New indices of arterial stiffness correlate with disease severity and mid-term prognosis in acute decompensated heart failure.

Arterial stiffness has been implicated in pathophysiology of heart failure (HF) since it is involved in the ventricular-vascular coupling. Recently, new indices obtained by a cuff oscillometric technique, the arterial velocity pulse index (AVI) for the stiffness of central arteries and the arterial pressure volume index (API) for the stiffness of peripheral arteries have been developed and validated. However, the AVI and API measurement has not been attempted in HF population. This study aimed to investigate the relationship between the AVI, API and clinical parameters and outcomes in HF patients. A prospective cohort of patients with acute decompensated HF were enrolled within 6 months, who were admitted to a tertiary referral hospital in China. Measurement of the AVI and API (AVE-1500, Shisei Datum, Tokyo, Japan) was performed on the day of admission and discharge. Patients were followed up to 6 months for the composite endpoint of all-cause death and rehospitalization for HF. A total of 127 patients were recruited for analysis (60 ± 15 years, 70% male). 80% of the patients were in New York Heart Association (NYHA) Class III or IV at admission with mean left ventricular ejection fraction (LVEF) of 34 ± 9%. During hospitalization, all patients received guideline-directed medical therapy if not contraindicated. The AVI (27.3 ± 5.0 vs. 28.6 ± 6.7, P = 0.002) and API (24.9 ± 4.9 vs. 26.0 ± 6.5, P = 0.05) were lower at discharge than at admission. By dividing the patients into mild to severe group based on systolic blood pressure (SBP) and LVEF or into tertiles according to the amino-terminal pro-brain natriuretic peptide (NT-proBNP), transmitral E velocity over mitral annular e' velocity (E/e' ratio), it was observed that the AVI increased with a higher level of NT-proBNP (P for trend < 0.001), a larger E/e' (P for trend < 0.001) and a lower LVEF (P for trend = 0.0183), while the API increased as the E/e' and systolic blood pressure became higher (both P for trend < 0.05). The improvement in AVI at discharge was correlated with LVEF (R = - 0.3024, P < 0.05) and NT-proBNP improvement (R = 0.3118, P < 0.05), while the change in API was positively correlated with SBP change (R = 0.3897, P < 0.001). In 6 months after discharge, there were 52 predefined events including 15 deaths and 44 rehospitalization for HF. Apart from the level of NT-proBNP, the AVI at discharge of ≥ 26 showed a trend of being associated with the composite outcome (HR 2.747, 95% CI 1.411-5.349, P < 0.001 for univariate analysis; HR 1.864, 95% CI 0.892-3.893, P = 0.09761 for multivariate analysis). New noninvasive arterial stiffness indices as the AVI and API reflected severity of illness and midterm prognosis in admitted HF patients. Further studies are warranted for understanding its mechanisms and developing clinical applications.