Hospital Medicine

Pulmonary embolism

Jump to Section

I. What every physician needs to know.

Pulmonary embolism and deep venous thrombosis are two manifestations of the same underlying disorder, venous thromboembolism (VTE). Venous thrombosis usually occurs as the result the combination of three potential causative factors, specifically venous endothelial damage (e.g., prior clot), hypercoaguability (e.g., malignancy or hereditary thrombophilia) and venous stasis (e.g., prolonged immobilization or congestive heart failure).

Pulmonary embolism results when a thrombus forms somewhere in the venous system, most commonly in the deep veins of the lower extremities, and dislodges, traveling through the right side of the heart before becoming lodged in the pulmonary arterial circulation. Pulmonary embolism causes symptoms and occasionally hemodynamic instability through several mechanisms including obstruction of pulmonary blood flow and liberation of vasoconstrictive cytokines. Thus, although deep venous thrombosis and pulmonary embolism are closely related pathophysiologically, they differ significantly in presentation and means of diagnosis.

The diagnosis of pulmonary embolism generally requires some form of imaging the results of which must be interpreted in conjunction with the clinical likelihood of the disease.

II. Diagnostic Confirmation: Are you sure your patient has Pulmonary Embolism?

The single most important factor in evaluating a patient for suspected pulmonary embolism is determination of the clinical probability of disease as no single laboratory or imaging test can rule-in or rule-out the diagnosis.

�-�Important: No test done in the evaluation of a patient with suspected pulmonary embolism can be interpreted without first considering the clinical likelihood of disease.

Clinical criteria for diagnosis should include clinical probability of disease as determined by clinical gestalt or the Wells Score. The diagnosis is confirmed in the following situations:

  • - High or moderate clinical probability of PE and positive computed tomography (CT) scan completed under pulmonary embolism protocol

  • - Low clinical probability with a positive d-dimer and positive CT scan

  • - Positive pulmonary angiogram

  • - High or moderate clinical probability and a high probability VQ scan scan result

  • - Positive lower extremity duplex ultrasonography for deep venous thrombosis in the proper clinical setting

SeeFigure 1 for Pulmonary Embolism Diagnosis Algorithm.

Figure 1.

Pulmonary Embolism Diagnosis Algorithm

A. History Part I: Pattern Recognition:

Patients with pulmonary embolism defy a "typical presentation" as many of the manifestations are non-specific and protean. The classic presentation of pleuritic chest pain, dyspnea and tachycardia is both uncommon and non-specific. The most important aspect of the history is determining whether an alternate condition that may explain the symptoms and signs is present.

Most Common Symptoms:


Pleuritic chest pain*



Leg pain

Most Common Signs:




Accentuated P2

All patients in whom pulmonary embolism is a consideration should undergo a determination of the clinical likelihood of disease. Although clinical gestalt may be used, it is generally more reliable to use one of the validated clinical prediction tools. Many such tools exist, although the simplified Wells score is one of the easiest to use.

Simplified Wells Score

Score points for each of the following clinical findings:

  • - Clinical signs/symptoms of DVT = 3 points

  • - Alternative diagnosis less likely than PE = 3 points

  • - Heart rate>100 beats/minute = 1.5 points

  • - Immobilization for more than 3 days or surgery in prior 4 weeks = 1.5 points

  • - Previous VTE = 1.5 points

  • - Hemoptysis = 1 point

  • - Cancer (active treatment within 6 months or palliative care) = 1 point

If total point score is 5 or greater, PE is likely

If total point score is 4 or less, PE is unlikely

B. History Part 2: Prevalence:

Most patients will have at least one risk factor for VTE. Common risk factors for pulmonary embolism include:

Prior venous thromboembolic disease

Active malignancy (especially adenocarcinoma)

Recent surgery (especially orthopedic surgery)

Hereditary thrombophilia (Factor V Leiden, Protein C/S deficiency, ATIII deficiency)

Prolonged immobilization (including air travel over 4-6 hours duration)



Oral contraceptive use (especially in conjunction with tobacco use)

C. History Part 3: Competing diagnoses that can mimic Pulmonary Embolism.

Many diagnoses present in a similar fashion to PE. Among the most common are the following:



Acute coronary syndrome

Acute pericarditis

Aortic dissection



Differentiation of the above from pulmonary embolism is difficult and depends on how strong the diagnostic case is for the alternative diagnosis. Each of the above has salient findings that make it at least relatively likely (e.g., infiltrate on CXR for pneumonia, ECG changes and enzyme leak with ACS) and it is the lack of any of these distinctive findings that makes PE more likely.

