Portal de Periódicos da CAPES

Right Versus Left Ventricular Failure

2014; Lippincott Williams & Wilkins; Volume: 129; Issue: 9 Linguagem: Inglês

10.1161/circulationaha.113.001375

1524-4539

Mark K. Friedberg, Andrew N. Redington,

Cardiovascular Function and Risk Factors

HomeCirculationVol. 129, No. 9Right Versus Left Ventricular Failure Free AccessResearch ArticlePDF/EPUBAboutView PDFView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticlePDF/EPUBRight Versus Left Ventricular FailureDifferences, Similarities, and Interactions Mark K. Friedberg, MD and Andrew N. Redington, MB BS, MRCP(UK), MD, FRCP(UK), FRCPC Mark K. FriedbergMark K. Friedberg From the Labatt Family Heart Center, Department of Paediatrics, Hospital for Sick Children and University of Toronto, Toronto, ON, Canada. and Andrew N. RedingtonAndrew N. Redington From the Labatt Family Heart Center, Department of Paediatrics, Hospital for Sick Children and University of Toronto, Toronto, ON, Canada. Originally published4 Mar 2014https://doi.org/10.1161/CIRCULATIONAHA.113.001375Circulation. 2014;129:1033–1044IntroductionVentricular failure manifests in many forms, its underlying physiology ranging from overt left ventricular (LV) systolic dysfunction to isolated right ventricular (RV) diastolic dysfunction, and the wide portfolio of resulting symptoms vary from chronic fluid retention to acute multiorgan dysfunction and death. In this review, we discuss the morphological, functional, and molecular similarities and differences in RV and LV responses to adverse loading and failure. We further discuss whether LV and RV function and failure can truly be discussed as separate entities and thereby examine interactions between the ventricles that on one hand contribute to ventricular dysfunction but on the other may be harnessed for therapeutic benefit.Embryological and Physiological Differences Between the RV and LVThe RV and LV have different embryological origins.1 The LV originates from the primary heart field; the RV, from the secondary heart field. Consequently, several genes specifically control RV formation, including, among others, Hand2 and Tbx20.2 During gestation, the RV functions as the systemic ventricle (Figure 1A). During fetal life, in addition to supplying the modest amount of pulmonary blood flow, the RV pumps blood to the lower body and placenta and contributes more than half of the combined cardiac output.3 With the transition from fetal to postnatal physiology and with the reduction in pulmonary vascular resistance, the subpulmonary RV transforms its morphology and geometry, becoming a thin-walled chamber to adopt its postnatal physiological characteristics.4 Because it faces a low impedance pulmonary circulation, the normal postnatal RV maintains a cardiac output equal to that of the LV at approximately a fifth of the energy cost. The trapezoidal RV pressure-volume loop reflects this difference, with few if any isovolumic periods. Consequently, RV output starts early during pressure generation and is later maintained by a "hangout period" when antegrade flow continues into the pulmonary artery despite the onset of RV relaxation.5 In contrast, the rectangular LV pressure-volume loops reflect the LV square-wave pump function with distinct and well-developed isovolumic contraction and relaxation periods.6 Likewise, RV myocytes display faster twitch velocities than LV myocytes.7Download figureDownload PowerPointFigure 1. A, Fetal echocardiogram in the short-axis view. Note the equal wall thicknesses of the right (RV) and left (LV) ventricles. The septum is in the neutral position as a result of equal pressures in the ventricles in normal fetal physiology. B, Parasternal short-axis view in a patient with pretricuspid Eisenmenger syndrome. Note the similarities to the fetal echocardiogram with equally thick walls of the LV and RV and the neutral position of the interventricular septum.The physiological differences are reflected in morphological differences between the ventricles (Table 1). The low-pressure RV is triangular in the sagittal plane and crescent-shaped in cross section as a result of the concave RV free wall and convex interventricular septum wrapping around the high-pressured, thick-walled, bullet-shaped LV. Consequently, although the normal RV has a lower ratio of volume to surface area and a thinner wall than the LV,8 the low cavity pressure determines a lower wall stress and lower oxygen demands.Table 1. Differences Between the Left and Right Ventricles Under Normal ConditionsLeft VentricleRight VentricleEvolutionary developmentEarlyLateEmbryological originPrimary heart fieldSecondary heart fieldMorphological characteristicsBullet shape; prolate ellipsoidComplex, crescenticMyocardial characteristicsThick smooth walls; fine trabeculationsThin, heavily trabeculated wallsMyocardial architecturePredominant radial myocyte orientation in the midlayers; subendocardial myocytes follow right-hand helix configuration; subepicardial myocytes form