Written by Samuel Rouleau
Spoon Feed
A post-hoc analysis of three studies found that using D-dimer alone would have missed zero pulmonary embolisms (PEs) in patients with a high pretest probability for PE, but this strategy should not change practice yet.
Don’t throw away the CTPA…yet
The main question addressed is whether D-dimer alone can be used to rule out PE in patients who have a high pretest probability for PE. The authors conducted a post-hoc analysis of 3 prior studies based in Europe: PROPER – an RCT of PERC; MODIGLIANI – an RCT of YEARS; and TRYSPEED – a retrospective study on CTPA use). Out of 12,300 patients screened, 651 patients were included; importantly, 584 of them were from TRYSPEED. In the study group, the prevalence of PE was 31.3%, median Wells score 6, and median revised Geneva score 13. 70 patients had D-dimer values less than the age-adjusted cut-off, and 48 were below the fixed 500 ng/mL cut-off. The failure rate for D-dimer only, with both the age-adjusted strategy and fixed cut-off would have been 0.0% (95%CI 0.0 – 6.5%) and (95%CI 0.0% – 7.4%), respectively. Using Bayesian statistics, the authors estimated a posterior probability of 76% that the failure rate of this strategy would be lower than 2% in this cohort.
How will this change my practice?
I like the idea, but I have too many concerns about the study design to forgo CTPA in patients with a high pretest probability of PE. Simply, the assembled cohort does not reflect the question or population of interest. The two RCTs did not contribute many patients, and the retrospective study had only 7.6% of patients (n=686) in the high pretest probability category. Plus, there was no patient follow-up. Admittedly, I have a high bar for evidence to change current practice patterns, especially when I have concerns about study design. This study won’t change my practice, but I am looking forward to future work on this idea.
Source
Failure rate of D-dimer testing in patients with high clinical probability of pulmonary embolism: Ancillary analysis of three European studies. Acad Emerg Med. 2024 Nov 1. doi: 10.1111/acem.15032. Epub ahead of print. PMID: 39487597.
