Yield of computed tomography pulmonary angiogram in the emergency department in cancer patients suspected to have pulmonary embolism

AbstractObjective/backgroundThe use of computed tomography pulmonary angiography (CTPA) in the emergency department (ED) for patients suspected to have pulmonary embolism (PE) has been steadily rising in the last 2decades. However, there are limited studies that specifically address the use of CTPA in the ED for cancer patients suspected to have PE. The objective of this study is to assess the rate of positive PE by CTPA in the ED in cancer patients and the variables that are associated with positive results.MethodsA retrospective review of electronic medical records for 208 consecutive patients with cancer who presented to the ED and received a CTPA for suspected PE over a 12-month period. The review included demographics, type and status of cancer, presenting symptoms, CTPA results, calculation of Wells Score, management based on CT findings, and outcome of patients.ResultsAmong the 208 patients who met the inclusion criteria during our study period (mean age 57±13.37years, 73% women, 59% African American, and 32% Caucasians), 5.7% were diagnosed with PE. One hundred and eighty-two (83.7%) had a Wells Score ⩽4, of which 2.2% were found to have to have PE, 22 (16.3%) patients had a Wells Score >4, of which 36.4% were found to have PE (p<.0001). Sensitivity and specificity of Wells >4 was 66.7% and 92.9%, respectively, with an odds ratio of 27 (95% CI 6.6–113.6). Receiver operator characteristics area under the curve for Wells Score was 0.868. Age, race, sex, malignancy type, stage, status, clinical presentation, D-dimer, and a previous history of venous thromboembolism were not found to have statistically significant predictive values.ConclusionThe yield of CTPA to rule out PE in patients with cancer presenting in the ED is low. Following a validated decision-making protocol such as Wells Criteria may significantly decrease the number of CTPA used in the ED