Clinical presentation and diagnostic evaluation of the nonpregnant adult with suspected acute pulmonary embolism

INTRODUCTION — 

Acute pulmonary embolism (PE) is a common and sometimes fatal disease. The diagnostic approach to PE should be efficient while simultaneously avoiding unnecessary testing so that therapy can be prompt and potential morbidity and mortality avoided.

This topic discusses the clinical manifestations and diagnostic evaluation of PE. The pathophysiology, treatment, and prognosis of PE, as well as the diagnosis of PE during pregnancy, are reviewed separately.

●(See "Pulmonary embolism: Epidemiology and pathogenesis in adults".)

●(See "Acute pulmonary embolism in adults: Treatment overview and prognosis".)

●(See "Pulmonary embolism in pregnancy: Clinical presentation and diagnosis".)

Our diagnostic approach is, in general, consistent with strategies outlined by several international societies including the American College of Physicians, American Society of Hematology, European Society of Cardiology, European Respiratory Society, American College of Emergency Physicians, American College of Radiology, and others [1-5].

CLINICAL PRESENTATION — 

PE has a wide variety of presenting features, ranging from no symptoms to shock or sudden death [6-9] (table 1). Maintaining a high level of suspicion is critical so that clinically relevant cases are not missed.

History and examination — Presenting features are as follows:

●Symptoms – Symptoms in patients with PE (table 1) include the following [7,10-12]:

•Dyspnea at rest or with exertion (73 percent)

•Pleuritic pain (66 percent)

•Cough (37 percent)

•Orthopnea (28 percent)

•Calf or thigh pain and/or swelling (44 percent)

•Wheezing (21 percent)

•Hemoptysis (13 percent)

•Hoarseness from a dilated pulmonary artery (Ortner syndrome; <1 percent)

Importantly, symptoms may be mild or absent in up to one-third of patients, even in the case of large PE [6,9,13].

Dyspnea onset is frequently (but not always) rapid, usually within seconds (46 percent) or minutes (26 percent) [9]. Dyspnea may be less frequent in older patients with no previous cardiopulmonary disease. Dyspnea is more likely present in patients with PE in the main or lobar vessels. However, some patients have a delayed or evolving presentation over days or weeks, which may be associated with large, centrally located PE [14]. (See "Pulmonary embolism: Epidemiology and pathogenesis in adults", section on 'Pathogenesis and pathophysiology'.)

Pleuritic pain is classic in this population due to inflammation of the pleura, especially when due to small, peripherally located infarction. Hemorrhage from the infarcted lung is also thought to be responsible for hemoptysis and small bloody pleural effusions. (See "Pulmonary embolism: Epidemiology and pathogenesis in adults", section on 'Pathogenesis and pathophysiology'.)

●Examination – Common presenting signs on examination include [9]:

•Tachypnea (54 percent)

•Calf or thigh swelling, erythema, edema, tenderness, palpable cords (47 percent)

•Tachycardia (24 percent)

•Rales (18 percent)

•Decreased breath sounds (17 percent)

•An accentuated pulmonic component of the second heart sound (15 percent)

•Jugular venous distension (14 percent)

•Transient or persistent arrhythmias (eg, atrial fibrillation <10 percent)

•Presyncope or syncope (<10 percent)

•Acute shock or hemodynamic collapse (<10 percent)

•Pleural effusion (<10 percent)

•Fever (3 percent)

•Upper extremity pain or swelling (upper extremity deep vein thrombosis [DVT] infrequently embolizes)

Among those presenting with syncope, rates of PE range from 1 to 17 percent [15-23], with the highest rates occurring in those hospitalized with syncope without a clear etiology [20,23]. Up to two-thirds of patients with PE who present with syncope have large thrombi located in the mainstem or lobar arteries [10,11]. The reasons for syncope in patients with PE are poorly understood but may be partially explained by transient arrhythmias as thrombus travels through the heart or pulmonic valve.

Patients who present with shock tend to be <65 years old [9,24,25]. These patients may also have signs of acute right ventricular (RV) failure manifested by increased jugular venous pressure, a right-sided third heart sound, a parasternal lift, and cyanosis. However, shock is not always due to large PE but, in fact, may also develop in patients with small PE who have severe underlying lung disease or pulmonary hypertension (PH). A transition from tachycardia to bradycardia, or from a narrow complex to a broad complex tachycardia (ie, right bundle branch block), is an ominous sign of RV dysfunction and impending shock. (See "Definition, classification, etiology, and pathophysiology of shock in adults".)

Laboratory tests — Laboratory tests are not diagnostic but alter the clinical suspicion for PE, confirm the presence of alternative diagnoses, and provide prognostic information if PE is diagnosed. In patients with suspected PE, we perform the following:

●Routine laboratory tests – Complete blood count, serum chemistries, liver function tests, and coagulation studies have limited diagnostic value. Abnormalities include leukocytosis, increased erythrocyte sedimentation rate, elevated serum lactate, elevated serum lactate dehydrogenase, and aspartate aminotransferase. Serum creatinine and the estimated glomerular filtration rate help determine the safety of administering contrast for angiography.

●Arterial blood gas (ABG) and pulse oximetry – ABGs are often abnormal among patients with PE but are normal in up to 18 percent [26]. Abnormal gas exchange may be due to, and/or worsened by, underlying cardiopulmonary disease [27]. Common abnormalities seen on ABGs include one or more of the following [7,26,28] (see "Arterial blood gases"):

•Hypoxemia (74 percent)

•Widened alveolar-arterial gradient for oxygen (62 to 86 percent)

•Respiratory alkalosis and hypocapnia (41 percent)

Hypercapnia, respiratory, and/or lactic acidosis are uncommon but can be seen in patients with PE associated with obstructive shock and respiratory arrest.

●D-dimer – D-dimer is typically elevated in PE, but is not diagnostic due to a high rate of false positive results. However, D-dimer can exclude PE when the suspicion is low. Details are provided below. (See 'D-dimer with standard fixed cutoff' below.)

●Cardiac biomarkers of RV dysfunction – Brain natriuretic peptide (BNP), N-terminal proBNP [29,30], and troponin levels [31-36] are sometimes elevated when RV dysfunction is present. They are limited diagnostically but are useful for stratifying the risk of death from PE. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Assess mortality risk (low, intermediate, high)'.)

Electro- or echocardiography — Electrocardiographic (ECG) abnormalities are common but nonspecific in patients with PE [37-41]. The most common findings are tachycardia and nonspecific ST-segment and T-wave changes (70 percent) [8].

RV abnormalities are uncommon (<10 percent) but often indicate poor prognosis in PE. ECG findings include S1Q3T3 pattern, new incomplete right bundle branch block, bradycardia, and inferior Q waves [37-40,42,43].

Echocardiography is not routinely obtained but helps identify patients at risk of death from PE. (See 'Echocardiography' below.)

Chest radiograph — Nonspecific abnormalities on chest radiography are common in PE (eg, atelectasis, effusion), but a normal chest radiograph can be seen in 12 to 22 percent of patients [7,8,44]. A chest radiograph is typically performed in most patients to look for an alternative cause of the patient's symptoms. It is also performed to determine eligibility for ventilation-perfusion (V/Q) scanning (see 'V/Q scan' below). However, it is not necessary if computed tomographic (CT) pulmonary angiography (CTPA) is planned.

A Hampton hump, Westermark sign, and Fleischner or Palla sign are rare but should raise the suspicion for PE when present [45,46].

●A Hampton hump is a shallow, hump-shaped opacity in the periphery of the lung, with its base against the pleural surface and hump towards the hilum (image 1).

●A Westermark sign demonstrates hypoperfusion due to the abrupt cutoff of vascularization of the lung beyond the occluded vessel (image 2) [45].

