The five-beat averaged method of assessing aortic stenosis in atrial fibrillation significantly underestimates stenosis severity compared to sinus rhythm. (25th November 2020)

• Record Type: Journal Article
• Authors: Tomlinson, S; Mandala, A; Hayman, S; Scalia, G
• Is Part Of: European heart journal. Volume 41:(2020)Supplement 2
• Publication Date: 2020-11-25
• Subjects: Echocardiography: Valve Disease; Cardiology -- Periodicals; Heart -- Diseases -- Periodicals; 616.12005
• DOI: 10.1093/ehjci/ehaa946.0068
• Languages: English
• ISSNs: 0195-668X

Abstract

Abstract: Background: Doppler echocardiographic assessment of aortic stenosis (AS) is dependent on trans-valvular flow rate (TVFR) and is most accurately performed in physiologically normal conditions (TVFR >200ml/s). Atrial fibrillation (AF) complicates AS assessment with heart rate variability and reduced averaged indexed stroke volume (SVi) compared to sinus rhythm (SR). We hypothesize that the 5-beat averaging technique in AF leads to underestimation of severity of AS compared to SR, and that diagnostic accuracy may be improved by maximum SVi beat selection. Methods: Patients with at least mild AS (mean aortic gradient (MG) >15mmHg) with separate echocardiograms demonstrating SR and AF within a six-month period were identified. Patients with interval cardiac surgery, change in left ventricular ejection fraction (LVEF) of >15%, aortic valve prostheses, infective endocarditis, paced rhythm, severe concomitant valvular dysfunction, LVEF <35%, heart rate >100 beats per minute (bpm) or TVFR <200ml/s on their SR examination were excluded. In AF, individual beat SVi was calculated from individual beat aortic velocity time integral (VTI) and the reported aortic valve area (AVA) via the continuity equation. Results: Seventeen patients were identified (mean time between echocardiograms 132 days, mean difference in LVEF 4%, mean difference in heart rate 11bpm). Using the 5-beat averaged model, compared to SR, the presence of AF significantly reduced SVi (−12.68ml/m 2, p=0.00007) and TVFR (−69ml/s, p=0.00014). In AF, 9/17 cases demonstrated one or more beats with TVFR <200ml/s, with 6/17 producing a 5-beat averaged TVFR <200ml/s – consistent with a low flow state. Correspondingly, flow derived severity assessment parameters and calculated aortic valve area (AVA) in AF were reduced compared to SR (aortic peak velocity (Vmax) −0.42m/s, p=0.00005, MG −7.01mmHg, p=0.002, AVA −0.11cm 2, p=0.025). As a proposed correction in AF, the single beat with the largest SVi was compared with SR. This measure improved concordance with flow conditions and severity assessment in SR (difference SVi −8.77ml/m 2, p=0.004, TVFR −46ml/s, p=0.001, Vmax −0.21, p=0.003, MG −3.55mmHg, p=0.02) and equalized calculated AVA (−0.07cm 2, p=0.09). Conclusion: In individual patients with AS observed in SR and AF within a six-month period, AF was associated with significantly lower SVi and TVFR, with underestimation of AS severity assessment. AS assessment in AF using maximum observed stroke volume beat (as opposed to 5-beat average) improves concordance with SR assessment significantly. Funding Acknowledgement: Type of funding source: None