A pulmonary embolism (PE) is a blockage in one of the pulmonary arteries in the lungs, most often caused by a blood clot that originates in the deep veins of the legs—a condition known as deep vein thrombosis (DVT)—and travels through the bloodstream to lodge in the lung vasculature, obstructing blood flow and potentially leading to life-threatening complications such as respiratory failure or right heart strain.[1][2] This acute form of venous thromboembolism (VTE) affects 39 to 115 individuals per 100,000 population annually in the United States, with an estimated 100,000 deaths attributed to PE each year, though prompt diagnosis and treatment can significantly improve outcomes.[3][4]Symptoms of PE typically arise suddenly and include shortness of breath, sharp chest pain that worsens with deep breathing or coughing, rapid heartbeat, coughing up blood (hemoptysis), dizziness, and leg pain or swelling if associated with DVT; however, smaller emboli may cause subtler signs like anxiety or low-grade fever, while massive PE can lead to fainting, low blood pressure, or shock.[1][2] The condition's severity varies based on the size and location of the clot, as well as the patient's underlying health, with untreated cases carrying a mortality rate of about one-third due to risks like pulmonary infarction or chronic thromboembolic pulmonary hypertension.[1][4]Key risk factors for developing PE include prolonged immobility (such as during long flights, bed rest, or post-surgery), major surgery (especially orthopedic procedures), trauma, cancer, heart failure, obesity, smoking, pregnancy, oral contraceptive or hormone replacement therapy use, and inherited clotting disorders like factor V Leiden.[1][2] Additional triggers encompass vein damage from injury or inflammation from infections, with severe COVID-19 identified as a notable recent risk amplifier due to hypercoagulability.[1][3]Diagnosis of PE relies on clinical suspicion combined with tests such as D-dimer blood assays to detect clot breakdown products, followed by imaging like computed tomography pulmonary angiography (CTPA)—the gold standard for confirmation—or ventilation-perfusion (V/Q) scans in cases where contrast is contraindicated.[5][2] Treatment primarily involves anticoagulation therapy with medications like heparin or direct oral anticoagulants to prevent clot extension, alongside thrombolytics for massive PE; in select high-risk scenarios, procedures such as catheter-directed clot removal or inferior vena cava filter placement may be employed to avert recurrence.[5][2] Prevention strategies emphasize early mobilization, prophylactic anticoagulants in at-risk patients, and lifestyle modifications like weight management and smoking cessation, which can substantially reduce incidence in vulnerable populations.[2][3]
Pathophysiology
Embolus formation and migration
Pulmonary embolism primarily arises from deep vein thrombosis (DVT), where a blood clot forms in the deep veins, most commonly in the lower extremities or pelvic region.[4] The initial thrombus formation in these veins is governed by Virchow's triad, encompassing venous stasis, endothelial injury, and hypercoagulability, which collectively promote the aggregation of platelets and fibrin.[6]Venous stasis reduces blood flow, allowing procoagulant factors to accumulate; endothelial injury exposes subendothelial collagen and tissue factor, triggering the coagulation cascade; and hypercoagulability shifts the hemostatic balance toward clot formation through various molecular pathways, such as increased thrombin generation.[7]Once formed, portions of the DVT can dislodge, becoming free-floating emboli that migrate through the venous system.[8] These emboli typically travel via the femoral or iliac veins into the inferior vena cava, entering the right atrium and right ventricle before lodging in the pulmonary arteries.[1] The migration path may involve single large thrombi or multiple smaller fragments, with the latter often originating from progressive fragmentation of the primary clot in the deep veins.[4]Emboli can vary in size and distribution, leading to distinct patterns such as multiple small peripheral emboli, which obstruct smaller branches of the pulmonary vasculature, or a saddle embolus, where a large thrombus straddles the bifurcation of the main pulmonary artery, potentially extending into both left and right main pulmonary arteries.[9] This classification by size and location contributes to the hemodynamic impact, with emboli assessed based on their effect on right ventricular function and systemic blood pressure.[10]Pulmonary emboli are stratified into massive (high-risk), submassive (intermediate-risk), and low-risk categories according to hemodynamic consequences.[4] Massive emboli cause sustained hypotension or shock, often from extensive obstruction like a saddle embolus; submassive emboli occur in hemodynamically stable patients with evidence of right ventricular dysfunction; and low-risk emboli involve minimal hemodynamic perturbation, typically from smaller or peripheral clots.[10]
Pulmonary vascular and cardiac effects
When an embolus lodges in the pulmonary arteries, it mechanically obstructs blood flow, leading to an acute increase in pulmonary vascular resistance (PVR). This obstruction, combined with hypoxic vasoconstriction in unaffected lung regions, elevates pulmonary artery pressure and imposes a sudden rise in right ventricular (RV) afterload.[4] The RV, which is thin-walled and accustomed to low-pressure circulation, faces impaired ejection as a result, potentially reducing cardiac output if the occlusion affects more than 30-50% of the pulmonary vascular bed.[4]The obstruction disrupts the normal matching of ventilation and perfusion (V/Q mismatch), creating regions of high ventilation-to-perfusion ratios (dead space) where alveoli are ventilated but not perfused, while blood is redirected to non-obstructed areas. This imbalance impairs gas exchange, resulting in hypoxemia, often exacerbated by a decrease in mixed venous oxygen levels due to reduced cardiac output.[11] In severe cases, the hypoxemia can progress to respiratory failure, as the compensatory hypoxic pulmonary vasoconstriction further increases PVR and perpetuates the cycle of impaired oxygenation.[4]The elevated RV afterload from increased PVR causes acute pressure overload, leading to RV dilation and wall stress. This dilation can flatten the interventricular septum, impairing left ventricular filling and contributing to systemic hypotension.[11] Prolonged strain may result in RV ischemia due to heightened myocardial oxygen demand and reduced coronary perfusion, culminating in RV dysfunction or failure if untreated.[4]Additionally, platelet activation at the site of embolism releases vasoactive mediators such as serotonin and thromboxane A2, which induce further pulmonary vasoconstriction and bronchoconstriction of small airways.[11] These mediators exacerbate PVR elevation and contribute to systemic hypotension by promoting platelet aggregation and local vasospasm, intensifying the overall hemodynamic instability.[4]
