Tracking progression of aortic stenosis with echocardiography

Transthoracic echocardiography (TTE) is used to assess aortic stenosis (AS) severity and track disease progression. As the field moves to study medical therapies to halt disease progression, reliable non-invasive imaging markers that are sensitive to small changes in disease progression are needed to enable efficient trial designs. The signal-to-noise ratio of commonly obtained TTE-based measures of progressive (non-severe) AS severity is unknown.

This is a retrospective study of TTEs done at a tertiary referral centre (Tufts Medical Center, Boston MA). A cohort of patients with progressive AS who had two TTEs done within 30 days (in the absence of valve intervention) and a cohort of progressive AS patients with TTEs ≥ 1 year apart, also without valvular intervention, were assembled. Limits of agreement (LOA) and intraclass correlation (ICC) were calculated for aortic valve area (AVA) by continuity equation, peak velocity, and mean gradient. Cohen's d-statistic (d) was calculated for each hemodynamic assessment and a composite marker to assess sensitivity for detecting disease progression normalised to measurement variability.

The reproducibility cohort included 24 patients. The progression cohort included 35 patients. The median age was 70 years (interquartile range [IQR] 13). 22 patients (37.3%) were female. In the progression cohort, the median time between TTEs was 2.2 years (IQR 3.1 years). In the reproducibility cohort, AVA LOA were -0.7 to 0.8, ICC = 0.61; peak velocity LOA were -149.0 to + 126.7, ICC = 0.29; and mean gradient LOA were -16.2 to 12.2, ICC = 0.06. The d-statistic for annualised change in AVA was -0.29, the d-statistic for annualised change in maximum velocity was 0.46, the d-statistic for mean gradient was 0.55. The d-statistic for a composite, including all three hemodynamic markers, was 0.45.

Standard TTE markers of AS severity have variable sensitivity for detecting AS progression. For patients with progressive (non-severe) AS, mean gradient has the highest signal-to-noise ratio and may be the most reliable TTE-based assessment of disease progression.