D. Physical Examination Findings.

As with the history, many findings on physical examination may suggest the diagnosis but none establish it. In large part, the value of the physical examination in patients with suspected pulmonary embolism lies in uncovering findings that suggest an alternate diagnosis (such as unilateral absent breath sounds in pneumothorax and a pericardial friction rub in acute pericarditis).

E. What diagnostic tests should be performed?

The diagnostic approach to the patient with suspect pulmonary embolism is entirely dependent on the clinical probability of disease.

In the patient with alow clinical probability of disease (or PE "unlikely" if the Wells Score is used), the initial test should be the d-dimer. If the d-dimer is negative, the diagnosis of PE may be discounted.

If the patient has a postive d-dimer or amoderate/high clinical probability of disease (or PE "likely" by the Wells Score), then the inital test should be a CT scan of the chest under a rule-out pulmonary embolism protocol.

1. What laboratory studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

There is no laboratory test that establishes the diagnosis of pulmonary embolism and there is no single test that should be done in every patient.

D-dimer:The rapid ELISA d-dimer level is very useful in excluding the diagnosis of pulmonary embolismin the proper patient population. Bedside latex agglutination assays do not have sufficiently strong test characteristics to be used for this purpose. In patients with a low clinical probability of pulmonary embolism, a negative d-dimer safely excludes the diagnosis. The test, however, should not be used in patients with a moderate or high probability of pulmonary embolism as there remains a significant chance of pulmonary embolism in these patients even with a negative d-dimer.

Arterial blood gas: Arterial blood gases are of little utility in evaluating the patient with suspected pulmonary embolism. Although many patients with pulmonary embolism have an increased alveolar-arterial oxygen gradient with a concomitant respiratory alkalosis, these findings are not specific to patients with pulmonary embolism and do little to distinguish PE from other causes of dyspnea and chest pain. They should be done only when they may help in the evaluation of other diagnostic possibilities, such as chronic obstructive pulmonary disease.

Troponin: Troponin levels may be elevated in acute pulmonary embolism secondary to right heart strain but are non-specific. An elevated troponin may indicate a worse prognosis but is not helpful in directing the diagnostic or therapeutic plan.

Thrombotic Risk Factor Panel: These panels are of little utility and are not necessary in most patients unless a clear indication for testing exists.

2. What imaging studies (if any) should be ordered to help establish the diagnosis? How should the results be interpreted?

Chest X-ray:The majority of patients with pulmonary embolism have an abnormal chest film, although the findings are rarely caused by the embolism itself. Only 12% of patients have a normal chest film. The major utility of the chest film is in finding other causes of the patient's symptoms (i.e. an infiltrate in pneumonia or absent lung markings in pneumothorax). Hampton's Hump and Westermark's Sign are the only findings with any specificity to pulmonary embolism.

CT scan of chest under pulmonary embolism protocol (helical or spiral CT):The CT scan is the most commonly utilized imaging study in patients with suspected pulmonary embolism. It has good test characteristics (positive likelihood ratio of ~20; negative likelihood ratio of <0.10), can be completed quickly and often provides information related to other potential diagnoses (e.g., pneumonia, lung cancer). Drawbacks include exposure to ionizing radiation and the need for intravenous contrast dye, limiting utility in patients with kidney disease. It is theinitial test of choice in patients with a moderate or high probability of disease and the most appropriate next test in patients with a low clinical probability of disease but a positive d-dimer assay.