Causes and risk factors
Virchow's triad components
Virchow's triad, first described by Rudolf Virchow in 1856, provides the foundational framework for understanding the pathogenesis of thrombosis, including the formation of deep vein thrombi that can embolize to cause pulmonary embolism (PE).[12] The triad consists of three interrelated components—stasis of blood flow, endothelial injury, and hypercoagulability—that synergistically promote thrombus development, particularly in the venous system where blood flow is slower and oxygen tension lower, facilitating clot initiation and propagation leading to venous thromboembolism (VTE) and subsequent PE.[6][7]Stasis refers to the slowing or interruption of normal blood flow, which allows procoagulant factors to accumulate and impairs the clearance of activated clotting components.[6] This component is often induced by immobility, such as in prolonged bed rest or long-haul travel, venous compression from tumors or pregnancy, or reduced cardiac output in conditions like heart failure, all of which diminish venous return and create hypoxic pockets at valve cusps where thrombi preferentially form.[6][7] In the context of PE, stasis in the lower extremities promotes deep veinthrombosis (DVT), the primary source of pulmonary emboli, by concentrating red blood cells and platelets near the vessel wall, thereby enhancing fibrin formation and platelet adhesion.[11]Endothelial injury involves damage to the intimal lining of blood vessels, exposing subendothelial collagen and tissue factor that trigger the coagulation cascade and platelet activation.[6] Common causes include direct trauma from indwelling catheters or surgery, inflammatory processes such as vasculitis, or chronic insults like smoking and hypertension that disrupt endothelial integrity.[6][13] In VTE and PE, this injury alters local hemodynamics, promotes leukocyte recruitment, and upregulates prothrombotic molecules like P-selectin on the endothelium, creating a nidus for thrombus growth that can dislodge and travel to the pulmonary arteries.[7]Hypercoagulability describes an imbalance in hemostatic mechanisms favoring clot formation, arising from either inherited or acquired alterations in coagulation factors, anticoagulants, or fibrinolysis.[6] Transient states, such as during pregnancy due to elevated levels of factors VII and VIII or with oral contraceptive use, temporarily increase clotting risk, while persistent conditions like factor V Leiden mutation sustain a prothrombotic milieu by resisting inactivation of factor Va.[6][7] This component amplifies VTE susceptibility by enhancing thrombin generation, which not only stabilizes clots but also impairs their dissolution, increasing the likelihood of embolization to the lungs in PE.[11]The components of Virchow's triad do not act in isolation but interact synergistically to escalate thrombosis risk; for instance, stasis exacerbates endothelial dysfunction through hypoxia-induced inflammation, while hypercoagulability potentiates the effects of even minor vessel wall perturbations.[7][13] This interplay is particularly evident in VTE, where multiple triad elements often converge—such as postoperative stasis combined with surgical endothelial trauma and transient hypercoagulability from inflammation—driving the majority of PE cases.[6]
Acquired and inherited predispositions
Acquired predispositions to pulmonary embolism encompass a range of modifiable and non-modifiable conditions that heighten the risk of venous thromboembolism (VTE), of which pulmonary embolism is a major manifestation. Major surgery, particularly orthopedic procedures such as hip or knee replacement, significantly elevates VTE risk due to endothelial injury and stasis, with incidence rates up to 40-60% without prophylaxis.[14]Trauma, including multiple injuries or spinal cord damage, similarly promotes clot formation through vessel damage and immobility, contributing to VTE in up to 50% of severe cases.[15]Hospitalization and prolonged immobility, such as bed rest exceeding one week, lead to venous stasis with autopsy studies showing DVT incidence up to 80%, often progressing to pulmonary embolism.[14]Cancer, especially adenocarcinomas of the lung, pancreas, or breast, induces a hypercoagulable state via tumor-released procoagulants, raising VTE risk 2- to 4-fold.[15]Pregnancy and the postpartum period amplify risk through elevated clotting factors and venous compression, with postpartum VTE rates 5- to 20-fold higher than in non-pregnant women.[14]Oral contraceptive use, particularly estrogen-containing formulations, boosts VTE risk 3- to 4-fold by increasing procoagulant factors.[16]Obesity, defined by BMI over 30 kg/m², contributes via chronic inflammation and stasis, with a 2- to 3-fold elevated risk in morbid cases.[14]Smoking exacerbates endothelial dysfunction and hypercoagulability, independently raising VTE odds by approximately 1.5-fold.[15]Inherited predispositions, or thrombophilias, involve genetic variants that impair natural anticoagulation mechanisms, often requiring environmental triggers to manifest as pulmonary embolism. Factor V Leiden mutation, present in 3-7% of the population, confers activated protein C resistance and increases first VTE risk 3- to 7-fold in heterozygotes.[15]Prothrombin G20210A gene mutation, affecting 1-3% of individuals, elevates prothrombin levels and VTE risk by 2- to 3-fold.[16]Antithrombin deficiency, a rare condition (0.02-0.04% prevalence), severely limits thrombin inhibition, heightening VTE risk up to 17-fold and often leading to early-onset events.[15]Protein C deficiency (0.2-0.5% prevalence) disrupts the protein C pathway, increasing risk 7- to 10-fold, while protein S deficiency (0.1-0.7% prevalence) similarly impairs this system, with 5- to 32-fold elevated VTE odds.[16]Antiphospholipid syndrome, an acquired autoimmune disorder rather than strictly inherited, features antibodies that promote thrombosis and is associated with 5- to 10-fold higher VTE risk, often recurrent.[15]Combined risks arise when acquired factors interact with inherited thrombophilias, substantially amplifying pulmonary embolism likelihood; for instance, immobility superimposed on Factor V Leiden can increase recurrence odds 2- to 3-fold beyond either alone.[16] The American Society of Hematology 2023 guidelines emphasize selective thrombophilia testing in such high-risk scenarios, like nonsurgical transient factors with family history, to inform extended management.[16]Recent data highlight COVID-19 as a prothrombotic acquired predisposition, inducing endothelial damage and inflammation that elevate pulmonary embolism risk 3- to 6-fold during acute infection, with elevated odds persisting up to 2 years post-recovery in some cohorts, as of 2025.[17][18] This aligns with post-2020 observations of COVID-19 as a major transient hypercoagulable state, comparable to severe trauma or cancer in VTE provocation.[19]
