Summary
Left ejection fraction (LVEF) – resulting from the difference between end-diastolic volume (EDV) and end-systolic volume (ESV), divided by EDV – is a poor index of left ventricular (LV) systolic performance due to its dependency on load conditions, inotropic state and LV remodelling. The characteristic impedance of the ascending aorta (Zc) integrates factors opposing LV ejection during the early ejection period when arterial wave reflection can be neglected. Zc is related to the pressure wave velocity (C) and the cross-sectional area of the aorta. The aim is to demonstrate that LV performance and geometry are closely related to the physical properties of the arterial system. LV pressure-volume loops were obtained from simultaneous measurements of LV (or aortic) pressure and LV volume. The slope Ees (also called LV end-systolic elastance) of the ESP-ESV relationship was assessed. Aortic diameters, pressure and flow measurements were synchronized to evaluate C, aortic forward and backward pressure waves, the elasticity of the aorta (Ep) and thereby Zc. In contrast to LVEF, LV end-systolic elastance (Ees), which reflects the stiffness of the chamber at maximal myofilament activation, is relatively insensitive to load conditions and may be considered as an index of ventricular chamber contractility. For a given Ees value, the end-systolic pressure (ESP) determines the LV end-systolic volume. Ees is determined by cardiac myocytes contractility and density, and thereby concentric remodelling. A tight correlation between Zc and the degree of LV concentric remodelling was found in hypertensive and in normal subjects. Zc was found to increase throughout the full lifespan and also with hypertension. Both Zc and wave reflections determine aortic input impedance estimated from the aortic pressure-flow relationship. Increased arterial stiffness resulted in increasing C and overlap of forward and backward waves and thereby in greater pulse pressure and ESP and a greater difference between ESP and diastolic pressure. Ees is an accurate index of LV systolic performance. Besides the inotropic state of myofibers, Ees depends on the concentric remodelling and thereby on the characteristic impedance of the aorta.
Résumé
La fraction d’éjection (FE) du ventricule gauche (VG), rapport entre le volume systolique (VS) (différence entre le volume télédiastolique [VTD] et le volume télésystolique [VTS]) et le VTD, est un mauvais indice de fonction systolique VG. La FE dépend certes de la qualité de pompe VG mais aussi de son remodelage et est étroitement dépendante des conditions de charge du VG. L’élastance télésystolique (ETS), rapport entre pression télésystolique (PTS) et VTS, est indépendante des conditions de charge du VG. Elle exprime la qualité du VG à développer de la pression. ETS est fonction de la qualité des fibres myocardiques et de leur nombre (remodelage concentrique). Le remodelage concentrique est étroitement en rapport avec les propriétés physiques du système artériel et en particulier avec l’impédance caractéristique de l’aorte (Zc). Zc exprime les « forces » qui s’opposent à l’éjection VG en début de systole avant le retour des ondes de réflexion (OR). Zc est d’autant plus élevée que la rigidité aortique est grande (augmentation de la vitesse d’onde C) et que la section aortique est petite. Une relation étroite relie Zc et le remodelage concentrique aussi bien chez des sujets normotendus lors du vieillissement que chez les patients hypertendus indépendamment de l’âge. En fin de systole le chevauchement des ondes de pression incidentes (OI) et réfléchies (OR) détermine la PTS. Le chevauchement des OI et OR est d’autant plus important que l’amplitude des ondes est importante et la vitesse d’onde est grande. Pour une valeur de ETS, PTS détermine le VTS. Zc, OI et OR interviennent ainsi tout au long de l’éjection. Zc et les ondes de réflexion sont les principaux éléments de l’impédance aortique. Cette dernière relie de façon quantitative pression et débit dans l’aorte. La performance et le remodelage ventriculaire gauche sont étroitement liés aux propriétés physiques du système artériel.
Background
Everything that is simple is theoretically false, everything that is complicated is pragmatically unusable .
