Residency · Residency · Internal Medicine

Pulmonary Embolism: Diagnosis, Risk Stratification, and Treatment

Epidemiology and Pathophysiology

Pulmonary embolism is the third most common cause of cardiovascular death after myocardial infarction and stroke, with an annual incidence of approximately 1 to 2 per 1,000 persons that increases with age. Most pulmonary emboli originate from deep venous thrombosis of the proximal lower extremities, with iliofemoral clots posing greater risk than popliteal. The primary mechanism of death in massive PE is acute right ventricular failure caused by a sudden increase in pulmonary vascular resistance. The conceptual foundation for understanding VTE risk remains Virchow's triad: stasis, endothelial injury, and hypercoagulability.

Risk Factors

Strong risk factors include recent surgery or trauma, immobilization exceeding three days, lower extremity fracture, prior VTE, active malignancy, and spinal cord injury. Moderate risk factors encompass hormonal therapy (oral contraceptives, hormone replacement therapy), pregnancy and the postpartum period, central venous catheters, obesity, and inherited or acquired thrombophilia. Weaker associations include advanced age, prolonged travel, varicose veins, and smoking.

Clinical Presentation

The classic triad of dyspnea, pleuritic chest pain, and hemoptysis is present in fewer than 20% of cases. Acute onset dyspnea is the most common symptom at 73%, while tachypnea is the most common sign at 54%. Other findings include tachycardia, pleuritic pain, cough, and syncope, which suggests massive PE. The condition may also present as unexplained hypoxemia, sinus tachycardia, or pulseless electrical activity arrest. Massive PE presents with hypotension (systolic blood pressure below 90 mmHg) or cardiac arrest. Clinicians should consider PE in patients with unexplained acute exacerbations of COPD, new-onset atrial fibrillation, or syncope.

Diagnostic Approach

Pre-Test Probability Assessment

Wells Score (Modified)

The Wells score assigns points for clinical signs of DVT (+3), PE as the most likely diagnosis (+3), heart rate above 100 (+1.5), immobilization or surgery in the prior four weeks (+1.5), previous DVT or PE (+1.5), hemoptysis (+1), and active cancer (+1). Scores are interpreted as low probability (below 2), moderate (2-6), or high (above 6), or dichotomized as PE unlikely (4 or less) versus PE likely (above 4).

YEARS Algorithm (Simplified)

The YEARS algorithm uses three criteria: clinical signs of DVT, hemoptysis, and PE as the most likely diagnosis. If zero YEARS criteria are present and the D-dimer is below 1000 ng/mL, PE is excluded. If one or more YEARS criteria are present and the D-dimer is below 500 ng/mL, PE is also excluded. This approach increases the proportion of patients in whom PE can be safely excluded without imaging.

PERC Rule (Pulmonary Embolism Rule-Out Criteria)

The PERC rule is applied only when clinical gestalt suggests low probability (below 15%). All eight criteria must be met to rule out PE without D-dimer testing: age below 50, heart rate below 100, SpO2 of 95% or higher, no hemoptysis, no estrogen use, no prior DVT or PE, no unilateral leg swelling, and no recent surgery or trauma. If any single criterion is positive, D-dimer testing should proceed.

D-Dimer

D-dimer has high sensitivity (approximately 95%) but low specificity (40-50%) and is useful for excluding PE in patients with low-to-moderate pre-test probability. Age-adjusted D-dimer cutoffs (age multiplied by 10 for patients over 50) improve specificity without sacrificing sensitivity. D-dimer is unreliable in hospitalized patients, post-surgical patients, pregnant patients, those with active cancer, and patients with known DVT, as it is elevated in numerous conditions including infection, inflammation, malignancy, pregnancy, trauma, and disseminated intravascular coagulation.

CT Pulmonary Angiography (CTPA)

CTPA is the primary imaging modality, with sensitivity of 83-100% and specificity of 89-97%. It identifies clot location and burden as well as right ventricular enlargement, with an RV/LV ratio exceeding 0.9 suggesting RV strain. Limitations include contrast allergy, renal insufficiency, and radiation exposure. The clinical significance of subsegmental PE remains debated, though bilateral subsegmental emboli in symptomatic patients are generally treated.

