Overutilisation of imaging studies for diagnosis of pulmonary embolism: are we following the guidelines?

To evaluate if imaging studies such as CT pulmonary angiography (CTPA) or ventilation-perfusion (V/Q) scan are ordered according to the current guidelines for the diagnosis of pulmonary embolism (PE).

We performed a retrospective observational cohort study in all adult patients who presented to the Sparrow Hospital Emergency Department from January 2014 to December 2016 and underwent CTPA or V/Q scan. We calculated the Wells' score retrospectively, and d-dimer values were used to determine if the imaging study was justified.

A total of 8449 patients underwent CTPA (93%) or V/Q scan (7%), among which 142 (1.7%) patients were diagnosed with PE. The Wells' criteria showed low probabilities for PE in 96 % and intermediate or high probabilities in 4 % of total patients. Modified Wells' criteria demonstrated PE unlikely in 99.6 % and PE likely in 0.4 % of total patients. D-dimer was obtained in only 37 % of patients who were unlikely to have a PE or had a low score on Wells' criteria. Despite a low or unlikely Wells' criteria score and normal d-dimer levels, 260 patients underwent imaging studies, and none were diagnosed with PE.

More than 99 % of CTPA or V/Q scans were negative in our study. This suggests extraordinary overutilisation of the imaging methods. D-dimer, recommended in patients with low to moderate risk, was ordered in only one-third of patients. Much greater emphasis of current guidelines is needed to avoid inappropriate utilisation of resources without missing diagnosis of PE.