Utilization of CT-PA in an emergency department with readily available V/Q scintigraphy

To describe the role of computed tomographic pulmonary angiography (CT-PA) in the emergency department of an institution which utilizes ventilation-perfusion (V/Q) scintigraphy as its primary imaging modality for the diagnosis of pulmonary embolism.

We retrospectively identified and reviewed the records of 24 consecutive patients evaluated in the emergency department from October 1998 to September 2000 who were suspected of having pulmonary embolism. The inclusion criterion was that the images from the emergency department and work-up included CT-PA, which was the initial imaging test after chest radiograph. There were 10 men and 14 women with a mean age of 63. Results of CT-PA were categorized as positive, negative, limited negative (no main or lobar pulmonary emboli but not all segmental arteries visualized), or nondiagnostic. Each chart was reviewed with reference to clinical presentation, relevant history, results of Doppler ultrasonography of the legs, V/Q scan and pulmonary angiography, and discharge diagnosis. During the same study period, approximately 400 V/Q scans were performed from the emergency department.

Each patient had a clinical presentation consistent with pulmonary embolism. CT-PA diagnosed pulmonary emboli in 21% (5/24), was negative in 33% (8/24), was limited negative in 38% (9/24), and was nondiagnostic in 8% (2/24). Chest radiographs were abnormal in 71% (17/24). V/Q scans were performed in 17% (4/24; 1 near normal, 2 low probability, 1 intermediate probability). None of these four patients was discharged with a diagnosis of pulmonary embolism. Doppler ultrasound leg exam was performed in 38% (9/24). Among the 5 patients diagnosed with pulmonary embolism, 1/3 examined had Doppler evidence of deep vein thrombosis (DVT). Among the 19 patients not diagnosed with pulmonary embolism, 3/6 examined had Doppler evidence of DVT. No patient with a negative or limited negative CT-PA was discharged with a diagnosis of pulmonary embolism. CT-PA provided alternative diagnoses explaining the patient's symptoms in 68% (13/19) of those not diagnosed with pulmonary embolism. During the most recent 12 months of the study period, 210 V/Q scans were performed from the emergency department, with results available in 194 cases as follows: normal/near normal 32% ( n=62), low probability 47% ( n=92), intermediate probability 14% ( n=28), high probability 6% ( n=12).

V/Q scintigraphy is the primary imaging modality for suspected pulmonary embolism in our emergency department. However, when utilized, CT-PA played an important role in patient management by confirming or excluding pulmonary embolism or providing an alternative diagnosis in the majority of patients suspected of having pulmonary embolism.