●A Fleischner and Palla sign are synonymous. Fleischner described a dilated central pulmonary artery [45] while Palla described an enlarged right descending pulmonary artery that tapers off abruptly and may have a 'sausage' appearance (image 3) [46,47].

Differential diagnosis — For patients who present with signs and symptoms of PE, the major competing diagnoses include heart failure, myocardial ischemia or infarction, pneumothorax, pneumonia, pericarditis, acute exacerbations of chronic lung disease, and musculoskeletal pain. Most conditions can be distinguished from PE by clinical features and ECG, echocardiographic, laboratory, and chest radiographic testing. However, PE can coexist with these conditions; therefore, an alternate diagnosis does not completely exclude PE.

The differential diagnosis of PE signs and symptoms includes the following:

●Dyspnea – Dyspnea that is abrupt in onset or disproportionate to the patient's underlying lung function or dyspnea that occurs with hypoxemia, hemoptysis, and/or pleuritic chest pain may favor a diagnosis of PE. (See "Pulmonary embolism: Epidemiology and pathogenesis in adults" and "Approach to the patient with dyspnea" and "Approach to the adult with dyspnea in the emergency department".)

●Chest pain – Acute chest pain, especially pleuritic pain, is highly suspicious for PE but may also be due to other etiologies, such as pneumonia, pericarditis, pleuritis, and rib fracture.

●Hemoptysis – Hemoptysis that occurs with pleuritic pain and hypoxemia should prompt consideration of acute PE but can also be secondary to pneumonia or heart failure (often frothy and pink). (See "Evaluation of nonlife-threatening hemoptysis in adults".)

●Leg pain and swelling – Unilateral leg swelling should raise the suspicion for PE due to DVT while bilateral swelling may be more supportive of heart failure.

●Syncope – Syncope in patients without a clear precipitant should raise suspicion for PE, particularly in hospitalized patients [20]. (See "Syncope in adults: Clinical manifestations and initial diagnostic evaluation" and "Approach to the adult patient with syncope in the emergency department".)

●Hypoxemia – Hypoxemia (partial pressure of oxygen in arterial blood on room air <80 mmHg [10 kPa]) in the setting of a normal chest radiograph or hypoxemia that is disproportionate to the chest radiograph appearance should prompt consideration of PE as well as the following alternate diagnoses:

•Other pulmonary vascular diseases (eg, chronic venous thromboembolism, PH, anatomic shunt, arteriovenous malformations) (see "Epidemiology, pathogenesis, clinical manifestations and diagnosis of chronic thromboembolic pulmonary hypertension" and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults" and "Pulmonary arteriovenous malformations: Epidemiology, etiology, and pathology in adults")

•Interstitial lung disease (eg, interstitial pneumonitis) (see "Approach to the adult with interstitial lung disease: Clinical evaluation")

•Congenital heart disease (eg, shunt, septal defect, left ventricular outlet obstruction, chronic mitral stenosis, Eisenmenger syndrome) (see "Pulmonary hypertension with congenital heart disease: Clinical manifestations and diagnosis")

•Lower-airway disease (eg, pneumonia, asthma, bronchiectasis, acute or chronic bronchitis, foreign body aspiration, tracheobronchomalacia) (see "Evaluation of wheezing illnesses other than asthma in adults")

•Upper-airway disease (eg, paradoxical vocal cord dysfunction, upper-airway obstruction syndromes, tumors) (see "Inducible laryngeal obstruction (paradoxical vocal fold motion)" and "Asthma in adolescents and adults: Evaluation and diagnosis", section on 'Clinical features')

•Neuromuscular disease (eg, hypoventilation, drugs, multiple sclerosis, diaphragmatic paralysis, myasthenia gravis) (see "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation")

●Tachycardia – Unexplained tachycardia, especially in a patient with risk factors for PE, should prompt clinicians to consider PE but is relatively nonspecific. (See "Sinus tachycardia: Evaluation and management".)

●Shock – Unexplained shock should prompt the clinician to consider acute PE. Although the presence of shock and a normal chest radiograph increases the suspicion for PE, this can be found in many forms of distributive shock (eg, anaphylaxis, shock from drugs and toxins, neurogenic shock, myxedema coma). (See "Evaluation of and initial approach to the adult patient with undifferentiated hypotension and shock", section on 'Differential diagnosis'.)

The differential diagnosis of common conditions that mimic PE include the following:

●Heart failure – The combination of dyspnea and leg swelling due to heart failure may mimic PE. Crackles and bilateral infiltrates on chest radiography may support evidence of pulmonary edema. While elevated BNP can support heart failure, this can also be seen in acute PE. (See "Heart failure with preserved ejection fraction: Clinical manifestations and diagnosis".)

●Pneumonia – Fever, consolidation on chest radiography, and leukocytosis may favor infection over PE but can also be the presenting features of an acute lobar pulmonary infarct secondary to PE, particularly as it evolves over the first few days or weeks. The presence of risk factors for PE, persisting symptoms or poor response to antibiotics, or abrupt onset of new symptoms during the course of subacute illness should prompt the clinician to investigate for PE. (See "Clinical evaluation and diagnostic testing for community-acquired pneumonia in adults" and "Epidemiology, pathogenesis, and microbiology of community-acquired pneumonia in adults" and "Nonresolving pneumonia".)

●Myocardial ischemia or infarction – Cardiac chest pain is typically not pleuritic and evidence of myocardial ischemia or infarction can be seen on ECG. While elevated troponin can suggest cardiac chest pain, this can also be seen in acute PE. (See "Diagnosis of acute myocardial infarction".)

●Pericarditis – Pericarditis pain can be pleuritic and therefore mimic PE. The presence of a viral prodrome, pre-existing inflammatory disease, and ST segment elevation on ECG increase the likelihood of pericarditis. (See "Acute pericarditis: Clinical presentation and diagnosis".)

●Exacerbation of underlying chronic lung disease – Patients with chronic lung disease often present with dyspnea. Conversely, PE can complicate acute pulmonary illness (eg, emphysema, pneumonia). Thus, another diagnosis does not completely exclude the possibility of PE. Wheezing is uncommon in PE and may suggest an exacerbation of pre-existing lung disease, such as asthma or chronic obstructive pulmonary disease. However, hypoxemia or respiratory distress out of proportion to obstructive symptoms or wheezing should prompt consideration of PE. (See "COPD exacerbations: Management".)

●Pneumothorax – While acute pleuritic chest pain and dyspnea due to pneumothorax may mimic PE, pneumothorax should be apparent on chest imaging. (See "Pneumothorax in adults: Epidemiology and etiology" and "Treatment of secondary spontaneous pneumothorax in adults".)

●Vasculitis – Unexplained dyspnea, pleuritis, and hemoptysis can be presenting symptoms of both PE and pulmonary vasculitis. The presence of an interstitial pattern on chest radiograph in a patient with an underlying rheumatologic condition (eg, scleroderma) may distinguish vasculitis from PE. (See "Overview of and approach to the vasculitides in adults", section on 'Differential diagnosis'.)

●Musculoskeletal pain – Acute chest wall pain may mimic the pleuritic pain of PE but is often tender on palpation. However, in the absence of a clear history of injury, musculoskeletal pain should be considered a diagnosis of exclusion when PE remains on the differential diagnosis. (See "Approach to the adult with nontraumatic chest pain in the emergency department".)

HEMODYNAMICALLY UNSTABLE PATIENTS (HIGH-RISK PE) — 

Less than 10 percent of patients present with hemodynamic instability (ie, hypotension, obstructive shock, and cardiac arrest). Hemodynamic instability represents high risk of death from PE due to right ventricular (RV) dysfunction. In this population, diagnosis and therapy are often approached simultaneously. In this section, we focus on diagnosis; the approach to therapy is discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Initial resuscitation, risk assessment, empiric anticoagulation' and "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'High-risk PE (unstable)'.)

