Pulmonary Embolism (PE) Workup

Clinical signs and symptoms for pulmonary embolism (PE) are nonspecific; therefore, patients suspected of having pulmonary embolism—because of unexplained dyspnea, tachypnea, or chest pain or the presence of risk factors for pulmonary embolism—must undergo diagnostic tests until the diagnosis is ascertained or eliminated or an alternative diagnosis is confirmed. Further, routine laboratory findings are nonspecific and are not helpful in pulmonary embolism, although they may suggest another diagnosis.

A hypercoagulation workup should be performed if no obvious cause for embolic disease is apparent. This may include screening for conditions such as the following:

• Antithrombin III deficiency

• Protein C or protein S deficiency

• Lupus anticoagulant

• Homocystinuria

• Occult neoplasm

• Connective tissue disorders

More clinical studies are needed to evaluate the utility of new approaches to the condition’s diagnosis. The availability of diagnostic tests, as well as cost-effectiveness analysis, local traditions, and the expertise of radiologists involved in the diagnosis, are considerations in the workup of a patient in whom pulmonary embolism is suspected.

Evidence-based literature supports the practice of determining the clinical probability of pulmonary embolism before proceeding with testing. ^[3] One study assessed the performance of four4 clinical decision rules in addition to D-dimer testing to exclude acute PE. All four4 rules, Wells rule, simplified Wells rule, revised Geneva score, and simplified revised Geneva score, showed similar performance for excluding acute PE when combined with a normal D-dimer result. ^[46]

See the Guidelines section and the article Pulmonary Embolism Clinical Scoring Systems.

When clinical prediction rule results indicate that the patient has a low or moderate pretest probability of pulmonary embolism, D-dimer testing may be the next step. ^[3]

D-Dimer, a degradation product produced by plasmin-mediated proteases of cross-linked fibrin, is measured by a variety of assay types, including quantitative, semiquantitative, and qualitative rapid enzyme-linked immunosorbent assays (ELISAs); quantitative and semiquantitative latex; and whole-blood assays. A systematic review of prospective studies of high methodologic quality concluded that the ELISAs—especially the quantitative rapid ELISA—dominate the comparative ranking among the D-dimer assays for sensitivity and negative likelihood ratio. ^[47] The quantitative rapid ELISA has a sensitivity of 0.95 and negative likelihood ratio of 0.13; the latter is similar to that for a normal to near-normal lung scan in patients with suspected pulmonary embolism.

Negative results on a high-sensitivity D-dimer test in a patient with a low pretest probability of pulmonary embolism indicate a low likelihood of venous thromboembolism and reliably exclude pulmonary embolism. A large, prospective, randomized trial found that in patients with a low probability of pulmonary embolism who had negative D-dimer results, forgoing additional diagnostic testing was not associated with an increased frequency of symptomatic venous thromboembolism during the subsequent 6 months. ^[48]

In a 2012 prospective cohort study, a Wells score of 4 or less combined with a negative qualitative D-dimer test was shown to safely exclude pulmonary embolism in primary care patients. ^[49]

D-dimer testing is most reliable for excluding pulmonary embolism in younger patients who have no associated comorbidity or history of venous thromboembolism and whose symptoms are of short duration. ^[4] However, it is of questionable value in patients who are older than 80 years, who are hospitalized, who have cancer, or who are pregnant, because nonspecific elevation of D-dimer concentrations is common in such patients.

