Computed tomography for suspected pulmonary embolism results in a large number of non-significant incidental findings and follow-up investigations

Anjum O, Bleeker H, Ohle R. Computed tomography for suspected pulmonary embolism results in a large number of non-significant incidental findings and follow-up investigations

Emerg Radiol 2019;26:29-35

Commentary

Diagnostic algorithms for suspected pulmonary embolism (PE) aim at safely excluding PE without the need for radiological imaging in as much as possible patients, to reduce costs, exposure to ionizing radiation and potential complications of contrast administration. In this paper of the month, Canadian researchers report of another complication of performing CT pulmonary angiography (CTPA): incidental findings. The investigators screened the radiological reports of 1708 CTPA scans performed in two academic urban tertiary care emergency wards in 2015. Of all evaluated patients, 14% was diagnosed with acute PE. The CTPA revealed an alternative diagnosis to explain the clinical presentation in 10% of patients, mostly pneumonia. Notably, 77% of these patients had the same non-PE diagnosis on chest radiograph reports. Incidental findings were identified in as much as 13% of patients, both in those with PE confirmed and with PE ruled out. These included mostly pulmonary nodules, pulmonary masses, mediastinal masses and adenopathy. The majority of incidental findings (68%) resulted in additional imaging test, for an average of 1.3% additional tests per incidental finding. Overall, cancer was confirmed in 12% of patients with an incidental finding. The main limitations of this study were the retrospective design, lack of patient-level outcome data and the lack of adjudication of the original CTPA reports. Moreover, the authors do not report details of the clinical presentation and full diagnostic management of the studied patients. It would for instance be very interesting how many patients with relevant or ‘irrelevant’ incidental findings could have been managed without CTPA, i.e. had a non-high pre-test probability and a normal D-dimer level. Further, information on other diagnostic tests performed (biomarkers, biopsies) were missing, as well as an assessment of costs and the psychosocial impact of an incidental finding on a patient.

What does this study teach us?  

Despite its methodological limitations, it clearly provides another strong argument to refrain from performing CTPA scans if the combination of a clinical decision rule and D-dimer testing can safely rule out acute PE.