Expert consultation — In many academic centers, including ours, the initial evaluation and resuscitation of hemodynamically unstable patients with suspected PE are often performed with the aid of PE response teams (PERTs). These teams comprise cardiothoracic surgeons, pulmonary and intensive care unit clinicians, cardiologists, emergency clinicians, and interventional radiologists. However, in centers with limited resources, the responding clinician must perform initial resuscitation, order appropriate diagnostic tests, and rely upon clinical judgment to assess the risk-benefit ratio of empiric anticoagulation and/or thrombolysis without definitive testing. Data discussing PERTs are described separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Pulmonary embolism response teams'.)

Hemodynamic stability restored following resuscitation — For patients in whom hemodynamic stability is restored following brief resuscitation (eg, intravenous fluids and/or vasopressors for 15 minutes) (see "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Hemodynamic support'), we suggest the following approach:

●High suspicion for PE – For patients with high suspicion for PE, we prefer immediate anticoagulation (provided there is no contraindication) followed by definitive diagnostic imaging, usually CT pulmonary angiography (CTPA).

Rarely, patients may directly undergo CTPA for diagnostic evaluation and catheter-directed therapy (if indicated).

This approach is contingent upon prompt access to imaging and the presence of staff that can administer cardiopulmonary resuscitation and/or empiric thrombolytic therapy if the patient decompensates during testing.

●Low or moderate suspicion for PE – For patients with a low or moderate suspicion for PE, the same approach to diagnosis and empiric anticoagulation should be used for hemodynamically stable patients, although many will ultimately undergo CTPA. (See 'Hemodynamically stable patients' below.)

Hemodynamically unstable despite resuscitation — For patients who remain hemodynamically unstable (eg, systolic pressure <90 mmHg for ≥15 minutes, clear evidence of shock or cardiac arrest), definitive testing is typically considered unsafe. In these circumstances, bedside tests that may be useful include the following:

●Echocardiography – Bedside echocardiography may provide a definitive diagnosis when thrombus is seen in the proximal pulmonary artery. However, pulmonary artery thrombus is rarely seen on transthoracic echocardiography and often needs transesophageal echocardiography, which is not universally available.

A presumptive diagnosis of PE may be obtained when thrombus is seen in the right atrium or RV (ie, thrombus-in-transit) and is sufficient to justify therapy. Similarly, if new RV dysfunction is found on echocardiography, this may also be enough to justify empiric anticoagulation pending definitive diagnosis. Further details regarding echocardiographic findings supportive of PE are provided separately. (See 'Echocardiography' below.)

●Lower extremity ultrasonography – Bedside lower extremity compression ultrasonography (CUS) does not diagnose PE. However, it does diagnose deep vein thrombosis (DVT), which is sufficient to initiate anticoagulation and may justify the administration of potentially life-saving therapies, such as thrombolytic therapy, when PE is highly suspected. While formal whole-leg CUS is the most sensitive test for DVT, the time and expertise required to perform it make it less suitable in an urgent situation [48]. (See 'Lower-extremity ultrasound' below.)

●Perfusion studies – A portable perfusion scanner that identifies perfusion defects may provide a definitive diagnosis. However, such scanners are not universally available.

Treatment of hemodynamically unstable PE is discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'High-risk PE (unstable)'.)

HEMODYNAMICALLY STABLE PATIENTS — 

Most patients with PE are hemodynamically stable on presentation [9]. Allowing sufficient time to adopt a systematic diagnostic approach. We selectively integrate clinical evaluation, a three-tiered pretest probability (PTP) assessment, PE rule-out criteria (PERC), D-dimer testing, and imaging (algorithm 1 and algorithm 2 and algorithm 3). CT pulmonary angiography (CTPA) is the imaging modality of choice. However, algorithms that use a ventilation-perfusion (V/Q) scan are appropriate when CTPA is contraindicated, not feasible, or inconclusive. (See 'CTPA' below and 'Alternate or additional imaging' below.)

Empiric anticoagulation — Empiric anticoagulation while waiting for test results should be individualized according to the clinical suspicion for PE, the anticipated timing of definitive testing, and the risk of bleeding, the details of which are discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Empiric anticoagulation or thrombolysis'.)

Determine pretest probability — The PTP for PE should be estimated by clinical gestalt assessment or calculated using a validated PTP score, such as the Wells score, modified Wells score (table 2) (calculator 1), or Modified Geneva score (table 3) (calculator 2) [49-57]. The YEARS criteria have been used in conjunction with D-dimer. Although gestalt estimates and probability scores have comparable sensitivity when combined with D-dimer testing, meta-analyses suggest that probability scores may have higher specificity [49,57] and increase the diagnostic yield of CTPA [58].

●Three-tiered score (Wells) – We use the Wells score (table 2) (calculator 1) based upon extensive validation, our clinical experience, and its three-tiered system of PE probability:

•Low (score <2) (see 'Low probability of pulmonary embolism' below)

•Intermediate (score 2 to 6) (See 'Intermediate probability of pulmonary embolism' below.)

•High (score >6) (see 'High probability of pulmonary embolism' below)

Some experts, further subdivide the intermediate-risk category into intermediate-low and intermediate-high probability, which has yet to be formally tested and is discussed below. (See 'Intermediate probability of pulmonary embolism' below.)

We prefer the three-tiered classification since it allows D-dimer testing to be applied to both low- and intermediate-probability patients (score ≤6), further reducing unnecessary testing. It can also be used to interpret the results of V/Q scans more accurately [6].

Despite validation of the Wells score, for unclear reasons, clinicians either do not use them or misuse them in up to 80 percent of patients [59,60]. In addition, they may not be as accurate in older or hospitalized patients [55,61,62] since they were mostly validated in the outpatient setting.

●Two-tiered score (modified Wells) – The modified Wells score classifies patients into a two-tiered system: PE likely (score >4) or PE unlikely (score ≤4).

●YEARS – The YEARS criteria include three items from the Wells score: are there clinical signs of deep vein thrombosis (DVT)?; does the patient have hemoptysis?; and is PE the most likely diagnosis?, all scored as a yes or no. PE is excluded in patients with no YEARS items and a D-dimer level <1000 ng/mL and patients with one or more YEARS items and a D-dimer <500 ng/mL. The YEARS algorithm is best studied with adjusted D-dimer in a protocolized approach to PE diagnosis, which is described below. (See 'Alternative (adjusted) D-dimer cutoffs' below.)

Low probability of pulmonary embolism

Nonhospitalized patients: PERC rule — For nonhospitalized patients with a low gestalt probability of PE, we apply the PERC rule (table 4) (calculator 3) to determine whether diagnostic evaluation with D-dimer is indicated (algorithm 3).

●For patients who fulfill all eight criteria, PE is considered excluded (<1 percent) and no further testing is required.

●For patients who do not fulfill all eight criteria or when PERC cannot be applied (eg, hospitalized patients), we obtain high-sensitivity D-dimer testing. (See 'D-dimer with standard fixed cutoff' below.)

PERC should not be used in patients with an intermediate or high suspicion for PE or for inpatients suspected as having PE since it is only valid in clinical settings (typically the emergency department [ED]) with a low prevalence of PE (<15 percent) [63]. In clinical settings with a higher prevalence of PE (>15 percent), PERC has a poorer predictive value [63].

While some experts measure D-dimer in all low-probability patients [64], our preference to use PERC is based upon the validity of this approach in low-risk patients in an outpatient setting and the likely reduction (approximately 20 percent) of unnecessary testing (ie, D-dimer and imaging) associated with its use [2,63,65-68].

●A crossover cluster-randomized noninferiority trial of 1916 ED patients with a low gestalt PTP for PE (eg, 15 percent probability of PE) reported similar rates of PE when PERC was compared with a conventional assessment using a D-dimer level to determine further testing (0.1 versus 0 percent) [69]. The application of PERC reduced the proportion of patients undergoing CTPA (13 versus 23 percent) and reduced the ED stay by 36 minutes.

