Pulmonary Embolism

Pulmonary embolism (PE) is an obstruction of the pulmonary arteries by thrombus, most commonly originating from the deep veins of the lower extremities. It is a leading cause of maternal mortality in the developed world and a frequent cardiovascular emergency ^[1]A1c.

Also Called / Synonyms

• Pulmonary thromboembolism, the preferred pathophysiologic term
• Venous thromboembolism (VTE), the umbrella term encompassing both PE and deep vein thrombosis (DVT)
• Massive PE, historical term for high-risk PE with sustained hypotension
• Submassive PE, historical term for intermediate-risk PE with right ventricular (RV) dysfunction but preserved blood pressure
• Chronic thromboembolic pulmonary hypertension (CTEPH), a late sequela of acute PE in which the embolus remodels into a chronic scar in the pulmonary arteries, causing vascular obstruction and pulmonary hypertension ^[13]D5

Key Terms Used in This Article

• Acute PE, symptomatic obstruction within days to weeks of embolization; the focus of the acute management sections
• High-risk (massive) PE, acute PE with sustained hypotension (systolic BP <90 mmHg for ≥15 min or requiring vasopressors); the 2014 European Society of Cardiology (ESC) algorithm places these patients in the highest risk category ^[11]B2b
• Intermediate-risk (submassive) PE, normotensive acute PE with evidence of RV dysfunction on echocardiography (e.g., increased RV/LV ratio, abnormal tricuspid annular plane systolic excursion [TAPSE]) or elevated cardiac biomarkers (e.g., high-sensitivity troponin T, NT-proBNP) ^{[11]B2b[14]B2a}. The ESC algorithm further subdivides into intermediate-high (both RV dysfunction and elevated biomarkers) and intermediate-low (one or neither)
• Low-risk PE, normotensive, no RV dysfunction, normal biomarkers; can be considered for early discharge or outpatient management
• Subsegmental PE, embolism confined to subsegmental pulmonary arteries; the 2016 CHEST guideline suggests clinical surveillance over anticoagulation when the risk of recurrent VTE is low ^[4]A1c
• Right ventricular dysfunction (RVD), echocardiographic finding that predicts increased risk of death in acute PE; risk ratio for death in all-comers was 1.49 (95% CI 1.24-1.79) ^[14]B2a

Classification of PE

Clinical Significance

PE is a common cardiovascular emergency with a case-fatality rate that remains substantial if untreated. It is a leading cause of maternal death ^[1]A1c and, in patients with COVID-19 , occurs more frequently than in seasonal influenza ^[12]B3b. For patients who survive an acute PE, up to 4% may develop CTEPH, a progressive condition characterized by chronic inflammation, smooth muscle cell proliferation, and pulmonary vascular remodeling ^[13]D5. In cardiac arrest suspected to be due to PE, thrombolytic therapy may be considered ^[18]A1c.

The sections that follow cover the pathophysiology and mechanisms that drive the clinical presentation and risk stratification outlined above.

Pearl: The single most critical decision in acute PE is the initial hemodynamic assessment: a patient with sustained hypotension (systolic BP <90 mmHg) is high-risk and warrants immediate reperfusion, while every normotensive patient requires systematic risk stratification using RV imaging and cardiac biomarkers to guide the intensity of management.

From the classification of acute and chronic thromboembolic disease, the pathophysiologic cascade begins with embolic occlusion of the pulmonary arteries. The hemodynamic consequence depends on the degree of obstruction and the interplay of mechanical and neurohumoral factors ^[37]D5.

Acute Pulmonary Embolism

An embolus lodged in the pulmonary vasculature produces two immediate effects: mechanical obstruction of the arterial lumen and neurohumorally mediated pulmonary vasoconstriction triggered by local release of thromboxane A₂ and serotonin from activated platelets ^[37]D5. The combined obstruction and vasoconstriction increase pulmonary vascular resistance, abrupt right ventricular (RV) afterload, and widen the ventilation-perfusion (V/Q) mismatch. Increased physiological dead space proportion (VD/VT) correlates with the extent of unmatched perfusion defects on V/Q scans ^[31]B3b.

The right ventricle, a thin-walled chamber poorly adapted to acute pressure overload, responds with dilation, increased wall stress, and subendocardial ischemia. If the afterload is sufficiently high, RV failure ensues and cardiac output falls, the hallmark of high-risk (massive) PE ^[37]D5.

Chronic Thromboembolic Pulmonary Hypertension (CTEPH)

In a subset of patients, the acute embolus does not resolve completely. Fibrin from patients who develop CTEPH is resistant to plasmin-mediated lysis, with significantly slower cleavage of all three polypeptide chains (p < 0.05) ^[33]B3b. The release of N-terminal fragments from the beta-chain, which promote cell signaling, migration, and angiogenesis, is also retarded (p < 0.01), suggesting that persistent fibrin structural motifs may drive the transition from acute thromboemboli to chronic obstructive vascular scars ^[33]B3b. The exact mechanism transforming red clots into fibrotic remnants remains unknown ^[34]A1c.

Single-cell RNA sequencing of CTEPH thrombus removed at pulmonary endarterectomy has identified a complex cellular ecosystem: macrophages (with a predominant subcluster upregulating inflammatory signaling), CD4⁺ and CD8⁺ T cells, and heterogeneous smooth muscle cells (SMCs) ^[13]D5. A subset of SMCs transitions into myofibroblasts expressing fibrosis markers, and cultured cells from CTEPH tissue show abnormal angiogenic potential, proliferation, and apoptosis compared with control cells ^[13]D5. PAR1 (protease-activated receptor 1) emerged as a potential therapeutic target: PAR1 inhibition decreased SMC and myofibroblast proliferation and migration, linking thrombosis directly to chronic vascular remodeling ^[13]D5. This model parallels atherosclerosis, with chronic inflammation driving SMC modulation and fibrotic occlusion ^[13]D5.

In addition to proximal arterial obstruction, a secondary microvasculopathy of vessels <500 µm develops, further increasing pulmonary vascular resistance and contributing to progressive right heart failure ^[34]A1c.

Nonthrombotic Pulmonary Embolism

Nonthrombotic pulmonary embolism (NTPE) encompasses embolization of diverse materials, adipocytes, amniotic fluid, trophoblasts, tumour cells, bacteria, fungi, or foreign bodies, and triggers a severe inflammatory reaction in the pulmonary circulation, often presenting with atypical clinical and radiographic features such as tree-in-bud opacities or micronodules ^[35]D5.

Pearl: In acute PE, the degree of RV overload is the central determinant of hemodynamic instability; in chronic PE, the transition from a lytic-resistant fibrin scaffold to a macrophage-driven inflammatory milieu marks the pathobiological shift toward CTEPH, and PAR1 inhibition may offer a future pharmacologic target to interrupt this cascade ^{[13]D5[33]B3b}.

The pathophysiologic triad of venous stasis, endothelial injury, and hypercoagulability translates into a measurable disease burden that varies sharply across populations, clinical contexts, and biomarker profiles.

Incidence and Prognosis

Among all-comers with acute pulmonary embolism, the pooled incidence of subsequent chronic thromboembolic pulmonary hypertension (CTEPH) is 0.56% (95% CI 0.1-1.0%); in survivors of the acute event, it rises to 3.2% (95% CI 2.0-4.4%) ^[47]A1a. The short-term risk of adverse outcomes depends on hemodynamic stability and biomarker status. In normotensive patients, a complicated 30-day course (death, hemodynamic collapse, or recurrent PE) occurred in 7.4% of a derivation cohort and 4.5% of a validation cohort ^[44]B2b. Among patients with intermediate-risk PE (right ventricular strain plus elevated troponin), the composite of PE-related death, cardiorespiratory decompensation, or symptomatic recurrence within 7 days was 10.3% with anticoagulation alone and 4.0% with catheter-directed fibrinolysis ^[52]A1b. In cancer-associated PE managed as outpatients, 30-day all-cause mortality was 1.74% (95% CI 0.99-3.03%) ^[39]A1a.

Risk Factors

Risk factors for PE and its complications span clinical, biomarker, and imaging domains. The table below summarizes the strongest predictors identified in meta-analyses and prospective cohorts.

Elevated growth differentiation factor-15 (GDF-15) independently predicts 30-day complications, with a c-statistic of 0.84 (95% CI 0.76-0.90), outperforming troponin T (0.72) and NT-proBNP (0.65) ^[43]B2b.