D-dimer testing should not be used when the clinical probability of pulmonary embolism is high, because the test has low negative predictive value in such cases. ^[50]

Combining D-dimer results with measurement of the exhaled end-tidal ratio of carbon dioxide to oxygen (etCO₂/O₂) can be useful for diagnosis of pulmonary embolism. Kline et al found that, in moderate-risk patients with a positive D-dimer (>499 ng/mL), an etCO₂/O₂< 0.28 significantly increased the probability of finding segmental or larger pulmonary embolism on computed tomography multidetector-row pulmonary angiography, while an etCO₂/O₂) >0.45 predicted the absence of segmental or larger pulmonary embolism. ^[51]

Because of the poor specificity, positive D-dimer measurements are not helpful in confirming the diagnosis of venous thromboembolic disease. However, a positive D-dimer measurement may lead to consideration of venous thromboembolic disease in the differential diagnosis in selected patients. In addition, the use of D-dimers in children is not well studied. A small pediatric series reported that D-dimer measurements are negative in 40% of patients. ^[29] A retrospective series reported an elevated D-dimer in 86% of patients at presentation. ^[17]

A potential alternative to D-dimer testing is assessment of the ischemia-modified albumin (IMA) level, which data suggest is 93% sensitive and 75% specific for pulmonary embolism. ^[52] Notably, in a study comparing the prognostic value of IMA to D-dimer testing, IMA assessment in combination with Wells and Geneva probability scores appeared to positively impact overall sensitivity and negative predictive value. ^[52] The positive predictive value of IMA, in particular, is better than D-dimer. However, it should not be used alone. ^[53]

The white blood cell (WBC) count may be normal or elevated in patients with pulmonary embolism, with a WBC count as high as 20,000 being not uncommon in patients with this condition.

Arterial blood gas determinations characteristically reveal hypoxemia, hypocapnia, and respiratory alkalosis; however, the predictive value of hypoxemia is quite low. The PaO₂ and the calculation of alveolar-arterial oxygen gradient contribute to the diagnosis in a general population thought to have pulmonary embolism. Nonetheless, in high-risk settings such as patients in postoperative states in whom other respiratory conditions can be ruled out, a low PaO₂ in conjunction with dyspnea may have a strong positive predictive value.

The PO₂ on arterial blood gases analysis (ABG) has a zero or even negative predictive value in a typical population of patients in whom pulmonary embolism is suspected clinically. This is contrary to what has been taught in many textbooks, and even though it seems counterintuitive, it is demonstrably true. This is because other etiologies that masquerade as pulmonary embolism are more likely to lower the PO₂ than pulmonary embolism. In fact, because other diseases that may masquerade as pulmonary embolism (eg, chronic obstructive pulmonary disease [COPD], pneumonia, CHF) affect oxygen exchange more than does pulmonary embolism, the blood oxygen level often has an inverse predictive value for pulmonary embolism.

In most settings, fewer than half of all patients with symptoms suggestive of pulmonary embolism actually turn out to have pulmonary embolism as their diagnosis. In such a population, if any reasonable level of PaO₂ is chosen as a dividing line, the incidence of pulmonary embolism will be higher in the group with a PaO₂ above the dividing line than in the group whose PaO₂ is below the divider. This is a specific example of a general truth that may be demonstrated mathematically for any test finding with a Gaussian distribution and a population incidence of less than 50%.

Conversely, in a patient population with a very high incidence of pulmonary embolism and a lower incidence of other respiratory ailments (eg, postoperative orthopedic patients with sudden onset of shortness of breath), a low PO₂ has a strongly positive predictive value for pulmonary embolism.

The discussion above holds true not only for arterial PO₂ but also for the alveolar-arterial oxygen gradient and for the oxygen saturation level as measured by pulse oximetry. In particular, pulse oximetry is extremely insensitive, is normal in the majority of patients with pulmonary embolism, and should not be used to direct a diagnostic workup.

Serum troponin levels can be elevated in up to 50% of patients with a moderate to large pulmonary embolism, presumptively due to acute right ventricular myocardial stretch. ^[50]

Although troponin assessment is not currently recommended as part of the diagnostic workup, studies have shown that elevated troponin levels in the setting of pulmonary embolism correlate with increased mortality. ^[54] However, further studies need to be performed to identify subsets of patients with pulmonary embolism who might benefit from this testing.

