A Center-Wide Study of D-Dimer Testing for Pulmonary Embolism in the ED Setting

Abstract

Abstract Pulmonary embolism (PE) is a life-threatening condition associated with a very high mortality. While its diagnosis remains challenging, the use of diagnostic tests has increased tremendously. There are many different guidelines used to work up suspected PE. The Wells and/or Geneva scores are used to estimate the pretest probability (PTP) of PE before any complementary testing is done. While a CT pulmonary angiography (CTPA) should not be performed in individuals with a low/intermediate PTP and a negative D-dimer (DD), some data suggest that up to 55% of patients with low PTP undergo a CTPA, even though they might not have had a DD or had a negative DD. As such, in this study, we wanted to evaluate the use of DD in our emergency department (ED). A pilot study was conducted using an interactive software developed in-house for the evaluation of health care quality, efficiency, and effectiveness. The EMR database (EPIC) we use consists of laboratory, clinical, pathological, and radiological reports, allowing us to use different search criteria and to outline cutoff values for continuous and categorical variables while permitting us to define a timeframe for the different observations. We looked at all ED visits in the period from 2014 to 2018 in which CTPAs were done for suspected PE. A total of 8,873 ED visits from 2014 to 2018 had a CTPA done for suspected PE. Of these, 6,117 cases had no DD done and 1,812 had a DD done along with CTPA. The high percent of visits with imaging done without a DD (69%) is justified if these are all cases with a high PTP (DD not indicated). If this assumption was true, then we expected a high percent of positive confirmation of PE by imaging. However, only 11% of these cases showed positive findings of PE. This rate is similar to published studies in many US hospitals that demonstrate overuse of CT owing to a lack of using the PTP-based approach. On the other hand, a vast majority of cases in our study with normal DD (<0.5 μg/L) had negative CT results (98%). Interestingly, all four cases that had positive PE findings by CT despite a negative DD (2%) were women (three below the age of 62 with chronic and/or subsegmental PE and one 47-year-old with a DD = 0.49 μg/L). This finding calls for consideration of age and possibly gender-based DD reference ranges. Lastly, while CTPA is considered the gold standard for PE diagnosis, it has limitations: cost, increased length of stay, unnecessary exposure of radiation, contrast, and unnecessary treatment of subsegmental PE. Therefore, an approach using PTP and using DD appropriately is warranted in order to prevent patients from being exposed to unnecessary risks and also to reduce costs.