V/Q Scan (Ventilation-Perfusion Scintigraphy)

V/Q scanning serves as an alternative when CTPA is contraindicated due to severe contrast allergy, chronic kidney disease, or as a preference in pregnancy. Results are interpreted as normal (excludes PE), high probability (treat), or non-diagnostic (requiring further workup). It performs best in patients with a normal chest X-ray and no underlying lung disease. SPECT V/Q imaging improves diagnostic accuracy over planar imaging.

Lower Extremity Ultrasound

When DVT is confirmed in a patient with suspected PE, treatment is the same regardless of PE diagnosis, potentially avoiding the need for CTPA. This approach is particularly useful when CTPA is contraindicated but has limited sensitivity for isolated PE without DVT.

Echocardiography

Echocardiography is not diagnostic for PE but is essential for risk stratification. Findings include RV dilation, RV hypokinesis, McConnell sign (RV free wall hypokinesis with apical sparing), tricuspid regurgitation, and a D-shaped septum. It can also identify alternative diagnoses such as tamponade, myocardial infarction, or aortic dissection.

Other Studies

Arterial blood gas typically shows hypoxemia, respiratory alkalosis, and an elevated A-a gradient, though values may be normal in 20% of PE cases. The ECG most commonly shows sinus tachycardia; the classic S1Q3T3 pattern is insensitive. Right axis deviation, right bundle branch block, and T-wave inversions in V1 through V4 indicate RV strain. Troponin and BNP are used for risk stratification rather than diagnosis.

Risk Stratification

Massive (High-Risk) PE

Massive PE is defined by sustained hypotension (systolic below 90 for more than 15 minutes) or the need for vasopressors, cardiac arrest, or obstructive shock. Mortality ranges from 25 to 65%, and immediate reperfusion therapy is required.

Submassive (Intermediate-Risk) PE

Submassive PE describes hemodynamically stable patients with evidence of RV dysfunction on echocardiography or CT and/or elevated cardiac biomarkers (troponin, BNP). It is further subdivided into intermediate-high risk (both RV dysfunction and elevated biomarkers) and intermediate-low risk (either finding alone but not both). Mortality ranges from 3 to 15%.

Low-Risk PE

Low-risk PE involves hemodynamically stable patients with no RV dysfunction and normal biomarkers, carrying mortality below 1%. These patients are candidates for outpatient management.

PESI / sPESI Scores

The Pulmonary Embolism Severity Index incorporates age, sex, cancer, heart failure, chronic lung disease, heart rate, systolic blood pressure, oxygen saturation, temperature, and altered mental status. The simplified PESI considers age above 80, cancer, chronic cardiopulmonary disease, heart rate at or above 110, systolic pressure below 100, and SpO2 below 90%. An sPESI of zero identifies low-risk patients with 30-day mortality of 1% who are candidates for outpatient treatment. An sPESI of one or higher indicates patients who are not low risk.

Hestia Criteria

The Hestia criteria form a checklist to identify patients suitable for outpatient PE management, incorporating hemodynamic stability, no need for supplemental oxygen, no bleeding risk, no severe pain, and no medical or social reasons for admission.

Treatment

Anticoagulation

Acute Phase

Unfractionated heparin is administered as an IV bolus of 80 U/kg followed by an infusion of 18 U/kg/hr, titrated to an aPTT of 1.5 to 2.5 times control. It is preferred for massive PE (where thrombolysis may be needed), high bleeding risk, renal failure, obesity, and hemodynamic instability. Low-molecular-weight heparin (enoxaparin 1 mg/kg subcutaneously twice daily or 1.5 mg/kg daily) is preferred for most stable patients and does not require monitoring, though anti-Xa levels may be checked for obesity or renal impairment. Fondaparinux, given as weight-based subcutaneous daily dosing, serves as an alternative for heparin-induced thrombocytopenia. Rivaroxaban and apixaban can be initiated as monotherapy without initial heparin after their respective loading periods.

