Quantitative fluid overload in severe aortic stenosis refines cardiac damage and associates with worse outcomes.

Abstract

Cardiac decompensation in aortic stenosis (AS) involves extra-valvular cardiac damage and progressive fluid overload (FO). FO can be objectively quantified using bioimpedance spectroscopy. We aimed to assess the prognostic value of FO beyond established damage markers to guide risk stratification. Consecutive patients with severe AS scheduled for transcatheter aortic valve implantation (TAVI) underwent prospective risk assessment with bioimpedance spectroscopy (BIS) and echocardiography. FO by BIS was defined as ≥1.0 L (0.0 L = euvolaemia). The extent of cardiac damage was assessed by echocardiography according to an established staging classification. Right-sided cardiac damage (rCD) was defined as pulmonary vasculature/tricuspid/right ventricular damage. Hospitalization for heart failure (HHF) and/or death served as primary endpoint. In total, 880 patients (81 ± 7 years, 47% female) undergoing TAVI were included and 360 (41%) had FO. Clinical examination in patients with FO was unremarkable for congestion signs in >50%. A quarter had FO but no rCD (FO+/rCD-). FO+/rCD+ had the highest damage markers, including N-terminal pro-B-type natriuretic peptide (NT-proBNP) levels. After 2.4 ± 1.0 years of follow-up, 236 patients (27%) had reached the primary endpoint (29 HHF, 194 deaths, 13 both). Quantitatively, every 1.0 L increase in bioimpedance was associated with a 13% increase in event hazard (adjusted hazard ratio 1.13, 95% confidence interval 1.06-1.22, p < 0.001). FO provided incremental prognostic value to traditional risk markers (NT-proBNP, EuroSCORE II, damage on echocardiography). Stratification according to FO and rCD yielded worse outcomes for FO+/rCD+ and FO+/rCD-, but not FO-/rCD+, compared to FO-/rCD-. Quantitative FO in patients with severe AS improves risk prediction of worse post-interventional outcomes compared to traditional risk assessment.