Special Populations

Pregnancy: PE is a leading cause of maternal mortality in the developed world, and the diagnostic algorithm must balance accuracy with minimizing cumulative radiation exposure ^[1]A1c.

Cancer: Active malignancy confers a persistent thrombotic risk. In high-risk ambulatory patients (Khorana score ≥2), rivaroxaban 10 mg daily reduced VTE during the intervention period (HR 0.40; 95% CI 0.20-0.80) ^[54]A1b. Extended reduced-dose apixaban (2.5 mg twice daily) was noninferior to full-dose for preventing recurrent VTE and reduced clinically relevant bleeding (12.1% vs 15.6%) ^[50]A1b.

Testosterone therapy: In men with hypogonadism and preexisting cardiovascular disease, testosterone-replacement therapy was associated with a higher incidence of pulmonary embolism ^[51]A1b.

ICU patients: Low-molecular-weight heparin reduces DVT risk compared with control (OR 0.59; 95% CrI 0.33-0.90) and is probably more effective than unfractionated heparin (OR 0.72; 95% CrI 0.46-0.98) ^[41]A1a.

Pearl: The risk of PE-related death is most strongly predicted by right ventricular dysfunction (OR 7.35) and elevated BNP (OR 7.6), these biomarkers are essential for refining risk stratification beyond clinical factors alone.

From the risk factors discussed above, the clinical presentation of pulmonary embolism emerges as a syndrome of acute cardiorespiratory compromise, driven by sudden obstruction of pulmonary arteries and the resulting right ventricular (RV) pressure overload. The spectrum ranges from asymptomatic incidental findings to fulminant obstructive shock, with the severity determined by clot burden, cardiopulmonary reserve, and the host's inflammatory response.

Presenting Symptoms

• Dyspnea - the most common symptom, often acute and pleuritic; persistent dyspnea after PE is frequent and may be the chief complaint in survivors ^[57]A1b.
• Pleuritic chest pain - sharp, localized pain that worsens with inspiration, indicating pleural irritation from distal emboli.
• Cough and hemoptysis - less common; hemoptysis is a YEARS criterion (clinical signs of deep-vein thrombosis, hemoptysis, and pulmonary embolism as the most likely diagnosis) ^[30]B2b.
• Syncope or presyncope - a red flag for hemodynamic compromise, often associated with large central emboli.
• Signs of deep-vein thrombosis - unilateral leg swelling, warmth, or cord-like veins.

Neurological Examination Findings

• Tachypnea (respiratory rate >20 breaths/min) is nearly universal.
• Tachycardia (heart rate ≥110 beats/min) is a predictor of adverse outcomes (adjusted odds ratio [aOR] 1.87, 95% CI 1.31-2.69) ^[58]B2b.
• Hypotension - systolic blood pressure (SBP) 90-100 mmHg carries an aOR of 2.45 (95% CI 1.50-3.99) for PE-related complications; SBP <100 mmHg is a mortality predictor in cancer patients ^{[58]B2b[61]B2b}.
• Hypoxia - basal oxygen saturation <90% is associated with increased 30-day mortality in cancer-associated PE ^[61]B2b.
• RV strain signs - elevated jugular venous pressure, a right ventricular heave, a loud P2 component of the second heart sound, and tricuspid regurgitation murmur.

Phenotypic Variants

Red Flags

• Systolic blood pressure <90 mmHg or a drop of ≥40 mmHg for >15 min.
• Heart rate ≥110 beats/min with evidence of RV strain.
• Oxygen saturation <90% on room air.
• Syncope - suggests large central embolus with transient hypotension.

Atypical Presentations

• In COPD exacerbation: PE may present with worsening dyspnea without classic pleuritic pain or hemoptysis; prevalence of PE in hospitalized COPD exacerbations is 24.7% (95% CI 17.9-31.4%) ^[64]B2a.
• In cancer patients: unsuspected PE (UPE) is often asymptomatic or minimally symptomatic; the 30-day mortality for truly asymptomatic UPE is 3% versus 20-21% for symptomatic PE ^[61]B2b.
• In children: symptoms are similar to adults (dyspnea, chest pain) but the PERC-Peds rule can safely rule out PE when negative ^[68]B2b.

The clinical suspicion raised by these features then guides the selection of diagnostic tests, detailed in the next section.

Pearl: A normotensive patient with PE and heart rate ≥110 bpm, SBP 90-100 mmHg, or oxygen saturation <90% is at intermediate risk and warrants urgent risk stratification ^{[58]B2b[61]B2b}.

The clinical suspicion raised by dyspnea , pleuritic chest pain , or hemoptysis enters a stepwise diagnostic algorithm that begins with assessing pre-test probability and ends with imaging. The goal is to avoid unnecessary radiation and contrast while identifying the ~20% of patients who truly have PE ^[21]A1b.

Step 1: Clinical Probability Assessment

Formal scoring systems ( Wells criteria , revised Geneva score ) or a single clinical question, "Is PE the most likely diagnosis?", stratify patients into low, moderate, or high pre-test probability ^[75]B2b. The single-question approach has been prospectively validated and simplifies the decision tree: when PE is not the most likely diagnosis, a higher D-dimer threshold of 1000 ng/mL can safely rule out the disease ^[75]B2b.

Step 2: D-Dimer Testing

D-dimer is a sensitive but non-specific fibrin degradation product. Its utility depends on the threshold chosen:

• Low pre-test probability: D-dimer < 1000 ng/mL rules out PE without further testing (diagnostic failure rate 0.00%, 95% CI 0.00-0.34) ^[75]B2b.
• Moderate pre-test probability: a threshold of 500 ng/mL is used; if negative, PE is ruled out ^[82]B2b.
• High pre-test probability: D-dimer is not routinely recommended because a negative result does not safely exclude PE; proceed directly to imaging.

The PEGeD strategy (low C-PTP + D-dimer < 1000 ng/mL, moderate C-PTP + D-dimer < 500 ng/mL) resulted in a 3-month VTE event rate of 0.00% (95% CI 0.00-0.29) and avoided chest imaging in 65.7% of patients ^[82]B2b. The simplified single-question strategy reduced chest imaging from 50% to 32% ^[75]B2b.

Step 3: Imaging (Gold-Standard Test)

Computed tomography pulmonary angiography (CTPA) is the reference standard for PE diagnosis. In the randomized trial comparing strategies, CTPA identified PE in 20.7% of patients, and the 3-month VTE risk in patients with a negative CTPA was 0.0%) ^[21]A1b.

Alternative imaging modalities:

• Ventilation-perfusion scan (V/Q): Planar V/Q had a PE diagnosis rate of 16.0% and a 3-month VTE risk of 0.8% after negative workup ^[21]A1b.
• SPECT V/Q : Although SPECT V/Q is widely used, the randomized trial did not demonstrate non-inferiority to CTPA/planar V/Q (p for non-inferiority = 0.75). However, the 3-month VTE risk after a negative SPECT V/Q was only 0.5% (1/221), suggesting it is a safe alternative in practice ^[21]A1b.

When to use V/Q over CTPA: Patients with contraindications to iodinated contrast (e.g., severe allergy, renal impairment) or young women (to reduce breast radiation) are candidates for V/Q imaging.

Step 4: Point-of-Care Ultrasound (Supplementary)

In patients with suspected PE, point-of-care ultrasound can rapidly identify those with high specificity who may not require immediate radiation imaging ^[66]B2a. The following signs have been prospectively evaluated:

Interpretation: A normal anterior lung profile with evidence of DVT on compression ultrasound has a specificity of 99% for PE ^[67]B2b. However, sensitivity is low; a negative ultrasound does not rule out PE.

