Recent studies have reported unexpectedly poor outcomes in patients with moderate aortic stenosis (AS).1, 2, 3, 4, 5, 6, 7, 8 Many of these studies are derived from echocardiographic databases, which rely on import of echocardiographic measurements contained within Digital Imaging and Communications in Medicine (DICOM) structured reporting (SR) files into various report generation systems. DICOM is a universal standard for picture archiving and communication systems that provides a digital platform for the exchange and storage of images between devices and medical applications.9

To encode and archive the imaging diagnostic report from DICOM, SR is used, which is a structured document that consists of text, individual measurements, and coded information important in electronic health records as well as allowing queries and data mining.10 DICOM-SR can be stored in multiple ways. Some store only the summary values of all measurements (such as average, maximum, and minimum), while others store individual measurements so that they can be revised individually.

Depending on the curation of echocardiographic measurements and the file transfer settings, erroneous and submaximal measurements may be inadvertently exported. Validation of big data to ensure data quality is not widely performed. This study evaluated the quality of an echocardiographic data set primarily used for the delivery of clinical care of patients with AS in a teaching hospital. We hypothesized that the use of DICOM-SR outputs may lead to misclassification of patients with severe AS as moderate.