A meta-analysis by Jimenez et al suggested that in acute symptomatic pulmonary embolism, elevated troponin levels do not distinguish between patients who are at high risk for death and those who are at low risk. Pooled results from studies including 1366 normotensive patients with acute symptomatic pulmonary embolism showed that elevated troponin levels were associated with a 4.26-fold increased odds of overall mortality (95% confidence interval [CI], 2.13-8.50; heterogeneity chi² = 12.64; degrees of freedom = 8; P = .125). Summary receiver operating characteristic curve analysis showed a relationship between the sensitivity and specificity of troponin levels to predict overall mortality (Spearman rank correlation coefficient = 0.68; P = .046). Pooled likelihood ratios (LRs) were not extreme (negative LR, 0.59 [95% CI, 0.39-0.88]; positive LR, 2.26 [95% CI,1.66-3.07]). ^[55]

Serum troponin, although seemingly marginal for purposes of diagnosis of pulmonary embolism, may contribute significantly to the ability to stratify patients by risk for short-term death or adverse outcome events when they reach the ED. In patients with pulmonary embolism and normal blood pressure specifically, elevated serum troponin level has been associated with right ventricular overload. ^{[54, 56, 57, 58]}

Leptin is another cardiovascular risk factor that may be associated with outcome in acute pulmonary embolism. Dellas et al conducted a prospective analysis of 264 patients with acute pulmonary embolus and found that serum leptin levels were inversely associated with the risk of adverse outcomes. Further study will be needed to confirm these findings and determine the clinical utility of leptin measurement. ^[59]

Although brain natriuretic peptide (BNP) tests are neither sensitive nor specific, patients with pulmonary embolism tend to have higher BNP levels. BNP testing had a sensitivity and specificity of only 60% and 62%, respectively, in a case-control study of 2213 hemodynamically stable patients with suspected acute pulmonary embolism. ^[60]

Elevated levels of BNP or of its precursor, N -terminal pro-brain natriuretic peptide (NT-proBNP), do correlate with an increased risk of subsequent complications and mortality in patients with acute pulmonary embolism. One meta-analysis revealed that patients with a BNP level greater than 100 pg/mL or an NT-proBNP level greater than 600 ng/L had an all-cause in-hospital mortality rate 6- and 16-fold higher than those below these cutoffs, respectively. ^[36] In a second smaller observational study, serum BNP levels greater than 90 pg/mL were associated with a higher rate of complications, such as the need for cardiopulmonary resuscitation, need for mechanical ventilation, need for vasopressor therapy, and death. ^[61]

BNP testing is not currently recommended as part of the standard evaluation of acute pulmonary embolism, and future studies may aid in defining its role in this setting.

Elevated levels of brain-type natriuretic peptides (BNP) may also provide prognostic information. ^[57] A meta-analysis demonstrated a significant association between elevated N-terminal–pro-BNP (NT-pro-BNP) and right ventricular function in patients with pulmonary embolism (P< .001), leading to an increased risk for complicated in-hospital course (odds ratio [OR] 6.8; 95% confidence interval [CI], 9.0-13) and 30-day mortality (OR 7.6; 95% CI, 3.4-17). ^[62] Importantly, increased NT-pro-BNP alone does not justify more invasive treatment.

A recent study by Scherz et al analyzed a large sample of patients hospitalized with acute pulmonary embolism. Hyponatremia at presentation was common and was associated with a higher risk of 30-day mortality and readmission. ^[63]

Venography is the criterion standard for diagnosing DVT. With the advent of noninvasive imaging, it has become less common in pediatric and adult practice.

Pulmonary angiography is the historical criterion standard for the diagnosis of pulmonary embolism. Following injection of iodinated contrast, anteroposterior, lateral, and oblique studies are performed on each lung. Positive results consist of a filling defect or sharp cutoff of the affected artery (as shown in the image below). Nonocclusive emboli are described as having a tram-track appearance. Abnormal findings on V/Q scans performed prior to angiography guide the operator to focus on abnormal areas. Angiography generally is a safe procedure. The mortality rate for patients undergoing this procedure is less than 0.5%, and the morbidity rate is less than 5%. Patients who have long-standing pulmonary arterial hypertension and right ventricular failure are considered high-risk patients. Negative pulmonary angiogram findings, even if false negative, exclude clinically relevant pulmonary embolism.