left-hand helixPredominant longitudinal myocyte orientation; angulated intrusion of superficial myocytes toward the endocardiumPhysiological pump conditionsHigh-resistance, high pressure pump; dominant radial thickening and contraction during ejectionLow-resistance, low-capacitance pump; peristaltic-like motion from inflow to outflow during ejectionFlow characteristicsWell-defined isovolumic contraction and relaxation; no hangout periodNo or minimal isovolumic periods; hangout periodAnatomic differences are also apparent in myocardial architecture. LV subepicardial and subendocardial fibers are oblique and helical with fiber angles ranging from 30° to 80° with a mean of ≈60°, whereas the midmyocardial myocytes are oriented predominantly in the short-axis plane of the equator with fiber angles of 20° to −20°.9 As a result, LV contraction is predominantly circumferential and radial with additional rotational and twisting motion. RV myocytes are predominantly longitudinal, creating a peristaltic contraction from the inlet to outlet and a bellows-like motion of the free wall toward the septum.10 That the ventricles differ in their anatomy and physiology is irrefutable, but as we discuss later, morphologically and functionally, they are inextricably linked not only in health but also as they react to disease.Does the RV Differ From the LV in Its (Mal) Adaptation to Adverse Loading?Although the RV is not immune to the direct effects of coronary disease with resulting global or regional ischemia, in clinical practice, RV physiology and failure are most frequently affected by increased preload or afterload. Adverse loading also affects LV function, but the RV is exquisitely dependent on, in particular, afterload. Even small changes in total pulmonary vascular resistance, as demonstrated by modest increases in mean airway pressure during positive pressure ventilation, can reduce RV contractile performance and lower cardiac output even when RV preload is maintained.11 In animal models, even modest increases in afterload lead to profound decreases in RV stroke volume.12 In contrast, much larger changes in LV afterload induced only modest changes in LV stroke volume.12 These experimental differences are reflected in clinical practice. Although patients with acute changes in systemic vascular resistance can compensate over a wide range, patients with acute pulmonary arterial hypertension (PAH), for example, in the setting of acute lung failure, frequently develop overt right heart failure and compromised cardiac output.Although when acute even modest increases in RV afterload cause dramatic reductions in RV output in most clinical scenarios, including PAH and RV outflow obstruction, changes in afterload are chronic and occur progressively. Indeed, in the chronic setting, the relative increase in RV afterload is much greater in PAH than the increase in LV afterload in systemic hypertension. The question becomes, Can the RV adapt to increased afterload, and is this response adaptive or maladaptive? This question relates to another question: What is RV failure? If RV output is maintained but the RV myocardium suffers detrimental remodeling and injury that ultimately affect its long-term function, the border between adaptation and failure becomes indistinct.Although we have emphasized the vulnerability of the RV to increased afterload, especially when acute,13 it is clear from clinical experience (eg, congenitally corrected transposition of the great arteries, Eisenmenger syndrome, systemic pulmonary shunts) that the RV can maintain function and adequate output in the face of systemic pressure over prolonged periods.Eisenmenger syndrome is an interesting example of RV adaptation to continuous system-level resistance from fetal life and throughout postnatal life. In contrast to the normal RV, the Eisenmenger RV myocardium never thins. Consequently, RV and LV wall thicknesses are similar from fetal to adult life (Figure 1B).14 Patients with Eisenmenger syndrome have better RV function, higher cardiac index, and lower mortality than patients with other causes of PAH, despite a higher pulmonary vascular resistance.15 Therefore, it would seem that the RV has some capacity to retain fetal characteristics and to maintain adequate function in the face of systemic resistance over many years. Likewise, patients with a systemic RV after an atrial switch procedure for transposition of the great arteries or congenitally corrected transposition of the great arteries can live for decades. At the same time, it is also apparent that these patients are prone to RV failure and increased mortality, although arrhythmias and tricuspid regurgitation are important drivers of morbidity and mortality beyond RV myocardial failure per se (Figure 2).16,17Download figureDownload PowerPointFigure 2. Color-flow Doppler in a patient with congenitally corrected transposition of the great arteries. In this abnormality, the left atrium (LA) connects to the right ventricle (RV), which