●Another multicenter prospective cohort study of 8138 ED patients with a low clinical suspicion for PE reported that among those who fulfilled all eight criteria, <1 percent were diagnosed with DVT or PE within the subsequent 45 days [65].

Hospitalized patients

D-dimer with standard fixed cutoff — In low-probability patients (table 2) (calculator 1) where PERC cannot be applied (ie, typically hospitalized patients) or PERC is positive, we perform high-sensitivity D-dimer testing:

●When the D-dimer level is <500 ng/mL (fibrinogen equivalent units), PE is considered excluded (<2 percent) and no further testing is required.

●When the D-dimer level is ≥500 ng/mL (fibrinogen equivalent units), diagnostic imaging should be performed, preferably with CTPA. (See 'CTPA' below.)

For patients with low-probability PE (including patients with prior PE and hospitalized patients), studies have consistently shown that a normal D-dimer level effectively excludes PE (ie, <500 ng/mL [fibrinogen equivalent units]) [14,62,70]. However, the sensitivity may be lower in subsegmental PE compared with patients with large main, lobar, or segmental PE (53 versus 93 percent) [71-73]. Additionally, the proportion of patients with negative results is lower among those with recurrence than among those with their first presentation [70].

We prefer "high-sensitivity" testing that uses quantitative or semiquantitative newer-generation immunoturbidimetric assays, latex agglutination-based assays, or rapid enzyme-linked immunosorbent assays (ELISA). These assays have a higher sensitivity (96 versus 90 percent) and negative predictive value (98 versus 95 percent) compared with older "low-sensitivity" assays (eg, qualitative rapid ELISA, first-generation latex, and erythrocyte agglutination) [74]. In addition, results are available quickly (10 to 30 minutes). The main disadvantage of D-dimer is that it is nonspecific since many other conditions increase the level (table 5).

For these assays, a level ≥500 ng/mL (fibrinogen equivalent units) is usually considered positive and <500 ng/mL (fibrinogen equivalent units) is considered negative [74,75]. The performance of this cutoff in a protocol using clinical probability and CTPA (the "Christopher study" [50]) is discussed below. (See 'Diagnostic performance' below.)

D-dimer adjusted for clinical factors, such as age or clinical probability, is discussed below. (See 'Alternative (adjusted) D-dimer cutoffs' below.)

The performance of D-dimer in intermediate- and high-probability PE is discussed below. (See 'D-dimer' below and 'High probability of pulmonary embolism' below.)

Alternative (adjusted) D-dimer cutoffs — Adjusted D-dimer levels based on certain criteria have been validated and may be considered an alternative in patients with a low or intermediate probability of PE. They should not be used in those with a high probability for PE. While adjusted D-dimer cutoffs decrease the use of imaging, they may miss some cases of PE.

●Age-adjusted D-dimer – Age-adjusted D-dimer assessments are increasingly used with significant institutional variation. D-dimer levels rise with age such that using the traditional cutoff value of <500 ng/mL (fibrinogen equivalent units) results in reduced specificity of D-dimer testing in older patients (>50 years), a population in whom PE is common. For example, the specificity of D-dimer in patients >80 years old is approximately 10 percent [76].

The most commonly used formula for age adjustment is:

One meta-analysis of six trials reported that in patients unlikely to have PE by the Wells criteria (score ≤4 (table 2) (calculator 1)), a negative age-adjusted D-dimer compared with a negative fixed-level D-dimer was associated with a 5 percent increase in the proportion of patients in whom imaging can be safely withheld [77].

The performance of this cutoff in a protocol using clinical probability and CTPA ("ADJUST-PE" [78]) is discussed below. (See 'Diagnostic performance' below.)

●D-dimer adjusted to clinical probability – D-dimer cutoffs adjusted to the clinical PTP have also been validated. However, they have not been tested in hospitalized patients or populations with a high prevalence of PE. In addition, some protocols are complex, which may limit their practicality in a busy setting, such as the ED. Furthermore, while they may reduce the number of patients who receive CTPA, they may miss more cases of PE and underestimate the value of CT in identifying other etiologies for a patient's presenting symptoms. One meta-analysis of 16 studies reported that protocols using PTP-dependent D-dimer thresholds excluded more cases of PE without imaging than protocols using standard D-dimer cutoffs (ie, 500 ng/mL) [79]. However, protocols using adjusted D-dimer levels missed more cases of venous thromboembolism (VTE) in the subsequent three months and performed poorly in patients ≥80 years old and those with cancer. Examples include the following:

•YEARS criteria – In one prospective study, 3465 patients with suspected PE from an outpatient setting underwent D-dimer testing and PTP assessment using the YEARS criteria [80]. Among those in whom PE was excluded, 0.6 percent had symptomatic PE confirmed at three-month follow-up [80], a rate that was similar to that reported in studies that utilize fixed D-dimer level testing <500 ng/mL [50]. It was estimated that this algorithm would result in a 14 percent reduction in the number of CT scans performed compared with the Wells score and a fixed D-dimer level <500 ng/mL. Using age-adjusted D-dimer had no additional value to this algorithm [81].

A similar prospective observational study of 1134 ED patients with suspected PE reported a 14 percent reduction in those imaged using the YEARS approach to D-dimer testing [82]. The sensitivity and specificity of this strategy were 93 and 55 percent, respectively. This algorithm has been externally validated for the safety of excluding PE; however, caution was advised in patients with no YEARS items and a D-dimer level that was <1000 ng/mL but above the age-adjusted cutoff [83].

•PEGeD protocol – In a prospective study of 2017 outpatients with suspected PE, patients with a low clinical probability (calculated per the Wells score (calculator 1)) plus a negative D-dimer <1000 ng/mL and patients with a moderate clinical probability and a negative D-dimer <500 ng/mL did not undergo CTPA [84]. All other patients underwent imaging. During a three-month follow-up, no patients with a low or moderate clinical probability plus a negative D-dimer developed symptomatic VTE. Using the PEGeD protocol, 34 percent of patients were imaged compared with an estimated 52 percent, had the traditional parameters been used to exclude PE.

•PERC-positive patients plus YEARS and age-adjusted D-dimer – One noninferiority trial randomized PERC-positive patients with a low clinical probability of PE and patients with an intermediate probability for PE to further triage using YEARS criteria plus age-adjusted D-dimer (intervention group) or age-adjusted D-dimer alone (control group) [85]. In patients with zero YEARS criteria plus a D-dimer <1000 ng/mL and in patients with one or more YEARS criteria plus a D-dimer level less than the age-adjusted threshold, PE was considered excluded and no imaging was performed. Imaging was performed if patients did not meet these criteria. This D-dimer strategy resulted in a 10 percent reduction in patients who underwent chest imaging and a 1.6-hour reduction in the ED stay. There was also a nonsignificant reduction in the rate of VTE at three months (1 versus 5 patients; 0.15 versus 0.8 percent).

•4-Level Pulmonary Embolism Clinical Probability Score (4PEPS) – 4PEPS, which has 13 clinical variables, was recently validated as a PE stratification tool to reduce unnecessary imaging. However, it requires additional study before considering it for routine use [86].

Intermediate probability of pulmonary embolism

D-dimer — For most patients in whom the suspicion for PE is intermediate, we perform high-sensitivity D-dimer testing (algorithm 2).

●When the D-dimer level is <500 ng/mL (fibrinogen equivalent units), PE is considered excluded (<2 percent) and no further testing for PE is required.

For most patients with an intermediate probability for PE, a normal D-dimer (<500 ng/mL [fibrinogen equivalent units]) effectively excludes PE [87,88]. However, some experts believe that a subset of patients in the intermediate-high probability category (ie, those in the upper zone of the intermediate range [eg, Wells score 4 to 6 or Modified Geneva score 8 to 10]) should undergo imaging based upon the higher probability of PE and a D-dimer sensitivity that is not as good as for low-probability PE.

