Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Global and Regional Systolic Function of the Left Ventricle
Case Study
An 82-year-old male presents to the office complaining of several months of shortness of breath and cough after walking two blocks or up a flight of stairs. He has also noted leg swelling and occasional palpitations. Past medical history is remarkable for hypertension, type 2 diabetes, lumbar spinal stenosis, and carpal tunnel syndrome. On physical exam, the blood pressure (BP) is 150/90 mmHg and the heart rate (HR) is 84 bpm. He is in no distress. The head exam is unremarkable with a normal tongue on inspection. Lungs are clear, and cardiac auscultation demonstrates a regular rhythm, nondisplaced point of maximum impulse (PMI), normal S 1 , S 2 with no murmur, and an S 4 gallop. No jugular venous distention is observed, and all central and distal pulses are felt bilaterally. The abdominal exam is normal, and there is no lower extremity edema.
A surface electrocardiogram (ECG) revealed sinus rhythm, left axis deviation, and minimal voltage criteria for left ventricular hypertrophy (LVH). Echocardiographic images are shown ( Figs. 29.1, 29.2, and 29.3 ). The clinical problem is one of dyspnea on exertion in the setting of multiple risk factors that bring into question the presence of an ischemic equivalent or overt heart failure (HF).
Introduction
Since the latter half of the 20th century, cardiovascular science has been searching for the still elusive ideal measurement of human left ventricular (LV) function; to that end, ejection fraction (EF) was, and still is, the most commonly used tool to assess what many times has been erroneously described as “myocardial performance,” “global systolic function,” or “LV contractility.”
The (almost intuitive) recognition that ventricular function directly impacts clinical outcomes was confirmed when seminal studies conducted during and after the 1970s demonstrated that EF was indeed an important and often fundamental parameter, both in the selection and adjudication of therapies in conditions such as coronary artery disease (CAD), valvular lesions (particularly aortic stenosis and mitral/aortic regurgitation), and more recently when considering device implantation in HF patients.
Furthermore, the measurement of LV function surpassed the horizon of its historical understanding based on knowledge of systole when, largely thanks to technology, the study of diastole came into play as a critical determinant of cardiac physiology. Presently, practitioners and researchers alike dispose of myriad methodologies that allow quick, reliable, and clinically relevant assessment of LV function making it an almost essential index in modern cardiology and medical practice at large, particularly in the setting of suspected or known HF.
The multiple phenotypes of HF, a syndrome conformed by symptoms of dyspnea, fatigue, and exercise intolerance coupled with signs of volume overload (edema, jugular venous distention, or pulmonary congestion) caused by structural/functional impairment of proper diastolic filling and/or cardiac output, have further catapulted basic science and clinical research in this field. Today, and paradoxically as the result of improved outcomes in acute and chronic ischemic heart disease, among other achievements, the diagnosis of HF has reached epidemic proportions both in the United States and worldwide, with several remaining key issues to resolve specially in what has been known as diastolic heart failure (DHF) or HF with preserved ejection fraction (HFpEF), arguably the most common form of HF in the general population, terms that may be used interchangeably through this chapter.
Traditionally, HF has been understood as systolic when the EF is less than 40% (HF with reserved EF [HFrEF]) or diastolic when greater than 50% (HFpEF); however, as it will be discussed, questions remain on the proper identification of these patients beyond a simplistic separation based on a single measurement of cardiac activity since many subjects do not necessarily fit in this classification, nor can they always be adequately categorized. In addition, observational studies show inconclusive results as far as mortality differences between HFrEF and HFpEF are concerned, though a recent meta-analysis of several publications has demonstrated a lower risk of death among those with better EFs. These are some of the very reasons why many question the true nature of HF syndromes as defined based on EF, a parameter that, as this chapter will attempt to demonstrate, may no longer be adequate or sufficient for those purposes.
As it has been alluded to, multiple randomized clinical trials show the unequivocal benefit of several therapeutic modalities in patients affected by HFrEF, not only on functional class or hospitalization rates but also on survival. The same cannot be stated about patients classified as having symptomatic or asymptomatic HFpEF, a group where trials have been difficult to design and/or shown equivocal results at best.
This chapter will review aspects of diastolic parameters relevant to regional and global systolic LV function departing from (1) a critical and historical analysis of EF as an indicator of LV function, (2) current controversies surrounding EF as a fundamental determinant of HF, (3) a review of older and newer tools and technologies available in the pursue of diastolic function, (4) a discussion on the normal physiologic mechanisms of how systole and diastole interact with each other, (5) an analysis of the various hemodynamic parameters obtained by echocardiography and invasive assessment relevant to how systole talks to diastole, and (6) a critical review of the most current literature dealing with the impact of parameters of systolic function in patients with HFpEF. With this, we intend to shed light into the complex process of how regional systolic LV dysfunction and diastolic abnormalities can affect global systolic performance.