A pulmonary angiogram shows the abrupt termination of the ascending branch of the right upper-lobe artery, confirming the diagnosis of pulmonary embolism.

View Media Gallery

If multidetector-row computed tomography angiography (MDCTA) is unavailable, conduct pulmonary angiography. Long the criterion standard for pulmonary embolism diagnosis, pulmonary angiography is nevertheless more invasive and harder to perform than MDCTA, and for these reasons, it is rapidly being replaced. However, pulmonary angiography remains a useful diagnostic modality when MDCTA cannot be performed.

When pulmonary angiography has been performed carefully and completely, a positive result provides virtually a 100% certainty that an obstruction to pulmonary arterial blood flow exists. A negative pulmonary angiogram provides a greater than 90% certainty for the exclusion of pulmonary embolism.

A positive angiogram is an acceptable endpoint no matter how abbreviated the study. However, a complete negative study requires the visualization of the entire pulmonary tree bilaterally. This is accomplished via selective cannulation of each branch of the pulmonary artery and injection of contrast material into each branch, with multiple views of each area. Even then, emboli in vessels smaller than third order or lobular arteries are not seen.

Small emboli cannot be seen angiographically, yet embolic obstruction of these smaller pulmonary vessels is very common when postmortem examination follows a negative angiogram. These small emboli can produce pleuritic chest pain and a small sterile effusion even though the patient has a normal V/Q scan and a normal pulmonary angiogram.

In most patients, however, pulmonary embolism is a disease of multiple recurrences, with large and small emboli already present by the time the diagnosis is suspected. Under these circumstances, the V/Q scan and the angiogram are likely to detect at least some of the emboli.

Pulmonary angiography demonstrates subsegmental vessels in more detail than does CT scanning, although the superimposition of the small vessels remains a limiting factor. As a result, the interobserver agreement rate for isolated subsegmental pulmonary embolism is only 45%.

The routine use of CT pulmonary angiography for the detection of pulmonary emboli has led to overdiagnosis of the condition, according to a recent study. Overdiagnosing pulmonary embolism has resulted in possible inappropriate treatment with anticoagulation, a leading cause of medication-related death. ^{[64, 65]}

Between 1998, when CT pulmonary angiography was introduced, and 2006, there was an 80% increase in the incidence of pulmonary embolism but little subsequent drop in deaths, which suggests that many of the extra emboli being detected are not clinically important. During this period, the detection rate rose from 62.1 to 112.3 per 100,000 US adults and US deaths from pulmonary embolism dropped from 12.3 to 11.9 per 100,000. ^{[64, 65]}

Technical advances in CT scanning, including the development of multidetector-array scanners, have led to the emergence of CT scanning as an important diagnostic technique in suspected pulmonary embolism. ^{[3, 44]} Contrast-enhanced CT scanning is increasingly used as the initial radiologic study in the diagnosis of pulmonary embolism, especially in patients with abnormal chest radiographs in whom scintigraphic results are more likely to be nondiagnostic.

Computed tomography angiography (CTA) is the initial imaging modality of choice for stable patients with suspected pulmonary embolism. ^{[66, 67, 68]} The American College of Radiology (ACR) considers chest CTA to be the current standard of care for the detection of pulmonary embolism. ^[69] A study by Ward et al determined that a selective strategy in which CTA is used after compression ultrasonography is cost-effective for patients with a high pretest probability of pulmonary embolism. ^[70] This strategy may reduce the need for CTA and help eliminate adverse effects associated with CTA.