In addition, some experts, obtain CTPA (without D-dimer testing) if patients have limited cardiopulmonary reserve since PE might be poorly tolerated in that population.

●When the D-dimer level is ≥500 ng/mL (fibrinogen equivalent units), diagnostic imaging should be performed, preferably with CTPA. (See 'CTPA' below.)

High probability of pulmonary embolism

CT pulmonary angiography — For most patients in whom the probability of PE is high (table 2) (calculator 1) or in whom the suspicion is low or intermediate and D-dimer is elevated, we perform CTPA (algorithm 1). While a negative D-dimer reduces the likelihood of PE in this population, it is insufficient to exclude the diagnosis; data suggest a PE prevalence of 5 percent or more when the PTP is high and the D-dimer is negative [74,89-92]. Further details are provided below. (See 'CTPA' below.)

DIAGNOSTIC IMAGING

CTPA — CT pulmonary angiography (CTPA) is the first-choice diagnostic imaging modality for the following reasons:

●It is sensitive and specific for diagnosing PE, especially when incorporated into diagnostic algorithms. (See 'Diagnostic performance' below.)

●CTPA is also helpful in guiding management by identifying findings that confer worse prognosis, such as central location, large thrombus load, and right ventricular (RV) enlargement (image 4) [93-96]. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Assess mortality risk (low, intermediate, high)' and "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Prognostic factors'.)

●Alternate diagnoses may also be discovered using this modality (image 5) [97-101]. Observational studies report that up to one-third of CTPA examinations identify an alternate diagnosis, which requires immediate attention in a small proportion [100,101].

●CTPA technology is widely available and can be performed on an urgent basis in most settings.

●Some common contraindications to CTPA can be readily resolved (eg, premedication for a contrast allergy). Thus, CTPA may be performed after a short delay (eg, 8 to 12 hours) with consideration for empiric anticoagulation while waiting. Empiric anticoagulation is discussed separately. (See 'CTPA imaging protocol' below and "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Empiric anticoagulation or thrombolysis'.)

Despite the publication of several well-validated protocols and clinical decision rules to prevent the overuse of CTPA, real-world data suggest increased use of CTPA and diagnosis of low-risk PE [102-104].

A chest CT with contrast performed for other indications may incidentally detect PE, but is not an adequate examination for excluding suspected PE [97]. In these cases, CTPA imaging to confirm PE or diagnose residual PE is reasonable; if CTPA is negative, performing lower extremity compression ultrasonography (CUS) is also appropriate. (See 'Alternate or additional imaging' below.)

Interpreting the results — Our subsequent approach is the following:

●A positive CTPA confirms the diagnosis of PE, necessitating treatment. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis".)

A positive CTPA demonstrates a filling defect or abrupt cutoff in any branch of the pulmonary artery (main, lobar, segmental, subsegmental) (image 4) [105].

●A negative CTPA indicates that the likelihood of PE is low (<5 percent). Typically, no further testing is required, unless the suspicion for PE remains and an alternate diagnosis is not evident. (See 'Alternate or additional imaging' below.)

●An indeterminate CTPA may require alternate imaging or a repeat CTPA if the reason is resolvable (eg, the contrast bolus is poorly timed). The most common reasons for an indeterminate CTPA include the following [106]:

•Patient motion

•Large body habitus

•Artifacts from metallic foreign bodies

•Suboptimal enhancement of the pulmonary artery usually due to hypotension or low cardiac output

•Inadequate bolus administration of contrast

•Poor scanner technology

Repeat CTPA for more definitive results may be worthwhile if the factor causing poor image quality can be mitigated (eg, greater cooperation with positioning and breath-holding instructions, blood pressure has improved, larger intravenous access obtained). However, the risk of kidney function impairment due to repeated doses of intravenous contrast should be considered when determining whether and when to repeat CTPA. Repeat imaging is unlikely to prove useful if CTPA is nondiagnostic from factors, such as scanner technology, large body habitus, or indwelling metallic foreign bodies.

CTPA imaging protocol — A typical protocol is as follows:

●Image acquisition – CTPA (also called chest CT angiogram with contrast) acquires thin (≤1.25 mm) section volumetric images of the chest after a bolus administration of intravenous contrast that is timed precisely for maximal enhancement of the pulmonary arteries [107,108]. The protocol of contrast administration is often determined locally. Primary axial and multiplanar reformations of the pulmonary arteries are routinely reviewed. For optimal image quality, the patient should be able to remain still and hold their breath for about 30 seconds.

●Radiation dosing – The effective radiation dose from CTPA ranges from 1 to 10 millisieverts (mSv) and varies depending upon patient size, scanner type, and imaging protocol [108,109]. In young adults (age <30 years) or pregnant patients who are undergoing multiple chest CT exams, minimizing cumulative radiation dose may be a consideration for alternative imaging techniques if the necessary technology and expertise are available. (See 'Alternate or additional imaging' below.)

●CT venogram (CTV) – CTV of the lower extremities and pelvis with contrast to evaluate for deep vein thrombosis (DVT) is not routinely performed concurrently with CTPA. CTV, when added to CTPA, may marginally improve diagnostic yield. However, the added effective radiation dose from CTV is approximately 6 mSv, thereby significantly increasing the radiation dose [110,111]. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Alternative imaging'.)

Diagnostic performance — Most CTPA studies report sensitivities and specificities >90 percent for the diagnosis of PE, regardless of the pretest probability (PTP) for PE [50,78,105,112-114]. Nevertheless, there is a risk of PE in those with a negative CTPA and a high clinical suspicion for PE (up to 5 percent when a multidetector row CT [MDCT] pulmonary angiography (with 64 or fewer rows is used), such that additional testing may be needed [115]. (See 'Alternate or additional imaging' below.)

CTPA was traditionally considered most accurate for detecting large, main, lobar, and segmental PE, and less accurate for detecting smaller, peripheral subsegmental PE (SSPE). However, newer scanners with improved resolution have increased the detection of smaller PE [116-119]. For example, in one systematic review, MDCT pulmonary angiography detected more SSPE than single-detector row CTPA (9.4 versus 4.7 percent) [116]. Management of SSPE is discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Patients with subsegmental PE'.)

Support for our preference for CTPA-based algorithms is derived from several prospective cohort studies. As examples:

●In a prospective study of 3306 patients with clinically suspected PE, patients classified as PE "unlikely" per modified Wells (table 2) (calculator 1) underwent sensitive D-dimer testing (Christopher study). PE was excluded when the D-dimer level was <500 ng/mL (fibrinogen equivalent units) [50]. Patients with a D-dimer level ≥500 ng/mL (fibrinogen equivalent units) and those with PE "likely" underwent CTPA. At three months follow-up, the rates of venous thromboembolism (VTE) in untreated patients was <0.6 percent.

●A similarly designed prospective study of 3346 patients with suspected PE reported comparable results using age-adjusted D-dimer cutoffs (ADJUST-PE) [78]. Patients in whom PE was unlikely underwent age-adjusted D-dimer testing (see 'Alternative (adjusted) D-dimer cutoffs' above); if negative, no further testing was performed. All other patients underwent CTPA. Patients with inconclusive CTPA results or in whom CTPA could not be performed had additional imaging (eg, ventilation-perfusion [V/Q] scan, serial ultrasound, pulmonary angiogram) to diagnose or exclude PE. At three months follow-up, rates of VTE in untreated patients were <0.5 percent.