Historical Views of EF as an Indicator of LV Function
As already stated, there is little doubt that for many decades LV EF measurement has played a fundamental role in modern cardiology, particularly in the decision-making process of therapeutic interventions, to the extent that an unequivocal inverse relationship exists between EF and all-cause mortality once it reaches a point below 40%. The prognosis and management of patients with CAD, cardiomyopathies, HF, arrhythmias, or congenital heart disease rely heavily on the accurate determination of LV size and function. Furthermore, thresholds for the timing of surgical valvular intervention, coronary revascularization, and device therapies have been established based on the volumetric quantification of EF making its precise measurement a matter of critical importance. It is understood then how the preoccupation for an accurate determination of LV EF has haunted clinicians for so long to the point that currently there are at least six clinically available methods for its assessment in humans.
Echocardiography has been the mainstay of EF evaluation despite limitations on image quality, technical and interpretation expertise, equipment variability, and intrinsic errors related to assumptions of LV geometry. Regardless of the technology used, however, the quantification of EF still relies on subjective visual estimation analysis or human-driven quantitative measurements, although more recently machine learning algorithms utilizing either ultrasound, radioisotopes, tomographic data, or invasively obtained angiographic images are beginning to evolve each with their own set of advantages and disadvantages. Visual estimation of EF, while providing a rapid, bedside assessment of global systolic function that can assist on quick clinical decisions, is fraught with the previously stated inconsistencies and, perhaps more importantly, with significant limitations resulting from geometric assumptions, somewhat limited reproducibility, and interobserver variability reported as high as 19%. For these reasons, current guidelines recommend against its use, endorsing instead the biplane rule of discs (modified Simpson method) that although requires tracing of the LV cavity and endocardial border, provides a more accurate assessment of the contribution of longitudinal contraction. The understanding of EF as an indicator of LV performance began decades ago when investigators categorized the basic concept of its calculation as “the central measure of LV function.” However, the LV EF equation, defined as a fraction of the volume ejected in systole (stroke volume, calculated as the difference between end-diastolic and end-systolic volume) over the LV end-diastolic volume, obviates several important facts that significantly impact its accuracy and true representation of contractility. That is, the change in stroke volume over the end-diastolic volume at a given point assumes a constant load; preload and afterload, important modifiers of EF, are often ignored; and compensatory (or pathologic) structural changes on the left ventricle may significantly alter ventricular volumes. Recently, investigators have gone as far as calling for the removal of EF as the gold standard surrogate of systolic function, while others assert that it has exhausted its usefulness as a presumed marker of contractility. Konstam and Abboud have argued that structural changes in LV end-diastolic volume—provoked by conditions such as remodeling occurring after myocardial infarction, compensatory hypertrophy or dilatation, aging, diabetes, or metabolic syndrome—are characterized by reduction on myocyte contractility/lusitropy that does not necessarily reflect changes on EF.
The search for prognostic indicators of potentially devastating conditions such as cardiovascular (CV) disease carries the benefit of risk-stratifying patients and illnesses where either their improved well-being remains a challenge or there is more than one (imperfect) marker under discussion. That is the case of the association between reduction of LV EF and outcomes in the setting of hypertrophy where a large meta-analysis by Kalam et al. has brought some light into the matter by assembling evidence that global longitudinal strain (GLS) represents an accurate predictor of death, malignant arrhythmias, and HF hospitalization superior to EF. After reviewing 16 studies that included nearly 6000 patients with various CV conditions, the authors found that mortality was independently associated with each SD (standard deviation) change in the absolute value of baseline GLS (HR 0.50, 95% confidence interval [CI] 0.36–0.69; p < 0.002), and less strongly, with EF (HR 0.81, 95% CI 0.72–0.92; p = 0.572). The HR (hazard ratio) per SD change in GLS was associated with a reduction in mortality 1.62 (95% CI 1.13–2.33; p = 0.009) times greater than the HR per SD change in LV EF. Justifiably so, these investigators stated that GLS should not only be paired and quoted with LV EF but that perhaps it may even replace it specially on those affected by HFpEF, a population that represents half of all HF cases.
