# Point-of-care ultrasound improves the diagnosis of heart failure in patients with dyspnea in primary care

## TL;DR

Using point-of-care ultrasound, trained general practitioners can more accurately diagnose heart failure in patients with shortness of breath in primary care settings.

## Contribution

Demonstrates that brief training allows GPs to use PoCUS to significantly improve heart failure diagnostic accuracy in primary care.

## Key findings

- PoCUS improved diagnostic accuracy with 86.8% sensitivity and 88.2% specificity for heart failure.
- Agreement with cardiologist diagnoses increased significantly after PoCUS use.
- B-line quantification and LVEF assessments by GPs correlated strongly with cardiologist results.

## Abstract

Heart failure (HF) is one of the most frequent and clinically important causes of dyspnea, and its recognition in primary care is often hindered by limited access to advanced diagnostics. This study aimed to evaluate whether general practitioners (GPs), after brief training, can use point-of-care ultrasound (PoCUS) to improve the diagnostic accuracy of HF in patients with new-onset dyspnea.

In this prospective validation study, 112 consecutive adult patients with unexplained new-onset dyspnea were initially enrolled across four primary care practices. After clinical evaluation, 102 patients underwent a standardized lung and focused cardiac ultrasound by GPs, using handheld ultrasound machines. B-line quantification and visual assessment of left ventricular ejection fraction (LVEF) were performed. Diagnoses before and after PoCUS were compared to the final diagnosis established by a cardiologist using standard echocardiography. Agreement metrics and diagnostic accuracy measures were calculated.

Post-PoCUS assessments by GPs showed significantly improved diagnostic performance for HF (sensitivity: 86.8%, specificity: 88.2%, positive predictive value: 93.7%, and negative predictive value: 76.5%) compared to pre-PoCUS clinical judgment (sensitivity: 85.3%, specificity: 38.2%, positive predictive value: 73.4%, and negative predictive value: 56.5%). Agreement with the final diagnosis increased substantially (Cohen’s κ from 0.254 to 0.723). GP-assessed B-lines and LVEF correlated strongly with the cardiologist’s results.

A brief, focused training enables GPs to use PoCUS effectively for the detection of HF in patients with dyspnea. Integrating lung and focused cardiac ultrasound into routine primary care may substantially improve diagnostic accuracy, optimize patient management, and reduce unnecessary referrals.