Clinical presentation
Common symptoms
The most common symptom of pulmonary embolism (PE) is dyspnea, or shortness of breath, which typically has a sudden onset and may occur even at rest.[4] This symptom arises due to impaired gas exchange and ventilation-perfusion mismatch in the lungs.[20] Pleuritic chest pain, described as sharp and worsening with deep inspiration or coughing, is also frequent, often resulting from pulmonary infarction or pleural irritation.[1]Cough is another prevalent complaint, which may be dry or productive and occasionally accompanied by hemoptysis (coughing up blood).[4]In cases of massive PE, where a large embolus significantly obstructs pulmonary blood flow, patients may experience syncope or presyncope due to acutely reduced cardiac output and systemic hypotension.[2] The presentation can vary widely by embolus size and location; small peripheral emboli often cause mild or no symptoms, and a significant proportion of PEs may be discovered incidentally during imaging for other reasons.[21]Atypical symptoms occur in some patients and can mimic other conditions, including fever that suggests infection or abdominal pain, particularly with right-sided emboli causing hepatic congestion or diaphragmatic irritation.[20][22] These less common manifestations highlight the need for clinical vigilance, as symptom severity correlates with embolus burden but is not always predictive.[21]
Physical examination findings
Physical examination in patients suspected of pulmonary embolism often reveals nonspecific but suggestive vital sign abnormalities, including tachypnea (respiratory rate greater than 20 breaths per minute), which is the most frequent finding, occurring in up to 96% of cases.[21]Tachycardia (heart rate greater than 100 beats per minute) is also common, present in approximately 44% of patients, while low-grade fever and hypoxia detected by pulse oximetry may occur, reflecting systemic inflammation and impaired gas exchange.[21][23] These vital sign changes arise from the pathophysiologic increase in dead space ventilation and right ventricular strain caused by vascular obstruction.[24]In more severe cases, cardiac examination may disclose signs of acute right ventricular overload and pulmonary hypertension, such as a parasternal right ventricular heave or an accentuated pulmonic second heart sound (loud P2), noted in about 53% of patients.[21]Lung auscultation is frequently normal but can occasionally reveal crackles or decreased breath sounds in up to 58% of cases, particularly if infarction or pleural effusion complicates the embolism.[4][21]Signs of concomitant deep vein thrombosis, the source of many emboli, may be evident in the lower extremities; deep vein thrombosis is present in approximately 70% of confirmed PE cases, though clinical signs such as unilateral leg swelling, tenderness to palpation, and erythema are often subtle or absent.[4] Homan's sign—pain on dorsiflexion of the foot—can be elicited but is nonspecific and infrequently emphasized due to low diagnostic utility.[23]Notably, the physical examination can be entirely normal in a notable proportion of pulmonary embolism cases, contributing to diagnostic disparities, particularly in underrepresented populations where subtle presentations may lead to delayed recognition.[25]
Diagnosis
Clinical probability tools
Clinical probability tools are validated scoring systems designed to stratify the pretest probability of pulmonary embolism (PE) in patients with suspected disease, guiding the need for further diagnostic testing such as D-dimer assays or imaging. These tools incorporate clinical features like symptoms, signs, and risk factors to categorize patients into low, intermediate, or high risk groups, thereby optimizing resource use and reducing unnecessary investigations. Widely adopted systems include the Wells criteria, the Geneva score, the Pulmonary Embolism Rule-out Criteria (PERC) rule, and the YEARS algorithm, each developed through prospective studies to improve diagnostic accuracy in emergency and outpatient settings.[26]The Wells criteria, originally derived in 2000 from a cohort of over 1,300 patients, assign points based on clinical features to estimate PE likelihood. Key components include clinical signs and symptoms of deep vein thrombosis (3 points), pulmonary embolism as the most likely diagnosis or equally likely to alternatives (3 points), heart rate greater than 100 beats per minute (1.5 points), immobilization or surgery within the previous 4 weeks (1.5 points), previous deep vein thrombosis or PE (1.5 points), hemoptysis (1 point), and malignancy (1 point). A total score of less than 2 indicates low probability (approximately 3% PE prevalence), 2 to 6 indicates moderate probability (approximately 20%), and greater than 6 indicates high probability (approximately 66%). This model has been prospectively validated and demonstrates good interrater reliability, with kappa values ranging from 0.41 to 0.75 across studies.[27]The Geneva score, first introduced in 2001 and revised in 2005 for enhanced simplicity and objectivity, evaluates pretest probability using objective variables to avoid subjective elements like the "alternative diagnosis" assessment in Wells criteria. The revised version includes age over 65 years (1 point), previous deep vein thrombosis or PE (3 points), surgery or lower limb fracture within 1 month (2 points), active malignancy (2 points), unilateral lower limb pain (3 points), hemoptysis (2 points), heart rate 75-94 beats per minute (3 points) or 95 or more (5 points), and pain on lower limb deep venous palpation and unilateral edema (4 points). Scores categorize patients as low risk (0-3 points, about 8% PE prevalence), intermediate risk (4-10 points, about 28%), or high risk (11 or more points, about 74%). Validation studies confirm its equivalence to Wells criteria in diagnostic performance, with sensitivity and specificity around 77% and 78%, respectively.