Toward a goal of reducing unnecessary CTA and associated radiation exposure, Drescher et al studied the effect of implementing a computerized decision support system for pulmonary embolism evaluation in the ED. Before implementation, the rate of positive pulmonary embolism diagnosis for CTAs performed was 8.3%; after, the positivity rate rose to 12.7%. The positive yield would have been higher (16.7%) had emergency physicians adhered in all cases to the outcome of the decision support system; in 27% of cases they did not. ^[71]

Like pulmonary angiography, CT scanning shows emboli directly, but it is noninvasive, cheaper than pulmonary angiography, and widely available. CT scanning is the only test that can provide significant additional information related to alternate diagnoses ^[72] ; spiral (helical) CT scanning results may suggest an alternative diagnosis in up to 57% of patients. This is a clear advantage of CT scanning over pulmonary angiography or scintigraphy.

A study of multidetector computed tomography (MDCTA) for detection of right ventricular dysfunction in 457 patients with acute pulmonary embolism found reasonable correlation with echocardiography, the reference standard. The criterion selected, a right-to-left ventricular dimensional ratio of 0.9 or more at MDCTA, had 92% sensitivity for right ventricular dysfunction. ^[73] The combination of quantitative assessment of ventricular dimensions by CT and measurement of biomarkers may provide additional diagnostic accuracy for the presence of right ventricular dysfunction. ^[74]

Chest radiographs are abnormal in most cases of pulmonary embolism, but the findings are nonspecific. Common radiographic abnormalities include atelectasis, pleural effusion, parenchymal opacities, and elevation of a hemidiaphragm. The classic radiographic findings of pulmonary infarction include a wedge-shaped, pleura-based triangular opacity with an apex pointing toward the hilus (Hampton hump) or decreased vascularity (Westermark sign). These findings are suggestive of pulmonary embolism but are infrequently observed.

The abrupt tapering or cutoff of a pulmonary artery secondary to embolus (knuckle sign), cardiomegaly (especially on the right side of the heart), and pulmonary edema are other findings. In the appropriate clinical setting, these findings could be consistent with acute cor pulmonale. A normal-appearing chest radiograph in a patient with severe dyspnea and hypoxemia, but without evidence of bronchospasm or a cardiac shunt, is strongly suggestive of pulmonary embolism.

The ACR recommends chest radiography (see the images below) as the most appropriate study for ruling out other causes of chest pain in patients with suspected pulmonary embolism. ^[69] Initially, the chest radiographic findings are normal in most cases of pulmonary embolism. However, in later stages, radiographic signs may include a Westermark sign (dilatation of pulmonary vessels and a sharp cutoff), atelectasis, a small pleural effusion, and an elevated diaphragm. Generally, chest radiographs cannot be used to conclusively prove or exclude pulmonary embolism; however, radiography and electrocardiography may be useful for establishing alternative diagnoses. (See Electrocardiography.)

Posteroanterior and lateral chest radiograph findings are normal, which is the usual finding in patients with pulmonary embolism.

View Media Gallery

A chest radiograph with normal findings in a 64-year-old woman who presented with worsening breathlessness.

View Media Gallery

A posteroanterior chest radiograph showing a peripheral wedge-shaped infiltrate caused by pulmonary infarction secondary to pulmonary embolism. Hampton hump is a rare and nonspecific finding. Courtesy of Justin Wong, MD.

View Media Gallery

V/Q scanning of the lungs is an important modality for establishing the diagnosis of pulmonary embolism. V/Q scanning may be used when CT scanning is not available or if the patient has a contraindication to CT scanning or intravenous contrast material. Children generally have a more homogenous perfusion scan; thus, deficits in perfusion are more likely to represent real or significant pulmonary embolism than they are in adults.

The PIOPED II trial provided high-, intermediate-, and low-probability criteria for V/Q scanning diagnosis of pulmonary embolism (see the images of high-probability scans below).

High-probability perfusion lung scan shows segmental perfusion defects in the right upper lobe and subsegmental perfusion defects in right lower lobe, left upper lobe, and left lower lobe.

View Media Gallery

A normal ventilation scan will make the noted defects in the previous image a mismatch and, hence, a high-probability ventilation-perfusion scan.