Ventilation-perfusion (VQ) scan:Despite being supplanted by the CT in the evaluation of most patients with suspected PE, the VQ scan remains a viable diagnostic option. It is most often used when a patient may not receive intravenous contrast dye, specifically those patients with allergy to dye or significant kidney injury/disease. A normal chest X-ray greatly increases the likelihood of a definitive VQ scan result. It has well-characterized test characteristics and is safe, but requires substantial time to complete, is not widely available and, most importantly, its results are commonly misinterpreted by clinicians.To properly interpret the results of a VQ scan, one must combine the pre-test clinical probability and the VQ scan result to determine the post-test probability of pulmonary embolism.

Lower extermity duplex ultrasonography:Lower extremity ultrasonography is a second-line test in patients with suspected pulmonary embolism. The rationale for use rests on the supposition that a positive scan for deep venous thrombosis indicates the clinical findings are secondary to concomitant pulmonary embolism (as occurs at least 50% of the time with deep venous thrombosis). It, however, has limited utility in ruling out pulmonary embolism (negative likelihood ratio of ~0.73) except when used in serial fashion in patients with good cardiopulmonary reserve.

Lower extremity ultrasonography should not be used as a first-line test except in those circumstances where a CT may not be performed and VQ scanning is not available (such as in patients with kidney disease in institutions where a VQ scan is unobtainable). It is useful as a second-line test in those patients in whom diagnostic uncertainty remains after the CT scan (i.e. high clinical probability but a negative CT) or VQ scan.

Pulmonary angiography: Although pulmonary angiography is nominally the standard test for pulmonary embolism, it has substantial drawbacks including its invasive nature, the need for intravenous contrast dye and limited availablity. Use is generally limited to those situations where diagnostic uncertainty remains despite extensive evaluation with other methods of testing and the establishment of a definitive diagnosis is critical.

Echocardiography: Echocardiography is a poor first-line test for pulmonary embolism. Its use is limited to help determine prognosis as a dilated right ventricle portends increased mortality, although the effect of this information on treatment options is limited.

Magnetic resonance imaging (MRI): Because of issues relating to technical quality of study, difficulty in interpretation and accessibility, MRI is rarely used in the evaluation of patients with suspected pulmonary embolism. Although it was initially heralded as a potentially useful test in patients with kidney disease, the emergence of nephrogenic systemic fibrosis has further limited its utility in this situation.

Interpreting a VQ scan result

A VQ scan result can only be interpreted in light of the predetermined clinical probability. Given that the studies with VQ scan were completed prior to the Wells Score, clinical gestalt is used and patients are stratified into low, moderate and high probabilities of PE based on the clinical findings. The VQ scan result is reported out as normal, low probability, indeterminate and high probability.

Ahigh probability scan result with a high or moderate clinical probability secures the diagnosis of PE

Ahigh probability scan result with a low clinical probability, means PE is likely but further testing may be considered.

Anindeterminate scan result means anywhere from a 66% (high clinical probabilty) to 15% (low clinical probability) chance of PE and more testing is indicated in most patients.

Alow probability scan result with a low clinical probability rules out PE.

Alow probability scan result with a moderate or high clinical probability yields a 15-40% chance of PE, necessitating more testing.

Anormal scan result excludes the diagnosis regardless of clinical probability.

F. Over-utilized or “wasted” diagnostic tests associated with this diagnosis.

All tests must be ordered after specific consideration of the clinical probability of disease in the individual patient. In patients with a moderate or high probability of diease, a d-dimer level should not be ordered. Similarly, a CT scan may be avoided in patients with a low clinical probability of disease if a d-dimer assay is negative.

III. Default Management.

Management of pulmonary embolism in all patients includes consideration of:

1) Thrombolytic therapy

2) Immediate systemic anticoagulation

3) Inferior vena caval interruption

A. Immediate management.

1) Thrombolytic therapy: Thrombolysis has a limited role in the treatment of pulmonary embolism but may be life saving in select patients. Given in the form of intravenous tPA, thrombolysis should be administered to patients without contraindications who are in shock secondary to pulmonary embolism. Although other indications for thrombolysis including right ventricular strain on echocardiography and elevated troponin levels have been suggested, there is no clear data to demonstrate it improves outcomes in these situations.

2) Immediate anticoagulation:Unless undergoing thrombolytic therapy, all patients with confirmed pulmonary embolism and without strong contraindications should be immediately anticoagulated. In addition, patients with a moderate to high likelihood of pulmonary embolism without a high risk of bleeding should be anticoagulated while undergoing testing for PE even before the diagnosis is confirmed.

