More robust educational interventions may be necessary to encourage appropriate imaging, as a single lecture and Q & A session failed to deliver results in one institution’s efforts to limit unnecessary CT pulmonary angiography (CTPA) tests in the emergency department (ED), according to a study published in the September issue ofAcademic Radiology .
“In this small single-center study, we found that clinicians employ a very low threshold to test for PE [pulmonary embolism], resulting in a very high rate of negative testing and a low rate of PE diagnosis,” wrote Yassine Kanaan, MD, of Texas Scottish Rite Hospital for Children, Dallas, and colleagues from the University of Michigan, Ann Arbor.
Prevalence rates for PE at CTPA have been shown by prior studies to be in the 10 percent to 23 percent range; however, at the authors’ institution, positive rates were 5 percent or less. This could mean unnecessary use of imaging, which in turn increases radiation dose and false-positive rates.
Clinical diagnosis of PE can be unreliable as asymptomatic patients do not have clinical signs, explained Kanaan and colleagues. Clinical probability can be estimated in the emergency setting using the Wells score, with guidelines suggesting this probability be determined before imaging is ordered. D-dimer testing is recommended for patients with low and intermediate pretest probability for PE before imaging.
These recommendations regarding ED imaging use for suspected PE were relayed to ED faculty at the University of Michigan through a lecture, and Kanaan and colleagues tested this intervention’s impact by looking at records for 100 consecutive CTPA studies before the lecture and another 100 after the lecture.
Pre-intervention, 1 percent of patients suspected of PE had Wells scores performed and 40 percent of patients who underwent CTPA had D-dimer testing. Of these patients, 15 percent had a negative D-dimer test, 17 percent had an alternative explanation for chest pain and 76 percent had low or intermediate pretest probability, reported the authors. The overall appropriateness rate for CTPA was 7 percent, with 8 percent of the tests resulting in positive findings.
The intervention failed to achieve significant improvements, as the appropriateness rate for CTPA post-intervention was 6 percent, according to Kanaan and colleagues. The positive CTPA rate was 10 percent. Fewer patients had D-dimer testing after the lecture and 84 percent had low or intermediate pretest probability.
The authors speculated that a number of factors besides knowledge play into the decision to order CTPA in the ED. Residents may request the study for a lesser indication because they know the attending physician will want CTPA imaging subsequently anyway. Also, since CTPA has a high negative predictive value, the authors suggested clinicians may get sufficient reassurance from a negative CTPA exam to discharge patients from the ED sooner. This may reduce costs in the long run, but no formal comparative effectiveness analysis has been conducted.
Kanaan and colleagues called for repeated and sustained educational interventions to improve imaging ordering, and also suggested incorporating decision support could increase effectiveness.