Diagnostic Algorithm (Simplified Single-Question Strategy)

Step-by-step:

1. Assess clinical probability. Use either the single question or a validated score. ^[75]B2b
2. If PE is NOT the most likely diagnosis: perform D-dimer. If < 1000 ng/mL, stop, PE is ruled out. If ≥ 1000 ng/mL, proceed to age-adjusted D-dimer (or fixed 500 ng/mL) and then to imaging if positive. ^[75]B2b
3. If PE IS the most likely diagnosis: go directly to age-adjusted D-dimer (or fixed 500 ng/mL). If the result is below the threshold, PE is ruled out; otherwise, obtain CTPA or V/Q. ^[75]B2b
4. Imaging: CTPA is the reference standard. V/Q (planar or SPECT) is an alternative when contrast is contraindicated. ^[21]A1b

Special Populations: Pregnancy

In pregnant women with suspected PE, the pregnancy-adapted YEARS algorithm is validated: apply the three YEARS criteria (clinical signs of DVT, hemoptysis, PE as most likely diagnosis). If 0 criteria, use D-dimer threshold of 1000 ng/mL; if ≥1 criteria, use 500 ng/mL. All patients with a positive D-dimer undergo CTPA. This algorithm avoided CTPA in 39% of patients overall (65% in the first trimester) with a 3-month VTE failure rate of 0.21% (95% CI 0.04-1.2) ^[30]B2b.

Pearl: The single question "Is PE the most likely diagnosis?" can replace complex scoring systems and safely allow a 1000 ng/mL D-dimer threshold, reducing chest imaging by nearly 20 percentage points ^[75]B2b.

Once the diagnosis of acute pulmonary embolism is confirmed by computed-tomographic pulmonary angiography (CTPA) or ventilation-perfusion scintigraphy, the next imperative is to stratify the patient's risk of early death or adverse outcome. This classification drives the triage decision, home, ward, or intensive care, and the intensity of initial therapy.

Clinical Risk Scores: PESI and sPESI

The Pulmonary Embolism Severity Index (PESI) and its simplified version (sPESI) are the most extensively validated clinical prediction rules for 30-day mortality. The sPESI assigns one point each for age >80 years, cancer, chronic cardiopulmonary disease, heart rate ≥110 beats/min, systolic blood pressure (SBP) <100 mmHg, and arterial oxyhemoglobin saturation <90%. A score of 0 identifies low-risk patients with a 30-day mortality of approximately 1% ^[85]D5. Use of sPESI to select patients for outpatient treatment has been associated with a low risk of adverse events ^[40]A1b. In a randomized trial, patients without any Hestia criteria (a 11-item checklist that includes hemodynamic instability, hypoxia, bleeding risk, and comorbidity) had a 30-day adverse outcome rate of 1.1% (95% CI 0.2-3.2%) and could be safely discharged directly ^[40]A1b.

Biomarkers for Risk Stratification

Biomarkers of myocardial stretch and injury refine risk assessment. Brain-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP) are markers of right ventricular (RV) dysfunction. In a meta-analysis of 13 studies, elevated BNP was associated with a 6.8-fold increased odds of a complicated in-hospital course (95% CI 4.4-10) and a 7.6-fold increase in 30-day mortality (95% CI 3.4-17). Patients with high NT-proBNP had a 10% risk of dying and a 23% risk of adverse clinical outcome ^[42]A1a. Cardiac troponin elevation indicates myocardial injury and independently predicts adverse outcome (adjusted OR 2.49, 95% CI 1.71-3.69) ^[58]B2b. Copeptin, the C-terminal fragment of vasopressin released during stress, adds prognostic value: a threshold of ≥24 pmol/L conferred a 5.4-fold increased risk of adverse outcome (95% CI 1.68-17.58) and further stratified normotensive patients with elevated hsTnT and NT-proBNP into intermediate-low and intermediate-high risk groups ^[86]B2b.

Imaging for Right Ventricular Dysfunction

RV dilation on CTPA (right-to-left ventricle diameter ratio) is a strong predictor of mortality. In a meta-analysis, CT-detected RV dilation was associated with increased 30-day mortality in all-comers (OR 2.08, 95% CI 1.63-2.66) and in hemodynamically stable patients (OR 1.64, 95% CI 1.06-2.52), and with a 7.35-fold increase in PE-related death (95% CI 3.59-15.09) ^[46]A1a. Echocardiographic RV dysfunction, together with right-to-left ventricle diameter ratio, also independently predicts 30-day adverse events (adjusted OR per 0.1 increase 1.2, 95% CI 1.1-1.4) ^[83]B2b.

Multimarker Models and Integrated Algorithms

The combination of clinical scores, biomarkers, and imaging improves classification. The Bova score integrates SBP 90-100 mmHg, heart rate ≥110 bpm, elevated troponin, and RV dysfunction into a 7-point index identifying three stages with 30-day PE-related complication rates of 4.2%, 10.8%, and 29.2% ^[58]B2b. The PREP model combines altered mental state, shock on admission, cancer, BNP, and right-to-left ventricle ratio, achieving an area under the ROC curve of 0.84 (95% CI 0.78-0.90) ^[83]B2b. A multimarker model from the PROTECT study incorporating sPESI, cardiac troponin I, BNP, and lower limb ultrasound showed a negative predictive value for a complicated course of 99.1% in derivation and 100% in validation cohorts ^[44]B2b.

Special Populations: Cancer-Associated PE

Patients with active cancer require careful risk assessment. A systematic review of 19 studies (1675 patients) found that selected cancer patients managed as outpatients had a 30-day all-cause mortality of 1.74% (95% CI 0.99-3.03%), major bleeding rate of 2.71% (95% CI 1.51-4.83%), and recurrent VTE rate of 1.26% (95% CI 0.53-3.00%) ^[39]A1a. These rates suggest that low-risk cancer patients can be considered for outpatient management, though dedicated prospective validation is needed.

Outpatient Management Criteria

Validated selection tools for early discharge include the Hestia criteria and sPESI = 0. In a randomized noninferiority trial, 550 patients without Hestia criteria were randomized to direct discharge or additional NT-proBNP testing; the primary endpoint (PE- or bleeding-related mortality, CPR, or ICU admission) occurred in no patient in the NT-proBNP group and in 1.1% in the direct discharge group (P=0.25) ^[40]A1b. The Hestia criteria alone appear sufficient for safe outpatient triage. Adding NT-proBNP to a negative Hestia assessment may not further reduce adverse events, but the low event rate precludes definitive conclusions ^[40]A1b.

Controversies and Guideline Disagreement

Pearl: The single most important decision is distinguishing normotensive patients with RV dysfunction from those without; this separates intermediate-risk from low-risk PE and determines whether escalation to reperfusion therapy may be warranted.

Risk stratification, completed in the previous section, directly drives the first therapeutic decision. The clinician must decide within minutes whether the patient is high-risk (persistent hypotension or shock), intermediate-risk (normotensive with right ventricular [RV] dysfunction and elevated cardiac biomarkers), or low-risk (normotensive, no RV strain, normal biomarkers), each category warrants a distinct management pathway.

Step 1: Triage and Disposition

For low-risk patients, outpatient or early discharge is safe. The Hestia clinical decision rule identifies candidates: if none of the 11 criteria (e.g., hemodynamic instability, need for oxygen, active bleeding, severe renal impairment) are present, direct discharge is feasible. In a randomized trial, 550 patients selected by Hestia criteria alone had a 30-day adverse event rate of 1.1% (95% CI, 0.2-3.2%) in the direct-discharge group, and no patient with an elevated NT-proBNP (>500 ng/L) who was treated in hospital experienced the primary endpoint (0%; 95% CI, 0-10.2%) ^[40]A1b. A structured prognostic assessment and management pathway further reduces length of stay: median 4.0 days (IQR 3.7-4.2) vs. 6.1 days (IQR 5.7-6.5) with usual care, without increasing 30-day mortality (2.4% vs. 2.0%) or readmission (4.0% vs. 4.8%) ^[88]A1b. All intermediate- and high-risk patients require in-hospital management, at least initially in an intensive care unit for high-risk patients.

Step 2: Anticoagulation, Initiation

Anticoagulation should be started immediately once the diagnosis is confirmed, unless a contraindication exists. Direct oral anticoagulants (DOACs) are the agents of choice for most patients because of their fixed dosing, no routine monitoring, and superior safety profile.

Among DOACs, apixaban carries a lower risk of clinically relevant bleeding than rivaroxaban. In the COBRRA trial (N=2760), the primary outcome (major or clinically relevant nonmajor bleeding) occurred in 3.3% of apixaban-treated patients vs. 7.1% of rivaroxaban-treated patients (relative risk 0.46, 95% CI 0.33-0.65; P<0.001) ^[92]A1b. NNT = 26 to prevent one clinically relevant bleed with apixaban. For patients with cancer-associated VTE, apixaban is noninferior to dalteparin and does not increase major bleeding (HR 0.82, 95% CI 0.40-1.69) ^[94]A1b. Rivaroxaban is also an option (cumulative incidence of recurrent VTE 6.4% vs. 10.1% with dalteparin; SHR 0.75, 95% CI 0.21-2.66) ^[91]A1b, though the trial was underpowered for noninferiority.

