Perspective Chapter: Evolution of Techniques to Assess Vascular Impedance in Patients with Aortic Stenosis

Abstract

Aortic stenosis (AS) once was conceptualized as a mechanical problem with a fixed left ventricular (LV) afterload because of an obstructive valve. With time, there has been growing recognition that AS functions more like a series circuit, with important contributions from the ventricle through to the vasculature. Emerging evidence suggests that higher blood pressure and increased arterial stiffness, synonymous with vascular aging, increases global LV afterload in patients with AS. This in turn, has adverse consequences on quality-of-life measures and survival. Although traditional methods have emphasized measurement of the transvalvular pressure gradient, focusing on valvular hemodynamics alone may be inadequate. By definition, total vascular load of the human circulation includes both steady and pulsatile components. Steady load is best represented by the systemic vascular resistance whereas pulsatile load occurs because of wave reflections and vascular stiffness, and is often referred to as the valvulo-arterial impedance. In the following Review, we evaluate existing and upcoming methods to assess vascular load in patients with AS in order to better understand the effects of vascular aging on this insidious disease process.

1. Introduction

Aortic valve stenosis (AS) is a progressive disease in which the end-stage is characterized by an increase in global left ventricular (LV) load, resulting in inadequate cardiac output, decreased exercise capacity, congestive cardiac failure and death [1]. Without correction, the rate of death is more than 50% at 2-years for patients with severe AS and symptomatic disease [2]. Patients with AS are frequently elderly, with concomitant hypertension and increased arterial stiffness. Higher global LV load – as reflected in blood pressure, resistive (i.e. systemic vascular resistance) and pulsatile (i.e. vascular impedance) load – is known to be associated with poorer quality-of-life measures and survival outcomes. As such, AS is no longer regarded as an isolated valvular disease, but rather a pathological process involving the left ventricle, aortic valve and large conducting arteries (Figure 1) [3].

Figure 1.

Factors that contribute to the global left ventricular load in elderly patients with aortic valve stenosis. Abbreviations: LV, left ventricular.

The past decade has seen a rapid uptake of device technologies to treat patients with AS. Despite the obvious effects of vascular aging, uncoupling intrinsic properties of the left ventricle and arterial tree from the degree of valvular stenosis remains challenging. This is particularly true of patients with severe AS with low-flow/low-gradient (LF-LG) (aortic valve area [AVA] ≤1 cm²; ejection fraction [EF] ≤30–45%; and mean transvalvular gradient ≤30–40 mmHg) or paradoxical LF-LG severe AS (LVEF ≥50%; indexed stroke volume [SVi] ≤35 mL/m²; and mean gradient ≤40 mmHg). This is important, however, as acute interventions on either compartment may cause reciprocal changes in the other. Patients with severe LF-LG or paradoxical LF-LG AS, for example, are at considerably higher risk of ongoing exertional intolerance, even after relief of valvular obstruction, due to presumed mismatch between ventricular filling, valvular stenosis, and vascular stiffness.

Pulsatile pressure-flow relationships to describe vascular impedance (the relationship of pressure to flow) of the human circulation were first reported over half-a-century ago in (then) pioneering invasive studies [4, 5]. The clinical use of this technique for determination of vascular impedance has remained limited however, as high-fidelity catheters are considered cumbersome, expensive and may fail to appreciate the eccentricities of pressure and flow dynamics in the ascending aorta. Over the intervening years, measurements of steady-state load (such as systolic arterial pressure, systemic arterial compliance, pulse pressure and systemic vascular resistance [SVR]) have erroneously been taken to represent the vascular impedance. The advent of transcatheter aortic valve implantation (TAVI) has re-ignited this conversation.

