P928Pseudo-discordance mimicking low-flow low-gradient AS in TAVR patients with severe symptomatic aortic stenosis

Abstract

Abstract Background The combination of a small aortic valve area valve area (AVA) and a low mean gradient (MG) severity pose a serious clinical challenge in the diagnosis of severe aortic stenosis (AS). While this discordance is frequently labeled “low-flow low-gradient AS”, there are two additional potential causes: underestimation of MG and underestimation of AVA. Purpose To investigate prevalence and root causes of discordant echocardiographic findings in symptomatic patients with AS and normal LV function by comparing Doppler data, invasive hemodynamic data, CT LV outflow tract size and calcium score. Methods and results We studied 67 severely symptomatic patients with AS and an LVEF>50%. Thirty patients (45%) had discordant echocardiographic findings (MG<40 mmHg, AVA≤1cm2). Of those, 16 had also discordant findings by invasive measurements (True Discordants), compatible with low-gradient severe AS. In 14, catheterization data were actually concordant and hence discrepant to the echocardiographic findings (False Discordants): In 7 of them, the echocardiographic MG was >35 mmHg; in 5 it was >30 mmHg and only in 2 between 25 and 29 mmHg. In 6 of the 14 patients with underestimated MG, no right parasternal Doppler examination had been performed, LVOT VTI tracings were clearly suboptimal in 3 patients and 1 case was deemed inadequate due to poor imaging quality. LVOT area by echocardiography or by CT could not differentiate between Concordants and Discordants (p=0.3 and p=0.8 respectively) or between True and False Discordants (p=0.5 and p=0.6 respectively). While calcium score was similar in Concordants (2711±1159 AU) and False Discordants (2692±1136AU, p=0.96), it tended to be higher in Concordants (2711±1159 AU), when compared to True Discordants (1906±1284 AU, p=0.07). In patients with concordant echocardiographic findings calcium score levels of >3000 AU in men and >1600 AU in female had a positive predictive value (PPV) of 90% for the correct diagnosis of severe aortic stenosis. In patients with discordant findings the PPV was 80%. Conclusions 1) The majority of severely symptomatic AS patients with normal LV systolic function an echocardiographic AVA≤1.0 cm2 and a MG>30mmHg considered for TAVR have severe AS by calcium score. 2) In this patient population, discordant echocardiographic findings are in about half of the cases due to technical factors (“pseudo-discordance”) rather than due to true low flow low-gradient stenosis. 3) Pseudo-discordance is mainly due to mild-moderate underestimation of gradients, caused by a lack of reliable right parasternal tracings, rather than due to underestimation of valve area due to the echocardiographic circularity assumption in the presence of an elliptical LVOT. 4) The diagnosis of true low-flow AS cannot be established without a clear Doppler tracing from the right parasternal window. 5) Absent an adequate right parasternal window, patients should undergo CT assessment of calcium score.