Despite the limitations and controversies just discussed, the understanding and therapeutic approach to HF, the predominant epidemic in North America and one of the most costly conditions in current health care, are precisely largely dependent on a single measurement: the LV EF. We should bear in mind that HF phenotypes are defined based on this rather simplistic index of LV function where preserved EF is only 10 percentage points apart from reduced EF with all the implications that such binary approach implies, let alone the uncertain characterization of those falling within the gray zone of 40% to 49%.
Clearly, the answer to these challenges relies on a more precise understanding of the underlying disease responsible for suspected/confirmed LV systolic dysfunction, the multiple mechanisms involved, and certainly the application of newer tools and evolving concepts on its day-to-day approach.
Current Controversies on EF as the Fundamental Determinant of HF
EF by definition is within the normal range in patients labeled as suffering from HFpEF while contractility may not; whether related to microvascular abnormalities, autonomic dysfunction, chronotropic incompetence, calcium metabolic derangements, or abnormal myocardial energetics, it is generally accepted that LV systolic function (contractility) impairment in HFpEF can be demonstrated using volume-dependent and volume-independent markers such as tissue Doppler imaging (TDI) and speckle tracking strain imaging, among others. Even in those with normal or only mildly impaired systolic function it is important to recognize that systolic reserve and peak cardiac output during exercise may also be limited, resulting from abnormal elastic recoil that in turn further impairs diastolic performance.
From the early times of myocardial deformation and diastole imaging, Aurigemma, Gaasch, and others not only alerted to the fact that EF is notably influenced by loading conditions and contractility both acutely and on chronic bases, but also called for a fundamental understanding of the process through which regional systolic function impacts the contractile behavior of the left ventricle in the setting of what was then known as DHF. Simply stated, abnormal diastolic behavior is not the exclusive mechanism of DHF since regional and global derangements of LV shortening are present as well, even when the EF is still measured as normal. It has been clear since then that a full comprehension of the contractile behavior of the ventricle requires knowledge of all indices that reflect systolic and diastolic performance, function, and contractility both globally and regionally ( Fig. 29.4 ).
It is worthwhile to revisit the definitions of the various indices of contractile behavior provided by these authors. Ventricular performance refers to the pumping ability of the LV assessed by the quantification of the pressure developed by this chamber, the stroke volume ejected, and the stroke work generated by it. Ventricular function includes shortening parameters, such as EF and fractional shortening of the minor axis or the apex-to-base long axis. Ventricular contractility refers to the inotropic state of the whole ventricle and includes isovolumic and ejection phase indices, as well as those determined at the end of ejection such as end-systolic elastance. Lastly, myocardial contractility addresses the intensity of cross-bridge activity, a basic propriety of the human heart muscle that is expressed on the extent and velocity of force development and fiber shortening, independent of loading and remodeling.
For all the previous reasons, it is quite appropriate to question if EF indeed constitutes the fundamental determinant of HF or not; discussing the following dilemmas intends to serve as a departing point that aims at illustrating the understanding of the complex interplay between systole and diastole.
Can the EF Be Normal and Impaired Contractility Still Be Present?
The LV EF can indeed be normal despite changes in cavity dimensions, hypertrophy, or regional systolic dysfunction; although earlier studies had reported the presence of normal systolic performance in HFpEF patients based on EF preservation plus normal systolic elastance and preload-recruitable stroke work, subsequent investigators showed that in reality many patients actually have abnormal longitudinal LV systolic function (as detected by reduced mitral annular systolic velocities and longitudinal strain [LS] on TDI and speckle tracking). Such preservation of EF exists thanks to compensatory conservation of twist that counteracts reduced shortening of the subendocardial fibers and to some extent, at least early on before the appearance of systolic dysfunction, by the compensatory effect of increased radial contraction.
Are There Distinct HFpEF Phenotypes That Obviate the Importance of EF?
Whether utilizing clinical individual parameters/comorbidities or bioinformatics methodologies, grouping and classification of HFpEF patients is currently a rapidly involving area of research that is even applying machine learning algorithms. The hope is that by establishing more precise categorizations, targeted interventions could be attempted in large-scale clinical trials to hopefully shed light into the question of whether HF phenotypes represent a spectrum of the same disease or altogether separate expressions requiring different therapies. Furthermore, phenotyping may not explain all the circumstances surrounding the initial process of asymptomatic diastolic dysfunction nor how it evolves into HFpEF and later into HFrEF, a topic that remains largely undefined.
