Incidental findings on computed tomography angiography in patients evaluated for pulmonary embolism.

RATIONALE

To investigate how often computed tomography (CT) pulmonary angiography contributes to establishing a diagnosis in patients presenting to the emergency department and how it performs compared to chest radiograph.

OBJECTIVES

The objective of this study was to measure the ability to identify a diagnosis and to investigate the prevalence and significance of incidental findings in patients evaluated with computed tomography pulmonary angiography in the emergency department.

METHODS

All adult patients evaluated with CT angiography over a 2-year period (January 1, 2011 to December 31, 2012) were included in the analysis. A total of 641 records were identified. Chest radiographs and CT angiography reports were reviewed to determine whether they could provide a diagnosis in patients without pulmonary embolism (PE). Studies negative for PE were stratified into three categories according to significance: type I prompted immediate action, type II required follow up, and type III had findings of limited significance.

MEASUREMENTS AND MAIN RESULTS

CT angiography identified a diagnosis in 22.46% of the patient population and in 14.31% of patients without PE. In patients who had CT angiography with chest radiograph, diagnoses were provided in 14.01 and 9.86% of patients, respectively. When analysis was isolated to patients with low probability for PE, CT angiography provided a diagnosis in 20% and chest radiography in 10.23% of patients. The majority of missed cases represented infiltrates too small to be detected by radiography and were believed to represent lung infections by the interpreting radiologist. Among studies negative for PE, 15% were type I, 17.07% were type II, 48.1% were type III, and the rest were normal.

CONCLUSIONS

CT angiography is superior to chest radiography at providing a diagnosis in patients investigated for PE, even when no PE is present. However, in patients at low risk for PE, the clinical benefit of the additional diagnoses is questionable.