acts as the systemic pumping chamber to eject blood into the aorta. The RV commonly fails over time. Structural tricuspid abnormalities and progressive RV dysfunction commonly lead to tricuspid regurgitation, which is a major factor, perhaps more important than myocardial contractile dysfunction per se, driving morbidity and mortality in this condition. LV indicates left ventricle; and RA, right atrium.How Does the RV Adapt to Chronically Increased Afterload, and Does This Differ From the LV?In chronically increased afterload, the RV pressure-volume relation shifts from a trapezoidal shape, reflecting its high efficiency/low impedance (Figure 3), to a square or rectangular shape, with well-developed isovolumic contraction and relaxation periods, indistinguishable from the normal LV pressure-volume loop (Figure 3).Download figureDownload PowerPointFigure 3. Right ventricular (RV) pressure-volume (P-V) loops obtained by a conductance catheter. The diagram on the right demonstrates the use of P-V loops. The white solid lines pass tangential to the end-systolic P-V points of a "family" of loops produced by varying the loading conditions. The slope of this line gives the RV end-systolic elastance. A steeper slope depicts higher end-systolic elastance. Loop a depicts a normal RV P-V loop. Loop b represents a compensated, chronically hypertensive RV. Loop c is obtained from a decompensated hypertensive RV. Note the decrease in RV end-systolic elastance from the compensated RV depicted in loop b to the decompensated RV depicted by loop c.This compensated RV adaptation to chronic afterload occurs through increased contractility, manifested by a steeper end-systolic pressure-volume relation (Figure 3),18,19 thereby preserving cardiac output.19 However, with disease progression, the RV ultimately fails, leading to further RV dilatation, a rightward shift in the pressure-volume curve, and a resultant decrease in end-systolic elastance and compromised cardiac output (Figure 3).18These physiological changes are mirrored by changes in myocardial contraction patterns. The systemic RV demonstrates increased circumferential contraction relative to decreased longitudinal shortening in a pattern indistinguishable from the normal LV.20 This observation is somewhat surprising in that diffusion tensor magnetic resonance imaging shows that fiber orientation in RV hypertrophy is not fundamentally different from normal.21 Likewise, the functionally single systemic RV in hypoplastic left heart syndrome may adapt a more circumferential versus longitudinal contraction pattern after stage 1 of surgical palliation (a particularly vulnerable period for these patients).22 Yet, the RV in hypoplastic left heart syndrome continuously faces systemic resistance from fetal through postnatal life, and it is difficult to attribute changes in RV contraction patterns to increased afterload alone.At a molecular level, recent literature has highlighted differences between the RV and LV in the expression of genes involved in the response to pressure loading and failure.23 Some of these differences are detailed in the following text and are summarized in Table 2.24–34 Likewise, there are differences in the RV response to certain effectors, including adrenergic hormones. Although α1-adrenergic agonists increase LV contractility, they may decrease RV contractility.24 Long-term infusion of norepinephrine induces LV but not RV hypertrophy.25Table 2. Molecular Differences Between the Left and Right Ventricles in Response to Adverse Loading*Molecular ResponseRight VentricleLeft VentricleWnt pathway activation and glycolysis-to-glucose oxidation metabolism in afterloadHigher activation; potentially inefficient energy metabolism26Lower activation; potentially improved energy metabolism26Fibrotic response to volume loadingStronger27Weaker27Irx2 transcription factor expression in afterloadNot expressed28Expressed28Atrial natriuretic peptide expressionNot expressed29Expressed29miRNA 133a expression in experimental PAHDecreased28…Expression in afterload of clusterin, neuroblastoma suppression of tumorigenicity 1, Dkk3, Sfrp2, formin binding protein, annexin A7, lysyl oxidaseIncreased30Not increased30Response to α-1 adrenergic receptor agonistsDecrease contractility24Increase contractility24Response to long-term norepinephrine infusionNo hypertrophy25hypertrophy25miRNA 28, 148a, and 93 expression in failureIncreased30Decreased30Response to dichloroacetate in hypertrophyIncreased inotropy26Unchanged inotropy26Response to PDE5 inhibitors in hypertrophyIncreased inotropy31Unchanged inotropy31Response to recombinant BNP infusionUnchanged inotropy33,34Increased inotropy32BNP indicates brain natriuretic peptide; Irx2; Iroquois homeobox 2; miRNA, microRNA; PAH, pulmonary arterial hypertension; and PDE5, phosphodiesterase type-5.