Introduction and Objectives Pulmonary embolism (PE) is a common cardiovascular emergency and is the most common preventable cause of hospital deaths. Various factors can impact on the diagnostic accuracy of CTPA, which is now considered the 1^st line investigation for PE diagnosis. Mis-timing of contrast medium administration or cardiac impairment can result in an indeterminate scan through hidden or mimicked emboli. This study focuses on the accuracy of CTPAs in excluding a diagnosis of PE.

Method This is a single centre study looking at all inpatient and outpatient CTPAs carried out over a 12-month period with matched CTPA and lower limb ultrasound Doppler events selected. Of these matched studies, all positive CTPAs for PE were excluded. Of the remainder, only those studies done within 6 months of the original CTPA were included in the study. CT pulmonary arterial (PA) opacification with intravenous contrast medium was objectively measured in Hounsfield units (HU) using an oval region of interest in the main pulmonary trunk. PA opacification was categorised as very poor (<100 HU), suboptimal (< 200 HU) and optimal (= 200HU) as measured at the PA trunk based on the estimation that a minimal opacification of 100 HU is required for identification of acute emboli and 200 HU is required for identification of chronic emboli.

Results From the 32 CTPAs included in the study, 28 had a negative initial CTPA with subsequent negative CT or US follow-up. 4 of these cases had an initial negative CTPA with a subsequent positive CTPA or lower limb US Doppler study. Of these latter cases only one initial CTPA was deemed poorly suboptimal (average 174HU) with optimal pulmonary arterial opacification in all other three initial CTPAs.

Conclusion This study demonstrates that despite a large number of initially negative CTPAs undergoing subsequent follow-up imaging due to presumed persistent or recurring patient symptoms, the majority (87.5%) of these negative scans were still negative and only 12.5% were positive for PE. No major pitfall has been identified in local CTPA acquisition technique that may have led to subsequent misses of potentially life-threatening PEs.