Objectives Computerised tomography pulmonary angiography (CTPA) studies are commonly requested in the diagnostic workup of suspected pulmonary emboli (PE). CTPA involve significant contrast and radiation exposure. There is also a risk of over or incidental diagnosis that may prompt further harmful investigations or treatment. Efforts should therefore be made to avoid use of CTPA when possible. Pretest probability scores – such as the Wells Score – and d-dimer can be used to reduce unnecessary CTPA requests. A low Wells Score and negative d-dimer almost excludes PE – without additional diagnostic workup. A diagnosis of PE in approximately 15% of CTPA is suggested as an acceptable yield by the British Royal College of Radiologists. We conducted a quality improvement project in our emergency department (ED) at Port Macquarie Base Hospital, to encourage judicious use of CTPA and improve the PE diagnostic yield.

Method We retrospectively collected data for patients that presented over a three month period. All patients that had a CTPA requested during their ED attendance were included. The patients were identified through a search of electronic records. After the initial audit posters were displayed and teaching sessions organised. It was envisaged this would improve use of the Wells Score and – ultimately – more judicious CTPA requesting. The audit cycle was repeated for the exact same time period one year later. Evidence of documented Wells Score use was sought in the medical records. We evaluated the formal radiologist report for each CTPA. All data were anonymised and input into a secure spreadsheet. As data were gathered retrospectively formal ethics approval was not deemed necessary.

Results For the initial audit – 1 st June to 31 st August 2016 – 57 patients were identified. Only 7%–4 – patient records had evidence of Wells Score use. The CTPA diagnostic yield was corresponding low with only 9%–5 – studies identifying a PE. 87 patients were identified for the repeat audit. Wells Score use increased to 13%–11 – and the CTPA PE yield to 16%–14. No patient – in either audit – had a low Wells Score and PE on CTPA.

Conclusions The interventions to encourage more judicious requesting of CTPA appear to have been successful. The absolute number of CTPA requested increased but so did the diagnostic yield for PE. This suggests a more appropriate use of CTPA, likely due to increase use of the Wells Score. In the repeat audit our ED exceeded British College of Radiologist standards on the expected PE yield from CTPA. The increase in CTPA requests may be explained by a larger overall number of ED attendances. We recognise some patients may have had a Wells Score applied but not documented. This is an inherent weakness of retrospective audit. The sample size was also relatively small. Nonetheless we recommend other ED evaluate their use of Wells Score in the context of suspected PE. It is a simple – and effective – way to minimise over investigation for this common presentation.