Evaluation of Pulmonary Embolism in the Emergency Department and Consistency With a National Quality Measure: Quantifying the Opportunity for Improvement

Study Questions:

To what extent is imaging for evaluation of possible pulmonary embolism (PE) in the emergency department (ED) appropriate, and what factors are associated with overuse?


The authors reported the results of a prospective, multicenter, observational study of ED patients undergoing testing for suspected PE, at 12 US hospitals from July 2003 through March 2007. Patients who underwent any objective diagnostic test for PE (D-dimer, CT scan, VQ scan, or pulmonary angiogram) were included in the study and underwent a prospective collection of data—including the elements of the Wells score—through a combination of patient interview and medical record review. The primary outcome was avoidable imaging according to the National Quality Forum (NQF) measure. Imaging was performed in low-risk patients with either no D-dimer testing or negative D-dimer testing. Multivariable logistic regression analysis was performed to identify factors associated with inappropriate imaging.


The authors enrolled 5,940 subjects, of whom 4,113 (59%) had a clinical low pretest probability of PE. Imaging was performed in 2,238 low-risk patients (38%), of whom 811 (36%) had no D-dimer testing, and 394 (18%) had negative D-dimer test results. Overall, by NQF measure, 1,205 patients underwent avoidable imaging studies (32%; 95% confidence interval [CI], 31%-34%). Age was associated with avoidable imaging when no D-dimer test was done (odds ratio [OR], 1.15 per decade; 95% CI, 1.10-1.21); and a history of inactive malignant disease was associated with avoidable imaging after a negative D-dimer test result (OR, 1.66; 95% CI, 1.11-2.49). Pregnancy was also associated with avoidable imaging when no D-dimer test was performed (OR, 1.96; 95% CI, 1.29-2.99), but not with avoidable imaging after a negative D-dimer test result.


The authors concluded that one third of the imaging performed for suspected PE may be categorized as avoidable. They further opined that improving adherence to established diagnostic protocols is likely to result in significantly fewer patients receiving unnecessary irradiation and substantial savings.


Although this study is observational in nature, it carries particular weight because of the prospectively collected data and the attention to corroborating the results with well-established, validated clinical decision tools that effectively identify patients with sufficiently low risk of PE to forgo further testing. Given the wealth of clinical research validating the NQF appropriateness measures, and the documentation of inappropriate imaging associated with historical features likely to inspire heightened concern for PE despite low overall clinical risk (age, inactive malignant disease, pregnancy), this study not only characterizes the magnitude of inappropriate imaging in patients with suspected PE (roughly one third), but also offers guidance regarding potential pitfalls or red flags that might entice ED physicians to perform unnecessary tests. It should be noted that only patients cared for by board-certified emergency physicians were included in the study, thereby decreasing the likelihood that inexperience or poor qualifications are to blame. Data such as these could be used to identify patient groups in whom to be wary of over testing.

Keywords: Risk, Diagnostic Tests, Routine, Emergency Medical Services, Fibrin Fibrinogen Degradation Products, Tomography, Cardiology, Pulmonary Embolism, Diagnostic Imaging

< Back to Listings