In the 1970s, Merillon and Gourgon performed landmark haemodynamic studies on the evaluation of left ventricular (LV) performance in the intact human circulation. A few years earlier, Sagawa et al. had shown similar results in dogs, using variably loaded LV pressure-volume loops . The pioneers from Beaujon University Medical Centre soon recognized that LV ejection fraction (LVEF) – resulting from the difference between end-diastolic volume (EDV) and end-systolic volume (ESV), divided by EDV – was a poor index of LV systolic performance due to its dependency on load conditions, inotropic state and LV remodelling. Merillon and Gourgon consistently demonstrated that LV performance and geometry were tightly related to the physical properties of the arterial system in normal, hypertensive and heart failure patients .
The LV pump is designed to generate pressure to eject a systolic volume of blood (the stroke volume [SV]) into a high-pressure system. Series of LV pressure-volume loops are obtained using variable loads in order to assess LV performance, which includes systolic properties and filling capacities. The forces that oppose LV ejection include: inertia forces related to blood mass acceleration; capacitive forces related to aortic wall distensibility, which opposes aortic volume variation; resistive forces related to wall and blood viscosity (which are negligible in large vessels); reflected waves arising from the arterial tree; and peripheral vascular resistance (PVR), which determines the mean arterial pressure.
The first four of these forces determine the aortic input impedance (Z) to pulsatile blood flow, while PVR represents the opposition to steady flow (i.e. cardiac output [CO]). The change in pressure in the proximal aorta resulting from a given change in flow in the absence of wave reflection defines the characteristic aortic impedance (Zc); this occurs within the first third of systole before the reflected pulse wave has returned to the ascending aorta.
Methods
Left ventricular variables
Patients were investigated invasively at a time when Doppler echocardiography had not been developed ( Fig. 1 ). The LV mass (m) was calculated by using the method of Rackley et al., from anteroposterior film of the opacified left ventricle . CO was measured by right heart catheterization, using the dye (indocyanine green) dilution technique. LV pressure-volume loops were obtained from simultaneous measurements of LV (or aortic) pressure and LV volume ( Fig. 2 ). The LV pressure was measured with high-fidelity micromanometer-tipped catheters (Millar 5F), which, unlike a fluid-filled catheter, yield accurate pressure measurements, without phase or frequency distortion. The LV volumes were measured by single-plane cineangiography (50 frames/second) at a right anterior oblique incident angle. The end-systolic pressure (ESP) is the pressure corresponding to the LVESV and not to the dicrotic notch of the aortic pressure (however, in routine practice both values are similar). The slope Ees (also called LV end-systolic elastance) of the ESP-ESV relationship was assessed.
Aortic input impedance
The elastic properties or stiffness of the proximal aorta were comprehensively studied by these pioneers before pulse wave velocity (PWV) measurement devices or high resolution wall tracking systems became available. The aortic pressure was measured above the Valsalva sinuses with a Millar micromanometer. Aortic diameters were measured on cineangiograms (50 frames/second) in an oblique left anterior view. Blood flow velocity was measured by an electromagnetic catheter-tip velocity transducer placed in the ascending aorta. Aortic diameters, pressure and flow measurements were synchronized to evaluate PWV, aortic forward and backward pressure waves, the elasticity of the aorta (Ep) and thereby Zc, obtained by averaging values of the moduli of impedance (|Zn|) above 4 Hz. Zc as a measure of arterial wall stiffness correlates with pressure wave velocity (C) and aortic radius (R).
