Chapter 9: Heart Failure



The heart normally accepts blood at low filling pressures during diastole and then propels it forward at higher pressures in systole. Heart failure is present when the heart is unable to pump blood forward at a sufficient rate to meet the metabolic demands of the body (forward failure), or is able to do so only if the cardiac filling pressures are abnormally high (backward failure), or both. Although conditions outside the heart may cause this definition to be met through inadequate tissue perfusion (e.g., severe hemorrhage) or increased metabolic demands (e.g., hyperthyroidism), in this chapter, only cardiac causes of heart failure are considered.

Heart failure may be the final and most severe manifestation of nearly every form of cardiac disease, including coronary atherosclerosis, myocardial infarction, valvular diseases, hypertension, congenital heart disease, and the cardiomyopathies. More than 500,000 new cases are diagnosed each year in the United States, where the current prevalence is approximately 5 million. The number of patients with heart failure is increasing, not only because the population is aging, but also because of interventions that prolong survival after damaging cardiac insults such as myocardial infarction. As a result, heart failure now accounts for more than 12 million medical office visits annually and is the most common diagnosis of hospitalized patients aged 65 and older.

Heart failure most commonly results from conditions of impaired left ventricular function. Thus, this chapter begins by reviewing the physiology of normal myocardial contraction and relaxation.


Experimental studies of isolated cardiac muscle segments have revealed several important principles that can be applied to the intact heart. As a muscle segment is stretched apart, the relation between its length and the tension it passively develops is curvilinear, reflecting its intrinsic elastic properties ( Fig. 9.1A, lower curve). If the muscle is first passively stretched and then stimulated to contract while its ends are held at fixed positions (termed an isometric contraction), the total tension (the sum of active plus passive tension) generated by the fibers is proportional to the length of the muscle at the time of stimulation (see Fig. 9.1A, upper curve).

That is, stretching the muscle before stimulation optimizes the overlap and interaction of myosin and actin filaments, increasing the number of cross bridges and the force of contraction. Stretching cardiac muscle fibers also increases the sensitivity of the myofilaments to calcium, which further augments force development.

This relationship between the initial fiber length and force development is of great importance in the intact heart: within a physiologic range, the larger the ventricular volume during diastole, the more the fibers are stretched before stimulation and the greater the force of the next contraction. This is the basis of the Frank–Starling relationship, the observation that ventricular output increases in relation to the preload (the stretch on the myocardial fibers before contraction).

Figure 9.1. Physiology of normal cardiac muscle segments.

Physiology of normal cardiac muscle segments.

A. Passive (lower curve) and total (upper curve) length–tension relations for isolated cat papillary muscle. Lines ab and cd represent the force developed during isometric contractions. Initial passive muscle length c is longer (i.e., has been stretched more) than length a and therefore has a greater passive tension. When the muscle segments are stimulated to contract, the muscle with the longer initial length generates greater total tension (point d vs. point b). B. If the muscle fiber preparation is allowed to shorten against a fixed load, the length at the end of the contraction is dependent on the load but not the initial fiber length; stimulation at point a or c results in the same final fiber length (e). Thus, the muscle that starts at length c shortens a greater distance (?Lc) than the muscle at length a (?La). C. The uppermost curve is the length–tension relation in the presence of the positive inotropic agent norepinephrine. For any given initial length, an isometric contraction in the presence of norepinephrine generates greater force (point f) than one in the absence of norepinephrine (point b). When contracting against a fixed load, the presence of norepinephrine causes greater muscle fiber shortening and a smaller final muscle length (point g) compared with contraction in the absence of the inotropic agent (point e).
(Adapted from Downing SE, Sonnenblick EH. Cardiac muscle mechanics and ventricular performance: force and time parameters. Am J Physiol. 1964;207:705–715.)

A second observation from isolated muscle experiments arises when the fibers are not tethered at a fixed length but are allowed to shorten during stimulation against a fixed load (termed the afterload). In this situation (termed an isotonic contraction), the final length of the muscle at the end of contraction is determined by the magnitude of the load but is independent of the length of the muscle before stimulation (see Fig. 9.1B).

That is, (1) the tension generated by the fiber is equal to the fixed load; (2) the greater the load opposing contraction, the less the muscle fiber can shorten; (3) if the fiber is stretched to a longer length before stimulation but the afterload is kept constant, the muscle will shorten a greater distance to attain the same final length at the end of contraction; and (4) the maximum tension that can be produced during isotonic contraction (i.e., using a load sufficiently great such that the muscle is just unable to shorten) is the same as the force produced by an isometric contraction at that initial fiber length.

This concept of afterload is also relevant to the intact heart: the pressure generated by the ventricle, and the size of the chamber at the end of each contraction depend on the load against which the ventricle contracts, but are independent of the stretch on the myocardial fibers before contraction.

A third key experimental observation relates to myocardial contractility, which accounts for changes in the force of contraction independent of the initial fiber length and afterload. Contractility reflects chemical and hormonal influences on cardiac contraction, such as exposure to catecholamines. When contractility is enhanced pharmacologically (e.g., by a norepinephrine infusion), the relation between initial fiber length and force developed during contraction is shifted upward (see Fig. 9.1C) such that a greater total tension develops with isometric contraction at any given preload.

Similarly, when contractility is augmented and the cardiac muscle is allowed to shorten against a fixed afterload, the fiber contracts to a greater extent and achieves a shorter final fiber length compared with the baseline state. At the molecular level, enhanced contractility is likely related to an increased cycling rate of actin–myosin cross-bridge formation.

