Point-of-care ultrasound for differential diagnosis of dyspnea in primary care: A pilot study

Róbert Kiss-Kovács, Blanka Morvai-Illés, András Mohos, Roland Tóth-Szeles, Ildikó Bakó, Albert Varga, Gergely Ágoston

Keywords: point-of-care ultrasound; lung ultrasound; focused cardiac ultrasound; dyspnea

Background:

Shortness of breath is a common clinical complaint that often arises as a diagnostic challenge due to its multifactorial aetiology. However, imaging tools to facilitate diagnosis are increasingly becoming more available at the primary care level.

Research questions:

The primary aim of this validation pilot study was to determine whether general practitioners (GPs), after a short training period, are able to detect the number of B-lines and qualitatively assess the left ventricular systolic function in patients presenting with dyspnea.

Method:

Four GPs received training in point-of-care ultrasound (PoCUS). Patients underwent an eight-sector lung and a qualitative focused cardiac ultrasound examination to estimate the number of B-lines and left ventricular systolic function using a portable ultrasound device. A cardiologist validated the GPs' findings. Additionally, the suitability of PoCUS for monitoring diuretic therapy was evaluated.

Results:

A total of forty-two patients were enrolled (69.0% women), with a mean age of 68.1±12.9 years. There was a significant correlation between the number of B-lines detected by GPs and those measured by the cardiologist in patients not receiving diuretic therapy (r=0.972; p<0.001). A substantial interrater agreement was found between the GPs' and cardiologist's assessments of left ventricular systolic function (κ=0.627, 95% CI: 0.241-1.000; p<0.001). In patients with heart failure who received diuretic therapy (n=10), the last intake of loop diuretic before the expert's examination did not significantly affect the number of B-lines (p=0.093). The diagnostic accuracy of GPs considerably improved after PoCUS, with moderate agreement with the validating cardiologist's diagnoses before PoCUS (κ=0.460; 95% CI: 0.258-0.662; p<0.001), and substantial agreement after PoCUS (κ=0.731; 95% CI: 0.562-0.900; p<0.001).

Conclusions:

Our study suggests that the semi-quantitative assessment of B-lines and qualitative estimation of left ventricular systolic function with point-of-care ultrasonography can be rapidly acquired by GPs and used reliably for the differential diagnosis of dyspnea.

Points for discussion:

effectiveness of short training programs

reproducibility of PoCUS examination techniques

improvement in diagnostic accuracy

#27