View Media Gallery

Anterior views of perfusion and ventilation scans are shown here. A perfusion defect is present in the left lower lobe, but perfusion to this lobe is intact, making this a high-probability scan.

View Media Gallery

This perfusion scan shows bilateral perfusion defects. The ventilation scan findings were normal; therefore, these are mismatches, and this is a high-probability scan.

View Media Gallery

The high-probability criteria are as follows:

• Two large (>75% of a segment) segmental perfusion defects without corresponding ventilation or chest radiographic abnormalities

• One large segmental perfusion defect and two moderate (25-75% of a segment) segmental perfusion defects without corresponding ventilation or radiographic abnormalities

• Four moderate segmental perfusion defects without corresponding ventilation or chest radiographic abnormalities

• The intermediate-probability criteria are as follows:

• One moderate to fewer than two large segmental perfusion defects without corresponding ventilation or chest radiographic abnormalities

• Corresponding V/Q defects and radiographic parenchymal opacity in lower lung zone

• Single moderate matched V/Q defects with normal chest radiographic findings

• Corresponding V/Q and chest radiography small pleural effusion

• Difficult to categorize as normal, low, or high probability

Low probability criteria are as follows:

• Multiple matched V/Q defects, regardless of size, with normal chest radiographic findings

• Corresponding V/Q defects and radiographic parenchymal opacity in upper or middle lung zone

• Corresponding V/Q defects and large pleural effusion

• Any perfusion defects with substantially larger radiographic abnormality

• Defects surrounded by normally perfused lung (stripe sign)

• More than three small (< 25% of a segment) segmental perfusion defects with normal chest radiographic findings

• Nonsegmental perfusion defects (cardiomegaly, aortic impression, enlarged hila)

The normal finding is the presence of no perfusion defects and perfusion outlines the shape of the lung seen on a chest radiograph.

The very low–probability criterion is the presence of three small (< 25% of a segment) segmental perfusion defects with normal chest radiographic findings.

In the PIOPED II study, very low–probability V/Q scans in patients whose Wells score indicated low pretest probability of pulmonary embolism reliably excluded pulmonary embolism. ^[75]

The most common ECG abnormalities in the setting of pulmonary embolism are tachycardia and nonspecific ST-T wave abnormalities. The finding of S₁ Q₃ T₃ is nonspecific and insensitive in the absence of clinical suspicion for pulmonary embolism. The classic findings of right heart strain and acute cor pulmonale are tall, peaked P waves in lead II (P pulmonale); right axis deviation; right bundle-branch block; an S₁ Q₃ T₃ pattern; or atrial fibrillation. Unfortunately, only 20% of patients with proven pulmonary embolism have any of these classic electrocardiographic abnormalities. If electrocardiographic abnormalities are present, they may be suggestive of pulmonary embolism, but the absence of such abnormalities has no significant predictive value.

With magnetic resonance imaging (MRI), evidence of pulmonary emboli may be detected by using standard or gated spin-echo techniques. Pulmonary emboli demonstrate increased signal intensity within the pulmonary artery. By obtaining a sequence of images, this signal that is originating from slow blood flow may be distinguished from pulmonary embolism. However, this remains a problem in pulmonary hypertension.

Magnetic resonance angiography is performed following intravenous administration of gadolinium. Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or magnetic resonance angiography scans.

NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness.

MRI has a sensitivity of 85% and specificity of 96% for central, lobar, and segmental emboli; MRI is inadequate for the diagnosis of subsegmental emboli.

Few data are available regarding the use of MRI in children suspected of having a pulmonary embolism. Its use in these patients should be considered investigational at this time.

Few investigators have reported the feasibility of MRI in the evaluation of pulmonary embolism. However, the role of MRI is mostly limited to the evaluation of patients who have impaired renal function or other contraindications for the use of iodinated contrast material. ^{[76, 77]} Newer blood-pool contrast agents and respiratory navigators may enhance the role of MRI in the diagnosis of pulmonary embolism.

