filter migration. Most models of IVC filters are retrievable, typically within several months of insertion. While this may alleviate some of the late complications of IVC filter placement, complications may also occur with retrieval. Clearly, more data are needed examining the indications for IVC filter placement and the pros and cons of retrieval. At present, the evidence base is inadequate to recommend either retrieval or leaving an IVC filter in place.

26. Can PE be treated as an outpatient?

    Although the use of LMWH as outpatient therapy for DVT is well established, the data for outpatient treatment of acute PE are less robust. Recent data suggest that early discharge in acute PE can be done safely if patients are carefully screened. Initial admission, even if brief, is clearly the most common practice today.

27. Are there any advances in oral anticoagulation therapy for PE?

    Several new oral anticoagulant drugs have been developed. These direct (i.e., antithrombin-independent) inhibitors of factor Xa (e.g., rivaroxaban, apixaban) or thrombin (e.g., dabigatran) avoid most of the drawbacks of heparin and could replace vitamin K antagonists and heparins in many patients. These drugs are administered in fixed doses, do not need coagulation monitoring in the laboratory, and have very few drug–drug or drug–food interactions. Dabigatran and rivaroxaban are U.S. Food and Drug Administration (FDA)-approved for treatment of nonvalvular atrial fibrillation. Rivaroxaban is also approved for prophylaxis for both total hip and total knee replacement. Both drugs have proven noninferior to warfarin in acute treatment of DVT and PE, after a short period of bridging with LMWH. Oral rivaroxaban was FDA-approved for the treatment of acute DVT and/or PE based upon the EINSTEIN-DVT and EINSTEIN-PE studies. In EINSTEIN-PE, oral rivaroxaban was noninferior to the standard approach of parenteral anticoagulation together with warfarin therapy with regard to recurrent VTE events. Importantly, major bleeding rates were significantly less with rivaroxaban.

Bibliography, Suggested Readings, and Websites