Introduction CT Pulmonary Angiography (CTPA) is the gold standard investigation for suspected pulmonary embolism (PE). Low or intermediate probability clinical prediction (e.g. Wells score) combined with a negative D-dimer effectively rules out PE in over 97% of cases, avoiding the need for CTPA and its inherent risks (radiation exposure and contrast induced nephropathy). This is the recommendation of the BTS guidelines. We undertook a study to examine if the BTS guidelines were being adhered to in our Trust, and whether increasing the D-dimer threshold may safely reduce the need for CTPA.

Methods We obtained a list of CTPAs performed within the Trust between September 2009 and September 2011 and searched our pathology system for a contemporary D-dimer result (HemosIL latex immuno-assay). For all patients with a negative D-dimer (≤230ng/mL), we looked for a documented pre-test probability score in the clinical notes or calculated a Wells score if not documented. We then analysed CTPA results with D-dimer between 230–500ng/mL.

Results There were 1645 CTPAs performed during the study period, of which 15% had confirmed PE 903(54.9%) had a contemporary D-dimer result, and of these 57(6.3%) had a negative D-dimer, and 193(21.4%) were between 230–500ng/mL. In the negative D-dimer group, 3 (5.3%) had confirmed PE’s on CTPA. One was on tranexamic acid, which can falsely lower D-dimer, and 2 had prolonged admissions in whom D-dimer testing was not appropriate. We were able to examine the notes of 39/57 cases and only 3 (8%) had a pre-test probability documented. On review 31/39 (79.5%) had a low to intermediate Wells score and should not have had a CTPA.

In the 230–500ng/mL D-dimer group, there were only 4/193 (2.1%) positive CTPAs with a negative predictive value 98%.

Conclusions In our Trust, the lack of pre-test probability scoring combined with D-dimer is leading to inappropriate CTPAs. The rate of PE in the patients with a D-dimer between 230–500ng/mL is also very low. A protocol recommending initial treatment pending an urgent respiratory team review prior to CTPA, could safely reduce the number performed in this group, if combined with a low to intermediate pre-test probability score.