As discussed further below, simultaneous high-fidelity pressure and flow catheters have recently been used to describe the acute changes that occur following TAVI [6]. The valvulo-arterial impedance (Z_VA) index obtained by Doppler echocardiography (TTE) is one of the most widely adopted non-invasive methods to assess vascular impedance in patients with AS [7]. Valvulo-arterial impedance is assessed using brachial systolic pressure, mean aortic pressure gradient and indexed stroke volume within the LV outflow tract. It has been found to be useful in patients with AS because it incorporates stenosis severity, volume flow rate, body size, SVR and vascular impedance. More recent studies have utilized non-invasive pressure (from carotid or radial applanation tonometry [AT]) and flow velocity (from the LV outflow tract on cardiac magnetic resonance [CMR]) to estimate vascular impedance in the time or frequency domain [8, 9].

Determination of both the steady-state and pulsatile components of the vascular tree are expected to play an increasingly important role in the clinical evaluation of patients with severe LF-LG or paradoxical LF-LG AS states moving forward, as well as in the prognostication of adverse clinical outcomes following TAVI. The following Review provides an overview of key concepts, as well as invasive and non-invasive methods to measure global LV load in individuals with AS.

2. Defining hydraulic load of the human circulation

Hydraulic load of the systemic circulation includes both steady-state and pulsatile components. Steady-state load is best represented by the SVR, although systolic arterial pressure, systemic arterial compliance, and pulse pressure are frequently used. Systemic vascular resistance is calculated as:

SVR = BAm (mmHg) - RAPm (mmHg) ÷ CO (L/min) x 80,

where BAm represents mean left brachial arterial pressure, RAP represents mean right atrial pressure and CO represents cardiac output by the standard direct Fick method.

Pulsatile load occurs because of wave reflections and vascular stiffness and is best described using the term vascular impedance (Z), or the relationship of pressure to flow. When the general term ‘impedance’ is applied to a vascular bed, it is usually referring to “input impedance” (Z_in), this being the relationship between pulsatile pressure and pulsatile flow recorded in an artery feeding a particular vascular bed. Z_in can be estimated with the following complex equation:

$$Z_{\mathit{in}}=\frac{\left|P\right|}{\left|Q\right|}cos\left(\beta -\varphi \right)\text{,}$$

where |Z_in| = |P|÷|Q| is the modulus and θ = (β – ϕ) is the phase of the impedance [10, 11]. Both the steady-state and pulsatile load contribute to the total hydraulic load of the systemic circulation [12].

3. Effects of vascular aging in patients with aortic stenosis

With aging, the central elastic aorta progressively dilates, elongates and becomes tortuous with stiffened, thickened walls [13]. Characteristic age-related changes in aortic flow velocity, pressure waveform and vascular impedance now well described [11, 14]. Stiffer, older vessels lead to a faster velocity of pressure pulse and earlier timing of reflected pulse wave from the periphery, augmenting central aortic systolic pressure and yielding a greater afterload on the heart [10, 11]. Systemic vascular resistance is higher, while the systemic impedance phase typically shows similar values for the first harmonic at all ages, then increases for all age groups, crossing zero to positive values later in elderly patients and also those with AS (around 3–4 Hz). Harmonic refers to the analysis of signals with respect to frequency (Hz) rather than time. Put simply, a frequency-domain graph shows how much of the signal lies within each given frequency band over a range of frequencies. The effect of age-related alterations to vascular impedance is to cause further mismatch between energy expenditure of LV ejection, and an increase in pulsatile energy lost in the circulation. The result is a direct increase in LV afterload and left ventricular mass. Additionally, mean aortic systolic pressure is increased, thereby increasing LV oxygen requirements and LV afterload, while mean aortic diastolic pressure is decreased, reducing coronary blood flow [15].

4. Effects of transcatheter aortic valve implantation on vascular impedance

Limited studies have explored the effect of aortic valve replacement on vascular impedance in elderly patients with AS. Residual LV afterload is more often than not assumed by assessing transvalvular pressure gradient, effective orifice area or the degree of valve patient-prosthesis mis-match [16]. However, focusing on valvular hemodynamics alone is clearly inadequate. In one study, Lindman et al., examined the effects of blood pressure, and indices of steady-state and pulsatile-load on outcomes following TAVI [17]. 2141 patients recruited to the PARTNER I trial (Placement of Aortic Transcatheter Valve) were included. Higher total and pulsatile arterial load were associated with increased mortality for all (p < 0.001), but resistive load was not. Patients with low 30-day blood pressure and high pulsatile load had a 3-fold higher mortality than those with high 30-day blood pressure and low pulsatile load [17]. Lindman et al. concluded that even after relief of valve obstruction in patients with AS, there was an independent association between post-TAVI blood pressure, elevated vascular load and mortality [17].

