The diagnosis of PE should incorporate clinical and pretest probability assessment, D-dimer testing, and imaging. Because most common symptoms include dyspnea, chest pain, and cough, other conditions such as pericarditis, acute myocardial infarction, acute aortic syndrome, and pleuritis should be considered in the differential diagnosis. Of note, the three most prevalent diagnoses associated with high risk of adverse outcomes in patients presenting with nontraumatic chest pain include acute coronary syndrome, thoracic aortic dissection, and PE. Accordingly, these three need to be considered in any individual presenting with chest pain and dyspnea.⁸ When imaging is indicated (ie, the patient is suspected of having a moderate or high probability of PE), computed tomographic pulmonary angiography (CTPA) is the imaging modality of choice with ventilation perfusion (V/Q) scanning used less often. For hemodynamically unstable patients, bedside echocardiography or venous ultrasound may be used for presumptive diagnosis of PE.

The combination of clinical symptoms, signs, and predisposing factors for thromboembolism helps categorize patients with suspected PE into low-, intermediate-, and high-risk categories for the likelihood of PE. The most frequently used
prediction rules are the revised Geneva rule (Table 85.1) and the Wells rule to assess the probability of PE.

Using the Geneva score, the proportion of patients with confirmed PE can be expected to be ˜10% in the low-probability category, 30% in the moderate-probability category, and 65% in the high-probability category.¹¹

For the Wells score, 3 points are assigned to clinical symptoms of deep vein thrombosis (DVT), 3 points if PE is the most likely diagnosis, 1.5 points for heart rate > 100 beats/min, 1.5 points for immobilization or surgery within 4 weeks, 1.5 points for previous DVT or PE, 1 point for hemoptysis, and 1 point for malignancy. Note that all the criteria can be obtained from the history and physical examination: heart rate is the only parameter that is measured. Everything else comes from history. The modified Wells criteria for PE categorizes suspected patients with PE to less than 2 as low risk, 2-6 as intermediate risk, and greater than 6 as high risk.¹² The simplified Wells criteria define PE as unlikely if the patient has a score of less than or equal to 4 whereas PE is likely for those with a score greater than 4. A further simplified Wells rule assigns one weight for the presence of all individual variables in the model with a score between 0 and 7. One study suggested that this simplification of the Wells rule by assigning only one point to each of the seven variables had similar diagnostic accuracy and clinical utility as the original Wells criteria.¹³

The Wells score has a slightly lower sensitivity (81% vs 84%) and a higher specificity (82% vs 77%) compared with the revised Geneva score. In the high-probability group, the positive predictive value for the Wells score is reported to be superior to that of the revised Geneva score (80% vs 58%).¹⁴

Electrocardiography is fairly nonspecific, and findings may include sinus tachycardia, anterior precordial T wave inversion, an S1Q3T3 pattern (ie, a deep S wave in lead I with Q wave and T wave inversion in lead III), and precordial ST-segment elevation.

The chest radiography is neither sensitive nor specific for PE. It is more useful in assessing for differential diagnostic possibilities such as pneumonia and pneumothorax rather than establishing the diagnosis of PE. Nevertheless, described chest radiographic signs include enlarged pulmonary artery; Hampton hump (ie, peripheral wedge of airspace opacity owing to lung infarction); Westermark sign (ie, regional oligemia); pleural effusion (35%); enlarged right descending pulmonary artery with sudden cutoff; and elevated diaphragm.

Performing definitive diagnostic imaging in every patient with dyspnea or pleuritic chest pain will lead to unnecessary testing with inherent risks and costs so a targeted approach is indicated. The pulmonary embolism rule-out criteria (PERC) were developed for patients presenting to the emergency department for selecting patients whose likelihood of having PE is so low that diagnostic imaging should not be considered.¹⁵ The criteria include eight clinical variables significantly associated with an absence of PE: age less than 50 years; pulse less than 100 beats/min; arterial oxygen saturation greater than 94%; no unilateral leg swelling; no hemoptysis; no recent trauma or surgery; no history of thromboembolism; and no oral hormone use.¹⁵ A prospective validation study and a randomized management study demonstrated that PE can be safely excluded in patients with low clinical probability who met all criteria of the PERC rule.¹⁶,¹⁷

Serum D-dimer levels are elevated in acute thrombosis because of activation of coagulation and fibrinolysis. The negative predictive value of D-dimer testing is high, with a normal serum D-dimer level making the diagnosis of PE quite unlikely. On the other hand, the positive predictive value of elevated serum D-dimer levels is low; hence, an elevated D-dimer is not useful for confirmation of PE. Of note, the specificity of D-dimer in establishing the diagnosis PE decreases with age to ˜10% in patients greater than 80 years of age, and the use of age-adjusted cutoff levels is recommended to improve the performance of D-dimer testing in the elderly.¹⁸ A prospective study evaluated an age-adjusted cutoff for serum D-dimer levels
(age × 10 µg/L, for patients aged >50 years) in 3346 patients and reported that its use increased the number of patients in whom PE could be excluded from 6.4% to 30%, without additional false-negative findings, thus avoiding many unnecessary imaging tests.¹⁹ The 2020 European Society of Cardiology (ESC) Guidelines for the diagnosis and management of acute PE specifically recommends that serum D-dimer cutoff values adjusted for age or adapted to clinical probability can be used as an alternative to fixed D-dimer cutoff values⁴ to better avoid unnecessary imaging.

CTPA is the diagnostic modality of choice for patients with suspected PE, as it allows excellent visualization of the pulmonary arteries down to the subsegmental level with a sensitivity of 83% and a specificity of 96% for PE diagnosis using older four-detector scans.²⁰ V/Q lung scintigraphy is an older established diagnostic test for suspected PE and is less often used these days. Compared to CTPA, V/Q scans use lower radiation and no iodinated contrast. Thus, it may be preferred in outpatients with a low clinical probability and a normal chest radiography, in young (especially female) patients, in pregnant individuals, in patients with a history of contrast-induced anaphylaxis, and patients with severe kidney disease.²¹ Finally, invasive pulmonary angiography—once the gold standard for the diagnosis of PE—is now rarely used because less-invasive CTPA offers similar diagnostic accuracy and information and is associated with significantly lower procedural risk.²²

Acute PE may lead to RV pressure overload and dysfunction, which can be detected by echocardiography and may be considered in patients with suspected PE and hemodynamic instability. Echocardiography has a reported negative predictive value of 40% to 50%; hence, a negative result does not exclude PE.²³ The combination of a pulmonary ejection acceleration time less than 60 ms with a peak systolic tricuspid valve gradient less than 60 mm Hg (the “60/60” sign), or depressed contractility of the RV free wall compared to the RV apex—known as the McConnell sign—is suggestive of PE but is present in only 12% to 20% of patients with PE.²⁴

PE typically originates from lower-limb thrombosis, and venous ultrasound showing proximal DVT in patients suspected of having PE should be adequate grounds for anticoagulation therapy. In patients with suspected PE and hemodynamic instability who are unable to undergo CTPA or a V/Q scan, venous ultrasound and/or echocardiography may help guide diagnostic and therapeutic choices.