*See text for details.In response to increased afterload, the RV reverts to a fetal gene pattern, re-expressing genes normally expressed in the fetal but not postnatal RV. This includes a shift from α- to β-myosin heavy chain expression and an increase in adrenergic receptors, calcineurin activation,35–37 and phosphodiesterase type-5 expression.31A detailed analysis of the accumulating experimental literature on the progression from adaptive to maladaptive hypertrophy and from hypertrophy to failure is beyond the scope of this review. However, suffice it to say that not all studies of RV afterload show RV failure. One study in rodents found that progressive pulmonary artery banding induces RV hypertrophy but not failure, as evidenced by increased contractility; although RV pressure was at only ≈60% of systemic levels, a degree of severity that usually does not induce symptoms in patients with pulmonary stenosis.38 Although it is clear that pressure overload alone does not induce RV failure, and even if contractility increases as an adaptive response to increased afterload, RV stroke volume and cardiac output may still decrease, fulfilling the definition of failure for some.39 Other rodent studies inducing systemic RV pressure have shown RV failure, RV dilation, decreased RV wall motion, elevated RV end-diastolic pressure, decreased cardiac output, clinical right heart failure, and decreased survival.40Microarray gene chip studies of mice with LV hypertrophy from aortic banding compared with mice with RV hypertrophy from pulmonary banding have demonstrated both similar and different LV and RV adaptive mechanisms.40 One pathway more activated in the pressure-loaded RV compared with the pressure-loaded LV is the Wnt pathway. Wnt regulates glycogen synthase kinase-3b activity, a serine/threonine protein kinase active in multiple intracellular signaling pathways, including cell proliferation, migration, inflammation, glucose regulation, and apoptosis.41,42 Therefore, there are potentially multiple differences between the RV and LV in their adaptation to increased loading and potential differences in metabolism, mitochondrial remodeling, and glycolysis-to-glucose oxidation coupling. These metabolic changes may subsequently lead to hyperpolarization of the mitochondrial membrane potential in RV hypertrophy, inefficient energy metabolism, and increased lactate production at an earlier stage of maladaptation compared with the LV.26 These molecular effects have potential therapeutic implications specific for the pressure-loaded RV. For example, dichloroacetate, which improves glycolysis-to-glucose oxidation coupling, may increase inotropy in the hypertrophied RV.26 Likewise, phosphodiesterase type-5 is not expressed in the normal RV but is upregulated in RV hypertrophy.31 Accordingly, phosphodiesterase type-5 inhibitors such as sildenafil increase RV contractility in experimental models of RV hypertrophy but not in the normal RV.31There may also be differences in the beneficial action of brain natriuretic peptide in the 2 ventricles. Brain natriuretic peptide is expressed in both ventricles, and serum levels correlate with the severity of both LV and RV failure.43 However, although recombinant brain natriuretic peptide (neseritide) improves cardiac output, symptoms, and ventricular performance in LV failure,32 similar beneficial effects have not been shown in PAH and RV failure.33,34In increased volume loading, the RV appears more prone than the LV to developing fibrosis, as demonstrated in an experimental pig model with an aortocaval shunt.27 Similarly, patients after surgical repair of tetralogy of Fallot who have long-standing RV volume load secondary to pulmonary insufficiency develop RV fibrosis, even remote from surgical incision sites.44 This is clinically important as a risk factor for increased propensity to arrhythmias, exercise intolerance, and RV failure.44,45 It has been suggested that these differences in response between the RV and LV to volume loading may stem from the different embryological origin of the 2 ventricles.27There has been increasing interest in the role of microRNAs (miRNA) as regulators of a wide range of cardiovascular processes and as possible therapeutic targets.46,47 Studies have highlighted both overlapping and varying expression between the failing RV and LV in various transcription factors, mRNA, and miRNA expression. Some transcription factors such as Iroquois homeobox 2 are expressed in the LV but not the RV. Others, including some nuclear receptors and insulin growth factor-1, are expressed in both ventricles but to different degrees.28 The lack of Iroquois homeobox 2 in the normal RV may explain why atrial natriuretic peptide is not expressed in the normal RV.29 Failing RVs from rodents with SU5416/hypoxia-induced PAH show overall increased miRNA but a specific decrease in miRNA 133a.28 miRNA 133a is thought to suppress cardiac fibrosis and is decreased in the failing LV secondary to aortic constriction.46,48 This aligns with the marked upregulation of connective tissue growth factor/CCN2 and other profibrotic signaling molecules during RV and LV fibrosis in our and other