Step 3: Escalation for High-Risk and Intermediate-Risk Patients

High-risk (massive) PE: Immediate systemic fibrinolysis is indicated unless contraindicated. Tenecteplase, given as a single weight-based bolus, is the most studied agent. In the PEITHO trial, normotensive patients with intermediate-risk PE (RV dysfunction plus elevated troponin) who received tenecteplase plus heparin had a lower rate of death or hemodynamic decompensation within 7 days than those receiving placebo plus heparin (2.6% vs. 5.6%; odds ratio 0.44, 95% CI 0.23-0.87; P=0.02) ^[93]A1b. However, major extracranial bleeding increased (6.3% vs. 1.2%; P<0.001) and stroke occurred in 2.4% (hemorrhagic in 10 of 12) vs. 0.2% (P=0.003). NNT = 33 to prevent death or decompensation, NNH = 20 for major bleeding. Therefore, systemic fibrinolysis is reserved for high-risk PE or intermediate-risk patients who deteriorate.

Intermediate-risk PE with indicators of cardiorespiratory distress: Ultrasound-facilitated, catheter-directed fibrinolysis (UF-CDT) with alteplase plus anticoagulation reduces the composite of PE-related death, cardiorespiratory decompensation, or symptomatic recurrence within 7 days compared with anticoagulation alone (4.0% vs. 10.3%; relative risk 0.39, 95% CI 0.20-0.77; P=0.005) ^[52]A1b. Major bleeding within 7 days was 4.1% vs. 2.2% (P=0.32), and no intracranial hemorrhage occurred. NNT = 16 to prevent a primary outcome event. This approach is an option for intermediate-risk patients with systolic BP ≤110 mm Hg, heart rate ≥100 bpm, or respiratory rate >20 breaths/min.

Step 4: Supportive Care

Oxygen therapy: Supplemental oxygen should be administered to achieve a target SpO₂ ≥90% (or ≥92% in patients with pre-existing lung disease). For nonhypoxemic patients with intermediate-risk PE, routine oxygen does not significantly improve RV recovery. In a pilot trial, normalization of RV size at 48 hours occurred in 42.4% of patients receiving oxygen. 21.6% receiving ambient air ^[90]A1b. Do not give supplemental oxygen to normoxemic patients solely for intermediate-risk status.

Hemodynamic support: For high-risk patients with hypotension, cautious fluid resuscitation (250-500 mL crystalloid boluses) is appropriate, followed by vasopressors (e.g., norepinephrine) if shock persists. Avoid excessive fluid administration that could worsen RV distension.

Step 5: Monitoring and Transition to Long-Term Therapy

All patients should be monitored for bleeding (especially gastrointestinal and intracranial) and for signs of recurrent PE or hemodynamic deterioration. For patients receiving anticoagulation, check hemoglobin, platelet count, and renal function at baseline and periodically. The decision to transition to long-term anticoagulation (beyond 3 months) is guided by the presence of provoking factors, cancer, and risk of recurrence (see Section 8). In patients with active cancer who have completed ≥6 months of anticoagulation, extended therapy with reduced-dose apixaban 2.5 mg twice daily is noninferior to full-dose (5 mg twice daily) for preventing recurrent VTE (cumulative incidence 2.1% vs. 2.8%; adjusted subhazard ratio 0.76, 95% CI 0.41-1.41; P=0.001 for noninferiority) and reduces clinically relevant bleeding (12.1% vs. 15.6%; adjusted subhazard ratio 0.75, 95% CI 0.58-0.97; P=0.03) ^[50]A1b. NNT = 28 to prevent one clinically relevant bleed.

Treatment Failure Protocol

If a patient deteriorates (new hypotension, worsening hypoxia, or cardiopulmonary arrest) despite adequate anticoagulation:

• High-risk: Consider immediate systemic fibrinolysis (tenecteplase bolus) if not already given; if contraindicated, proceed to catheter-directed therapy or surgical embolectomy.
• Intermediate-risk: If cardiorespiratory distress develops, escalate to UF-CDT or, if not available, systemic fibrinolysis (with careful bleeding risk assessment).
• Suspected recurrence: Reimage (CT pulmonary angiography ) and switch anticoagulant class (e.g., from apixaban to edoxaban, or to LMWH).

What NOT to Do

• Do not use routine oxygen in normoxemic intermediate-risk patients; it does not improve RV function and may mask clinical deterioration ^[90]A1b.
• Do not give systemic fibrinolysis to all intermediate-risk patients; the bleeding risk (6.3% major bleeding, 2.4% stroke) outweighs the benefit in most ^[93]A1b.
• Do not discharge a patient with intermediate-risk PE to home without a clear plan for close follow-up and anticoagulation adherence.

Figure 1: Acute management algorithm for pulmonary embolism, stratified by risk.

Controversies and Guideline Disagreement

Pearl: For acute PE, choose a DOAC (apixaban or rivaroxaban) for most patients; apixaban causes less bleeding (NNT=26 vs. rivaroxaban ^[92]A1b); reserve fibrinolysis for high-risk or selected intermediate-risk patients, and consider catheter-directed therapy for those with cardiorespiratory distress (NNT=16 ^[52]A1b).

After initial stabilization, the focus shifts to preventing recurrent thromboembolism and managing the chronic sequelae of pulmonary embolism. The CHEST 2016 guideline and trial evidence provide a structured framework for anticoagulant duration, agent selection, and follow-up.

Step 1: Risk Stratification for Recurrence

Before deciding on anticoagulation duration, stratify the patient’s risk of recurrent VTE. The PADIS-PE trial identified independent predictors of recurrence after a first unprovoked PE: age >65 years (HR 4.70, 95% CI 1.78-12.40), pulmonary vascular obstruction index (PVOI) ≥5% at 6 months (HR 2.06, 95% CI 1.14-3.72), and antiphospholipid antibodies (HR 2.38, 95% CI 1.15-4.89) ^[22]A1b. PVOI ≥40% at diagnosis also conferred a twofold increased risk. These factors guide the decision for extended therapy.

Step 2: Anticoagulation Duration and Agent Selection

For acute VTE (first 3-6 months):

• Non-cancer patients: The CHEST 2016 guideline suggests dabigatran (Grade 2B), rivaroxaban (Grade 2B), apixaban (Grade 2B), or edoxaban (Grade 2B) over vitamin K antagonist (VKA) therapy, and suggests VKA over low-molecular-weight heparin (LMWH) (Grade 2C) ^[4]A1c.
• Cancer patients: The guideline suggests LMWH over VKA (Grade 2B), and also suggests dabigatran, rivaroxaban, apixaban, or edoxaban (each Grade 2C) ^[4]A1c. The Hokusai VTE Cancer study confirmed that clinical outcomes are similar for incidental and symptomatic VTE, supporting the same treatment approach ^[87]B2b.

Choice of DOAC - apixaban vs rivaroxaban: The COBRRA trial randomized 2760 patients with acute VTE to apixaban (10 mg BID ×7d, then 5 mg BID) or rivaroxaban (15 mg BID ×21d, then 20 mg daily) for 3 months. The primary outcome (clinically relevant bleeding) occurred in 3.3% with apixaban vs 7.1% with rivaroxaban (RR 0.46, 95% CI 0.33-0.65; P<0.001); NNT = 27 to prevent one clinically relevant bleed ^[92]A1b. Death from any cause was 0.1% vs 0.3% (RR 0.25, 95% CI 0.03-2.26). This suggests apixaban has a lower bleeding risk in the acute treatment phase.

Extended therapy (beyond 3-6 months): For patients with unprovoked PE and high recurrence risk (e.g., age >65, PVOI ≥5%, antiphospholipid antibodies), extended anticoagulation is recommended. The CHEST guideline suggests extended therapy without a scheduled stop date for unprovoked VTE ^[4]A1c.