Standard approach:

Low molecular weight heparin or intravenous unfractionated heparin concomitant with the initiation of longerm warfarin therapy is the best studied means of establishing anticoagulation in the setting of acute pulmonary embolism.

Low molecular weight heparins are likely modestly superior to unfractionated heparin for pulmonary embolism. Administed sub-cutaneously and dosed by patient weight, they result in reliable and immediate anticoagulation. They, however, are difficult to use in markedly obese patients and are contraindicated in advanced kidney disease (CrCl<30ml/min). Furthermore, there is no well-studied means of reversal and the effects are prolonged. Lastly, there is limited experience in patients with marginal hemodynamics as these patients were specifically excluded from the trials of low-molecular weight heparin in PE.

Intravenous unfractionated heparin begun as a bolus and followed by a continuous infusion is also effective. A weight-based protocol should be used and frequent monitoring of the activated partial thromboplastin time (aPTT) is required. Failure to achieve a therapeutic aPTT (1.5 times the control value) within 24-hours is associated with worse clinical outcomes and close attention must be paid to ensuring anticoagulation is adequate. Conversely supra-therapeutic anticoagulation in the first several days of therapy is not clearly associated with an increased risk of bleeding. Unfractionated heparin may also be adminstered via intermittent subcutaneous injection, although clinical experieince with this approach is limited.

Because of superior efficacy, low-molecular weight heparin should be used in most patients with pulmonary embolism. Unfractionated heparin, however, is preferred in the following settings:

  • - a surgical procedure is anticipated in the short-term,

  • - the patient is thought to be at very high risk of bleeding in the short-term,

  • - morbid obesity,

  • - significant renal dysfunction (CrCL<30ml/min), and

  • - hemodynamic instability.

Once systemically anticoagulated with heparin, most patients will be started on warfarin. If possible, a standard warfarin dosing algorithm should guide initiation of therapy. Heparin therapy should be continued through at least the first 4-5 days of warfarin therapy even if the INR becomes therapeutic more quickly. Once the INR has been in the therapeutic range for at least 2 consecutive days, the heparin may be discontinued. Warfarin therapy causes a rapid depletion of coagulant factor VII which results in elevated INR, although most of the true anticoagulant effect of warfarin comes from the depletion of other clotting factors which take longer to drop. Discontinuing the heparin coverage sooner, even with a therapeutic INR, may result in adverse events.

Other anticoagulants:

Fondaparinux: This agent has been shown to be as effective as intravenous unfractionated heparin in the treatment of acute pulmonary embolism with similar bleeding rates and is FDA-approved for this indication. Given the relatively limited experience, however, its use is reserved for limited situations such as prior heparin-induced thrombocytopenia.

Direct thrombin inhibitors (argatroban, bivalarudin, dabigitran): Although the intravenous direct thrombin inhibitors argatroban and bivalirudin result in immediate and reliable anticoagulation, there is little data on their use for treatment of pulmonary embolism. The orally available dabigitran has been shown to be non-inferior to warfarin for the longterm treatment of PE (but not as a substitute for heparin despite its immediate onset) but it is not FDA-approved for this indication and clinical experieince is limited.

Rivaroxaban/apixaban: The oral direct Factor Xa antagonists have also been shown to be non-inferior to the vitamin K antagonists in the treatment of venous thromboembolic disease. There is limited clinical experience in using these medications for venous thromboembolism and they are not FDA-approved for this indication.

3) Vena caval interruption:Although frequently employed, there is limited data to support the use of vena caval filters in acute pulmonary embolism. They are most commonly used to prevent recurrent events in patients with established pulmonary embolism who have significant contraindications to anticoagulation such as active bleeding. They are also used for other indications including sub-massive pulmonary embolism and significant concomitant significant lower extremity deep venous thrombosis, although there is no data to support this approach.

Of note, although IVC filters do decrease the risk of pulmonary embolism in the short-term, they are associated with a high risk of long-term deep venous thrombosis. As such, retrieveable filters may be placed although there is little data regarding the utility of this approach.

