Background: The prevalence of pulmonary embolism (PE) is approximately 11-17% in patients with an acute exacerbation of chronic obstructive pulmonary disease (AE-COPD). The optimal diagnostic strategy for PE in these patients remains undetermined.

Aims: To evaluate the safety and efficacy of standard (revised Geneva and Wells PE scores combined with fixed D-dimer cut-off) and computed tomography pulmonary angiogram (CTPA)-sparing diagnostic strategies (ADJUST-PE, YEARS, PEGeD, 4PEPS) in patients with AE-COPD.

Method: Post-hoc analyses of data from the multicenter prospective PEP study (NCT02035293). The primary outcome was the diagnostic failure rate of venous thromboembolism (VTE) during the entire study period. Secondary outcomes included diagnostic failure rate of PE and DVT, respectively, during the entire study period and the number of CTPA needed per diagnostic strategy.

Results: 740 patients were included. The revised Geneva and Wells PE scores combined with fixed D-dimer cut-off had a diagnostic failure rate of VTE of 0.7% (95%CI 0.3%-1.7%), but >70.0% of the patients needed imaging. All CTPA-sparing diagnostic algorithms reduced the need for CTPAs (-10.1% to -32.4%, depending on the algorithm), at the cost of an increased VTE diagnosis failure rate of up to 2.1% (95%CI 1.2%-3.4%)(Figure).

Conclusion: Revised Geneva and Wells PE scores combined with fixed D-dimer cut-off were safe, but a high number of CTPA remained needed. CTPA-sparing algorithms would reduce imaging, at the cost of an increased VTE diagnosis failure rate that exceeds the safety threshold. Further studies are needed to improve diagnostic management in this population.