Artificial intelligence (AI) has been proposed to help radiologists interpret medical images [120,121]. In one retrospective study of 1202 patients with suspected PE, 16 percent had true PE according to the study criterion standard that included both radiologist and AI results. AI suspected PE in 219 patients, of which 176 (80 percent) were true PE and 20 percent were false positives. Of the true PE, 19 cases were missed by radiologists. Sensitivity and negative predictive values were greater with AI while specificity and positive predictive values were greater with radiologists. Further data are needed to determine what role AI could play in diagnosing patients with suspected PE.

Patients with precautions or contraindications — Our approach to patients with potential contraindications to CTPA is the following:

●Contrast allergy – CTPA is relatively contraindicated in patients with a history of moderate to severe iodinated contrast allergy or kidney function impairment (estimated glomerular filtration rate <30 mL/minute per 1.73 m²). The risk of these contraindications must be weighed against the clinical importance of performing CTPA examination and the availability of alternative imaging approaches (eg, V/Q scan). If clinically feasible, we prefer to delay CTPA for premedication to ensure its performance while minimizing the adverse effects of contrast, especially when the suspicion for PE is high. The prevention of contrast allergy and kidney injury are provided separately. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography" and "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Prevention'.)

For patients with high PTP for PE and the time to obtain CTPA is expected to be prolonged (>2 hours), we typically initiate empiric anticoagulation and sometimes obtain lower extremity ultrasonography. Guidance on empiric anticoagulation is provided separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Empiric anticoagulation or thrombolysis' and 'Lower-extremity ultrasound' below.)

●Intolerance – For patients who cannot tolerate CT scanning (eg, due to class 2 or 3 obesity, difficulty lying flat, claustrophobia, or agitation), we typically perform alternate imaging modalities. (See 'Alternate or additional imaging' below.)

Alternate or additional imaging

V/Q scan — When CTPA cannot be performed or is indeterminate, we perform V/Q scanning. In some cases, we prefer to repeat CTPA if it was previously contraindicated but subsequently becomes feasible (eg, when kidney function improves or after premedication for a contrast allergy). (See 'Interpreting the results' above and 'Patients with precautions or contraindications' above.)

●Interpretation – A high-probability V/Q scan (typically segmental or subsegmental perfusion defect with normal ventilation) in a patient with a high-probability of PE is considered diagnostic.

V/Q scans are interpreted in the context of the clinical PTP for PE (Prospective Investigation of Pulmonary Embolism Diagnosis [PIOPED] criteria [6]) (table 6). We use Wells criteria to stratify probability:

•In patients with a normal V/Q scan and any clinical probability, no further testing is necessary.

•In patients with a low-probability V/Q scan and low clinical probability (eg, Wells score <2 (table 2) (calculator 1)), no further testing is necessary.

•In patients with a high-probability V/Q scan and high clinical probability (eg, Wells score >6 (table 2) (calculator 1)) (image 6), PE is diagnosed.

•All other combinations of V/Q scan results and clinical PTPs are indeterminate (inconclusive), and further testing is required. (See 'Other imaging' below.)

The Society of Nuclear Medicine has practice guidelines regarding the performance and interpretation of V/Q scans for lung scintigraphy [122].

●V/Q image acquisition – The patient is asked to lie still for 30 to 60 minutes for a V/Q scan. The approximate effective radiation dose is less than 2 mSv.

●Diagnostic performance – Support for our approach using V/Q scanning is based upon the following data:

•In PIOPED, the risk of PE was reported in combination with PTP assessment (table 6) [6] (see 'Determine pretest probability' above):

-Patients with a low clinical probability and a normal or low-probability V/Q scan had a <4 percent chance of having a PE while those with an intermediate and high probability scan had a 16 and 56 percent chance of having PE.

-Patients with a high clinical probability and a high-probability scan had a 96 percent chance of having a PE. Those with normal scan had a 0 percent chance of having PE, and those with a low- and intermediate-probability scan had a 40 and 66 percent chance of having PE, respectively.

-Patients with an intermediate probability of PE and a high-probability scan had an 88 percent chance of having PE while the probability of PE in all other combinations ranged from 6 to 28 percent.

Most patients have indeterminate scans, which is the major limitation of V/Q scanning and necessitates additional testing.

A normal chest radiograph improves diagnostic performance. Scans performed on patients with abnormal chest radiographs are more likely to result in false positives as the images rarely appear normal, increasing the odds of an indeterminate scan.

•One systematic review evaluated over 7000 patients from 25 prospective studies, 23 of which included V/Q scan-based algorithms [123]. Three diagnostic strategies were identified as safely excluding patients with PE over a three-month follow-up:

-Among patients with a low clinical probability of PE in whom normal D-dimer levels excluded PE, PE occurred in <3 percent.

-Among patients in whom clinical probability combined with D-dimer assessment was inconclusive, a normal perfusion scan (Q scan) safely excluded PE.

-Among patients with an intermediate-probability V/Q scan, holding therapy was safe until further testing was performed (eg, catheter-based pulmonary angiography or serial lower-extremity venous ultrasonography).

●Alternate types of V/Q scans – Although perfusion scanning alone is sometimes performed in limited cases, data to support its use in diagnosing PE are unavailable.

Portable V/Q scans have been described for use in the intensive care unit or emergency department, although their accuracy has yet to be compared with CTPA [124].

Lower-extremity ultrasound — We do not routinely perform lower extremity CUS as a first-line test since the sensitivity in patients with suspected PE is low [125,126]. When neither CTPA nor V/Q scanning can be performed or are indeterminate, we perform whole-leg lower extremity CUS with Doppler to evaluate for coexisting DVT [48]. A lower extremity ultrasound demonstrating proximal DVT can justify initiating anticoagulant treatment, although it is not diagnostic of PE itself.

We suggest the following subsequent approach:

●If lower-extremity ultrasound is positive, patients should be treated. (See "Overview of the treatment of lower extremity deep vein thrombosis (DVT)".)

●If lower-extremity ultrasound is negative, we use the following as a guide:

•Low PE suspicion – If the clinical suspicion for PE is low (calculator 1) (table 2), it is generally considered safe to withhold anticoagulation and monitor until chest imaging can be performed (eg, after treatment of contrast allergy) [123,127]. However, empiric anticoagulation may be appropriate in patients with poor cardiopulmonary reserve (ie, patients who would not tolerate another PE). Empiric anticoagulant therapy is discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Empiric anticoagulation or thrombolysis'.)

If chest imaging is not feasible, some experts monitor for DVT using serial imaging. Data to support this approach are limited. In one prospective study of 874 patients with suspected PE who had a low or intermediate-probability V/Q scan and negative serial lower-extremity venous ultrasounds over two weeks, <3 percent of patients developed PE at three months [127]. (See "Overview of the treatment of lower extremity deep vein thrombosis (DVT)", section on 'Low risk of extension: Serial ultrasound'.)

•Intermediate or high PE suspicion – If the suspicion for PE is intermediate or high, we obtain further imaging while empiric anticoagulation is ongoing, unless contraindications to anticoagulation are present. The rationale for this approach is that ultrasonography may be negative in the setting of PE, either because thrombus has traveled to the lung or because thrombus in the calf and/or iliac veins are not readily detected by proximal ultrasonography (figure 1) [128,129]. (See 'Other imaging' below and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Diagnostic compression ultrasonography (CUS)'.)

Other alternatives include whole-leg CUS if not already performed (to detect calf vein DVT; calf pain, tender cords) or Doppler ultrasound of the iliac vein (to detect pelvic vein DVT; massive swelling; abdominal, back, or buttock pain). Further details regarding lower extremity CUS and whether proximal or whole-leg CUS should be performed are provided separately. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Diagnostic compression ultrasonography (CUS)'.)

Other imaging — When CTPA, V/Q scan, or lower extremity CUS is not helpful, magnetic resonance pulmonary angiography (MRPA), contrast-enhanced pulmonary angiography, and echocardiography are options. (See 'Magnetic resonance pulmonary angiography' below and 'Catheter-based pulmonary angiography' below and 'Echocardiography' below.)