B. Physical Examination Tips to Guide Management.

None other than the standard examination.

C. Laboratory Tests to Monitor Response To, and Adjustments in, Management.

Laboratory tests required in follow-up include:

INR (daily; baseline must be drawn BEFORE initiation of warfarin)

Platelet count (in patients on low molecular weight and unfractionated heparin) every 3-4 days

CBC is not indicated unless there are clinical signs of bleeding

D. Long-term management.

Duration of anticoagulation in patients with pulmonary embolism varies according to the clinical presentation and presence of risk factors for recurrent disease:

First episode without clear risk factor (unprovoked): 6-12 months of warfarin (INR 2-3)

First episode with modifiable/transient risk factor (provoked): 3-6 months of anticoagulation (warfarin to INR 2-3)

Unmodifiable risk factor: At least 12 months of warfarin (INR 2-3) with consideration of lifelong therapy

Recurrent VTE: Lifelong anticoagulation is appropriate

D-dimer testing and thrombin generation after completion of therapy may help stratify the risk of recurrent disease, but more data is necessary before this approach can be recommended.

E. Common Pitfalls and Side-Effects of Management.

As the patient is transitioned to warfarin, heparin therapy should be continued for at least 4-5 days and until the INR is in the therapeutic range for at least 2 consecutive days. Warfarin therapy causes a rapid depletion of coagulation factor VII, which results in elevated INR although most of the true anticoagulant effect of warfarin comes from the depletion of other clotting factors which take longer to drop.

IV. Management with Co-Morbidities.


A. Renal Insufficiency.

The management of pulmonary embolism is significantly altered in the presence of significant renal disease. Low molecular weight heparin cannot be used with a CrCl <30ml/min necessitating use of unfractionated intravenous heparin followed by warfarin.

B. Liver Insufficiency.

Titration of warfarin therapy in patients with severe liver disease and an elevated INR is difficult and the baseline INR must be kept in mind when assessing the adequacy of anticoagulation. Most patients with an elevated INR secondary to liver disease are poor candidates for systemic anticoagulation.

C. Systolic and Diastolic Heart Failure.

No change in standard management.

D. Coronary Artery Disease or Peripheral Vascular Disease.

No change in standard management.

E. Diabetes or other Endocrine issues.

No change in standard management.

F. Malignancy.

Patients with malignancy and pulmonary embolism present a unique problem as in many cases the thrombosis is driven by the presence of malignancy (particularly adenocarcomas). Malignancy-associated pulmonary embolism may be "warfarin-resistant" and heparin is effective in this situation. Long-term anticoagulation with low molecular weight heparin is a better alternative as long as the patient is accepting of the need for daily injections. Anticoagulation should be continued lifelong unless the malignancy is cured.

G. Immunosuppression (HIV, chronic steroids, etc).

No change in standard management.

H. Primary Lung Disease (COPD, Asthma, ILD).

No change in standard management.

I. Gastrointestinal or Nutrition Issues.

In patients with active gastrointestinal bleeding, anticoagulation is contraindicated and IVC filter placement may be necessary. Chronic gastrointestinal blood loss is a relative contraindication to anticoagulation and therapy must be individualized to the patient.

J. Hematologic or Coagulation Issues.

Significant thrombocytopenia may render systemic anticoagulation very high risk and necessitate IVC filter placement.

K. Dementia or Psychiatric Illness/Treatment.

Longterm anticoagulation may be difficult in patients with dementia secondary to issues with medical adherance and fall risk. Therapy decisions must be individualized although at least short-term anticoagulation is usually indicated.

V. Transitions of Care.

A. Sign-out considerations While Hospitalized.

Ensure adequate anticoagulation (therapeutic aPTT) if on unfractionated heparin.

Monitor for evidence of bleeding.

Consider thrombolytic therapy if becomes hemodynamically unstable.

Warfarin dosing contingency if not already dosed for the day.

B. Anticipated Length of Stay.

If on unfractionated heparin, length-of-stay is 4-6 days.

If on low molecular weight heparin, length-of-stay may be 2-3 days if close interval follow-up of INR and home administration of heparin can be arranged.