Extended therapy in cancer patients: The API-CAT trial tested reduced-dose apixaban (2.5 mg BID) vs full-dose (5 mg BID) for 12 months in patients with active cancer who had completed at least 6 months of anticoagulation. Recurrent VTE occurred in 2.1% vs 2.8% (adjusted sub-HR 0.76, 95% CI 0.41-1.41; P=0.001 for noninferiority). Clinically relevant bleeding was lower with reduced dose: 12.1% vs 15.6% (adjusted sub-HR 0.75, 95% CI 0.58-0.97; P=0.03); NNT = 28 to prevent one clinically relevant bleed ^[50]A1b. Thus, reduced-dose apixaban is an option for extended secondary prevention in cancer-associated VTE.

Dosing Table

Doses are from the CHEST 2016 guideline and pivotal trials; refer to full prescribing information for complete dose modifications.

Step 3: Monitoring and Follow-Up

After discharge, patients should be followed at 1, 3, and 6 months, then annually while on anticoagulation. Assess for bleeding, recurrent symptoms, and adherence. The Hestia criteria can identify low-risk patients suitable for outpatient management: in the trial, the primary endpoint (30-day adverse outcome) occurred in 0% (95% CI 0-1.3%) of patients managed with Hestia criteria alone ^[40]A1b. A prognostic assessment pathway reduced hospital stay from 6.1 to 4.0 days (P<0.001) without increasing 30-day mortality (2.4% vs 2.0%) ^[88]A1b.

Step 4: Management of Chronic Sequelae

Persistent dyspnea and functional limitation: A randomized trial of exercise-based rehabilitation (8 weeks, 2 sessions/week) in patients with persistent dyspnea after PE (6-72 months prior) showed a mean improvement in the Incremental Shuttle Walk Test of 53.0 m (95% CI 17.7-88.3) vs usual care, and a small improvement in PE-specific quality of life (mean difference -4%, 95% CI -0.09 to 0.00) ^[57]A1b. No adverse events occurred. Rehabilitation should be considered in patients with persistent dyspnea.

Chronic thromboembolic pulmonary hypertension (CTEPH): Patients with persistent dyspnea, exercise intolerance, or signs of right heart failure 3-6 months after PE should undergo echocardiography and ventilation-perfusion scanning. If CTEPH is confirmed, surgical pulmonary endarterectomy is the treatment of choice, with medical therapy (riociguat, bosentan) for inoperable cases. This is beyond the scope of this section but is a key long-term consideration.

Step 5: Special Populations

Cancer-associated VTE: As above, DOACs (edoxaban, rivaroxaban, apixaban) or LMWH are options. The Planquette trial comparing rivaroxaban vs dalteparin in 158 patients with cancer-associated VTE found a primary outcome (recurrent VTE at 3 months) of 6.4% vs 10.1% (SHR 0.75, 95% CI 0.21-2.66), but the trial was underpowered for noninferiority ^[91]A1b. An updated meta-analysis within that paper supports DOACs as effective alternatives.

Antiphospholipid syndrome: The CHEST 2016 guideline suggests VKA therapy (INR 2-3) rather than DOACs for VTE in antiphospholipid antibody-positive patients, as DOACs have shown higher recurrence rates in this setting. The PADIS-PE trial identified antiphospholipid antibodies as a strong predictor of recurrence (HR 2.38) ^[22]A1b.

Pregnancy: Anticoagulation with LMWH is preferred; VKAs and DOACs are contraindicated.

Controversies and Guideline Disagreement

No major guideline disagreements identified for the core anticoagulation duration recommendations beyond the above.

Pearl: For long-term management after PE, risk-stratify using age, pulmonary vascular obstruction index, and antiphospholipid antibodies (PADIS-PE) to guide extended anticoagulation; apixaban appears to have a lower bleeding risk than rivaroxaban in the acute phase (COBRRA), and reduced-dose apixaban is a noninferior option for extended therapy in cancer-associated VTE (API-CAT).

Building on the principles of long-term anticoagulation, the treatment of pulmonary embolism has undergone a dramatic transformation over the past three decades, shifting from a one-size-fits-all approach of parenteral heparin followed by vitamin K antagonists (VKAs) to a risk-stratified paradigm dominated by direct oral anticoagulants (DOACs) and outpatient management.

The Era of Heparin and Vitamin K Antagonists

For decades, the standard of care for acute PE was initial parenteral anticoagulation with unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH), overlapped with a VKA such as warfarin . The 9th edition of the CHEST guidelines (2012) codified this approach, recommending initial parenteral anticoagulant therapy (Grade 1B) and suggesting LMWH or fondaparinux over IV UFH (Grade 2C) ^[5]A1c. Thrombolytic therapy was reserved for PE with hypotension (Grade 2C) ^[5]A1c. For cancer-associated VTE, LMWH was preferred over VKAs (Grade 2B) ^[5]A1c. However, the guidelines still suggested VKAs or LMWH over dabigatran or rivaroxaban (Grade 2B) ^[5]A1c, reflecting the nascent evidence for DOACs.

The DOAC Revolution

The 2016 CHEST update marked a watershed: for VTE and no cancer, the panel suggested dabigatran, rivaroxaban, apixaban , or edoxaban over VKA therapy (all Grade 2B) ^[4]A1c. For cancer-associated VTE, LMWH remained the preferred agent, but DOACs were listed as alternatives (Grade 2C) ^[4]A1c. The 2021 second update further solidified DOACs, generating 29 guidance statements, 13 of which were strong recommendations ^[3]A1c[7]A1c. The COBRRA trial (2026) provided head-to-head bleeding data: clinically relevant bleeding occurred in 3.3% of patients on apixaban versus 7.1% on rivaroxaban (RR 0.46, 95% CI 0.33-0.65; P<0.001; NNT = 26 to prevent one bleeding event) ^[92]A1b. This trial has refined DOAC selection, favoring apixaban for its lower bleeding risk.

Risk Stratification and Outpatient Management

Parallel to pharmacologic advances, risk stratification enabled safe outpatient treatment. The Hestia criteria, validated in a 2016 randomized trial, selected patients for direct discharge; the primary endpoint occurred in 0% (95% CI 0-1.3%) of those with NT-proBNP testing vs 1.1% (95% CI 0.2-3.2%) in the direct discharge group (P=0.25) ^[40]A1b. A 2022 trial showed that a prognostic assessment and management pathway reduced median length of stay from 6.1 days to 4.0 days (P<0.001) without increasing 30-day mortality or readmission ^[88]A1b. These data support the widespread adoption of risk-stratified outpatient management.

Special Populations: Cancer and Pregnancy

For cancer-associated VTE, the Hokusai VTE Cancer study (2020) confirmed that patients with incidental VTE have similar adverse outcomes to those with symptomatic VTE, primary outcome 12.7% vs 13.8%, supporting equal treatment ^[87]B2b. In pregnancy, the ATS/STR guideline (2011) recommended chest radiography as the first radiation-associated procedure, lung scintigraphy if the CXR is normal, and CTPA over digital subtraction angiography for nondiagnostic V/Q results ^[1]A1c. For heparin-induced thrombocytopenia (HIT), argatroban, lepirudin, or danaparoid are suggested (Grade 2C) ^[8]A1c.

Recent Evolution and Future Directions

The 2021 CHEST update added four new guidance statements and substantially modified eight from the 2016 update, reflecting ongoing uncertainty in limited disease and special populations ^[3]A1c[7]A1c. The PADIS-PE trial (2018) identified age >65 years (HR 4.70), pulmonary vascular obstruction index ≥5% at 6 months (HR 2.06), and antiphospholipid antibodies (HR 2.38) as independent predictors of recurrence after unprovoked PE ^[22]A1b. Rehabilitation has emerged as a treatment for persistent dyspnea: a 2023 trial showed a 53.0 m improvement in Incremental Shuttle Walk Test ^[57]A1b. Supplemental oxygen in nonhypoxemic intermediate-risk PE did not significantly increase RV normalization (42.4% vs 21.6%) in a pilot trial ^[90]A1b.

Pearl: The 2016 CHEST guideline's recommendation of DOACs over VKAs for non-cancer VTE marked a paradigm shift; the 2026 COBRRA trial further refined DOAC selection by demonstrating apixaban's lower bleeding risk (NNT = 26), reinforcing the need to individualize anticoagulant choice based on bleeding risk profile.