Magnetic resonance pulmonary angiography — We do not perform MRPA as a first-line test for the diagnosis of PE but use it as an imaging option when neither CTPA nor V/Q scan can be performed or when the cumulative radiation dose is a concern, provided the necessary technology and expertise are available.

●Interpretation – A filling defect or abrupt cutoff in any pulmonary artery branch is diagnostic of PE.

●Image acquisition – MRPA should only be performed at sites with the necessary technology and expertise. The patient is asked to lie still in a magnetic resonance scanner for >30 minutes before administering intravenous gadolinium. Technically inadequate images can result from patient motion, scanner technology, and poor contrast bolus timing [130-134]. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Indications for giving contrast with MRI'.)

●Diagnostic performance – MRPA is less sensitive and more dependent on the technologist's experience doing the scan than CTPA, resulting in a higher rate of indeterminate scans. Potential advantages of MRPA are that no ionizing radiation is involved and the examination can be combined with magnetic resonance venography (MRV) in the same sitting. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity", section on 'Alternative imaging'.)

MRPA was studied prospectively in 371 adults with suspected PE. Among the 75 percent of patients with technically adequate images, MRPA alone (ie, without MRV) showed a sensitivity of 78 percent and a specificity of 99 percent [131]. Among the 48 percent of patients with technically adequate images for MRPA with MRV, the sensitivity was 92 percent and the specificity 96 percent.

Additional prospective studies reported a similarly poor sensitivity for MRPA alone [130,131,135]. Sensitivity was greater for PE located in the main/lobar and segmental vessels (100 and 84 percent, respectively) than subsegmental vessels (40 percent; ie, PE that should be easily detected on CTPA).

Catheter-based pulmonary angiography — Pulmonary angiography, in which contrast is injected under fluoroscopy via a catheter introduced into the right heart, was the diagnostic gold standard historically. With the widespread emergence of CTPA, this procedure is infrequently used. Catheter-based pulmonary angiography is most often performed in patients in whom concurrent therapy is planned since it can combine diagnosis with therapeutic interventions aimed at clot lysis (eg, catheter-directed embolectomy and/or thrombolysis) (image 7). (See "Acute pulmonary embolism in adults: Thrombolytic therapy in intermediate- and high-risk patients".)

●Interpretation – A filling defect or abrupt cutoff in any pulmonary artery branch is diagnostic of PE (image 7).

●Image acquisition – Typical access sites are internal jugular and femoral veins. Catheters are placed under fluoroscopic guidance in the pulmonary artery with the patient supine. Contrast is injected to fill the branches of the pulmonary artery. We measure the pulmonary artery pressure before injection, which may modify the rate and amount of contrast injected. The pulmonary arteries usually require a large volume of contrast for complete filling. Several planes may be used to optimize imaging (eg, oblique views maximize evaluation of the basal branches).

●Diagnostic performance – Pulmonary angiography may be less accurate than CTPA since its diagnostic performance is highly variable and dependent on the operator's experience [123,136]. As the historical gold standard, the sensitivity and specificity of catheter-based pulmonary angiography for the diagnosis of PE have not been formally evaluated. However, one retrospective analysis of 20 cases from PIOPED II suggested that it may be less sensitive than CTPA for the detection of small PE [136,137]. Nonetheless, data suggest that in patients with a negative angiogram, the risk of subsequent symptomatic embolization is low (<2 percent) [6,123].

●Complications – Although pulmonary angiography is generally well-tolerated, data in patients with pulmonary hypertension (PH) suggest that the mortality of the procedure is approximately 2 percent, lower if the patient is hemodynamically stable [138,139]. Morbidity occurs in approximately 5 percent of patients and is usually related to catheter insertion, contrast reactions, cardiac arrhythmia, or respiratory insufficiency [138,140,141]. Radiation exposure depends upon the length and complexity of the procedure but is typically greater than that from CTPA [142,143].

Echocardiography — Echocardiography may be helpful in the following settings:

●When thrombus is visualized in the proximal pulmonary arteries, although this is a rare phenomenon. Transesophageal echocardiography or endobronchial ultrasonography are better tools than transthoracic echocardiography to identify thrombus in central pulmonary arteries [144-146].

●When right-sided cardiac thrombus or new RV dysfunction is seen in hemodynamically unstable patients, a presumptive diagnosis can be obtained to justify the emergency use of thrombolytic therapy or embolectomy [147-151]. However, identifying new right heart strain is not feasible without a prior echocardiogram.

Limited evidence in patients with intracardiac thrombus report that 35 percent have PE [152] and approximately 4 to 9 percent of those with known PE have RV thrombus [153,154].

RV findings include (see "Echocardiographic assessment of the right heart"):

●Increased RV size

●Decreased RV function

●Tricuspid regurgitation

●Abnormal septal wall motion

●McConnell sign

●Decreased tricuspid annular plane systolic excursion

In most cases, echocardiography is generally considered insensitive for diagnosing PE since abnormalities are frequently absent. Data suggest that only 30 to 40 percent of patients with PE have RV strain on echocardiography [155-158]. In addition, echocardiography is nonspecific since RV abnormalities can be seen in other conditions including chronic pulmonary disease, PH, and RV infarction. One meta-analysis reported a diagnostic sensitivity of 53 percent and specificity of 61 percent [151]. Regional wall motion abnormalities that spare the RV apex (McConnell sign) are also insensitive (77 percent), but it may be used to distinguish patients with RV strain due to acute PE from those with PH who tend to have global RV dysfunction [159].

When present, there appears to be a direct correlation between the extent of RV dysfunction and the degree of perfusion defects on lung scans [149,150]. In keeping with this observation, echocardiography is best used in patients with confirmed PE to identify those at greatest risk of death from PE. These details are discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Assess mortality risk (low, intermediate, high)' and "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Prognosis'.)

Investigational — Dual-energy CT, single-photon emission CT (SPECT), and multiorgan ultrasound are being developed as imaging tools to accurately and more safely diagnose PE.

●Dual-energy CT – Dual-energy CT could reduce the amount of iodinated contrast needed to perform CTPA examinations and increase the sensitivity for PE by imaging an iodine map, which serves as a surrogate for lung perfusion [160]. Large cohort studies have not yet been reported.

●SPECT – Technological advances in SPECT ventilation and perfusion imaging may allow for accurate diagnosis of PE without iodinated contrast administration. Other forms of SPECT imaging (perfusion SPECT, V/Q SPECT, SPECT or V/Q SPECT with non-enhanced CT) also have a lower radiation dose than CTPA. Another potential advantage is that SPECT may increase the detection of smaller PE. However, the optimal SPECT technique is unknown and its use may be limited to centers with the technological expertise to perform it. Preliminary studies suggest that SPECT is more sensitive than CTPA and V/Q scanning [161-164].

●Multiorgan ultrasound – Multiorgan ultrasonography (ultrasound of the heart, lung, and lower extremity) was prospectively examined in 357 patients suspected of having PE (Wells score >4) [165]. Compared with CTPA, a sensitivity of 90 percent and specificity of 86 percent were reported.

LOOKING FOR THE SOURCE — 

In most cases, we do not routinely image the extremities for the source based upon the assumption that most PEs are derived from the lower extremities and should resolve with therapy.

However, a select population may need testing. Indications for imaging are unknown and largely dependent upon the clinical suspicion for pathology that may need additional intervention. Examples include massive deep vein thrombosis that may need thrombolysis and extrinsic compression/mechanical obstruction (eg, pelvic or upper extremity tumor, May Thurner syndrome). The detection of right-sided intracardiac thrombus may also prompt a search for thrombus in the inferior vena cava by CT. (See "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors" and "Overview of thoracic central venous obstruction" and "May-Thurner syndrome" and "Overview of iliocaval venous obstruction".)