[28]The PERC rule, developed in 2004 as a screening tool for very low-risk patients, consists of eight binary criteria to identify individuals who can have PE safely ruled out without additional testing if all are negative: age less than 50 years, heart rate less than 100 beats per minute, oxygen saturation greater than 94% on room air, no unilateral leg swelling, no hemoptysis, no recent trauma or surgery, no prior deep vein thrombosis or PE, and no oral hormone use. In a prospective multicenter validation involving over 3,100 patients, a negative PERC in those with low clinical suspicion (e.g., via Wells score <2) had a negative predictive value of 99.5% for PE, allowing up to 20% of suspected cases to avoid further evaluation. However, its utility is limited to low-risk populations, as it does not apply to intermediate or high pretest probability.The YEARS algorithm, validated in 2017-2019 studies, uses three criteria (clinical signs of deep vein thrombosis, hemoptysis, and PE as the most likely diagnosis) combined with D-dimer testing to rule out PE without imaging. If none of the three criteria are present and D-dimer is negative, PE can be excluded with a low miss rate (<1%). A pregnancy-adapted version, incorporating selective D-dimer use and leg ultrasound, safely avoids imaging in up to 65% of suspected cases in pregnant patients.[29][30]Recent guidelines integrate these tools into diagnostic algorithms. For pregnant patients, the 2025 European Society of Cardiology (ESC) guidelines recommend using standard clinical probability scores such as Wells or Geneva for risk stratification, along with empiric anticoagulation in cases of high clinical suspicion pending confirmatory imaging, given the 4- to 5-fold increased baseline risk of venous thromboembolism during pregnancy. These updates prioritize safety in special populations, with validation showing maintained high negative predictive values (over 99%) in adjusted models.[31]
Tool
Key Features
Risk Categories
Source
Wells Criteria
Subjective element (alternative diagnosis); includes DVT signs, immobilization, malignancy
Binary yes/no for 8 criteria; for very low-risk exclusion
All negative: Rule out PE (NPV 99.5%)
Kline et al., J Thromb Haemost 2004
Laboratory tests
Laboratory testing serves as an essential adjunct in the evaluation of suspected pulmonary embolism (PE), primarily aiding in ruling out the diagnosis in low-risk patients and providing prognostic insights in those with higher suspicion or confirmed disease. These tests are interpreted in conjunction with clinical probability assessment to guide further imaging.The D-dimer assay is the cornerstone laboratory test for excluding PE due to its high negative predictive value. This fibrin degradation product is elevated in acute thrombosis from concurrent coagulation and fibrinolysis activation. In patients with low or intermediate clinical pretest probability, a negative D-dimer (below 500 ng/mL or age-adjusted cutoff of age × 10 ng/mL for those over 50 years) safely rules out PE with sensitivity exceeding 95% and negative predictive value near 100%, avoiding unnecessary imaging. However, its specificity is limited (around 40-60%), as elevations occur in conditions like malignancy, infection, surgery, or advanced age, necessitating cautious interpretation in high-risk scenarios.Cardiac biomarkers, including troponin I or T and B-type natriuretic peptide (BNP) or N-terminal pro-BNP (NT-proBNP), are recommended for risk stratification in patients with acute PE, particularly to detect right ventricular (RV) dysfunction. Elevated troponin levels occur in approximately 50% of cases with moderate-to-large emboli, reflecting myocardial injury from RV strain, and are independently associated with adverse outcomes such as 30-day mortality (odds ratio up to 5.9 in normotensive patients). Similarly, BNP levels above 90 pg/mL or NT-proBNP above 900 pg/mL indicate RV overload and correlate with higher in-hospital mortality risk, helping classify patients as intermediate- or high-risk for hemodynamic deterioration. These markers lack diagnostic specificity for PE but enhance prognostic accuracy when combined with clinical scores.[32][33][34]Arterial blood gas (ABG) analysis typically demonstrates hypoxemia (PaO₂ <80 mmHg in about 80% of cases), hypocapnia from compensatory hyperventilation (PaCO₂ 30-35 mmHg), and an elevated alveolar-arterial oxygen gradient (>20 mmHg), reflecting ventilation-perfusion mismatch. While these patterns support PE suspicion in dyspneic patients, ABG findings are nonspecific and occur in other cardiopulmonary disorders, limiting their standalone diagnostic utility; normal oxygenation does not exclude PE.[32][35]Additional routine tests include complete blood count (CBC) to screen for thrombocytopenia (platelet count <150,000/μL), which may signal heparin-induced complications, and renal (creatinine >1.5 mg/dL) and liver function assessments to evaluate contraindications or dosing for anticoagulation. Coagulation profile, including prothrombin time/international normalized ratio (PT/INR) and activated partial thromboplastin time (aPTT), monitors therapeutic anticoagulation response. In post-acute PE follow-up, persistent troponin elevation may inform ongoing RV strain and long-term risk, as highlighted in recent management guides.[32][36]
Imaging modalities
Computed tomography pulmonary angiography (CTPA) serves as the gold standard for diagnosing pulmonary embolism (PE), offering high sensitivity and specificity by directly visualizing intraluminal filling defects in the pulmonary arteries.[37] This modality involves intravenous administration of iodinated contrast material to enhance vascular structures, allowing rapid assessment of clot location, extent, and potential alternative diagnoses such as pneumonia or aortic dissection.[38] However, CTPA exposes patients to ionizing radiation, with effective doses typically ranging from 2 to 10 mSv, raising concerns for radiation-sensitive populations including young women and pregnant individuals due to potential long-term cancer risks.[39] In pregnancy, fetal radiation exposure is minimized but still estimated at 0.01-0.66 mGy, prompting selective use after negative D-dimer testing to avoid unnecessary scans.[40]Ventilation-perfusion (V/Q) scintigraphy provides a valuable alternative when CTPA is contraindicated, such as in cases of severe renal impairment, contrast allergy, or high radiation risk.[41] This nuclear medicine test assesses regional lungventilation and perfusion using inhaled radioactive tracers (e.g., technetium-99m aerosol) and intravenous perfusion agents, respectively, to identify characteristic mismatch patterns indicative of PE.[42] A high-probability scan features multiple segmental or larger perfusion defects with normal ventilation, correlating with PE in approximately 85-90% of cases, while normal scans effectively rule out the diagnosis.[43] V/Q scans deliver lower radiation doses (about 2-4 mSv to the lungs) compared to CTPA and are particularly preferred in pregnant patients, where perfusion-only protocols further reduce exposure.