The evolution of endovascular and surgical techniques has expanded the procedural options for pulmonary embolism beyond anticoagulation and thrombolysis, creating a distinct interventional pulmonology toolkit that includes mechanical thrombectomy, bronchoscopic cryoextraction, bronchial artery embolization, and pulmonary endarterectomy. These procedures are particularly relevant in lung transplant recipients, who carry a high thromboembolic burden.

Mechanical Thrombectomy in Acute Massive PE

In lung transplant recipients with acute massive PE, endovascular mechanical thrombectomy can be life-saving. A case report describes a patient with a single left lung transplant and atrial septal defect who underwent successful endovascular mechanical thrombectomy with immediate improvement in oxygen saturation and hemodynamic status despite persistent instability after systemic thrombolysis; the procedure was performed without adverse outcomes or paradoxical embolization ^[118]C4. This supports the use of mechanical thrombectomy even in anatomically complex post-transplant patients.

Bronchoscopic Interventions for Airway Obstruction

Cryoextraction via flexible bronchoscopy can restore airway patency when PE-related thrombi cause tracheobronchial obstruction. In a pediatric series of 11 critically ill children (including those with pulmonary embolism), cryoextraction resulted in partial (n = 5) or complete (n = 6) restoration of airway patency with no procedure-related complications; 5 patients were decannulated from ECMO ^[110]C4. The technique is effective for blood clots, mucus plugs, and casts ^[110]C4.

Bronchial Artery Embolization for Hemoptysis

Hemoptysis from pulmonary infarction or PE can be managed with bronchial artery embolization (BAE). The Chinese expert consensus identifies BAE as "a pivotal interventional technique for managing hemoptysis due to its minimally invasive nature and robust hemostatic efficacy" ^[111]D5. Preoperative bronchial artery CT angiography is recommended to identify culprit vessels, and super-selective embolization using coaxial microcatheters is advised to avoid non-target embolization ^[111]D5.

ECMO and Invasive Procedures

Patients requiring extracorporeal membrane oxygenation (ECMO) for massive PE often need concurrent invasive procedures. A systematic review of 46 studies found that airway and bronchoscopic procedures generally showed high procedural success, though registry-level data reported hemorrhagic complications in 26.0% and surgical-site bleeding in 13.0% ^[105]B2a. Lung transplantation studies on ECMO reported that hemothorax , transfusion, and thromboembolism remain central safety issues ^[105]B2a.

Lung Transplantation and PE

Thromboembolic complications develop in 8.6% of lung transplant recipients (4-24 months post-transplant), often associated with elevated fibrinogen, factor VIII/IX/XI, and antiphospholipid antibodies ^[99]B3b. DVT after lung transplantation is common (incidence 53.6%) and linked to higher postoperative mortality (33.17% vs. 12.9%) and a fourfold increase in PE (22.4% vs. 5.4%) ^[113]B3b. Lower extremity DVT (OR 2.14) and elevated preoperative platelet count predict PE, while lung transplantation for pulmonary artery hypertension is protective (OR 0.36) ^[113]B3b.

IVC filters

Retrievable inferior vena cava filters are placed in lung transplant recipients for VTE with high success. In a cohort of 95 patients, 63 filters (66.3%) were retrieved after a median dwell time of 7.7 months; only one complication (filter migration, successfully removed) occurred ^[112]C4.

Donor lungs with PE

Lungs from donors with acute bilateral PE and chronic thromboembolic disease can be salvaged by backtable pulmonary thromboendarterectomy. Two recipients of such grafts had graft dysfunction scores of 0 at 48 and 72 hours and remained free of rejection at follow-up, demonstrating utility in expanding the donor pool ^[117]C4.

Biomarker surveillance

Serum Aspergillus galactomannan antigen (AGA) positivity, detected in 24% of lung transplant recipients, is associated with a higher rate of pulmonary embolism (23.9% vs. 12.6%) and may serve as an early biomarker for thrombotic risk ^[114]B3b.

Pulmonary Endarterectomy for CTEPH

Pulmonary endarterectomy (PEA) is the definitive surgical treatment for chronic thromboembolic pulmonary hypertension. Immediate postoperative pulmonary vascular resistance (PVR) is the only predictor of long-term survival/freedom from lung transplantation ^[104]B2b. Patients with an immediate postoperative PVR <590 dynes·s·cm⁻⁵ have significantly better long-term outcomes than those with PVR ≥590 dynes·s·cm⁻⁵ ^[104]B2b.

Pearl: In lung transplant recipients, a high index of suspicion for PE is warranted given the 8.6% incidence and associated hypercoagulable abnormalities; mechanical thrombectomy and IVC filters can be safely employed in this population, and donor lungs with PE can be used after backtable thromboendarterectomy.

The preceding section detailed advanced respiratory support and interventional options; even with optimal management, PE carries a substantial burden of acute and chronic sequelae that require structured surveillance and supportive care.

Acute Complications

Right ventricular failure and haemodynamic collapse remain the most immediate life-threatening complications. The Bova score identifies intermediate-risk patients using systolic blood pressure 90-100 mmHg (aOR 2.45), heart rate ≥110 bpm (aOR 1.87), elevated cardiac troponin (aOR 2.49), and right ventricular dysfunction (aOR 2.28); the 30-day PE-related complication rate rises from 4.2% (stage I) to 29.2% (stage III) ^[58]B2b. Right ventricular adaptation to pressure overload determines short- and long-term outcomes, yet validated RV-targeted therapies remain an unmet need ^[2]D5.

Bleeding complications are the most common treatment-related adverse events. In selected cancer-associated PE patients managed as outpatients, the 30-day major bleeding rate is 2.71% (95% CI 1.51-4.83%) ^[39]A1a. In medically ill patients, extended apixaban 2.5 mg twice daily for 30 days caused more major bleeding than short-course enoxaparin (0.47% vs 0.19%; RR 2.58, 95% CI 1.02-7.24) ^[124]A1b. Among patients with active cancer completing ≥6 months of anticoagulation, extended reduced-dose apixaban (2.5 mg twice daily) reduced clinically relevant bleeding compared with full-dose apixaban (5.0 mg twice daily) (12.1% vs 15.6%; adjusted subhazard ratio 0.75, 95% CI 0.58-0.97) while maintaining noninferior efficacy for recurrent VTE ^[50]A1b.

Recurrent venous thromboembolism occurs in 1.26% (95% CI 0.53-3.00%) within 30 days in cancer-PE outpatients ^[39]A1a and in 2.1% over 12 months with reduced-dose apixaban versus 2.8% with full dose (adjusted subhazard ratio 0.76, 95% CI 0.41-1.41) ^[50]A1b.

Heparin -induced thrombocytopenia (HIT) is infrequent but consequential. In a meta-analysis of 13 studies (5275 patients), thrombocytopenia rates did not differ between low-molecular-weight heparin (LMWH) and unfractionated heparin (UH) (1.2% vs 1.5%; p = 0.246), and confirmed HIT or HITT was too rare for adequate comparison ^[119]A1a. For patients receiving heparin in whom the risk of HIT is >1%, guideline recommends platelet count monitoring every 2-3 days from day 4 to day 14 (Grade 2C) ^[8]A1c.

Chronic Complications

Chronic thromboembolic pulmonary hypertension (CTEPH) is the most feared long-term complication. A meta-analysis of 16 studies (4047 patients) reported a pooled incidence of 0.56% (95% CI 0.1-1.0%) in all-comers, 3.2% (95% CI 2.0-4.4%) in survivors, and 2.8% (95% CI 1.5-4.1%) in survivors without major comorbidities. Both unprovoked PE (OR 4.1, 95% CI 2.1-8.2) and recurrent VTE (OR 3.2, 95% CI 1.7-5.9) were significantly associated with CTEPH ^[47]A1a.

Residual pulmonary vascular obstruction (RPVO) occurs in 28.4% and 25.3% of patients in two prospective cohorts, and fibrinogen Bβ-chain monosialylation independently predicts its development ^[125]B2b. Among symptomatic patients with chronic thrombi and absent or mild resting pulmonary hypertension, 40.2% develop exercise pulmonary hypertension on right heart catheterisation, identified by a multiparametric score incorporating peak tricuspid regurgitation velocity and TAPSE/PASP ratio ^[126]B2b.

