Pulmonary embolism (PE) kills 100,000 people in the United States each year making it the second most common cause of sudden, unexpected, nontraumatic death outside of the hospital. PE-related deaths can be unexpected because it can present with minimal symptoms, vague symptoms, difficulty breathing, syncope, or sudden-death. Among ED patients with PE, about 3%-4% have had a syncopal event. However, there is no clear data on what percentage of ED patients with syncope have PE. Currently ED providers consider PE as a possible etiology for syncope in ED patients and pursue further evaluation for PE based on clinical suspicion.

Clinical Question

What is the prevalence of PE among patients hospitalized after first-time syncope?


All patients > 18 years of age in 11 Italian hospitals (academic and non-academic) who presented with a first episode of syncope and required admission to the hospital. Syncope was defined as transient loss of consciousness with rapid onset, lasting less than 1 minute, spontaneous resolution, and without obvious causes such as seizure, stroke, or head trauma.


Primary: Prevalence of PE among patients hospitalized after first-time syncope — confirmed by computed tomography pulmonary angiography (CTPA), ventilation-perfusion (V/Q) scanning, or autopsy.


Multicenter cross-sectional study


Previous syncope
On anticoagulation therapy
Declined participation

Primary Results

  • 2584 patients were assessed for syncope
    • 1867 patients were discharged without hospital admission
    • 717 were admitted to the hospital
      • 157 admitted patients were excluded from the study
      • 560 patients were included in the study
  • Admitted patients included in analysis
    • Mean age = 76 (75% > 70 years old)
    • 330 patients had low pretest probability and a negative d-dimer and did not have additional imaging
    • 230 patients had either high pre-test probability of PE or a positive d-dimer, or both and went on to diagnostic imaging

Critical Results

  • PE prevalence among admitted patients with 1st time syncope: 17.3%
  • Of the 230 patients who underwent further testing:
    • 180 had CTPA, 49 VQ scans, and 1 Autopsy
    • CTPA: n = 180 + Autopsy: n = 1
      • PE diagnosed: 72/180 (40%) CTPA. 1/1 (100%) Autopsy
      • 49/73 (67.1%) had a proximal embolus in the main pulmonary artery or lobar artery.
    • VQ scans: n = 49
      • PE diagnosed: 24/49 (49%)
      • 4/24 patients with defect involving > 50% of lungs
      • 8/24 with defect involving 1 – 25% of lungs
  • PE found in 12.7% of admitted patients with a suspected alternate syncope etiology


  • Multi-center: increases external validity
  • Assessed all patients hospitalized after syncope: limits availability bias
  • Used standardized protocol for PE evaluation


  • Prevalence of PE among all patients presenting for syncope unknown as discharged patients were not included in the study
  • Mean age of patients was 76; likely older than that of all patients who are hospitalized for syncope in the U.S.
  • Subgroup of patients admitted to the hospital were high-risk syncope patients based on demographic information supplied. This limits external validity as may not be the norm for admitted syncope patients in other countries (i.e. The US)
  • Many of the patients admitted to the hospital had signs and symptoms concerning for PE (tachypnea, tachycardia, signs of DVT) but were not evaluated for PE prior to admission. This runs contrary to how patients would be assessed in most EDs
  • Evaluation for PE took place up to 48 hours after admission; unclear if all found PE were present on admission
  • Of diagnosed PE, 32.9% on CTPA and 50% of those on V/Q scanning were small. It is unclear if the small PE caused the syncope or an alternative cause may have been missed
  • Clinical significance of PE on CTPA, VQ scan, or autopsy unclear because treatment decision and patient follow-up after diagnosis was not done
  • Syncope evaluation was not standardized

Author's Conclusions

” Pulmonary embolism was identified in nearly one of every six patients hospitalized
for a first episode of syncope.”

Our Conclusions

This study demonstrated that there is a high prevalence of PE among high-risk first time syncope patients that are hospitalized. Many of the patients had clinical symptoms that should have prompted a workup for PE. However, expanding the workup to all patients with syncope even in the absence of these symptoms is non-sensical.

Potential Impact To Current Practice

These results should not encourage providers to evaluate all syncope patients for pulmonary embolism but rather to pursue evaluation and workup in those patients at risk for PE based on their presentation. In conclusion, this data does not change who we select to workup.

Bottom Line

PE was commonly found in a select, high-risk group of 1st time syncope patients. However, this finding is association only. It is unclear if the PEs found caused the syncope or were simply incidental findings. Emergency providers should continue to work up PE when clinically appropriate in syncope patients and not change to an approach of working up all syncope patients.

Read More


NU EM Blog: The PESIT Trial: Not All First Time Syncope Needs Testing for PE

EM Nerd: The Case of the Incidental Bystander

St. Emlyn’s: JC: Prevalence of PE in Patients with Syncope

REBEL EM: The PESIT Trial: Do All Patients with 1st Time Syncope Need a Pulmonary Embolism Workup?

FOAM Cast: Episode 59 – Syncope (and the PESIT study)