Long-Term Anticoagulation

Direct oral anticoagulants are preferred for non-cancer VTE. Rivaroxaban is dosed at 15 mg twice daily for 21 days followed by 20 mg daily. Apixaban is given as 10 mg twice daily for 7 days followed by 5 mg twice daily. Edoxaban (60 mg daily) and dabigatran (150 mg twice daily) both require at least 5 days of parenteral anticoagulation before initiation. Warfarin with an INR target of 2 to 3 requires bridging with heparin and is less preferred but may be necessary for antiphospholipid syndrome, mechanical valves, or severe renal impairment.

Duration of therapy depends on the clinical context: provoked VTE from a transient risk factor (surgery, immobilization) warrants 3 months; VTE provoked by an ongoing risk factor such as cancer requires indefinite treatment until the risk resolves; unprovoked VTE warrants at minimum 3 months followed by reassessment for indefinite therapy. Factors favoring extended treatment include male sex, positive D-dimer after stopping anticoagulation, persistent residual DVT, and thrombophilia.

Cancer-Associated VTE

LMWH was formerly the standard, but DOACs are now preferred for most cancer patients. The Hokusai VTE Cancer trial demonstrated edoxaban was non-inferior to dalteparin, and SELECT-D evaluated rivaroxaban against dalteparin. Caution is warranted with DOACs in gastrointestinal and genitourinary malignancies due to increased mucosal bleeding risk. Anticoagulation should continue while cancer is active or being treated.

Thrombolysis

Indications

Systemic thrombolysis has a clear indication in massive PE with hemodynamic instability. For submassive PE, it is not routinely recommended; the PEITHO trial showed that alteplase reduced hemodynamic decompensation but increased major bleeding and intracranial hemorrhage.

Agents

Alteplase at 100 mg IV over 2 hours is most commonly used. Tenecteplase as a single weight-based bolus offers easier administration and appears non-inferior in observational data. Half-dose alteplase (50 mg) may carry fewer bleeding complications with similar efficacy and is used in submassive PE or when bleeding risk is elevated, as demonstrated in the MOPETT trial.

Contraindications (Absolute)

Absolute contraindications include active internal bleeding, recent intracranial hemorrhage, intracranial neoplasm, ischemic stroke within three months, suspected aortic dissection, and recent cranial or spinal surgery.

Catheter-Directed Therapy (CDT)

Options include catheter-directed thrombolysis, aspiration thrombectomy, and ultrasound-assisted catheter-directed therapy (EKOS). These approaches are increasingly used for submassive and massive PE, particularly when systemic thrombolysis is contraindicated or has failed. The ULTIMA trial demonstrated that ultrasound-assisted CDT improved the RV/LV ratio compared to heparin alone in submassive PE. Large randomized controlled trial evidence remains limited, and a PE response team (PERT) approach aids multidisciplinary decision-making.

Surgical Embolectomy

Surgical embolectomy is reserved for massive PE when thrombolysis is contraindicated or has failed. It requires cardiopulmonary bypass and carries mortality of 20-50% in experienced centers, with best outcomes achieved when performed before cardiac arrest.

IVC Filters

IVC filters are indicated for acute VTE with an absolute contraindication to anticoagulation, or recurrent PE despite adequate anticoagulation. Retrievable filters are preferred and should be removed when anticoagulation can be resumed. Complications include filter thrombosis, migration, IVC perforation, and post-thrombotic syndrome. The PREPIC2 trial showed no benefit of routine filter placement as an adjunct to anticoagulation.

Special Populations

Pregnancy

D-dimer is physiologically elevated in pregnancy and therefore less useful. Both CTPA and V/Q scan are acceptable imaging modalities, with CTPA preferred by most guidelines due to lower fetal radiation dose. Treatment involves LMWH throughout pregnancy, as DOACs and warfarin are both contraindicated. Patients transition to UFH at 36 weeks or hold LMWH 24 hours before planned delivery.

Chronic Thromboembolic Pulmonary Hypertension (CTEPH)

CTEPH occurs in 2 to 4% of PE survivors when persistent organized thrombus causes pulmonary hypertension. It should be suspected in patients with persistent dyspnea after PE despite adequate anticoagulation. V/Q scan is the screening test of choice, showing mismatched perfusion defects. Treatment options include pulmonary thromboendarterectomy (potentially curative), balloon pulmonary angioplasty, and riociguat for inoperable disease.

