Background: The optimal diagnostic management of patients with chronic lung disease (CLD) and suspected pulmonary embolism (PE) is unclear.

Objectives: The aim of this study was to evaluate the performance of PE diagnostic strategies in patients with and without CLD.

Methods: This is a secondary analysis of an individual-patient data meta-analysis (PROSPERO CRD42018089366) of prospective or cross-sectional studies evaluating conventional (Wells or revised Geneva score with fixed or age-adjusted D-dimer) and newer (YEARS and the Pulmonary Embolism Graduated D-dimer Study algorithms) diagnostic strategies. Main outcomes were safety and efficiency. Safety was defined by the failure rate (proportion of patients diagnosed with venous thromboembolism during initial workup or follow-up among those in whom PE was considered ruled out at baseline without imaging). Efficiency was defined as the proportion of patients in whom PE was considered excluded without the need for imaging among all patients.

Results: Twelve studies, representing 16 990 patients (2201 patients with CLD) were included. The safety of each strategy was comparable in patients with and without CLD, whereas efficiency of the strategies was lower in patients with CLD. In patients with CLD, the predicted failure rate varied between 0.58% (95% CI, 0.10%-3.20%) and 1.06% (95% CI, 0.44%-2.53%), and between 2.54% (95% CI, 1.45%-4.39%) and 3.12% (95% CI, 2.04%-4.74%) for conventional and newer diagnostic strategies, respectively. The predicted efficiency was 19.0% to 33.2% and 35.8% to 43.9% for conventional and newer diagnostic strategies, respectively.

Conclusion: In patients with CLD, diagnostic failure rate seemed slightly lower with conventional diagnostic strategies, but more patients would need imaging to rule out PE, compared with newer diagnostic strategies.