[44]Echocardiography offers a noninvasive, bedside method to evaluate indirect signs of PE, particularly right ventricular (RV) dysfunction, which occurs in up to 30% of cases and signals hemodynamic compromise.[45] Transthoracic echocardiography may reveal RV dilation, hypokinesis, or elevated pulmonary artery systolic pressure, aiding risk stratification in unstable patients.[46] A specific finding, McConnell's sign—characterized by akinesia of the mid-RV free wall with preserved apical contraction—is highly suggestive of acute PE, with reported specificity exceeding 90% in some cohorts, though its sensitivity is lower (around 20-70%).[47] This modality does not directly visualize thrombi but is useful in emergency settings where immediate imaging is needed.Recent advancements include the 2025 European Society of Vascular Medicine (ESVM) guidelines, which emphasize catheter-based imaging techniques, such as intravascular ultrasound or optical coherence tomography, for guiding interventional therapies in high-risk PE during procedures like thrombectomy.[48] Additionally, magnetic resonance imaging (MRI) is emerging as a radiation-free alternative, particularly for young patients, with contrast-enhanced protocols demonstrating sensitivity and specificity comparable to CTPA (around 80-90%) in detecting central emboli through direct thrombus visualization and perfusionmapping.[49]
Prevention
Prophylactic anticoagulation
Prophylactic anticoagulation is a key strategy to prevent pulmonary embolism (PE) in individuals at high risk for venous thromboembolism (VTE), particularly those with transient risk factors such as recent surgery, hospitalization, or immobility.[50] Guidelines emphasize individualized assessment using validated tools like the Padua or IMPROVE scores to identify candidates, balancing thrombosis risk against bleeding potential.[50] For hospitalized medical patients at moderate to high VTE risk, low-molecular-weight heparin (LMWH), such as enoxaparin at 40 mg subcutaneously once daily, is recommended as first-line prophylaxis during the inpatient stay, reducing VTE incidence by approximately 50% compared to no prophylaxis.[51] Unfractionated heparin (UFH), typically 5000 units subcutaneously every 8-12 hours, serves as an alternative for patients with severe renal impairment or contraindications to LMWH, offering similar efficacy in shortening activated partial thromboplastin time without routine monitoring.[52]In surgical settings, particularly major orthopedic procedures like hip or knee replacement, LMWH remains the cornerstone for inpatient prophylaxis, initiated 12 hours pre- or post-operatively to minimize bleeding while preventing deep vein thrombosis that could embolize to the lungs.[53] For outpatient extended prophylaxis following hip fracture or arthroplasty, direct oral anticoagulants (DOACs) such as rivaroxaban (10 mg daily for 35 days post-hip surgery) or apixaban (2.5 mg twice daily for 12 days post-knee surgery) are preferred over LMWH due to oral administration, noninferior efficacy in reducing symptomatic VTE (relative risk 0.88 versus enoxaparin), and lower rates of major bleeding.[54] These agents inhibit factor Xa directly, providing predictable anticoagulation without dietary restrictions.[55]Vitamin K antagonists like warfarin are reserved for select cases where LMWH or DOACs are unsuitable, such as in patients with antiphospholipid syndrome or mechanical heart valves.[56] Prophylactic dosing targets an international normalized ratio (INR) of 2.0-3.0, with initial bridging using LMWH until therapeutic levels are achieved, though this approach is less favored due to monitoring requirements and higher intracranial hemorrhage risk compared to DOACs.[57]In cancer patients, prophylactic anticoagulation is tailored to setting and risk; for hospitalized individuals with active malignancy, LMWH or UFH is advised throughout admission.[58] For ambulatory patients receiving chemotherapy with a Khorana score ≥2, the 2023 European Society for Medical Oncology (ESMO) guidelines recommend primary prophylaxis with a DOAC (apixaban 2.5 mg twice daily or rivaroxaban 10 mg daily) or LMWH for up to 6 months, reducing VTE risk by 60% without significant survival impact.[59] Recent reviews (as of 2024) affirm extended use in high-risk scenarios, emphasizing discontinuation upon risk resolution to mitigate bleeding.[60]
Non-pharmacological measures
Non-pharmacological measures play a crucial role in preventing pulmonary embolism (PE), particularly by addressing venous stasis and immobilizing risk factors associated with deep vein thrombosis (DVT). Mechanical prophylaxis methods, such as graduated compression stockings (GCS) and intermittent pneumatic compression (IPC) devices, are recommended for immobile patients at risk of VTE. GCS apply graduated pressure to the lower extremities to promote venous return and reduce stasis, while IPC devices provide cyclic inflation to mimic muscle contractions and enhance blood flow. These interventions are especially useful in postoperative or bedridden patients where bleeding risk precludes pharmacological options, with guidelines suggesting their use in low-risk or high-bleeding-risk scenarios.[61][62][63]Early mobilization is another key strategy to mitigate PE risk, particularly following surgery or prolonged bed rest. Encouraging patients to ambulate as soon as clinically feasible helps counteract venous stasis by promoting circulation and reducing thrombus formation. Major guidelines, including those from the American College of Chest Physicians (ACCP), endorse early ambulation over strict bed rest in stable patients with DVT or at risk for VTE, noting its safety and efficacy in preventing progression to PE. This approach is often combined with mechanical methods for optimal effect in hospitalized individuals.[64][65]In cases where anticoagulation is contraindicated, such as active bleeding or recent surgery, inferior vena cava (IVC) filters serve as a mechanical barrier to prevent DVT propagation to the pulmonary arteries. Retrievable IVC filters are preferred over permanent ones due to their removability once anticoagulation can be safely initiated, minimizing long-term complications like filter migration or thrombosis. The Society of Interventional Radiology and ACCP guidelines recommend IVC filter placement in patients with acute PE or proximal DVT and absolute contraindications to anticoagulation, emphasizing individualized assessment. However, disparities in access persist; a 2025 American Heart Association scientific statement highlights that Black patients have lower odds of receiving IVC filters (OR 0.83, 95% CI 0.75–0.92) compared to White patients, often linked to socioeconomic barriers and unequal care access.[25][66]Lifestyle modifications further support PE prevention by reducing everyday risk factors. Avoiding prolonged sitting, such as during long-haul flights or sedentary work, is advised, with recommendations to stand and walk every 1–2 hours to maintain venous flow. The American Heart Association promotes changing positions frequently, avoiding leg crossing, and performing ankle flexion exercises during immobility. Staying hydrated with water is also encouraged to support blood flow, though direct evidence linking dehydration to VTE is limited; these measures are particularly vital for travelers at moderate risk.[67][8][68]