<image> A comprehensive diagnostic algorithm for suspected pulmonary embolism. Begin with clinical probability assessment (Wells score or gestalt). Branch into PE unlikely (apply PERC, then D-dimer with age adjustment) vs. PE likely (proceed directly to CTPA). Show decision endpoints: PE excluded, PE confirmed, or alternative diagnosis. Include a sidebar for V/Q scan indications (contrast allergy, CKD, pregnancy). Use color coding for risk pathways. </image>

<image> A risk stratification pyramid for pulmonary embolism showing three tiers: Low-risk PE at the base (normal vitals, sPESI 0, no RV dysfunction, normal biomarkers -- candidate for outpatient management), Submassive PE in the middle (hemodynamically stable but RV dysfunction and/or elevated troponin/BNP -- anticoagulation with close monitoring), and Massive PE at the top (hypotension, shock, cardiac arrest -- systemic thrombolysis or catheter-directed therapy). Include mortality rates and key management interventions at each level. </image>

<image> A treatment timeline diagram for PE anticoagulation. Show the acute phase (0-7 days) with options for UFH, LMWH, or DOAC loading. The early maintenance phase (7 days to 3 months) with DOAC standard dosing or warfarin (INR 2-3). The extended phase decision point at 3 months: provoked (stop) vs. unprovoked (consider indefinite at reduced dose -- EINSTEIN-CHOICE, AMPLIFY-EXT). Include cancer-associated VTE pathway with LMWH or DOAC options. </image>

Clinical Pearls

The PERC rule should only be used when clinical gestalt suggests low probability (below 15%) and must not be applied in moderate or high probability situations. Age-adjusted D-dimer (age multiplied by 10 for patients over 50) increases specificity and reduces unnecessary CTPAs without missing clinically significant PE. A normal D-dimer in a low-probability patient effectively rules out PE, but D-dimer should never be sent in a high-probability patient -- imaging should be pursued directly. An sPESI of zero with no contraindications identifies patients who can safely be managed as outpatients, and clinicians should know the Hestia criteria. In massive PE, thrombolysis should not be delayed for CTPA; if clinical suspicion is high and the patient is in shock, bedside echocardiography showing RV strain is sufficient to initiate treatment. DOACs are now preferred over warfarin for most non-cancer VTE, with rivaroxaban and apixaban offering the advantage of not requiring initial heparin bridging. For unprovoked PE, extended anticoagulation at reduced dose (apixaban 2.5 mg twice daily or rivaroxaban 10 mg daily) should be considered, as the EINSTEIN-CHOICE and AMPLIFY-EXT trials demonstrated continued benefit with minimal increase in bleeding. All PE survivors should be counseled about CTEPH symptoms, particularly persistent exertional dyspnea, and a V/Q scan should be considered if symptoms persist at 3 to 6 months.

References

• Konstantinides SV, et al. 2019 ESC Guidelines for the Diagnosis and Management of Acute PE. European Heart Journal. 2020.
• Kearon C, et al. Antithrombotic Therapy for VTE Disease: CHEST Guideline. Chest. 2016 (updated 2021).
• Meyer G, et al. PEITHO Trial: Fibrinolysis for Intermediate-Risk PE. NEJM. 2014.
• Raskob GE, et al. Hokusai VTE Cancer Trial: Edoxaban vs. Dalteparin. NEJM. 2018.
• Young T, et al. EINSTEIN-CHOICE Trial: Extended Rivaroxaban for VTE. NEJM. 2017.
• Agnelli G, et al. AMPLIFY-EXT Trial: Extended Apixaban for VTE. NEJM. 2013.
• Mismetti P, et al. PREPIC2 Trial: IVC Filter for Severe Acute PE. JAMA. 2015.
• Kucher N, et al. ULTIMA Trial: Ultrasound-Assisted CDT for Submassive PE. Circulation. 2014.
• van der Hulle T, et al. YEARS Algorithm. Lancet. 2017.

Read this lecture as Markdown