It is firmly established that the histologic composition of cardiomyocytes of patients with HFpEF versus HFrEF is functionally and structurally different as determined by changes in several parameters mostly related to thickness and filament characteristics. In the former, hypertrophy and preserved myofilament density predominates, whereas in the latter there is minimal or no hypertrophy and significant loss of myofilaments. Despite these differences, as Sanderson and Yip have stated, the separation of HF subjects into clear-cut phenotypes, is contentious,to say the least, for several reasons: Regardless of EF there is similar neuroendocrine activation (brain natriuretic peptide [BNP] and norepinephrine levels), impaired chronotropicity, diastolic dysfunction, and abnormal contractile reserve on exercise among patients with either HFpEF or HFrEF.
For these reasons it is thought that the real difference between HF phenotypes may be the degree of ventricular remodeling and volume changes paralleling the extent of ventricular fibrosis and scarring. Figs. 29.5, 29.6, and 29.7 show various two-dimensional (2-D), Doppler, and speckle tracking derived LV myocardial mechanic parameters found in two distinct HF phenotypes.
Is It More Appropriate to Classify HF Based on Etiology and/or Pathophysiology?
Some authors have postulated three distinct mechanisms of HFpEF based on resting parameters: (1) increased myocardial stiffness and impaired diastolic relaxation that leads to high left atrial (LA) pressure predisposing to pulmonary venous congestion and dyspnea; (2) disturbed ventriculo-arterial coupling with increased systolic pressure load, hypertensive response to stress, and load-dependent diastolic dysfunction plus sensitivity of BP to circulating volume (see Chapter 6 ); and (3) impaired renal function and renal atherosclerosis, particularly among the elderly, causing excessive fluid retention and rapid rise on BP.
Using this concept, one would have the potential advantage of recognizing the most appropriate therapeutic intervention for each physiologic subset. However, since no single, reliable, and reproducible test can fulfill all these requirements, multiple tools would have to be used in the diagnostic process, let alone the fact that such approach will require clinical validation in large-scale series. In consequence, it cannot be overemphasized that since HF shares multiple etiologies and pathophysiologic mechanisms on a given patient, categorizations such as these are quite impractical.
Are Myocardial Deformation Echocardiographic Tools the Answer to Phenotyping?
The recognition that the myocardium is composed of multiple anatomically and functionally separate layers not only brought an unparalleled new understanding of ventricular mechanics but also the promise of a deeper comprehension of HF beyond the use of simple diastolic parameters or EF quantification. As such, departing from anatomic mechanisms and pattern of disease may be a reasonable initial approach to the HF conundrum since it could explain, for example, why CV mortality among HF groups shows such a disparate pattern despite their sharing similar morbidities and predisposing conditions.
In this context, Salem Omar has reviewed three proposed alternative subgroups of HF patients based on LV deformation mechanics: (1) HF with predominant longitudinal dysfunction, where compensatory hypertrophy and unopposed contractile response of the subepicardial layers maintain a normal EF and overall systolic LV performance; (2) HF with transmural dysfunction (longitudinal and circumferential), caused either by simultaneous involvement of both layers or extension of the disease from subendocardium to subepicardium that eventually leads to a fall in EF and LV chamber dilatation; and (3) HF with predominant circumferential dysfunction, in which case the EF is preserved but diastole is impaired due to abnormal twist and suction of the LV.
This mechanistic categorization explains how early subendocardial impairment attenuates LV longitudinal performance, a finding present in many conditions (such as ischemia or hypertension) that initially is compensated by enhanced mechanical deformation in other orientation, in concrete, circumferentially and radially. Should the process continue, particularly with transmural involvement, exhaustion of these mechanisms will invariably lead to chamber dilation and reduction of EF, hallmarks of advanced systolic HF. Furthermore, this myocardial layer approach not only has the advantage of elucidating the various stages of HFpEF that are, at least in part, supported by histologic studies but also could potentially facilitate the identification of cohorts in whom more specific therapies can become effective.
The Use of EF to Distinguish Between Systolic and Diastolic Dysfunction Is Inaccurate
Applying EF measurements to classify HF patients in clinical studies and in the assignment of therapeutic algorithms does not seem to be an ideal conceptualization given the challenges so far described. In fact, some have warned against the common misconception that a reduced EF implies systolic dysfunction and that a preserved EF equates with diastolic dysfunction. That is, to assume that the status of the EF defines the underlying disease mechanism is fraught with potential errors and perhaps has even been responsible for the inconclusive results of many trials. Those authors have gone as far as stating that LV EF exhausted its utility as a presumed marker of contractility and that this practice has hampered the understanding and therapeutic progress of HF as a clinical entity.
