What guideline-discordant ordering for CT pulmonary angiography studies can reveal

Nearly a quarter of CT pulmonary angiography (CTPA) orders did not align with scoring system guidelines for evaluating potential pulmonary embolism (PE) in the emergency department, reported authors of a single-center study published in the Journal of the American College of Radiology.

Of that total, 18 studies included at least one patient-specific factor associated with an increased risk for PE that was not a part of the risk stratification scoring systems. The results may indicate a gap between a provider’s assessment and guidelines used to evaluate potential PE risk. 

“Although CTPA use provides the benefit of clear diagnosis, overuse exposes patients unnecessarily to risks associated with imaging (ie, radiation, intravenous contrast administration, incidental findings, costs, and anxiety) without any potential benefit,” wrote Emma Simon, BS, department of population health, NYU School of Medicine in New York, and colleagues.

Simon et al. performed a retrospective review of 212 consecutive encounters with 208 patients in the ED that led to CTPA orders between January 6 and February 25, 2016. For each patient the researchers calculated a revised Geneva score and two versions of the Wells criteria—each validated tools used by providers to help stratify patients by level of risk. Encounters were then classified as either PE unlikely or PE likely.

Overall, guideline-discordant studies ranged from 25% to 37% depending on which scoring system was used. Twenty-two percent did not abide by any of the three scoring systems.  The automatic Wells criteria produced 53 guideline discordant studies, followed by the Geneva score (73 studies) and nonautomatic Wells criteria (79 studies).

Importantly, the researchers found a majority of guideline-discordant cases were because the provider did not perform a d-dimer assessment in patients unlikely to have PE before ordering a CTPA study. The d-dimer test can be used to avoid CT scanning in low-risk and some intermediate-risk patients, and therefore those cases were defined as “potentially avoidable cases.”

However, the 43 cases that included at least one patient-specific factor for increased risk of PE that were not included in the risk stratification scores indicate it may have been correct to bypass d-dimer testing.

“Through analysis of CTPA ordering in our institution’s ED, we found that many guideline-discordant orders were placed on patients who presented with gestalt-elevating, evidence-based risk factors for PE that are not included in the risk stratification scores,” the authors wrote. “Therefore, guideline-discordant ordering may indicate that, in the presence of these factors, the assessment of risk made by current scoring systems and clinical suspicion can differ.”

Fixing the problem

The heart of this problem, according to the authors, is that stratification scoring such as the Wells system provide “an excellent approximation of risk, but they cannot and are not meant to capture the full clinical picture.”

Because of this, clinical-decision support tools have likely “failed” due to a mismatch between computer-generated opinions and provider opinions, the authors wrote.

In the future, Simon et al. plan to create a tool that asks the provider whether they agree with the computer-created risk level before offering a recommendation.

“By focusing interventions more on the difficulties involved in these decisions rather than further simplification of an inherently complex issue, it is possible that we will see an improvement in outcomes,” the authors concluded.