Left ventricular performance
Left ventricular systolic performance
LV pressure-volume diagrams were analysed with changes in loading conditions, inotropism and heart rate in normal subjects and in patients with cardiomyopathy . Infusion of sodium nitroprusside was used to reduce preload and afterload (10–15 mmHg LV end-diastolic pressure [EDP] decrease and 20 mmHg mean aortic pressure decrease). Angiotensin infusion was used to specifically increase afterload (20 mmHg mean aortic pressure increase) . Post extrasystolic potentiation was triggered by a right ventricular stimulus to alter inotropic state. Heart rate increase was obtained by right atrial pacing . The investigators found that Ees was not modified by nitroprusside ( Fig. 3 A) or angiotensin in either group, while mean velocity of fibre shortening (VCF) and LVEF increased after vasodilation and decreased after vasoconstriction in congestive heart failure patients . During post extrasystolic potentiation Ees, EF, +dP/dt max and VCF all increased in normal subjects and in patients with heart failure and reduced EF (HFrEF) ( Fig. 3 B). Right atrial pacing increased +dP/dt max and Ees, while VCF and EF were not altered in normal and congestive heart failure subjects. These studies showed that, in contrast to EF, Ees is relatively insensitive to load conditions and may be considered as an index of ventricular chamber (but not of myocyte) contractility. Thus, the variable Ees appears useful in discriminating a normal from a failing ventricle. The Ees is typically close to 2.0 mmHg/mL in normal hearts, < 1.0 mmHg/mL in failing dilated hearts and about 4.0 mmHg/mL in hypertrophic hearts.
The same authors found that Ees increases with age and in hypertensive subjects . The reduction in CO observed in older and/or hypertensive patients is not associated with a decrease in LV performance, but with lower LVESV and SV resulting from concentric LV remodelling. Besides indicating chamber contractility, Ees reflects the stiffness of the chamber at maximal myofilament activation. Concentric remodelling and Ees are thus keys in LV adaptation to chronic pressure overload. Consistent with this, Kawaguchi et al. found that Ees is higher in heart failure with preserved ejection fraction (HFpEF) and hypertensive patients compared with control subjects .
Clinical applications of left ventricular pressure-volume loops
Interestingly, the acute effects of vasodilating or inotropic therapy on LVESV and SV regulation may be predicted from serial LV pressure-volume loops ( Fig. 3 A).
Considering a given HFpEF patient with ESP 1 = 150 mmHg and Ees = 5 mmHg/mL, then LVESV 1 = 150/5 = 30 mL. A 20 mmHg reduction in ESP with a vasodilator produces a modest reduction in ESV: LVESV 2 = ESP 2 /Ees = 130/5 = 26 mL. The SV gain is therefore modest (ESV 1 – ESV 2 = 30 – 26 = 4 mL).
For a given patient with heart failure and systolic dysfunction, with ESP 1 = 100 mmHg and Ees = 0.5 mmHg/mL, LVESV 1 = 100/0.5 = 200 mL. A 10 mmHg afterload reduction produces a significant ESV decrease: LVESV 2 = 90/0.5 = 180 mL. The predicted SV increase is 200 – 180 = 20 mL. This may partially explain the contrasting results of renin-angiotensin blockade in HFrEF patients (CONSENSUS) and HFpEF patients (I-PRESERVE) . Similarly, inotropic therapy is unlikely to improve HFpEF patients due to the already heightened Ees, while patients with HFrEF with low Ees benefit from inotropism.
Hyperthyroidism is a known cause of congestive heart failure. Merillon et al. assessed LV function in seven hyperthyroid subjects in the basal state and after 15 mg intravenous propranolol. Higher CO and +dP/dt max , lower PVR, but similar Ees were found in hyperthyroidism compared to normal subjects. Propranolol administration resulted in reduced SV and CO. Higher LVESV and thereby lower Ees were found following propranolol administration. These findings confirm that, in clinical practice, beta-adrenergic blockade should be cautiously administered in patients with thyrotoxicosis.
Left ventricular filling capacities
The authors also demonstrated a reduction in LV distensibility with ageing and hypertension. LVEDP was found to increase with age and hypertension despite a reduction in LVEDV ( Fig. 4 A and B) . These findings suggest that the entire diastolic pressure-volume relationship is shifted leftward and upward with progressive concentric remodelling ( Fig. 5 ) . Whether diastolic stiffness is intrinsically heightened in HFpEF that affects patients with longstanding hypertension remains a subject of debate .
