Low-flow low-gradient aortic stenosis: aortic valve area estimated by continuity equation versus projected aortic valve area

Introduction Accurately stratifying severity in low-flow, low-gradient (LFLG) aortic stenosis (AS) can be clinically challenging. The projected aortic valve area (AVAproj), which estimates the aortic valve area (AVA) at normal transvalvular flow rate, has been proposed as a more reliable marker of true stenosis. The TOPAS study previously demonstrated the prognostic relevance of AVAproj in identifying true severe aortic stenosis. Purpose To compare AS severity classification based on AVA estimated by continuity equation at peak dobutamine stress echocardiography (AVA-CE) versus AVAproj, and evaluate their prognostic value in predicting the composite outcome of aortic valve replacement (AVR) or all-cause mortality at 2 years. Methods Single-center retrospective study including individuals with AS who underwent dobutamine stress echocardiography (DSE). AVA at rest (AVA-rest), AVA-CE, and AVAproj were assessed. Reclassification rates were analyzed. Logistic regression models were constructed to assess the association of AS severity (as defined by AVA-CE or AVAproj) with the composite outcome. Results From a total of 57 individuals with AS submitted to DSE, we included 24 patients with transvalvular flow rate at peak stress inferior to 250 mL/s. The mean age was 69±10 years, 83% were male. Echocardiographic parameters are summarized in Table 1. When comparing absolute AVA values, AVA-rest and AVA-CE were both significantly different from the AVAproj (p<0.001 and p=0.044, respectively). AVA-rest is also significantly different from AVA-CE (p=0.002). When using categorical severity classification (severe vs. moderate), 22 patients (92%) were classified as having true severe stenosis with AVA- CE and 14 patients (58%) were classified as having true severe stenosis with projected AVA. Between rest and AVA-CE, only 1 patient changed classification (1 from severe to moderate), with no statistically significant difference (p=1.000). In contrast, when comparing AVAproj to AVA-rest, 9 initially classified as severe were reclassified as moderate (p=0.004). When comparing AVAproj to AVA-CE, 8 classified as moderate by AVAproj were classified as severe at peak (p=0.008). Two logistic regression models were developed to predict the combined endpoint of AVR or death at 2 years. The first model, including rest LVEF and AVA-CE severity classification, did not reach statistical significance overall (χ²=5.75, p=0.056), and neither variable was individually predictive. The second model, which included rest LVEF and AVAproj classification, was statistically significant (χ²=10.70, p = 0.005), with AVAproj classification showing a strong association with the outcome (OR 16.17 [IC95%: 1,29–301,81], p=0.050). Conclusion Projected AVA differed significantly from AVA-CE and demonstrated superior prognostic value for predicting AVR or death at 2 years. These findings support the clinical relevance of incorporating AVAproj in the evaluation of AS.Table 1.Echocardiographic parameters.