5. Invasive methods to determine vascular impedance

As mentioned earlier, vascular impedance (Z_-INV) of the human circulation was first determined in pioneering catheter studies during the 1960’s and 1970’s [13, 18, 19, 20, 21]. From this early work, a typical Z_-INV pattern in the ascending aorta was demonstrated to have a relatively high modulus at zero frequency – which is the peripheral/systemic resistance or steady-state load – then experience a fall of modulus with increasing frequency to a minimal value around 3–4 Hz, before rising to a maximal value around twice the minimal frequency and continuing to fluctuate around its characteristic impedance at higher frequencies [5, 22]. The phase (or difference in angle) value was found to be negative at low frequencies – indicating arterial flow leading pressure – then crossed zero around the same frequency where modulus was at a minimum and became positive at higher frequencies [5, 22]. These impedance patterns are mainly influenced by ascending aorta distensibility and arterial pulse wave velocity [4], both of which are significantly altered in the presence of vascular aging and cardiovascular disease.

Studies of Z_-INV have not been actively pursued beyond the 1970s until recently, partly because of the requirement of invasive technique and costly sensors to register arterial pressure and/or flow waveform accurately. In a study by Yotti et al., measurement of Z-_INV was performed before and after TAVI in 23 patients using a high-fidelity 0.014-inch pressure wire introduced through a 6 Fr multi-purpose guiding catheter and placed in the ascending aorta approximately 5 cm above the annulus [6]. The Z_-INV impedance spectrum was calculated as: Z_-INV = P ÷ (SV ÷ TVi), where P (mmHg) represents peak ascending aortic pressure, and volumetric flow rate (mL/s) was calculated from linear flow velocity measurements (cm/s) by means of a calibration constant (cm²) obtained as SV (stroke volume) divided by TVi (time velocity integral). Input impedance spectrum was derived using Fourier decomposition of pressure and velocity signals up to 10 Hz, whilst Z_-INV was calculated as the average of Z moduli above 4 Hz, excluding outlier values of >3 times the median. Yotti et al. found that calcific degenerative AS was conditioned by the upstream valvular obstruction that dampened forward and backward compression waves in the arterial tree and that stiffer vascular behavior post TAVI occurred [6]. The short and long-term effects of TAVI on Z_-INV have not been studied beyond the immediate post-operative period (Figure 2).

Figure 2.

Methods to determine vascular impedance in patients with aortic valve stenosis. Abbreviations: DBP, diastolic blood pressure; PP, pulse pressure; SAC, systemic arterial compliance; SBP, systolic blood pressure; SVRI, systemic vascular resistance index; VAL, valvulo-arterial load; Z_INV, valvulo-arterial impedance invasive; Z_VA, valvulo-arterial impedance; Z_VA–INV, valvulo-arterial impedance instantaneous.

6. Echocardiographic methods to determine vascular impedance

Pulse pressure (PP), systemic arterial compliance (SAC), the systemic vascular resistance index (SVRi), and valvulo-arterial impedance (Z_VA) are the most widely applied echocardiographic methods to determine vascular impedance in patients with AS. Steady-state load is best represented by the SVRi, which is calculated as:

(DBP + 1/3 PP) × 80 ÷ CI,

where DBP represents diastolic blood pressure (mmHg), PP represents the pulse pressure (mmHg) and CI represents cardiac index (L/min/m²). Pulsatile load is often measured as either PP, SAC or Z_VA. Systemic arterial compliance is calculated as:

SAC = SVi ÷ PP,

where SVi represents stroke volume index (mL/m²/beat) and PP represents pulse pressure (mmHg) [3, 23]. Stroke volume index is determined by the Doppler method and calculated as (π[LVOT radius]² × LVOT velocity time integral)/body surface area, where LVOT indicates LV outflow tract. Valvulo-arterial impedance is acquired using brachial systolic pressure, mean aortic pressure gradient and SVi within the LV outflow tract on Doppler echocardiography. It is expressed as:

Z_VA = (bSBP + MeanG_-NET) ÷ SVi,

where bSBP (mmHg) represents brachial systolic arterial pressure, MeanG_-NET (mmHg) represents mean aortic pressure gradient and SVi (mL/m²) represents indexed stroke volume [7].