Treatment
Initial stabilization and anticoagulation
Initial stabilization of patients with acute pulmonary embolism (PE) begins with assessment and support of the airway, breathing, and circulation (ABCs). Supplemental oxygen is administered to any patient with hypoxemia to maintain oxygen saturation above 90%, as this improves oxygenation without routine need for intubation unless respiratory failure occurs.[4]Hemodynamic support is critical for patients with hypotension or shock. Intravenous fluids should be administered cautiously due to the risk of right ventricular (RV) overload in PE, with a modest challenge of 250-500 mL recommended over 15-30 minutes, guided by invasive monitoring such as central venous pressure if available; excessive fluids can exacerbate RV strain and worsen outcomes.[69][70] If hypotension persists despite fluids, vasopressors are initiated promptly, with norepinephrine as the first-line agent to maintain mean arterial pressure above 65 mm Hg, often in combination with dobutamine if RV dysfunction is evident.[71][72]Anticoagulation is the cornerstone of treatment for all hemodynamically stable patients with confirmed or high-suspicion PE and low bleeding risk, ideally initiated empirically while awaiting diagnostic confirmation. For most stable patients, direct oral anticoagulants (DOACs) or low-molecular-weight heparin (LMWH) are preferred over vitamin K antagonists (VKAs) during the initial treatment phase. Specifically, apixaban is administered at 10 mg twice daily for the first 7 days, followed by 5 mg twice daily thereafter, based on evidence from randomized trials demonstrating noninferior efficacy and safety compared to conventional therapy.[73] LMWH, such as enoxaparin at 1 mg/kg subcutaneously twice daily, is an alternative, particularly in patients with contraindications to DOACs.[73]In hemodynamically unstable patients (e.g., systolic blood pressure <90 mm Hg for >15 minutes), unfractionated heparin (UFH) is recommended as the initial anticoagulant due to its rapid onset, titratability, and reversibility. UFH is given as an intravenous bolus of 80 units/kg followed by an infusion of 18 units/kg per hour, adjusted to target an activated partial thromboplastin time (aPTT) of 1.5-2.5 times control.[69][74] Transition to LMWH or DOACs can occur once stability is achieved.[73]The minimum duration of anticoagulation is 3 months for PE provoked by transient risk factors, such as surgery or immobilization, with strong evidence supporting this approach to prevent recurrence while minimizing bleeding risk. For unprovoked PE or recurrent events, indefinite anticoagulation is often recommended after reassessing bleeding risk, using reduced-intensity DOACs (e.g., apixaban 2.5 mg twice daily) for extended therapy in select cases.[73][16]Recent updates emphasize outpatient management for low-risk PE (e.g., simplified Pulmonary Embolism Severity Index score of 0, no RV dysfunction on imaging, and normal biomarkers), allowing initiation of DOACs at home if social support and follow-up are adequate, as supported by 2021 CHEST guidelines and 2025 American College of Cardiology reviews.[73][75]
Thrombolysis and embolectomy
Systemic thrombolysis involves the administration of fibrinolytic agents to dissolve the clot in patients with high-risk pulmonary embolism (PE) characterized by hemodynamic instability, such as systolic blood pressure below 90 mm Hg. The standard regimen uses alteplase at a dose of 100 mg administered intravenously over 2 hours, with anticoagulation initiated near the end of or immediately following the infusion. This approach has been shown to reduce mortality in high-risk cases but carries a significant risk of major bleeding, estimated at 9-11% overall, including intracranial hemorrhage in approximately 1-2% of patients. Bleeding risks are heightened in elderly patients or those with recent surgery, prompting careful patient selection following initial stabilization measures like hemodynamic support.Catheter-directed thrombolysis (CDT) delivers lower doses of thrombolytic agents directly into the pulmonary arteries via a catheter, minimizing systemic exposure and potentially reducing bleeding complications compared to systemic methods. Typical regimens involve alteplase at 0.5-1 mg/hour per catheter for 12-24 hours, guided by imaging to target the thrombus. The 2025 European Society for Vascular Medicine (ESVM) guidelines endorse CDT for acute PE in intermediate-high-risk patients or those with contraindications to systemic therapy, highlighting its role in improving right ventricular function with lower major bleeding rates (around 2-5%). This technique is particularly useful when systemic thrombolysis risks outweigh benefits, often following initial anticoagulation.Surgical embolectomy is reserved for high-risk PE patients with absolute contraindications to thrombolysis, such as active bleeding or recent stroke, or in cases of thrombolytic failure. Performed via median sternotomy or thoracotomy, it involves direct removal of central pulmonary thrombi under cardiopulmonary bypass, frequently supported by extracorporeal membrane oxygenation (ECMO) for perioperative hemodynamic stability. Outcomes show in-hospital mortality rates of 10-25% in specialized centers, comparable to thrombolysis in selected cases, with improved long-term survival when performed early. The American Heart Association recommends this approach for patients with massive PE and persistent instability despite medical therapy.Percutaneous embolectomy employs catheter-based mechanical devices to aspirate or fragment thrombi without thrombolytics, ideal for submassive PE or high-risk cases with bleeding contraindications. Devices such as the FlowTriever system (Inari Medical) use large-bore aspiration catheters to retrieve clots, achieving significant reductions in pulmonary artery pressure and right ventricular strain, with major bleeding rates below 1% in trials. The Aspirex catheter (BD/Straub Medical) similarly enables rapid thrombus removal through rotational aspiration. Recent guidelines, including updates from the European Respiratory Society, have elevated aspirationthrombectomy to a class I recommendation for high-risk PE, emphasizing its efficacy in restoring hemodynamics with minimal adjunctive pharmacotherapy.
