Clinical scenario

A 24-year-old woman presents to the ED with shortness of breath and pleurisy. She is otherwise healthy and on no medications except the birth control pill for the past year. There are no other clinical signs or symptoms suggestive of DVT, and her heart rate is normal. You wonder whether as an emergency room physician with some training in bedside ultrasound, reliably confirm or refute the diagnosis of a pulmonary embolus using point-of-care ultrasound (POCUS)?

Three-part question

(In adult patients presenting with clinical suspicion of pulmonary embolus) is (non-invasive, multiorgan point of care ultrasound, when compared with CT pulmonary angiography (CTPA)) specific and sensitive enough (to confirm or refute the presence of pulmonary embolus)?

Search strategy

PubMed was used for papers of any date to the present (January 2017); all databases were searched.

Google Scholar was used to search for any publications that may have been missed in PubMed.

PubMed search: ((((((‘computed tomography angiography’[MeSH Terms]) OR CT[Text Word])) OR ((((CTPE[Text Word]) OR CTPA[Text Word]) OR CT pulmonary angiogram[Text Word]) OR CT pulmonary embolism[Text Word]))) AND ((((‘echocardiography’[MeSH Major Topic]) OR ultrasound[Text Word]) OR echo[Text Word]) OR echocardiography[Text Word])) AND (((pulmonary embol*[Text Word]) OR ‘pulmonary embolism’[MeSH Terms]) OR PE[Text Word])

Google Scholar: Searched for the first 100 results using the terms ‘ultrasound’, ‘pulmonary embolism’ and ‘diagnosis’.

Outcome

Six hundred and fourteen papers were found on the PubMed search, of which three looked specifically at bedside POCUS administered by non-radiologists or technicians. The reference investigation for all three papers was always CTPA, and criteria for patient enrolment was also based on Wells criteria calculations and/or D-dimer values. The remaining papers were excluded because of the following reasons:

1. Ultrasound administered by technician or radiologist.

2. Transoesophageal ultrasound was used.

3. Complex calculations of ultrasound findings requiring expertise beyond that of a minimally trained sonographer to interpret data, not making it a feasible tool in the busy environment of an ED.

Ninety-three thousand and eight hundred results were found on Google Scholar; no additional papers were identified.

Comments

The diagnosis and evaluation of patients presenting with shortness of breath has always been a challenge in the busy environment of an ED. Numerous evidence-based approaches have been developed over the years to aid the clinician in risk-stratifying patients based on clinical symptoms and history. However, the reference test to diagnose a PE remains CTPA, which holds with it some potential harm to the patient. It is a fair amount of radiation as well as even contraindicated in certain patient populations. Ultrasound provides a non-invasive diagnostic tool with virtually no harm or risk to the patient. The fact that more and more emergency medicine residents and staff physicians are being trained on the use of ultrasound makes it an even more attractive diagnostic modality.

There have been numerous publications on the use of ultrasound to help with the diagnosis of PE. However, only recently has there been a focus on a multiorgan approach by the physician at the bedside. The interpretation of data gathered by these ultrasound techniques is also simple enough that a non-radiologist or non-technician is able to make use of the information. From a thorough publication search strategy, two papers were found that examined specifically the multiorgan POCUS approach and its diagnostic power. There was also another paper that looked only at leg vein POCUS but is one of the few that actually used emergency room physicians (ERPs) as the ultrasonographers and explored the role of POCUS in determining patient risk in having a PE. Thus, it was also included in this analysis.

The three papers reviewed show that the ability for bedside POCUS to rule out PE in the absence of findings is rather poor. However, positive findings were shown to have a strong predictive value in diagnosing either PE or alternative diagnosis, if present. The two multiorgan trials agree that sensitivity is quite high for any positive findings in multiorgan POCUS (90% (82.8%–94.9%) in Nazerian et al and approaches 100% for Koenig et al). Additionally, the specificity for POCUS to predict an alternative diagnosis to explain patient symptoms was virtually 100% for both studies. The CT venography (CTV) versus emergency physician-performed ultrasound study also helped reinforce the utility of POCUS when positive findings were present. When compared with CTV, the sensitivity and specificity of an ERP conducted leg vein POCUS was 86% (42%–99%) and 100% (91%–100%), respectively.