Taking into consideration that myocardial deformation imaging has already reached the bedside and the clinical arena, and recognizing that its different parameters have shown an association with several molecular phenomena (genetic expression, stimulation of β-adrenergic receptors, and messenger ribonucleic acid [mRNA] expression of intracellular calcium ATPase, among others), it appears that the time has come to focus on the underlying myocardial deformation mechanisms involved in HF if a new paradigm is to be established on what arguably represents the most challenging CV epidemic of the millennium.
Should Diastolic Stress Testing Be the Gold Standard Tool When Searching for Diastolic Dysfunction?
Many patients with or without echocardiographically demonstrated diastolic abnormalities at rest may not exhibit symptoms until provoked by effort; evidently, exercise-induced dyspnea is the hallmark of HF, and as such it is only fair to wonder if this largely underutilized testing modality could represent the true index of myocardial performance when confronting a patient with suspected HFpEF.
Under normal physiologic conditions, systolic twisting and early diastolic untwisting of the left ventricle allow the creation of vital negative intracavitary gradients or suction, so that upon increased demand the heart can not only accelerate ejection but also enable rapid filling despite a shortened diastole while keeping a low filling pressure thanks to enhanced myocardial relaxation (shorter tau [τ]).
With exercise, multiple disruptions on the relationship between systole and diastole occur, including reduced mitral annular motion, reduced systolic longitudinal strain, and failure to increase ventricular systolic rotation. LV suction, untwist, and LA function are also compromised and correlate with elevated LV filling pressure, peak VO 2 max, and symptoms, findings that identify a subgroup that carries a higher mortality at least in retrospective series.
Whether using a supine bicycle or posttreadmill images, Doppler signals can be obtained with reasonable success especially after a brief delay following peak exercise; although merging of both mitral E and A waves may occur, tissue Doppler velocities and strain data can be grabbed after stress in most individuals (see Chapter 18 ). Tan et al. studied subjects diagnosed with HFpEF and with limitations on cardiopulmonary exercise testing, showing reduced radial and longitudinal myocardial systolic strain both at rest and on exercise, reduced systolic and diastolic longitudinal functional reserve (mitral annular velocities failure to rise), impaired ventricular systolic rotation that failed to increase on exercise, and delayed ventricular untwisting with further worsening on exercise associated with diminished LV suction and reduction on stroke volume rise during exercise.
The clinical implications of such findings are that HFpEF does not represent a disorder of diastolic dysfunction or altered stiffness alone, but perhaps more importantly, as the authors state, that given the complex physiologic derangements seen and their link to the most common etiologies behind this disorder (hypertension, diabetes, aging, and the ensuing myocardial hypertrophy/fibrosis) symptoms and outcomes of HF may not change until architectural function of the heart is restored.
Global Longitudinal Speckle Tracking Strain Versus EF?
The already mentioned limitations of EF and the relative ease of performing speckle tracking echocardiography have brought up the question of whether the latter, in concrete GLS, should be a substitute for EF when it comes to the assessment of global LV function. Validated formulae useful in the estimation of EF derived from GLS [EF = − 4.35 × (GLS + 3.9)4], and vice versa, in addition to being practical, bring up the notion that GLS could become a better and more reliable parameter of the systolic function.
The first study that compared GLS to conventional measures of LV systolic function, EF, and wall motion score index (WMSI), as a prognosticator of all-cause mortality, showed that in a population with a wide range of EF, it was superior in predicting death. In fact, based on their data, the authors were able to determine that a cutoff of GLS greater than −12% was equivalent to an EF less than 35% concerning outcomes, leading to the conclusion that a GLS less than −12% could be recommended as a threshold for diagnosing severe LV dysfunction. The systematic review of the literature published by Kalam et al. previously discussed found that strong evidence seems to support the notion that the prognostic value of GLS for major adverse cardiac events is indeed superior to EF in a wide range of conditions ( Fig. 29.8 ).
Normal Physiology: Mechanisms Showing How Systole and Diastole Talk to Each Other
The cardiac cycle is a complex sequence of interactive elements that complement each other and lead to the major events of systole and diastole. Systole begins with the closure of the mitral valve, extending into isovolumetric contraction, aortic valve opening, ejection, and aortic valve closure. Diastole starts when the aortic valve closes and consists of isovolumic relaxation (IVR), mitral valve opening (MVO), early rapid filling, diastasis, late atrial filling, and mitral valve closure. Each of these elements play an important role in the integration of ventricular performance with key fundamental processes that today, thanks to technology, we can assess with great certainty. Fig. 29.4 summarizes graphically the main phenomenon occurring during the various phases of the cardiac cycle relevant to diastole.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here