Association of Guideline-Concordant Echocardiographic Surveillance With Mortality and Aortic Valve Replacement in US Adults With Aortic Stenosis

BACKGROUND: Consensus-based clinical practice guidelines recommend echocardiographic surveillance intervals based on aortic stenosis (AS) severity; however, real-world surveillance practices and associated outcomes remain poorly characterized. METHODS: We evaluated AS echocardiographic surveillance patterns using validated natural language processing algorithms applied to echocardiogram reports from Kaiser Permanente Northern California between January 1, 2008, and September 30, 2017. Patients with prior aortic valve replacement (AVR) were excluded. Guideline concordance was based on the American College of Cardiology/American Heart Association (ACC/AHA) recommendations for surveillance frequency of patients with mild, moderate, and severe AS. Mild-to-moderate and moderate-to-severe AS were assigned to follow surveillance intervals of the more severe AS classification. We used multivariable Cox proportional hazards regression to examine associations between guideline concordance and all-cause death and receipt of AVR. RESULTS: Among 20 571 patients with AS (mean age 75.7±11.2 years, 48% women), rates of guideline concordance were 74% for mild AS, 51% for mild-to-moderate AS, 63% for moderate AS, 51% for moderate-to-severe AS, and 49% for severe AS. Male sex, younger age, having a cardiologist, and certain comorbid conditions were positively associated with guideline concordance. During median follow-up of 5.2 (interquartile range, 3.0-7.4) years, patients who were guideline concordant experienced 1929 (28.8%) deaths and 1783 (26.6%) AVRs, while those not concordant experienced 3683 (42.2%) deaths and 1266 (14.5%) AVRs. Guideline concordance was associated with lower adjusted all-cause death for those with at least moderate AS (moderate: adjusted hazard ratio [aHR], 0.76 [95% CI, 0.68-0.85]; moderate-to-severe: aHR, 0.70 [0.54-0.91]; and severe: aHR, 0.62 [0.51-0.76]), and higher rates of AVR across all AS severity levels (mild: aHR, 2.23 [1.93-2.57]; mild-to-moderate: aHR, 1.56 [1.22-1.99]; moderate: aHR, 1.45 [1.28-1.64]; moderate-to-severe: aHR, 2.09 [1.62-2.69]; and severe: aHR, 1.95 [1.60-2.36]). CONCLUSIONS: Implementation of the current American College of Cardiology/American Heart Association guideline-recommended echocardiographic surveillance frequency in adults with AS is associated with improved outcomes yet remains suboptimal in real-world settings.