Accuracy of Highly Limited Echocardiographic Screening Images for Determining a Structurally Normal Heart: The...

Area of Science:

• Cardiology
• Medical Imaging
• Diagnostic Ultrasound

Background:

• Increasing use of limited cardiac ultrasound as a screening tool.
• Advantage of reduced patient-sonographer exposure time during COVID-19 pandemic.
• Uncertainty regarding the sensitivity and negative predictive value of limited echocardiograms.

Purpose of the Study:

• To examine the accuracy of limited echocardiography in 203 de novo studies.
• To compare the interpretation of limited echocardiograms by experienced cardiologists and sonographers against a gold standard.

Main Methods:

• Utilized six standard echocardiographic views (parasternal long axis, apical four-chamber) with color Doppler.
• Evaluated interpretations from 12 experienced professionals (cardiologists and sonographers).
• Classified studies as normal or requiring a full study; reported sensitivity, specificity, PPV, NPV, and accuracy.

Main Results:

• Overall screening echocardiogram accuracy was 65.4%, with 71.2% sensitivity and 58.4% negative predictive value.
• Accuracy remained similar (64.6%) when excluding inadequate images.
• Interpreters' accuracy was comparable across groups (echocardiography specialists, other cardiologists, sonographers), though sonographers showed higher sensitivity and cardiologists higher specificity.

Conclusions:

• Limited cardiac ultrasound screening for normal hearts has a sensitivity of only 71%, questioning its safety.
• Caution is advised when using limited echocardiography in non-specialized settings due to potential inaccuracies.