Since Z_VA was first described in patients with AS, elevated values have been associated with poorer outcomes, a greater degree of myocardial dysfunction and reduced overall survival [24, 25]. Poorer outcomes are thought to be related to comorbidities which elevate vascular impedance – namely advanced age, hypertension, and obesity [26]. For patients with asymptomatic severe AS, who have a 3% mortality rate a 5-years if left untreated, a Z_VA index >5 mmHg/mL/m² is associated with a 2.5-fold increase in overall mortality, regardless of treatment type [24]. Paradoxical low-flow, low-gradient AS is known to account for one-third of patients with severe AS and preserved EF. Patients with paradoxical LF-LG AS and a Z_VA index >5.5 mmHg/mL/m² also have increased mortality [24]. Interestingly, in patients with LF-LG AS, elevated Z_VA has not yet been found to be associated with operative or long-term mortality [27]. The prognostic impact of Z_VA is not well studied in patients with symptomatic moderate AS.

Despite the accessibility of Z_VA and its widespread use, the technique has limitations including: (i) the potential for underestimation of flow velocity due to misalignment of the Doppler signal with flow direction; (ii) the risk of underestimation of LV outflow tract diameter due to inadequate quality and/or positioning of the imaging plane; (iii) measurement variability related to manual tracing of flow velocity contours, and; (iv) the calculation of mean pressure from brachial cuff pressure rather than direct measurement of central aortic pressure [12]. Furthermore, Z_VA is measured as the ratio of pressure to indexed volume (rather than pressure to flow) and may therefore be more accurately described as a resistance index rather than a true measure of vascular impedance (Figure 2).

7. Cardiac magnetic resonance methods to determine vascular impedance

With the introduction of CMR into routine clinical practise, several parameters of aortic stiffness in patients with AS have now been described - aortic compliance, distensibility, capacitance, elasticity, stiffness index and valvulo-arterial impedance. Two methods to assess vascular impedance in patients with AS have been described – valvulo-arterial impedance instantaneous (Z_VA-INS) and valvulo-arterial load (VAL). Soulat et al. [8] first described Z_VA-INS in 2017 using CMR and non-simultaneous carotid tonometry. Valvulo-arterial impedance instantaneous is estimated by acquiring CMR velocities above the aortic valve and within the LV outflow tract, and by carotid tonometry after CMR exam. It is calculated in the time domain by combining the incident LV pulse pressure to 95% of peak flow:

Z_VA-INS = (ΔP_-Q95 + MaxG_-NET) ÷ ΔQ_-95,

where ΔP_-Q95 is the LV pressure (taken to be the same as carotid tonometric pressure) change from its end-diastolic foot to time of 95% of peak flow (Q₋₉₅) and MaxG_-NET is the maximum gradient calculated in the aortic valve considering pressure recovery [8]. Hungerford et al., subsequently described the VAL index in 2020. Whereas Soulat et al. [8] calculate Z_VA-INS as the ratio between total arterial pressure (the summation of carotid tonometric pressure as a surrogate for central aortic pressure and maximum pressure gradient of the aortic valve) and CMR ascending aortic flow, VAL is estimated as the global LV afterload. That is, VAL is derived from the simultaneous relationship between aortic pressure and flow velocity. Data obtained forms a graph of modulus and phase, plotted against frequency and is expressed as:

$$VAL=\frac{Pn}{Qn}{e}^{i\left(\theta n-\phi n\right)}$$,

where VAL represents global LV load, Pn represents derived central aortic pressure, Qn represents aortic flow velocity product at the MPA level, and ei(θn − ϕn) represents both the harmonic component of pressure and phase of impedance [9].