PATIENTS WITH SUSPECTED RECURRENT PULMONARY EMBOLISM — 

The approach to patients with suspected recurrent PE (days to years) should be the same as for a first suspected event with some minor differences:

●In hemodynamically stable patients with suspected recurrence, D-dimer level is less likely to be negative, especially if the initial PE was recent. However, D-dimer can still be useful in a limited proportion (<15 percent) to distinguish those who should have imaging from those who should not. (See 'D-dimer' above.)

●Many patients present with similar symptoms to their initial PE. Prior imaging should be obtained when feasible, but should not delay empiric treatment when indicated. While previous imaging may help distinguish new from old thrombus, the interpretation of repeat imaging may be difficult since thrombus can migrate with time and the resolution rates are variable [166,167]. As examples:

•In a cohort of 79 patients with acute PE receiving anticoagulant therapy, complete resolution occurred in 40 percent of patients within one week, 50 percent within two weeks, 73 percent within four weeks, and 81 percent by four weeks or longer [166]. Resolution was quicker in larger (main and lobar) pulmonary arteries than smaller (segmental and subsegmental) vessels, particularly during the first week.

•Another cohort of 111 patients with acute PE reported similar results, but thrombus resolved more quickly in peripheral than larger pulmonary arteries [167].

Management of suspected recurrence is discussed separately. (See "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Management of recurrence on therapy'.)

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions worldwide are provided separately. (See "Society guideline links: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Pulmonary embolism (blood clot in the lung) (The Basics)")

●Beyond the Basics topics (see "Patient education: Pulmonary embolism (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical features – Pulmonary embolism (PE) has a wide variety of presenting features, ranging from no symptoms to shock or sudden death (table 1). (See 'Clinical presentation' above.)

•The most common presenting symptoms are (see 'History and examination' above):

-Dyspnea (at rest or on exertion); often sudden onset but may develop over days

-Chest pain, especially pleuritic pain; may be a dull ache

-Cough

Hemoptysis, orthopnea, and wheezing may also occur along with leg pain or swelling. Less common are arrhythmias (usually atrial fibrillation), syncope, and shock.

Common examination findings include tachypnea, tachycardia, and leg swelling.

•We obtain the following studies (see 'Laboratory tests' above and 'Electro- or echocardiography' above and 'Chest radiograph' above):

-Routine laboratory tests – Complete blood count, chemistries and kidney function, liver function tests, and coagulation studies (activated prothrombin time and international normalized ratio)

-D-dimer, brain natriuretic peptide, and troponin level

-Electrocardiography (sinus tachycardia, atrial fibrillation, S1Q3T3, right ventricular strain or block)

-Chest radiography (atelectasis, pleural effusion)

We assess oxygenation with peripheral pulse oximetry and may obtain a venous or arterial blood gas. Hypoxemia, widened alveolar-arterial oxygen gradient, and respiratory alkalosis are common in PE. Some patients have a normal blood gas.

Findings in PE are typically nonspecific; however, these tests may aid in differential diagnosis and in risk stratification. An elevated D-dimer is not diagnostic of PE (table 5).

•Many conditions present similarly including:

-Heart failure

-Myocardial ischemia

-Pneumothorax

-Pneumonia

-Pericarditis

-Musculoskeletal pain

In most cases, clinical features and initial testing distinguish these from PE. However, these can be comorbid with PE and the presence of an alternate diagnosis does not exclude PE. (See 'Differential diagnosis' above.)

●Hemodynamically unstable patients – (See 'Hemodynamically unstable patients (high-risk PE)' above.)

•For patients with a high clinical suspicion for PE who are hemodynamically unstable, we obtain expert consultation. Once successfully resuscitated, we immediately perform definitive diagnostic imaging and initiate empiric anticoagulation. (See 'Hemodynamic stability restored following resuscitation' above and "Acute pulmonary embolism in adults: Treatment overview and prognosis", section on 'Empiric anticoagulation or thrombolysis'.)

•For patients with a low or moderate suspicion of PE (table 2) (calculator 1) who are successfully resuscitated, we use the same approach to diagnosis and empiric anticoagulation as for hemodynamically stable patients. (See 'Hemodynamically stable patients' above.)

•For patients who remain unstable despite resuscitation, we perform bedside echocardiography and lower extremity compression ultrasonography (CUS) with Doppler for a presumptive diagnosis of PE to justify the administration of potentially life-saving therapies. (See 'Hemodynamically unstable patients (high-risk PE)' above and 'Electro- or echocardiography' above.)

●Hemodynamically stable patients – In this population, we estimate the pretest probability (PTP) for PE by clinical gestalt or a validated instrument, such as the Wells criteria (table 2) (calculator 1), and categorize the PTP as low (eg, Wells score <2), intermediate (eg, Wells score 2 to 6), or high (eg, Wells score >6).

•In nonhospitalized patients with a low gestalt PTP, we apply the PE rule-out criteria (PERC) (calculator 3) (table 4 and algorithm 3). In patients who fulfill all eight criteria, PE is excluded (<1 percent). Some experts measure D-dimer in all low-probability patients as an alternative. (See 'Low probability of pulmonary embolism' above and 'Nonhospitalized patients: PERC rule' above.)

•For patients with a low PTP who do not fulfill PERC criteria or in whom PERC cannot be applied (eg, hospitalized patients), obtaining high-sensitivity D-dimer testing is reasonable. PE is excluded (<2 percent) when the D-dimer level is negative (<500 ng/mL [fibrinogen equivalent units]), and we obtain diagnostic imaging when the D-dimer is positive (≥500 ng/mL [fibrinogen equivalent units]). Adjusted D-dimer is an alternative that may decrease the use of imaging but may also miss more cases of PE. (See 'Hospitalized patients' above.)

•In patients with an intermediate PTP (eg, Wells score 2 to 6), we obtain high-sensitivity D-dimer testing (algorithm 2). PE is excluded (<2 percent) when the D-dimer level is negative (<500 ng/mL [fibrinogen equivalent units]), and we obtain diagnostic imaging when the D-dimer is positive (≥500 ng/mL [fibrinogen equivalent units]). Diagnostic imaging without D-dimer testing is a reasonable alternative in patients with limited cardiopulmonary reserve or when PTP is in the upper intermediate range (ie, Wells score 4 to 6). (See 'Intermediate probability of pulmonary embolism' above.)

•In patients with a high PTP (eg, Wells score >6), we obtain diagnostic imaging with CT pulmonary angiography (CTPA). CTPA is sensitive and specific for PE, especially when incorporated into diagnostic algorithms (algorithm 1). It also identifies alternate diagnoses. A positive result (filling defect or abrupt vessel cutoff) confirms the diagnosis of PE while a negative result excludes significant PE in nearly all cases (<5 percent) (image 8 and image 6). (See 'High probability of pulmonary embolism' above and 'CTPA' above.)

●Alternate imaging approaches

•Ventilation-perfusion (V/Q) scanning – V/Q scanning is appropriate when CTPA is contraindicated, unavailable, or inconclusive. V/Q scan results are reported as high-, intermediate- (image 6), or low-probability (image 5); or normal. V/Q scan should be interpreted in conjunction with PTP (table 6) (see 'V/Q scan' above):

-A high-probability V/Q scan and high PTP confirm PE

-A normal or a low-probability scan and low PTP excludes PE

-Other combinations of V/Q results and PTP necessitate additional testing

•Others – Options for additional testing include lower-extremity CUS with Doppler and/or echocardiography; neither diagnose PE but can increase PTP for the diagnosis. Magnetic resonance pulmonary angiography is less sensitive than CTPA. Catheter angiography is invasive and usually reserved for therapeutic intervention (image 7). (See 'Other imaging' above.)

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Charles Hales, MD, now deceased, who contributed to earlier versions of this topic review.