Prognostic value of mean gradient to aortic valve area ratio (MG:AVA) in paradoxical low-flow low-gradient aortic stenosis (LFLGAS)

Introduction Patient selection for aortic valve replacement (AVR) in paradoxical LFLGAS (PLFLGAS) requires verification of aortic stenosis (AS) severity and symptoms attributable to the valvular lesion. Routine hemodynamic parameters (peak velocity, MG, and AVA) acquired during resting transthoracic echocardiograms (TTE) are unable to discriminate between pseudo-severe and true severe AS. Multislice computed tomography aortic valve calcium scoring (AVCS) is recommended as the adjudicator. However, current AVCS diagnostic thresholds were principally defined in cohorts of concordant AS, then extrapolated to PLFLGAS with limited validation, and may underestimate severity where valvular fibrosis predominates. The MG:AVA ratio of >=25mmHg/cm2 has been demonstrated to correctly identify true severe AS in classical LFLGAS and was independently associated with increased risk of mortality or AVR. The role of MG:AVA in PLFLGAS has not been assessed. Purpose To determine the prognostic utility of MG:AVA as a surrogate marker for hemodynamically significant stenosis in PLFLGAS. Methods Patients with TTE findings of PLFLGAS (defined by AVA<1.0cm2 or indexed AVA<0.6cm2/m2, peak velocity<4.0m/s, MG<40mmHg, indexed stroke volume =<35ml/m2, and left ventricular ejection fraction>=50%) were identified from the echocardiography laboratory databases within our health service. Included patients were grouped according to a MG:AVA of <25mmHg/cm2 and >=25mmHg/cm2. Outcome data was obtained from the statewide cardiac registry. Primary outcomes were the composite of cardiac mortality or AVR at 3 and 5 years. Results 171 PLFLGAS patients (median age 77.0 years [Interquartile range (IQR): 69.0-82.0 years]; 56.7% female) were included in the study with 48 (28%) in the MG:AVA<25mmHg/cm2 and 123 (72%) in the MG:AVA>=25mmHg/cm2groups. Median follow-up was 8.9 years [IQR: 6.2-11.2 years] with 105 (61.4%) patients meeting the composite endpoint of cardiac mortality or AVR. Freedom from the primary outcome at 3 and 5 years was significantly higher in those with a MG:AVA<25mmHg/cm2 (3 years: 88%, 95% confidence interval (CI): 79-97%; 5 years: 79%, 95% CI: 68-91%) in comparison to MG:AVA>=25mmHg/cm2 patients (3 years: 50%, 95% CI: 42-60%; 5 years: 36%; 95% CI: 29-46%). On multivariable Cox proportional regression analysis, MG:AVA>=25mmHg/cm2 independently predicted the composite endpoint at both 3 (Hazard ratio (HR):5.25, 95% CI:1.94-15.7; P=0.001) and 5 (HR:4.25, 95% CI:1.79-10.1; P=0.001) years. Conclusion MG:AVA ratio is a simple hemodynamic parameter that has prognostic value in undifferentiated PLFLGAS patients. MG:AVA>=25mmHg/cm2 predicts a poorer outcome or need for AVR, implying prognostically significant AS, and may help guide decision-making for aortic valve intervention in PLFLGAS.