Supportive Care and Prevention of Hospital-Acquired Complications

Respiratory monitoring is essential: acute respiratory failure occurs more frequently in PE patients with COVID-19 than with seasonal influenza ^[12]B3b; pulse oximetry , arterial blood gases, and clinical assessment guide escalation to noninvasive or invasive ventilation. Intraoperative management of PE patients requiring surgery demands caution: in the PROTHOR trial, a higher PEEP strategy (10 cm H₂O) during one-lung ventilation increased intraoperative hypotension (37.3% vs 14.3%) and new arrhythmias (9.9% vs 3.9%) compared with lower PEEP (5 cm H₂O) ^[24]A1b.

Autonomic complications such as haemodynamic instability and arrhythmias are common. The same trial reported that hypotension and arrhythmias were the most frequent intraoperative complications with aggressive lung expansion ^[24]A1b. Ileus and urinary retention are not specifically reported in the reviewed evidence but are recognised in critically ill patients.

Pain management of pleuritic chest pain typically requires nonsteroidal anti-inflammatory drugs or opiates; no specific dosing data from the reviewed sources.

Rehabilitation should begin after hospital discharge. A structured follow-up pathway for patients with COVID-19 pneumonia, which includes assessment for breathlessness, oxygen requirement, and rehabilitation, can serve as a model for post-PE care ^[127]D5.

Hospital-acquired complications (pneumonia, pressure injury, urinary tract infection) are not quantified in the reviewed evidence but are general risks of prolonged hospitalisation. Prevention bundles include early mobilisation, oral care, and catheter avoidance.

Complication Frequency Table

Pearl: The 0.56% CTEPH incidence in all-comers after PE is lower than often quoted, but the 3.2% incidence in survivors highlights the need for structured follow-up in patients with persistent dyspnea, especially after unprovoked or recurrent PE ^[47]A1a.

Having reviewed the complications of pulmonary embolism, the prognosis now warrants attention, short-term mortality is highly variable, driven by hemodynamic status, right ventricular strain, and underlying comorbidities. The untreated trajectory of acute PE carries a historically high mortality, but with modern risk-stratified management, outcomes have improved substantially.

Short-term Mortality and Adverse Outcomes

In contemporary cohorts, 30-day all-cause mortality in normotensive patients with acute symptomatic PE is approximately 1.9% (95% CI 0.79-4.6%) in low-risk patients managed as outpatients ^[129]A1a, and 7.4% (95% CI 5.5-9.8%) in unselected normotensive patients ^[83]B2b. The 30-day complicated course (death, hemodynamic collapse, or recurrent PE) occurs in 7.4% of normotensive patients in the derivation cohort and 4.5% in the validation cohort ^[44]B2b. Among patients with intermediate-risk PE (right ventricular dysfunction plus myocardial injury), the placebo group in the PEITHO trial had a 5.6% rate of death or hemodynamic decompensation within 7 days ^[93]A1b. For high-risk PE requiring extracorporeal membrane oxygenation support, overall mortality reaches 42.8% with strategies other than mechanical reperfusion, versus 22.6% with mechanical reperfusion (surgical embolectomy in most) ^[49]A1a.

Risk Stratification Models

Several validated tools stratify normotensive patients into low-, intermediate-, and high-risk groups:

• Simplified Pulmonary Embolism Severity Index ( sPESI ): Combined with BNP testing, it yields a negative predictive value of 99.1% (derivation) and 100% (validation) for a 30-day complicated course ^[44]B2b.
• Bova score: A seven-point index incorporating SBP 90-100 mmHg (aOR 2.45), heart rate ≥110 bpm (aOR 1.87), elevated troponin (aOR 2.49), and right ventricular dysfunction (aOR 2.28). The 30-day PE-related complication rate is 4.2% for stage I, 10.8% for stage II, and 29.2% for stage III ^[58]B2b.
• Hestia criteria: Patients selected for outpatient treatment using Hestia criteria alone have a 30-day adverse outcome rate of 1.1% (95% CI 0.2-3.2%) ^[40]A1b.

Prognostic Biomarkers

Elevated BNP or NT-proBNP levels are strongly associated with adverse outcomes. In a meta-analysis of 13 studies, 51% of patients had elevated levels; high NT-proBNP conferred a 10% risk of 30-day mortality (68/671; 95% CI 8.0-13%) and a 23% risk of adverse clinical outcome (209/909; 95% CI 20-26%) ^[42]A1a. The odds ratio for 30-day mortality with elevated BNP/NT-proBNP was 7.6 (95% CI 3.4-17) ^[42]A1a. Cardiac troponin elevation independently predicts complications (aOR 2.49) ^[58]B2b.

Treatment Effect on Prognosis

Effective anticoagulation dramatically reduces the risk of fatal recurrent PE. Among patients with acute VTE, apixaban shows a lower risk of clinically relevant bleeding than rivaroxaban (3.3% vs 7.1%; RR 0.46; NNT = 27 to prevent one clinically relevant bleeding event over 3 months) ^[92]A1b. For intermediate-risk PE, fibrinolytic therapy (tenecteplase) reduces the risk of death or hemodynamic decompensation from 5.6% to 2.6% (OR 0.44; NNT = 34) but increases major bleeding from 1.2% to 6.3% (NNH = 20) and stroke from 0.2% to 2.4% (NNH = 46) ^[93]A1b. Ultrasound-facilitated catheter-directed fibrinolysis reduces the composite of PE-related death, cardiorespiratory decompensation, or symptomatic recurrence within 7 days from 10.3% to 4.0% (RR 0.39; NNT = 16) ^[52]A1b. In patients with cancer-associated PE, outpatient management yields a 30-day mortality of 1.74% (95% CI 0.99-3.03%) and major bleeding of 2.71% ^[39]A1a. Extended reduced-dose apixaban (2.5 mg twice daily) is noninferior to full-dose for preventing recurrent VTE (2.1% vs 2.8%; adjusted subhazard ratio 0.76) and reduces clinically relevant bleeding (12.1% vs 15.6%; NNT = 29) ^[50]A1b.

Long-term Outcomes

Late complications include chronic thromboembolic pulmonary hypertension (CTEPH) and post-PE functional impairment. In the API-CAT trial, mortality at approximately 12 months was 17.7% in the reduced-dose apixaban group and 19.6% in the full-dose group among patients with active cancer ^[50]A1b. The risk of long-term adverse outcomes is influenced by the adequacy of initial anticoagulation, presence of residual thrombus, and underlying cardiopulmonary reserve.

Pearl: The 30-day mortality in normotensive PE is low (≈2-4%) with appropriate risk stratification, but intermediate-risk patients (Bova stage III) have a 29.2% complication rate, these patients may benefit from escalation of therapy beyond anticoagulation alone ^[58]B2b.

Controversies and Guideline Disagreement

Having outlined the natural history across unselected cohorts, the following sections address how pregnancy, comorbidities, and age modify the presentation, diagnosis, and management of pulmonary embolism.

Pregnancy

PE is a leading cause of maternal death in the developed world ^[1]A1c. The pregnancy-adapted YEARS algorithm safely reduces imaging: pulmonary embolism is ruled out if none of three criteria (clinical signs of deep-vein thrombosis, hemoptysis, PE as most likely diagnosis) are present and the D‑dimer level is <1000 ng/mL, or if ≥1 criterion is met and the D‑dimer level is <500 ng/mL ^[30]B2b. In a prospective study of 510 women, computed tomographic pulmonary angiography (CTPA) was avoided in 39% (95% CI 35-44), with the highest efficiency in the first trimester (65% avoided) and lowest in the third (32%) ^[30]B2b. No patient had PE during follow-up; one popliteal deep‑vein thrombosis occurred (0.21%; 95% CI 0.04-1.2) ^[30]B2b.

When imaging is required, the American Thoracic Society recommends chest radiography as the first radiation‑associated procedure; if the result is normal, lung scintigraphy (V/Q) is preferred over CTPA because of lower maternal‑fetal radiation exposure ^[1]A1c. Low‑dose perfusion scintigraphy (LDQ) has a negative predictive value of 100% compared with 97.5% for CTPA, with fewer nondiagnostic results ^[133]B2b. If V/Q is nondiagnostic, CTPA is recommended over digital subtraction angiography ^[1]A1c.