This modality generally has limited accuracy in the diagnosis of pulmonary embolism. Transesophageal echocardiography may identify central pulmonary embolism, and the sensitivity for central pulmonary embolism is reported to be 82%. Overall sensitivity and specificity for central and peripheral pulmonary embolism is 59% and 77%.

Echocardiography (ECHO) provides useful information. It may allow diagnosis of other conditions that may be confused with pulmonary embolism, such as pericardial effusion. ECHO allows visualization of the right ventricle and assessment of the pulmonary artery pressure. ECHO serves a prognostic function; the mortality rate is almost 10% in the presence of right ventricular dysfunction and 0% in the absence of right ventricular dysfunction. (Vanni et al reported that a right ventricular strain pattern is associated with a worse short-term outcome. ^[78] ) ECHO may be used to identify the presence of right-chamber emboli.

The subcostal view is preferred at initial screening for mechanical activity and pericardial fluid and for gross assessment of global and regional abnormalities. To obtain a subcostal view, place the transducer on the left subcostal margin with the beam aimed at the left shoulder.

The parasternal view allows visualization of the aortic valve, proximal ascending aorta, and posterior pericardium and permits determination of left ventricular size. It is particularly helpful when the subcostal view is difficult to obtain. To obtain a parasternal view, place the transducer in the left parasternal area between the second and fourth intercostal spaces. The plane of the beam is parallel to a line drawn from the right shoulder to the left hip.

Several echocardiographic findings have been proposed for noninvasive diagnosis of right ventricular dysfunction at the bedside, including right ventricular enlargement and/or hypokinesis of the free wall, leftward septal shift, and evidence of pulmonary hypertension. If right ventricular dysfunction is seen on cardiac ultrasonography, the diagnosis of acute submassive or massive pulmonary embolism is supported. While the presence of right ventricular dysfunction can be used to support the clinical suspicion of pulmonary embolism, prognostic information can be obtained by assessing the severity of right ventricular dysfunction.

The diagnosis of pulmonary embolism can be proven by demonstrating the presence of a DVT at any site. This may sometimes be accomplished noninvasively by using duplex ultrasonography. To look for DVT using ultrasonography, the ultrasonographic transducer is placed against the skin and pressed inward firmly enough to compress the vein being examined. In an area of normal veins, the veins are easily compressed completely closed, while the muscular arteries are extremely resistant to compression. Where DVT is present, the veins do not collapse completely when pressure is applied using the ultrasonographic probe.

A prospective observational study of 146 patients with suspected or confirmed pulmonary embolism indicates that identification of right ventricular dilatation on bedside echocardiography may aid diagnosis of pulmonary embolism. ^{[79, 80]} Bedside echocardiography showed right ventricular dilatation in 15 of the 30 patients who had pulmonary emboli, compared with 2 of the 116 patients without pulmonary emboli.

The presence of right ventricular dilatation on bedside echocardiography had a sensitivity of 50%, specificity of 98%, and positive and negative predictive values of 88% for the diagnosis of pulmonary embolism. ^{[79, 80]} Most of the 15 patients with confirmed pulmonary emboli and right ventricular dilatation had proximal clots, while most of those with confirmed pulmonary emboli and a normal right ventricular/left ventricular ratio had more distal clots. ^{[79, 80]}

Note that a negative ultrasonographic scan does not rule out DVT, because many DVTs occur in areas that are inaccessible to ultrasonographic examination. Before an ultrasonographic scan can be considered negative, the entire deep venous system must be interrogated using centimeter-by-centimeter compression testing of every vessel. In two thirds of patients with pulmonary embolism, the site of DVT cannot be visualized with ultrasonography, so a negative duplex ultrasonographic scan does not markedly reduce the likelihood of pulmonary embolism.

Ultrasonographic images of thrombi are seen below.

This ultrasonogram shows a thrombus in the distal superficial saphenous vein, which is under the artery.

View Media Gallery

Longitudinal ultrasound image of partially recanalized thrombus in the femoral vein at mid thigh.

View Media Gallery