VAL differs from Z_VA-INS as it permits (i) simultaneous acquisition of aortic pressure and flow; (ii) measures the combined LV afterload; (iii) samples the multiple flow profiles seen in patients with AS [28], and; (iv) estimates systemic impedance in the frequency domain [29]. As left ventricular hypertrophy, aortic valve stenosis and vascular stiffness represent elevated impedances in series, simple summation of these resistances (as in the case of Z_VA-INS) may lead to an overestimation of global LV load [9, 30]. Studies of CMR vascular impedance estimation in patients undergoing transcatheter aortic valve implantation (TAVI) are currently underway (Figure 3).

Figure 3.

Valvulo-arterial load in patients with aortic stenosis by simultaneous cardiac magnetic resonance/applanation tonometry. Abbreviations: VAL, valvulo-arterial load index.

8. Other methods

Although vascular impedance measures the pulsatile properties of the human circulation, aortic pulse wave velocity (PWV) is considered the ‘gold-standard’ measurement of arterial stiffness and an important tool to evaluate both arterial system damage, vascular adaptation, and therapeutic efficiency. Pulse wave velocity can be measured using non-invasive, reproducible, and relatively inexpensive techniques. The effect of elevated PWV has been studied in AS patients undergoing both surgical and transcatheter replacement, and found to be associated with poorer quality-of-life measures [31] and mortality [32] irrespective of stenosis severity. Waveform analysis post TAVI typically shows an acute increase in the forward compression wave, backward compression wave and forward expansion energies [33]. The duration that these changes persist post correction of AS remain unclear (Figure 2).

9. Future directions

Beyond techniques to determine vascular impedance in patients with AS, the question remains – what can be done to ameliorate the insidious effects of arterial stiffness in patients with AS before and after intervention? Indeed, arterial stiffness involves both arteriosclerosis and atherosclerosis and is influenced by structural and biochemical changes within vessel walls, involving changes in elastin/collagen ratio, elastin cross-linking, vascular calcification, vascular smooth muscle cell stiffness, endothelial dysfunction, and inflammation [34, 35]. Lifestyle modification may have a considerable impact. A diet rich in fruits and vegetables, polyunsaturated fatty acids, coca flavonoids, tea catechins and dairy products with a limited intake of salt and red meat have all been demonstrated to reduce arterial stiffness [36]. Anti-hypertensive treatment, anti-diabetic drugs and lipid lowering agents have also been shown to reverse vascular aging in middle aged individuals, and to a lesser degree, in older patients [37]. The effect of diet, lifestyle, and pharmaceutical interventions in elderly patients with AS remains to be determined.

10. Conclusions

The natural history of AS is dictated by a progressive decoupling of the aortic valve from the left ventricle and vascular system over time. Vascular function is conditioned by the upstream valvular obstruction, which in turn dampens forward and backward compression waves in the arterial tree. Abrupt correction of valvular stenosis may have unforeseen effects on the left ventricle and vasculature not adequately accounted for using traditional TTE imaging techniques. Simultaneous CMR and AT techniques are uniquely placed as they better account for steady-state and pulsatile components of flow. Incorporation of techniques to assess vascular impedance in patients with AS is expected to play a greater role in patient selection of severe AS sub-types and prognosis with time.

Funding

All other authors have reported that they have no funding disclosures relevant to the contents of this paper to disclose.

Disclosures

All authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Abbreviations

AA	ascending aorta
AS	aortic valve stenosis
AT	applanation tonometry
CMR	cardiac magnetic resonance
FFT	fast Fourier transformation
LV	left ventricular
MPA	main pulmonary artery
PA	pulmonary artery
Pn	derived central aortic pressure
PH	pulmonary hypertension
PWV	pulse wave velocity
Qn	aortic flow velocity product
RHC	right heart catheterisation
RV	right ventricular
TAVI	transcatheter aortic valve implantation
VAL	valvulo-arterial load
VTF	velocity transfer function
ZC	characteristic impedance
Zin	input impedance
ZVA-INS	valvulo-arterial impedance-instantaneous.