Determinants of Contractile Function in the Intact Heart

In a healthy person, cardiac output is matched to the body’s total metabolic need. Cardiac output (CO) is equal to the product of stroke volume (SV, the volume of blood ejected with each contraction) and the heart rate (HR):


The three major determinants of stroke volume are preload, afterload, and myocardial contractility, as shown in Figure 9.2.

Figure 9.2. Key mediators of cardiac output.

Key mediators of cardiac output.

Determinants of the stroke volume include contractility, preload, and afterload. Cardiac output = Heart rate × Stroke volume.


The concept of preload ( Table 9.1) in the intact heart was described by physiologists Frank and Starling a century ago. In experimental preparations, they showed that within physiologic limits, the more a normal ventricle is distended (i.e., filled with blood) during diastole, the greater the volume that is ejected during the next systolic contraction. This relationship is illustrated graphically by the Frank–Starling curve, also known as the ventricular function curve ( Fig. 9.3). The graph relates a measurement of cardiac performance (such as cardiac output or stroke volume) on the vertical axis as a function of preload on the horizontal axis.

As described earlier, the preload can be thought of as the amount of myocardial stretch at the end of diastole, just before contraction. Measurements that correlate with myocardial stretch, and that are often used to indicate the preload on the horizontal axis, are the ventricular end-diastolic volume (EDV) or end-diastolic pressure (EDP). Conditions that decrease intravascular volume, and thereby reduce ventricular preload (e.g., dehydration or severe hemorrhage), result in a smaller EDV and hence a reduced stroke volume during contraction. Conversely, an increased volume within the left ventricle during diastole (e.g., a large intravenous fluid infusion) results in a greater-than-normal stroke volume.

Table 9.1. Terms Related to Cardiac Performance

Term Definition
Preload The ventricular wall tension at the end of diastole. In clinical terms, it is the stretch on the ventricular fibers just before contraction, often approximated by the end-diastolic volume or end-diastolic pressure.
Afterload The ventricular wall tension during contraction; the resistance that must be overcome for the ventricle to eject its content. Often approximated by the systolic ventricular (or arterial) pressure.
Contractility (inotropic state) Property of heart muscle that accounts for changes in the strength of contraction, independent of the preload and afterload. Reflects chemical or hormonal influences (e.g., catecholamines) on the force of contraction.
Stroke volume (SV) Volume of blood ejected from the ventricle during systole.
SV = End-diastolic volume ? End-systolic volume.
Ejection fraction (EF) The fraction of end-diastolic volume ejected from the ventricle during each systolic contraction (normal range = 55% to 75%).
EF = Stroke volume ÷ End-diastolic volume.
Cardiac output (CO) Volume of blood ejected from the ventricle per minute. CO = SV × Heart rate.
Compliance Intrinsic property of a chamber that describes its pressure–volume relationship during filling. Reflects the ease or difficulty with which the chamber can be filled. Strict definition: Compliance = ? Volume ÷ ?

Figure 9.3. Left ventricular (LV) performance (Frank–Starling) curves relate preload, measured as LV end-diastolic volume (EDV) or pressure (EDP), to cardiac performance, measured as ventricular stroke volume or cardiac output.

Left ventricular (LV) performance (Frank–Starling) curves relate preload

On the curve of a normal heart (middle line), cardiac performance continuously increases as a function of preload. States of increased contractility (e.g., norepinephrine infusion) are characterized by an augmented stroke volume at any level of preload (upper line). Conversely, decreased LV contractility (commonly associated with heart failure) is characterized by a curve that is shifted downward (lower line). Point a is an example of a normal person at rest. Point b represents the same person after developing systolic dysfunction and heart failure (e.g., after a large myocardial infarction): stroke volume has fallen, and the decreased LV emptying results in elevation of the EDV. Because point b is on the ascending portion of the curve, the elevated EDV serves a compensatory role because it results in an increase in subsequent stroke volume, albeit much less than if operating on the normal curve. Further augmentation of LV filling (e.g., increased circulating volume) in the heart failure patient is represented by point c, which resides on the relatively flat part of the curve: stroke volume is only slightly augmented, but the significantly increased EDP results in pulmonary congestion


Afterload (see Table 9.1) in the intact heart reflects the resistance that the ventricle must overcome to empty its contents. It is more formally defined as the ventricular wall stress that develops during systolic ejection. Wall stress (?), like pressure, is expressed as force per unit area, and for the left ventricle, may be estimated from Laplace’s relationship:


where P is ventricular pressure, r is ventricular chamber radius, and h is ventricular wall thickness. Thus, ventricular wall stress rises in response to a higher pressure load (e.g., hypertension) or an increased chamber size (e.g., a dilated left ventricle). Conversely, as would be expected from LaPlace’s relationship, an increase in wall thickness (h) serves a compensatory role in reducing wall stress, because the force is distributed over a greater mass per unit surface area of ventricular muscle.

Contractility (also termed “Inotropic State”)

In the intact heart, as in the isolated muscle preparation, contractility accounts for changes in myocardial force for a given set of preload and afterload conditions, resulting from chemical and hormonal influences. By relating a measure of ventricular performance (stroke volume or cardiac output) to preload (left ventricular end-diastolic pressure or volume), each Frank–Starling curve is a reflection of the heart’s current inotropic state (see Fig. 9.3). The effect on stroke volume by an alteration in preload is reflected by a change in position along a particular Frank–Starling curve.

Conversely, a change in contractility actually shifts the entire curve in an upward or downward direction. Thus, when contractility is enhanced pharmacologically (e.g., by an infusion of norepinephrine), the ventricular performance curve is displaced upward such that at any given preload, the stroke volume is increased. Conversely, when a drug that reduces contractility is administered, or the ventricle’s contractile function is impaired (as in certain types of heart failure), the curve shifts in a downward direction, leading to reductions in stroke volume and cardiac output at any given preload.