Interventional and surgical options
Inferior vena cava (IVC) filter insertion serves as an interventional option for patients experiencing recurrent pulmonary embolism (PE) despite ongoing therapeutic anticoagulation, aiming to prevent further embolization from lower extremity deep vein thrombosis.[66] These devices, typically retrievable, are placed percutaneously via femoral or jugular access to mechanically trap clots within the IVC, thereby reducing the risk of additional PE in high-risk scenarios where anticoagulation alone is insufficient.[76] Guidelines recommend retrieval of these filters within 3 to 6 months post-insertion when the underlying thrombotic risk has resolved, to minimize long-term complications such as filter migration, fracture, or IVC thrombosis.[77] Retrieval success rates exceed 90% when performed within this timeframe, emphasizing the importance of scheduled follow-up imaging and procedural planning.[78]For chronic thromboembolic pulmonary hypertension (CTEPH), a sequela of unresolved PE, balloon pulmonary angioplasty (BPA) emerges as a key interventional strategy to alleviate pulmonary vascular obstruction in patients unsuitable for pulmonary endarterectomy. BPA involves selective catheterization of stenotic or occluded pulmonary artery branches, followed by controlled balloon inflation to fracture organized thrombi and improve blood flow, often performed in staged sessions to target multiple lesions safely.[79]European Society of Cardiology guidelines endorse BPA as a Class I recommendation for inoperable CTEPH or persistent pulmonary hypertension post-endarterectomy, with studies reporting mean pulmonary artery pressure reductions of 20-30% and enhanced exercise capacity in over 70% of treated patients.[80] In select cases where BPA yields suboptimal results due to elastic recoil or dissection, adjunctive pulmonary artery stenting may be employed to maintain vessel patency, particularly in peripheral lesions refractory to angioplasty alone.[81]Surgical thrombectomy represents a definitive operative intervention for massive PE cases unresponsive to medical therapies, including thrombolysis, particularly when hemodynamic instability persists due to central clot burden. This procedure entails median sternotomy, cardiopulmonary bypass, and direct extraction of thrombi from the main pulmonary arteries and branches, often under hypothermic circulatory arrest to facilitate precise clot removal.[82] It is reserved for patients with contraindications to fibrinolysis or those in cardiogenic shock, with perioperative mortality rates ranging from 10-20% in experienced centers, reflecting improved outcomes through rapid right ventricular unloading.[83] Long-term survival exceeds 70% at 5 years when combined with postoperative anticoagulation, underscoring its role in salvaging otherwise fatal presentations.[84]Recent advancements highlight the need for equitable access to these catheter-based interventions, as disparities in utilization persist across racial, socioeconomic, and geographic lines. The 2025 American Heart Association scientific statement emphasizes strategies to reduce barriers, such as standardized training programs and hospital quality improvement initiatives, to ensure broader implementation of devices like IVC filters and BPA, ultimately aiming to lower mortality gradients in underserved populations.[25]
Prognosis and complications
Mortality prediction models
Mortality prediction models for pulmonary embolism (PE) are essential clinical tools that stratify patients based on short-term fatality risk, guiding decisions on treatment intensity and disposition. These models integrate clinical variables, imaging findings, and biomarkers to categorize patients into low-, intermediate-, and high-risk groups, with validated associations to 30-day all-cause mortality rates. Widely adopted models emphasize simplicity and prognostic accuracy to facilitate rapid bedside assessment in acute settings.The Pulmonary Embolism Severity Index (PESI), developed in 2005, is a seminal prognostic model derived from a cohort of over 15,000 patients with acute PE. It assigns points based on 11 variables: age (1 point per year), male sex (10 points), cancer (30 points), heart failure (10 points), chronic lung disease (10 points), heart rate ≥110 bpm (20 points), systolic blood pressure <100 mm Hg (30 points), respiratory rate ≥30 breaths/min (20 points), temperature <36°C (20 points), altered mental status (60 points), and oxygen saturation <90% (20 points). Patients are classified into five risk classes (I-V), with class I (≤65 points) and II (66-85 points) indicating low risk and associated 30-day mortality rates of 0-1.6% and 1.7-3.5%, respectively.To enhance clinical usability, the simplified PESI (sPESI) was introduced in 2010, reducing the variables to six binary predictors without weighting: age >80 years, history of cancer, chronic cardiopulmonary disease, heart rate ≥110 bpm, systolic blood pressure <100 mm Hg, and oxygen saturation <90%. A score of 0 identifies low-risk patients (approximately 30-40% of cases), with a 30-day mortality of about 1.0-1.1%, while scores ≥1 denote high risk with mortality around 8.9-10.9%. The sPESI demonstrates comparable discriminative ability to the original PESI (c-statistic ≈0.75) and is recommended for identifying candidates for outpatient management.[85]The European Society of Cardiology (ESC) 2019 guidelines provide a comprehensive risk stratification framework that builds on clinical and diagnostic parameters. High-risk PE is defined by hemodynamic instability (systolic blood pressure <90 mm Hg or drop ≥40 mm Hg lasting >15 minutes), with untreated mortality up to 50% but reduced to approximately 15-25% with reperfusion therapy. Intermediate-risk PE involves normotensive patients with right ventricular (RV) dysfunction on imaging (e.g., echocardiography or CT) or elevated biomarkers, subcategorized as intermediate-high (both present, 3-15% mortality) or intermediate-low (either present, ≈3% mortality); low-risk PE lacks these features and carries <1% 30-day mortality. This approach integrates PESI/sPESI for initial assessment and refines it with objective tests for precise prognostication.Cardiac biomarkers, particularly troponin and B-type natriuretic peptide (BNP) or N-terminal pro-BNP (NT-proBNP), enhance risk prediction by detecting subclinical myocardial injury and RV strain. Elevated troponin levels (>0.01 ng/mL, manufacturer cutoff) in normotensive PE patients are associated with 30-day mortality rates of 10-15%, compared to 2% in those with negative results, due to their correlation with RV dysfunction. Similarly, BNP >90 pg/mL or NT-proBNP >900 pg/mL predicts adverse outcomes, with low levels (<50 pg/mL for BNP) identifying low-risk patients (negative predictive value >95% for in-hospital death). These markers are most valuable in intermediate-risk stratification, where their elevation signals up to 15% short-term mortality risk.