Ventricular dynamics is a major determinant of the augmentation index: An in vivo and in silico study

Research output: Chapter in Book/Report/Conference proceedingMeeting abstract


The role of pulse wave reflections in the increase of the augmentation pressure and, in turn, pulse pressure (PP) with ageing has been recently challenged as recent studies have highlighted the potential importance of ventricular ejection properties in determining blood pressure pulsatile components [1,2]. The first systolic shoulder (P1) of the pressure waveform is proportional to the product between the aortic pulse wave velocity (PWV) and aortic flow at time of P1 (U1) [3], and the usual peak pressure (P2) to the product between PWV and the volume of blood (V2) ejected at the time of P2 [4]. Assuming proportionality between proximal and distal PWV and noticing the close relationship between the augmentation index (AIx) and the ratio P2/P1, we investigate the relationship between AIx and a new index entirely based on ventricular mechanics, AIx, defined as V2/U1.
The study involved patients from a normotensive (n=164, 126 men, age 49±8 years, blood pressure 110±16/69±10 mmHg, means±SD) and hypertensive (n=156, 83 men, age 46±17 years, blood pressure 130±23/83±13 mmHg) cohort. Reflected waves were quantified using the reflection coefficient G, i.e. the ratio of backward to forward pressure component. A Least Absolute Shrinkage and Selector Operator (LASSO) analysis was performed to statistically identify the main contributors to AIx among a set of cardiac and arterial parameters (Age, PWV, G, QIx, MBP, PP). To determine the relative contribution to AIx of arterial (G) and cardiac (QIx) properties, variations of AIx with QIx for an approximately fixed G were assessed, and vice versa. A sensitivity analysis of changes in AIx to QIx and G was also performed using an in silico model of blood flow in the larger arteries of the upper thoracic aorta.
Results & Discussion
The LASSO analysis identified QIx and G as the main determinants of AIx with standardised coefficients of 0.24 and 0.49, respectively (p>0.001 in each case). AIx was found to increase with increasing QIx and G. In silico and in vivo studies were consistent as the coefficient of % increase in AIx per ml.s.m-1 increase in QIx was 0.18, for both normotensive and hypertensive subjects, compared with a theoretical value derived from the sensitivity analysis of 0.16 % change in AIx per change ml.s.m-1 in QIx. Change in AIx per change in G was 35% and 28% for normotensive and hypertensive cohorts, respectively, compared with a theoretical value derived from the sensitivity analysis of 30 % change in AIx per change in G. The sensitivity analysis also confirmed QIx had a greater impact on AIx since a 30% change in QIx from baseline resulted in a 35% increase in Aix, while the same increase in baseline G yielded a 27% increase in AIx. Conclusion We have proposed a new index based entirely on ventricular ejection dynamics and studied its relationship with AIx. The results of this part-in-silico/part-in-vivo study further challenge the role of reflection waves in the increase of AIx, as our new index was as correlated, if not more, to AIx than G. 
1. Torjessen et al., Forward and backward wave morphology and central pressure augmentation in men and women in the Framingham Heart Study, Hypertension, 64(2):259-265, 2014
2. Fok et al., Augmentation Pressure Is Influenced by Ventricular Contractility/Relaxation Dynamics, Hypertension, 63(5):1050-1055, 2014
3. Vennin et al., Noninvasive calculation of the aortic blood pressure wavefrom from the flow velocity waveform: a proof of concept, Am J Physiol Heart Circ Physiol, 309(5):H969-76, 2015
4. Vennin et al., Identifying hemodynamic determinants of pulse pressure: a combined numerical and physiological approach, Hypertension, 70(6):1176-1182, 2017
Original languageEnglish
Title of host publicationBioMedEng19 Conference Proceedings
Place of PublicationLondon
Number of pages1
ISBN (Electronic)978-1-9996465-2-3
Publication statusPublished - Sept 2019
EventBioMedEng19 - Imperial College London, London, United Kingdom
Duration: 5 Sept 20196 Sept 2019


Country/TerritoryUnited Kingdom
Internet address


Dive into the research topics of 'Ventricular dynamics is a major determinant of the augmentation index: An in vivo and in silico study'. Together they form a unique fingerprint.

Cite this