models of RV afterload and failure.28,49 In contrast, miRNA 21 and 34c* may increase during LV failure but decrease in RV failure.28 Reddy et al30 investigated miRNAs during the transition from RV hypertrophy to RV failure and compared these with miRNA expression in LV hypertrophy or failure. During RV hypertrophy, there was altered expression of miRNAs 199a-3p, which is associated with cardiomyocyte survival and growth. With the progression to RV failure and reactivation of the fetal gene program, there was increased expression of miRNA 208b, as well as miRNA 34, miRNA 21, and miRNA 1, which are associated with apoptosis and fibrosis. These patterns of miRNA expression are largely similar to LV hypertrophy and failure. However, there were several notable differences between RV and LV miRNAs linked to cell survival, proliferation, metabolism, extracellular matrix turnover, and impaired proteasomal function (miRNA 28, miRNA 148a, and miRNA 93), which were upregulated in RV hypertrophy or failure and downregulated or unchanged in LV hypertrophy or failure.30 Similarly, these investigators found that although the molecular responses of the RV and LV to increased afterload are mostly concordant, several key transcripts are increased in the afterloaded RV but not in the afterloaded LV. These included clusterin, neuroblastoma suppression of tumorigenicity 1, Dkk3, Sfrp2, formin binding protein, annexin A7, and lysyl oxidase. From these studies, it seems that although the ventricles share many common response mechanisms to stress, several key differences may warrant different management strategies in ventricular hypertrophy and the progression to failure. A better understanding of these subcellular events may lead to the development of novel, ventricle-specific treatments. Inhibition of miRNA 208a has recently been shown to abrogate LV dysfunction in rodent models of LV failure50; similar responses, albeit targeted toward a different miRNA signature, might be anticipated for RV failure.Are the RV and LV Really Different?Although there are undoubtedly differences in RV and LV responses to adverse loading and differences in response of the more complex heart failure syndrome to various therapies, it is also evident that the 2 ventricles share many common features in response to adverse loading and failure. The fetal gene pattern shift, particularly the myosin heavy chain shift from the α to β isoform, a hallmark of fetal gene reactivation, is also triggered in LV failure.35 Likewise, the progression from compensated to decompensated hypertrophy occurs in both ventricles. Common findings in both RV and LV hypertrophy are collagen deposition, fibrosis, and extracellular matrix remodeling.51 The mechanisms inducing fibrosis are multiple, and in the setting of increased ventricular afterload, putative triggers may include regional ischemia, necrosis, and apoptosis, among others.52 Our own experimental data suggest common injury pathways in both ventricles. In response to isolated RV afterload induced by pulmonary artery banding in a rabbit model, we found RV and LV fibrosis and upregulation of transforming growth factor-β1 signaling in both ventricles (Figure 4). Just as RV fibrosis is commonly seen in the setting of both severe RV afterload and chronic pulmonary regurgitation, LV fibrosis is common in both aortic stenosis and regurgitation.44,53,54 Pharmacological agents that decrease either pulmonary vascular resistance or systemic vascular resistance may attenuate the progression of fibrosis in the RV and LV, respectively. Likewise, mechanical unloading of the LV by LV assist devices can attenuate fibrosis in both ventricles.55 Microarray studies of thousands of genes have shown that although the RV response to pressure-induced hypertrophy is characterized by a stronger transcriptional response compared with the LV, there was no evidence of qualitatively distinct regulatory pathways in the RV compared with the LV.56Download figureDownload PowerPointFigure 4. Representative sections showing Masson trichrome staining for collagen content. The bar graph of the quantitative analysis shows increased collagen in response to pulmonary arterial banding (PAB) in both the right ventricle (RV) and the left ventricle. In association with increased fibrosis, there is upregulation of profibrotic signaling molecules, including transforming growth factor-β (TGFβ), connective tissue growth factor (CTGF), and matrix metalloproteinases (MMP) 2 and 9.49Mechanical and Functional Interdependence Between the RV and LVAlthough it has been customary to consider LV function and RV function as separate entities, this approach is flawed. The ventricles share common injury mechanisms and anatomically share fibers that encircle both ventricles. They are intimately attached through a common septum and share the pericardial space (Figure 5).57–60 Consequently, the function of the 2 ventricles is inextricably linked in both the structurally normal and abnormal heart.Download