C. When is the Patient Ready for Discharge.

Criteria for discharge include:

  • - Normal hemodynamics and lack of respiratory distress

  • - Adequate oxygenation on room air or low flow oxygen

  • - Solid plan for follow-up of anticoagulation status

  • - If on unfractionated heparin, INR greater than 2.0 for 2 consecutive days AND at least 4-5 days of heparin/warfarin overlap.

D. Arranging for Clinic Follow-up.

Solid plans for follow-up of anticoagulation status as an outpatient is CRITICAL and includes:

  • - If on warfarin, repeat INR in 1-3 days (depending on stability of INR and warfarin dose over last 3 days of hospitalization)

  • - If on low-molecular weight heparin, daily INR and clear communication to physician assuming care of when discontinuation of heparin is appropriate

  • - Communication of warfarin doses and resultant INRs to the physician assuming care of the patient

1. When should clinic follow up be arranged and with whom.

General Medical Clinical follow-up within one week is appropriate with most patients. This schedule must be accelerated if the need for active management of anticoagulation status (i.e. monitoring of INR and adjustment of warfarin dosing/discontinuation of heparin bridge therapy) is anticipated.

2. What tests should be conducted prior to discharge to enable best clinic first visit.


3. What tests should be ordered as an outpatient prior to, or on the day of, the clinic visit.

Follow-up INR if on warfarin.

Platelet count may be necessary if on heparin.

E. Placement Considerations.

None specific to pulmonary embolism.

F. Prognosis and Patient Counseling.

15% mortality rate untreated.

Drug-food interaction discussion/education is critical in patient being discharged on warfarin.

VI. Patient Safety and Quality Measures.

A. Core Indicator Standards and Documentation.

All patients must receive education on anticoagulant therapy (National Patient Safety Goal) including:

1) Importance of follow-up monitoring

2) Compliance

3) Drug-food interactions

4) Potential for adverse drug reactions and interactions

B. Appropriate Prophylaxis and Other Measures to Prevent Readmission.

Ensure close follow-up of anticoagulation status (INR and warfarin dosing).

VII. What's the evidence?

Stein, PD, Beemath, A, Matta, F, Weg, JG, Yusen, RD, Hales, CA. "Clinical characteristics of patients with acute pulmonary embolism: data from PIOPED II". Am J Med. vol. 120. 2007. pp. 871-879.

Stein, PD, Woodard, PK, Weg, JG, Wakefield, TW, Tapson, VF, Sostman, HD. "Diagnostic pathways in acute pulmonary embolism: recommendations of the PIOPED II investigators". Am J Med. vol. 119. 2006. pp. 1048-1055.

Stein, PD, Patel, KC, Kalra, NK, Petrina, M, Savarapu, P, Furlong, JW, Jr, Steele, RD, Jr., Check, FE. "Estimated incidence of acute pulmonary embolism in a community/teaching general hospital". Chest. vol. 121. 2002. pp. 802-805.

van Belle, A, Buller, HR, Huisman, MV, Huisman, PM, Kaasjager, K, Kamphuisen, PW. "Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography". JAMA. vol. 295. 2006. pp. 172-179.

Stein, PD, Fowler, SE, Goodman, LR, Gottschalk, A, Hales, CA, Hull, RD. "Multidetector computed tomography for acute pulmonary embolism". N Engl J Med. vol. 354. 2006. pp. 2317-2327.

"Results of the prospective investigation of pulmonary embolism diagnosis (PIOPED)". The PIOPED Investigators. JAMA. vol. 263. 1990. pp. 2753-2759.

Lucassen, W, Geersing, GJ, Erkens, PM, Reitsma, JB, Moons, KG, Buller, H, van Weert, HC. "Clinical decision rules for excluding pulmonary embolism: a meta-analysis". Ann Intern Med. vol. 155. 2011. pp. 448-460.

Todd, JL, Tapson, VF. "Thrombolytic therapy for acute pulmonary embolism: a critical appraisal". Chest. vol. 135. 2009. pp. 1321-1329.

Tapson, VF. "Acute pulmonary embolism". N Engl J Med. vol. 358. 2008. pp. 1037-1052.

You must be a registered member of Psychiatry Advisor to post a comment.

Sign Up for Free e-newsletters