Introduction Pulmonary Embolisms (PE) are clinically difficult to diagnose and associated with significant morbidity and mortality. Computerised Tomography Pulmonary Angiogram (CTPA) is routinely used to investigate suspected PE. Clinical concern and the increased availability of CTPA may mean that more patients may be receiving unnecessary radiation: a CTPA is approximately 15 mSv, equivalent to 750 chest radiographs.1 In addition, detection of other pathology by CTPA often has minimal clinical impact.2
Aims To investigate our compliance with NICE guidelines in ordering CTPAs, and whether detecting alternative diagnoses justifies their use.
Methods This prospective study, in a London teaching hospital, reviewed data in all medical and oncology patients who had a CTPA from January to February 2014. Clinical diagnoses and risk scores were recorded according to national guidelines.
Results A CTPA was carried out on 91 patients (63% female); 20 had confirmed PE (22%). In 50 (55%), guidelines were not followed: 43 did not have D-dimers, of these 15 (35%) had cancer. Of those with PE, 35% were detected despite low Wells Scores (n = 7).
In 47 (52%) alternative diagnoses made on CTPA accounted for the presenting symptoms: 18 diagnoses were newly made on CTPA but only 13 led to a change in management. Incidental findings requiring following were made in 9.
Conclusions PEs remain difficult to diagnose. Clinical skills may not be accurate; our detection rate was 22%. In 55% NICE guidelines were not followed. In 15 this was due to d dimers not being performed in patients with cancer. The testing of d dimers is not routinely performed in our trust in patients with cancer, due to reduced clinical usefulness. A better scoring system may be required especially in cancer patients.
Alternative diagnoses made on CTPA do not appear to alter management in the majority, suggesting that they should not be used to make other diagnoses. More research is required in diagnosing PE to minimise radiation and contrast risks, and ensure CTPAs are of maximum clinical benefit.
Davies et al, BMJ 2011;342:d947
van ES et al, Chest 2013;144(6):1893–9
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.