Because untreated pulmonary embolism (PE) can be rapidly fatal or cause permanent disability from pulmonary hypertension, an accurate and rapid diagnosis is essential. Deep venous thrombosis (DVT) and pulmonary embolism are closely associated. The approach to patients with signs and symptoms of PE needs to be developed from published research about new diagnostic modalities allowing the highest level of accuracy.
The American College of Emergency Physicians Clinical Policies Committee and the Clinical Policies Subcommittee on Suspected Pulmonary Embolism reviewed and analyzed the medical literature regarding the diagnosis of PE in adults. The committee placed emphasis on the following topics: (1) the diagnostic use of D-dimer, ventilation-perfusion (V/Q) lung scan, and spiral computed tomographic (CT) angiography in the evaluation of PE, and (2) therapeutic indications for fibrinolytic therapy in PE. Recommendations are classified by the strength of evidence contained in the reviewed clinical reports and research studies. Level A recommendations are strongly supported by good evidence and have a high degree of clinical certainty. Level B recommendations reflect a moderate level of clinical certainty. Level C recommendations are supported by preliminary, inconclusive, or conflicting evidence, or are based on panel consensus.
Initially, the pretest probability of PE can be determined by one of several sets of explicit criteria that divide patients into low-risk, moderate-risk, and high-risk categories (see accompanying table). The first critical question looked at the usefulness of a negative d-dimer test to exclude PE. Because D-dimer levels increase with fibrinolysis in the presence of endovascular thrombus, patients with a low probability by pretest determination who have a negative D-dimer result can be ruled out as having a PE (high sensitivity, relatively lower specificity) using a turbidimetric or enzyme-linked immunosorbent assay (ELISA) technique, or using whole blood cell qualitative assay with a Wells' pretest score of 2 or less (Level B recommendation). In patients with low probability by pretest determination, a negative whole blood d-dimer (when not used with a Wells' pretest score) or immunofiltration assay can exclude PE (Level C recommendation). V/Q scanning is frequently used but must be interpreted with careful consideration of pretest probability of PE.
The addition of duplex ultrasonography of the lower extremities to help identify the probability of PE in patients with nondiagnostic V/Q scan offers low sensitivity and does not exclude PE in patients with nondiagnostic V/Q scan and non-low pretest probability. A normal V/Q scan in patients with low-to-moderate pretest probability excludes clinically significant PE (Level A recommendation). In patients with low-to-moderate pretest probability who have a nondiagnostic V/Q scan, PE can be excluded by the following methods: a negative turbidimetric or ELISA d-dimer; a negative whole blood cell qualitative d-dimer assay in combination with a Wells' score of 4 or less; a negative single bilateral venous ultrasonographic scan; or a negative serial bilateral venous ultrasonographic scan (Level B recommendation). In patients with a low-to-moderate pretest probability of PE and a nondiagnostic V/Q scan, a negative whole blood d-dimer assay (when not used with Wells' scoring system) or immunofiltration d-dimer assay can exclude PE (Level C recommendation).
Spiral CT angiography with contrast is useful in patients who have a nondiagnostic V/Q scan, such as patients with cardiopulmonary disease, chronic obstructive pulmonary disease, or infiltrative lung disease. Newer CT scanners with 1- to 2-mm image reconstruction have a higher sensitivity and specificity for PE, allowing a negative CT angiogram to be used as an alternative to V/Q scan to exclude clinically significant PE (Level B recommendation). Spiral CT scan of the thorax with delayed CT venography may increase the detection of patients with significant thromboembolic disease (Level C recommendation).
Fibrinolytic therapy is most risky in the presence of diastolic hypertension when intracranial hemorrhage is more likely. It is useful only in patients who are hemodynamically unstable in whom PE is confirmed (Level B recommendation). Fibrinolytic therapy should be considered in the following patients: (1) hemodynamically stable patients with confirmed PE and right ventricular dysfunction on echocardiography, or (2) unstable patients with high clinical index suspicion of PE, especially if right ventricular dysfunction is demonstrated on echocardiography (Level C recommendation).