Cesarean section carries a fourfold higher risk of VTE than vaginal delivery (OR 3.7; 95% CI 3.0-4.6), with an absolute incidence of 2.6 per 1000 cesarean sections (95% CI 1.7-3.5) that rises to 4.3 per 1000 with longer follow‑up ^[130]A1a. Emergency cesarean section confers greater risk than elective ^[130]A1a. Women with sickle cell disease are at particularly high risk: pulmonary thromboembolism risk is 7.7‑fold higher (RR 7.74; 95% CI 4.65-12.89) than in women without SCD, and the prevalence of acute chest syndrome is 6.46% (95% CI 4.66%-8.25%) ^[128]A1a. Prophylactic blood transfusion in SCD pregnancies reduces pain crisis (RR 0.28; 95% CI 0.12-0.67) but does not clearly alter the risk of PE or acute chest syndrome ^[142]A1a.

For treatment, low‑molecular‑weight heparin (LMWH) is preferred over unfractionated heparin (UFH) during pregnancy. LMWH is associated with fewer adverse effects sufficient to stop treatment (RR 0.07; 95% CI 0.01-0.54) and fewer fetal losses (RR 0.47; 95% CI 0.23-0.95) compared with UFH ^[143]A1a. However, the overall certainty of evidence for thromboprophylaxis in pregnancy is very low ^{[136]A1a[143]A1a}. Direct oral anticoagulants are not recommended during lactation; rivaroxaban passes into breast milk in small amounts, and its safety is undetermined ^[134]C4.

Pediatrics

The retrieved evidence does not provide pediatric‑specific data. Clinical practice therefore extrapolates from adult guidelines, using weight‑based dosing of anticoagulants and accounting for developmental hemostasis. The need for dedicated pediatric trials remains.

Elderly

Age alters the performance of risk‑stratification biomarkers. The established high‑sensitivity troponin T (hsTnT) cut‑off of 14 pg/mL retains its prognostic value in younger patients but fails to predict adverse outcome in those aged ≥75 years ^[135]B2b. An age‑optimised cut‑off of 45 pg/mL is superior in this group, and an age‑adjusted strategy (≥14 pg/mL for age <75 years, ≥45 pg/mL for age ≥75 years) provides additive independent prognostic information beyond the simplified Pulmonary Embolism Severity Index and echocardiography (OR 4.56; 95% CI 1.30-16.01; C‑index 0.77) ^[135]B2b. A three‑step approach using sPESI , age‑adjusted hsTnT, and echocardiography identifies 16.6% of normotensive patients as higher risk, of whom 12.4% have an adverse 30‑day outcome ^[135]B2b.

Immunocompromised (Cancer)

Perioperative thromboprophylaxis in patients with cancer is the best‑studied scenario. Meta‑analyses comparing LMWH with UFH show no conclusive difference in mortality, pulmonary embolism, major bleeding, or minor bleeding ^{[138]A1a[140]A1a[141]A1a}. LMWH is associated with a lower incidence of wound hematoma (RR 0.70; 95% CI 0.54-0.92) ^[138]A1a. For acutely ill medical patients (a population that includes many immunocompromised individuals), extended‑duration anticoagulation reduces symptomatic VTE (RR 0.60; 95% CI 0.46-0.78; NNTB 204) but increases major bleeding (RR 2.05; 95% CI 1.51-2.79; NNTH 314), with little effect on all‑cause mortality ^[137]A1a.

Pearl: In pregnancy, the YEARS algorithm with a D‑dimer threshold of 1000 ng/mL (when no YEARS criteria are present) safely avoids CTPA in up to two‑thirds of women in the first trimester; in patients aged ≥75 years, use a hsTnT cut‑off of 45 pg/mL rather than 14 pg/mL for risk stratification.

Prevention of venous thromboembolism is a three-tiered strategy: primary prophylaxis in at-risk populations, secondary prevention of recurrence after an index event, and surveillance of patients who have recovered.

Primary Prevention

Thromboprophylaxis after major orthopedic surgery is the best-studied primary prevention strategy. The American College of Chest Physicians (CHEST) 2012 guidelines recommend low-molecular-weight heparin (LMWH), fondaparinux, dabigatran, apixaban, rivaroxaban, low-dose unfractionated heparin, adjusted-dose vitamin K antagonist, or aspirin (all Grade 1B) or an intermittent pneumatic compression device (Grade 1C) for a minimum of 10 to 14 days, with extension up to 35 days (Grade 2B) ^[6]A1c. They recommend against Doppler screening before discharge (Grade 1B) ^[6]A1c.

In trauma patients, VTE is common despite early chemoprophylaxis. Among 133 patients with hemodynamically unstable pelvic fractures, 24.1% developed VTE (12.8% pulmonary embolism alone) despite a median chemoprophylaxis start on hospital day 1 ^[159]B2b. After degenerative spine surgery, 61% of all VTE events occurred within the first four postoperative months, peaking in month 2 ^[157]B2b.

Statins have been investigated for primary prevention of VTE. A 2024 Cochrane meta-analysis of 27 randomized trials (122,601 participants) found that statins may slightly reduce VTE incidence (OR 0.86, 95% CI 0.76-0.98; low-certainty evidence), but no significant reduction in pulmonary embolism (OR 0.83, 95% CI 0.46-1.52) ^[149]A1a. Only rosuvastatin showed a signal, but the effect was very small and the evidence limited ^{[148]A1a[149]A1a}.

Drug-induced thromboembolic risk warrants awareness. Olanzapine use significantly increased the risk of VTE and pulmonary embolism (OR 2.07, 95% CI 1.37-3.14) ^[156]A1a.

Vaccine considerations: The BNT162b2 mRNA vaccine is not associated with elevated risk of most adverse events, while SARS-CoV-2 infection itself substantially increases the risk of pulmonary embolism ^[155]B2b. Rare cases of vaccine-induced thrombotic thrombocytopenia with pulmonary embolism have been reported 5 to 16 days after ChAdOx1 nCoV-19 vaccination ^[151]C4. The benefit-risk balance strongly favors vaccination given the higher thrombotic risk from COVID-19 itself.

Secondary Prevention (Preventing Recurrence)

CHEST 2016 guidelines provide a framework. For VTE without cancer, suggest dabigatran , rivaroxaban , apixaban , or edoxaban over vitamin K antagonist (VKA) therapy (Grade 2B), and VKA over LMWH (Grade 2C) ^[4]A1c. For VTE with cancer, suggest LMWH over VKA (Grade 2B) ^[4]A1c. Anticoagulation should be continued for at least 3 months, with extended therapy considered based on risk-benefit ^[4]A1c.

If recurrence occurs on a non-LMWH anticoagulant, switch to LMWH (Grade 2C); if on LMWH, increase the dose (Grade 2C) ^[4]A1c. For subsegmental pulmonary embolism without proximal DVT, clinical surveillance is suggested over anticoagulation when the risk of recurrence is low, and anticoagulation when risk is high (Grade 2C) ^[4]A1c. The guidelines recommend against inferior vena cava filter placement (Grade 1B) ^[4]A1c.

Screening and Surveillance

Surveillance for chronic thromboembolic pulmonary hypertension (CTEPH) after COVID-19-associated PE is of low yield. A UK national study found that 1.2% of confirmed CTEPH cases were potentially associated with COVID-19 PE; simple risk scoring using symptoms, ECG, and NT-proBNP is a potentially effective screening approach ^[145]B2b.

In patients with permanent IVC filters who are anticoagulated, annual ultrasound surveillance of the filter and lower extremity veins is reasonable; filter clot was detected in 30% of patients over follow-up, and symptomatic PE occurred in 5% ^[146]B2b.

Patient Education

Patients should be counseled to recognize symptoms of DVT (unilateral leg swelling, pain, warmth) and PE (sudden dyspnea, pleuritic chest pain, hemoptysis) and to seek urgent medical attention. Adherence to prescribed anticoagulation, avoidance of missed doses, and awareness of bleeding symptoms are essential. For those on extended prophylaxis, instruction on proper injection technique (for LMWH) and the importance of completing the prescribed duration should be reinforced.

Pearl: The most impactful prevention strategy is appropriate thromboprophylaxis in hospitalized surgical and medical patients, guided by validated risk scores; the CHEST guidelines recommend against routine Doppler screening before discharge after orthopedic surgery, emphasizing that pharmacologic prophylaxis, not surveillance, is the cornerstone of prevention.