[34]Recent advancements, as outlined in a 2025 systematic review and meta-analysis, incorporate artificial intelligence (AI) and machine learning (ML) models to refine post-acute PE predictions. These models, using variables like clinical scores, biomarkers, and imaging features, achieve superior performance with an area under the receiver operating characteristic curve (AUROC) of 0.91, outperforming traditional tools like PESI/sPESI (AUROC 0.75-0.85). Logistic regression and ensemble methods (e.g., random forests, XGBoost) demonstrate pooled sensitivity of 88% and specificity of 79% for 30-day mortality, enabling personalized risk assessment in practical guides for integrated care.[86]
Long-term sequelae
Survivors of acute pulmonary embolism (PE) often experience persistent symptoms collectively known as post-PE syndrome, which includes dyspnea and reduced exercise capacity affecting approximately 50% of patients at 3 months post-event.[87] This syndrome arises from incomplete resolution of thromboembolic material, right ventricular dysfunction, or deconditioning, leading to impaired quality of life and functional limitations.[88] Cardiopulmonary rehabilitation can improve exercise tolerance in these individuals, though symptoms may persist in up to 20-30% beyond one year.[89]A serious long-term complication is chronic thromboembolic pulmonary hypertension (CTEPH), occurring in 3-4% of PE survivors, characterized by organized thrombi causing pulmonary vascular obstruction and elevated pressures.[90] CTEPH typically develops within 2 years post-PE and is more common in unprovoked events or with recurrent venous thromboembolism (VTE).[91] The condition is potentially curable through pulmonary endarterectomy, a surgical procedure that removes chronic thrombi from pulmonary arteries, offering significant hemodynamic and symptomatic improvement in operable cases.[92]The risk of recurrent VTE remains elevated long-term, with cumulative incidence rates of 20-30% over 5 years in the absence of extended anticoagulation, particularly for unprovoked initial events.[93] Extended anticoagulation with direct oral anticoagulants (DOACs) reduces this risk by over 80% but introduces bleeding concerns, necessitating individualized assessment of benefits versus harms, such as a 1-2% annual major bleeding rate.[94]Current guidelines emphasize structured follow-up to detect and manage these sequelae, recommending echocardiography at 3-6 months for patients with ongoing dyspnea to evaluate for pulmonary hypertension, alongside functional assessments like the 6-minute walk test to gauge exercise capacity.[95] For those at high recurrence risk, indefinite DOAC therapy is advised after initial treatment, with periodic reassessment to balance thrombotic and hemorrhagic risks.[95]
Epidemiology
Incidence and prevalence
Pulmonary embolism (PE) occurs at an annual incidence of 60 to 120 cases per 100,000 individuals in Western populations, with rates escalating to over 500 per 100,000 among those aged 75 years and older.[96]The prevalence of PE remains underdiagnosed in clinical settings, as autopsy studies consistently reveal it in 10% to 30% of examined cases, often without prior suspicion; autopsy-confirmed fatal PE contributes to approximately 1 in 500 overall deaths.[97][98][99]Incidence trends have been largely stable in recent decades, but heightened awareness and occurrence post-COVID-19 have led to a reported 20% to 27% increase in hospitalized PE cases from 2020 to 2025.[100][101]Underreporting of PE is especially prevalent in low-resource settings due to limited diagnostic access, as emphasized in the 2025 American Heart Association scientific statement on disparities.[25]
Demographic variations
Pulmonary embolism (PE) incidence varies significantly by demographic factors, influencing both risk and outcomes. Age is a primary determinant, with PE being rare in younger populations. In individuals under 40 years, particularly those aged 18–34, the incidence remains low at approximately 18 cases per 100,000 person-years, reflecting limited exposure to cumulative risk factors. However, incidence rises exponentially after age 70, driven by age-related immobility, comorbidities, and vascular changes; rates reach 383 per 100,000 in those aged 75–84 and 417 per 100,000 in those over 85, underscoring the heightened vulnerability in advanced age.[102]Sex differences in PE epidemiology show patterns modulated by age and hormonal influences. Overall age-adjusted incidence is similar between sexes, but women exhibit higher rates during reproductive years (ages 20–40), at about 16 cases per 100,000 person-years compared to 7 in men, largely attributable to estrogen-related effects from oral contraceptives, hormone replacement therapy, and pregnancy. In contrast, men demonstrate a 20%–25% higher incidence in older age groups (60–80 years), contributing to a slight overall male predominance in PE cases among adults over 65. Outcomes also differ, with men facing elevated in-hospital mortality risks, potentially due to later presentation or more severe manifestations.[103]Ethnic and racial variations reveal disparities in PE incidence and mortality, often linked to socioeconomic factors, access to care, and genetic predispositions. Caucasians experience elevated rates partly due to higher prevalence of genetic thrombophilias like factor V Leiden, which increases VTE susceptibility. However, Black individuals face the highest overall incidence and severity, with 1.9 times higher odds of hospitalization for PE compared to Whites; this group also encounters diagnostic delays, resulting in substantially worse outcomes. According to 2025 American Heart Association data, in-hospital mortality among Black patients stands at 15.0%, approximately 40% higher than the 10.7% rate in White patients, highlighting persistent inequities in timely intervention. Hispanic and Asian populations generally report lower incidence than Whites and Blacks, though Hispanics show comparable mortality despite less severe presentations.[25]Comorbidities amplify PE risk across demographics.