Implementation of an electronic ordering algorithm based on the YEARS criteria to optimize pulmonary embolism diagnostic workup in the emergency department.

Abstract

Objective: The YEARS criteria combine D-dimer testing and clinical features (hemoptysis, signs of deep vein thrombosis, and pulmonary embolism as the most likely diagnosis) to risk stratify patients with symptoms of pulmonary embolism who may undergo CT pulmonary angiography in the emergency department (ED). Electronic clinical decision support can optimize CT pulmonary angiography utilization in EDs, yet its effectiveness with the YEARS criteria remains unstudied. Our goal is to increase the percentage of CT pulmonary angiograms performed with a D-dimer by 10% within six months of integrating the YEARS criteria into our electronic ordering system.

Methods: Single centre quality improvement initiative for all ED patients > 18 years investigated for pulmonary embolism with a D-dimer and/or CT pulmonary angiogram between Jan 2022 - Jan 2023. An electronic clinical decision support algorithm was created based on the YEARS criteria and a plan-do-study-act cycle was completed. Using an SPC chart, the percentage of CT pulmonary angiograms ordered with a D-dimer was identified. Process measures included the rate of CT pulmonary angiograms ordered, the rate of imaging ordered in accordance with the YEARS criteria, and diagnostic yield of imaging based on YEARS status. Balancing measures included pulmonary embolism identified on CT pulmonary angiogram with a D-dimer < 500 ug/L FEU or within 30 days of index visit after being ruled out with the YEARS criteria.

Results: Over 12 months, 2639 patients were investigated for pulmonary embolism and 906 underwent CT pulmonary angiogram. The percentage of CT pulmonary angiograms ordered with a screening D-dimer increased by 14.5% (63.8%, CI 59.5-68.1% vs 78.4%, CI 74.4-82.3%) without increasing the rate of imaging ordered (34.9%, CI 32.4-37.4% vs 33.7%, CI 31.1-36.3%). The YEARS criteria were properly applied in 60% (255/425) of patients who underwent CT pulmonary angiogram. The diagnostic yield of CT pulmonary angiogram was higher if patients were YEARS positive (13.2%, CI 9.0-17.4) compared to YEARS negative (0%, CI 0-28.3) or if not applied (8.1%, CI 3.9-12.4). There were two missed pulmonary embolisms at baseline and none post-intervention.

Conclusions: Implementation of the YEARS criteria using electronic ordering assistance increased the percentage of CT pulmonary angiograms ordered with a D-dimer without increasing imaging use in the ED.