Use of routinely captured echocardiographic data in the diagnosis of severe aortic stenosis.

OBJECTIVE: To determine the implications of applying guideline-recommended definitions of aortic stenosis to echocardiographic data captured in routine clinical care.
METHODS: Retrospective observational study of 213 174 patients who underwent transthoracic echocardiographic imaging within Allina Health between January 2013 and October 2017. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of echocardiographic measures for severe aortic stenosis were determined relative to the documented interpretation of severe aortic stenosis.
RESULTS: Among 77 067 patients with complete assessment of the aortic valve, 1219 (1.6%) patients were categorised as having severe aortic stenosis by the echocardiographic reader. Relative to the documented interpretation, aortic valve area (AVA) as a measure of severe aortic stenosis had the high sensitivity (94.1%) but a low positive predictive value (37.5%). Aortic valve peak velocity and mean gradient were specific (>99%), but less sensitive (<70%). A measure incorporating peak velocity, mean gradient and dimensionless index (either by velocity time integral or peak velocity ratio) achieved a balance of sensitivity (92%) and specificity (99%) with little detriment in accuracy relative to peak velocity and mean gradient alone (98.9% vs 99.3%). Using all available data, the proportion of patients whose echocardiogram could be assessed for aortic stenosis was 79.8% as compared with 52.7% by documented interpretation alone.
CONCLUSION: A measure that used dimensionless index in place of AVA addressed discrepancies between quantitative echocardiographic data and the documented interpretation of severe aortic stenosis. These findings highlight the importance of understanding the limitations of clinical data as it relates to quality improvement efforts and pragmatic research design. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2019. All rights reserved. No commercial use is permitted unless otherwise expressly granted.