Background: Helical computed tomography (CT) is a readily available tool for diagnosing pulmonary embolism (PE); however, its role in the management of patients with clinically suspected PE has not been fully established.
Objective: To determine the effectiveness and safety of using helical CT of the pulmonary arteries as the primary diagnostic test in patients with suspected PE.
Design: Multicenter, prospective clinical outcome study.
Setting: Two academic hospitals and one large teaching hospital in the Netherlands.
Patients: 510 consecutive inpatients and outpatients with clinically suspected PE followed for 3 months.
Interventions: Patients underwent helical CT of the pulmonary arteries within 24 hours after presenting with signs and symptoms of PE. If CT results were normal or inconclusive, compression ultrasonography was performed on the same day as CT and repeated on days 4 and 7 if findings on the first compression ultrasonography were normal. When CT or compression ultrasonography results were positive for thromboembolism, anticoagulation was started. Anticoagulation was not started when results of CT were negative for PE or indicated an alternative diagnosis that explained the clinical signs and symptoms, or when results on serial compression ultrasonography were normal.
Measurements: Patients received instructions to report any symptoms or signs of PE or deep venous thrombosis (DVT) during the 3-month follow-up period. The authors performed compression ultrasonography or phlebography for suspected DVT and pulmonary angiography for suspected PE.
Results: Computed tomography identified PE in 124 of 510 patients (24.3%) and an alternative diagnosis in 130 patients (25.5%); CT scans were normal in 248 patients (48.6%). The CT scan could not be interpreted in 8 patients (1.6%) and was not obtained in 2. Compression ultrasonography revealed DVT in 2 patients at the first examination; findings on repeated compression ultrasonography at days 4 and 7 were normal. Mortality in the patients with normal helical CT scans was 4.1% (10 of 246 patients). No patients in this group died of fatal PE, 1 patient developed nonfatal PE, and venous thromboembolism occurred in 0.4% of these patients (95% CI, 0% to 2.2%). In the patients with alternative diagnoses, 1 patient had DVT on objective testing during follow-up. Mortality in this group was 21.5% (28 of 130 patients); in 1 of these patients, PE could not be confidently ruled out as a contributing cause of death. Venous thromboembolism occurred in 1.5% of these patients (CI, 0.2% to 5.6%).
Conclusions: In patients with suspected PE, helical CT can be used safely as the primary diagnostic test to rule out PE. Serial compression ultrasonography has limited additional value.