Prevalence of PE in Syncope (Spoiler Alert: It’s lower than PESIT)
March 8, 2019
Written by Clay Smith
Spoon Feed
The prevalence of PE in this large prospective cohort of all-comers to the ED with syncope was 0.6%, much lower than PESIT seemed to indicate.
Why does this matter?
After PESIT showed 1 in 6 hospitalized patients with syncope had a PE, many of us found it implausible. A couple retrospective studies showed that the prevalence was probably much lower. This gives us another large, prospective look at this.
Serenity now…for syncope
Combining 2 large prospective cohorts from 17 centers, 9,091 patients presenting with syncope were included, and all had aspects of the PE workup (D-dimer, V/Q, CTPA) and 30-day serious outcomes recorded. Serious outcome was defined as PE or non-PE related (arrhythmia, myocardial infarction, serious hemorrhage, and death). Of the 9,091 patients, 547 (6%) had any PE workup, and 56 were diagnosed with PE, which means the prevalence of PE is this large prospective cohort was just 0.6% (95%CI, 0.5% to 0.8%). Four patients (0.04%) had death related to PE, and only 11 (0.1%) who had a non-PE serious outcome were found to also have PE. What this means in practice is that all-comers presenting to the ED with syncope rarely have PE. This is different than PESIT, which found the prevalence in patients hospitalized (up to 48 hours later) had PE prevalence 17.3%. Unlike PESIT, which mandated PE workup in syncope patients, this was an observational reflection of usual practice in an American and Canadian cohort. So, this may have introduced verification bias, as the same gold standard was not used for all patients. Also, since the U.S. study only included patients ≥60 years, the age of this cohort is skewed toward 60, which impacts generalizability but is similar to the patients in PESIT. At least we’re comparing apples to apples.
This should return sanity to the workup of patients presenting to the ED with syncope. While PE is in the differential diagnosis, it is not all that common, and the workup for PE should be initiated only if there are other symptoms which suggest it may be the cause of syncope.
Source
Prevalence of Pulmonary Embolism Among Emergency Department Patients With Syncope: A Multicenter Prospective Cohort Study. Ann Emerg Med. 2019 Jan 25. pii: S0196-0644(18)31535-X. doi: 10.1016/j.annemergmed.2018.12.005. [Epub ahead of print]
Open in Read by QxMD
Reviewed by Thomas Davis
Thanks for the spoon feed! What do you think about the exclusion of patients due to language barrier? In some parts of the country (like NM where I am), that can be a significant portion of patients. What about the differences between the CA and US study, specifically that the CA cohort excluded presyncopal patients (US included those), the US having a minimum age of 60yo (CA included all adults), and no data on DDimer usage in the US study (CA included Ddimer usage)? While this is reassuring compared to PESIT, I believe the true prevalence of PE is somewhere in between what this study quoted (0.6%) and PESIT’s conclusions (17%).