figureDownload PowerPointFigure 5. Diffusion tensor magnetic resonance image demonstrating orientation of myocytes. From the figure, it is evident that fibers cross from the parietal walls of the left ventricle to the right ventricle. Note also how myocyte pathways intrude from the epicardium to endocardium by means of a right-hand helix movement about a transverse circular axis.60The importance of LV-to-RV myocardial cross-talk was elegantly demonstrated in an experimental study of intact explanted hearts in which electric but not mechanical continuity between the RV and LV was interrupted.61 RV pacing led to little detectable mechanical activity (measured as developed pressure) in the LV. Conversely, however, pacing-induced contraction of the electrically isolated LV was associated with the development of an almost normal RV pressure trace and pulmonary blood flow.61 Santamore et al62 further elucidated the individual effects of LV volume loading and dysfunction on RV developed pressure. Reducing LV volume from its optimal volume to zero caused a 5.7% decrease in RV developed pressure, whereas ligating the coronary supply to the LV free wall resulted in an additional 9.3% decrease in RV developed pressure. Cutting the LV free wall to prevent any developed LV free wall force caused a further 45% decrease in RV developed pressure. Changes in RV developed pressure resulting from changes in LV volume and from coronary occlusion correlated with the degree of septal bulging into the RV cavity during systole, suggesting that the septum plays an important role in mediating ventricular-ventricular interactions. From these experiments, it was estimated that >50% of the normal RV mechanical work may be generated by LV contraction and that the LV free wall plays a pivotal role in RV function.62 Similarly, LV isovolumetric contraction results in simultaneous increases in RV stroke volume and developed pressure for a constant RV volume.63 Hoffman et al64 expanded on these observations in in vivo experiments. By replacing the RV myocardium with a noncontractile prosthesis, they were able to show virtually normal RV pressure generation as a consequence of normal LV shortening. Just as interesting was the observation that intact RV geometry is crucial for normal LV mechanical performance. During gradual enlargement of the noncontractile RV free wall, there was a progressive reduction in both RV and LV mechanical work; that is, as the RV dilated, LV pressure development and stroke work decreased. These experimental phenomena have been shown in vivo in the human heart during pre-excitation of 1 ventricle by pacing or during extrasystolic beats.65 Normally, LV electric activation and RV electric activation are temporally close enough that it is difficult to separate the peak dP/dt spike of 1 ventricle from the other. When LV activation is sufficiently separated from RV activation by a ventricular extrasystole or by left bundle-branch block, the contribution of LV contraction to RV dP/dt becomes apparent.65The RV also profoundly affects LV performance. Changes in RV volume lead to substantial changes in load-independent measures of LV function and a shift in the LV pressure-volume relation.66 These effects may be clinically relevant when the RV is volume unloaded by placement of a caval pulmonary shunt. Danton et al67 showed experimentally that acute RV ischemia induced by coronary artery ligation induced LV dysfunction as measured by end-systolic elastance, a load-independent measure of LV contractility. LV dysfunction secondary to RV ischemia was reversed by the addition of a caval pulmonary shunt with restoration of LV end-systolic elastance. These load-independent effects on LV contractility were likely mediated by the caval pulmonary shunt relieving RV volume load, thereby limiting RV dilation and restoring LV cavity geometry.67 One may envisage similar effects on LV function in the clinical setting of various congenital heart disease lesions, for example, when a caval pulmonary shunt is placed as part of a "1.5-ventricle repair" for RV hypoplasia or dysfunction.In PAH, in addition to decreased cardiac output that results directly from RV failure, leftward displacement of the interventricular septum impedes LV filling (Figure 6).68–73 This secondary LV geometric change is linearly related to cardiac output, whereas RV end-diastolic volume, in and of itself, is not related to cardiac output.69 Similarly, in patients with tetralogy of Fallot and conduit stenosis, the prolonged septal shift induced by RV afterload and prolonged RV contraction leads to reduced LV filling as the septum bulges into the LV in diastole.74 Relief of conduit stenosis reverses septal curvature, shortens RV contraction, synchronizes LV and RV contraction and relaxation, improves LV filling, and improves exercise capacity.74 The LV eccentricity index, a simple echocardiographic index that quantifies the anterior-posterior LV compression by the distended RV, correlates with