Diagnosing centrally located pulmonary embolisms in the emergency department using point-of-care ultrasound

Abstract

Objective: The study objective was to investigate the combined accuracy of right heart strain on focused cardiac ultrasound (FOCUS) and deep vein thrombosis (DVT) on compression ultrasound (CUS) for identification of centrally located pulmonary embolism (PE) diagnosed on computed tomography pulmonary angiography (CTPA).

Methods: This was a prospective observational study using a convenience sample of patients undergoing CTPA in the emergency department (ED) for evaluation of PE. Patients received a FOCUS looking for right heart strain (McConnell's sign, septal flattening, right ventricular enlargement or tricuspid annular plane systolic ejection (TAPSE) <17mm) and a CUS looking for DVT. Ultrasounds were interpreted by both the investigator performing the ultrasound and the principal investigator independently.

Results: There were 199 patients enrolled in the study, with 46/199 (23.1%) positive for a PE. Of these, 20/46 (43.5%) PE's were located centrally. Of those with a PE, 20/46 (43.5%) had an associated DVT identified on bedside ultrasound. Among patients with a proximal PE, 18/20 (90.0%) had evidence of right heart strain and the combination of lower extremity CUS and FOCUS was 100% sensitive. Diagnostic accuracy of ultrasound was much lower for peripherally located PEs.

Conclusions: Emergency physician-performed bedside ultrasound may be sufficient to exclude the presence of centrally located PE, as the sensitivity in this study was 100%. Additionally, several patients with PE may qualify for early anticoagulation when DVT is identified, and further research in indicated to determine whether these patients ultimately require CTPA given identical treatment algorithms in the absence of RV strain or biomarker elevation.