Altered cardiac function

L. STEELE and N.R. WEBSTER
Anaesthesia and Intensive Care, University of Aberdeen, Aberdeen, U.K.

Introduction

Myocardial contractility

Pressure - volume relationships and Starling's law.

The failing heart

Investigation

Treatment strategies for systolic failure

Treatment strategies for diastolic failure

Changes in heart function during sepsis

Conclusions

Suggested further reading

An understanding of the normal functioning of the heart and how it fails is important since it allows rational treatment. Pre-existing cardiac disease and myocardial dysfunction is common in the surgical patient. Moreover, the stress response of surgery and the alterations in body physiology seen in the post-operative period may further aggrevate any cardiac compromise. The end result may be a patient who has cold peripheries due to vasoconstriction and hypoperfusion with dyspnoea due to congested lungs and a heart that cannot function adequately. The postoperative patient is at risk of these changes because of the stress response, analgesic therapy, inappropriate fluid management, hypoxia and previous cardiac compromise. The patient with sepsis is at further risk because of alterations in both systolic and diastolic function, which may be the result of inadequate fluid resuscitation and also release of a variety of inflammatory mediators. Until treatments, which are aimed at correcting the effects of these mediators, are proven to be beneficial then the septic patient will continue to be managed according to the physiological principles as outlined by Starling.

Cardiac failure is a condition in which the output of the heart is not adequate to meet the needs of the body, either at rest or with exercise. This is usually accompanied by an increased filling pressure and/or volume. The condition requires prompt recognition and management since tissue oxygen supply and hence organ function can both be readily compromised. The hallmarks are fatigue, dyspnoea and oedema.

Congestive heart failure is the presence of heart failure and oedema in the presence of normal systolic function. In these patients, it is important to exclude other diseases such as valvular disease, recurrent ischaemia, pericardial disease, cor pulmonale and congenital heart disease as the cause of congestive heart failure. Often, these conditions arise because of diastolic dysfunction. Acute heart failure is not a single entity, occurring during diastole or systole. To determine the type of cardiac failure, it is necessary to understand the normal physiology and the factors, which regulate myocardial contraction.

Ventricular function is decreased during sepsis. Patients with septic shock have been documented to have lowered ejection fractions - mean of 32% - despite an increase in cardiac output. This returned to normal within 10 days in survivors. Similar findings have been observed in human volunteers given endotoxin. The causes of these changes have not been fully elucidated but clearly knowledge of the patho-physiology of this process will enable a logical approach to the management of these patients.

Ejection fraction is not a pure measure of systolic contractility of the heart but is a measure of ventricular function which also depends on diastolic compliance, preload and afterload. Each will be considered in the text.

The heart is regulated by two different systems - intrinsic regulation which involves the contractile properties inherent to the muscle itself; and extrinsic regulation which is the response of the heart to conditions imposed from outside such as neural stimulation, hormones, drugs and disease.

Intrinsic regulation involves the myocardial response to stretching of the myofibres prior to contraction. This is called the preload and involves the filling of the heart during diastole i.e. the end diastolic volume. The myocardial response to increased load after contraction has begun is called the afterload, e.g. for the left ventricle; the aortic pressure can be regarded as equivalent of its afterload. Finally, myocardial contractility alters with variation in heart rate.

Originally described by Frank and Starling, the law states that under normal physiological conditions the force or tension generated by the contracting muscle is greater if the muscle is previously stretched. This implies that during diastole a greater influx of blood into the ventricle will cause the following contraction to be more forceful.

Starling used isolated canine heart lung preparations in which he controlled the right atrial pressure (and, thereby, the right ventricular diastolic pressure or preload) and also controlled the aortic pressure (or afterload). From a series of experiments he showed that with increased diastolic volume, the isolated heart generated a higher peak systolic volume (Figure 1). In the normal heart, during diastole, the ventricle increases in volume from an end systolic volume of 60mls to an end diastolic volume of 130mls, with an increase in pressure for the left ventricle from around 5mmHg at A to about 10mmHg at B. The atrioventricular valve opens at A and closes at B and during ventricular contraction the pressure increases to C. At C the aortic valve opens and the actual pressure at C depends on the diastolic pressure in the aorta. The aortic valve closes at D, ventricular relaxation occurs and the pressure drops from D to A at which point the atrioventricular valve opens again since ventricular pressure has dropped below atrial pressure.

Figure 1: Left ventricular volume-pressure curves depicting the normal cardiac cycle (ABCDA) and the effect of increasing end diastolic volume from B to B'

According to Starling's law, an increase in end diastolic volume (preload) causes the heart to begin it's contraction at a higher pressure and volume (position B'); thus, the new pressure - volume loop is larger but with the pressures at C', D' and A' only slightly increased. The end systolic volume is slightly increased but the heart now operates at a larger end diastolic volume and hence it ejects a larger stroke volume.

Thus, the heart has it's own intrinsic ability to control stroke volume, i.e. the greater the diastolic stretch of myofibrils the greater the force of contraction. The upper limit of this control is reached when a particular end diastolic volume is achieved which results in optimal myocardial fibre length. It is theoretically possible for this process to progress further and cardiac output falls when the myofibrils are stretched too far (Figure 2). This seems to be due to over stretching of the actin-myosin cross bridges formed in myofibrils.

Figure 2: The Starling curve. The relationship between end diastolic volume and stroke volume (providing afterload is constant)

Congestive heart failure is often seen in postoperative patients who have been on medication previously or in new patients who have an acute illness on top of a progressively failing heart. It is a clinical syndrome of dyspnoea on exercise or worse at rest, oedema and fatigue caused by a failing left ventricle and activation of neurohormonal mechanisms that promote fluid retention. Many cardiac conditions may cause the syndrome of congestive cardiac failure and yet the patient can still retain normal systolic function. While the majority of manifestations of congestive heart failure are due to pulmonary and systemic venous hypertension and congestion, symptoms of a low cardiac output can be prominent. Many studies have been undertaken to try and redefine congestive heart failure into systolic and diastolic ventricular heart failure. Systolic and diastolic dysfunctions have different physiological mechanisms but can present similarly as congestive cardiac failure - however, and importantly, they require different treatments.

Here there is impaired left ventricular contractility. By looking again at the Starling curve it is possible to express what happens to preload when ventricular contractility worsens (Figure 3). The ability to increase stroke volume via an increase in preload is diminished and the curve shifts to the left of the normal curve if afterload and inotropy remain constant. As left ventricular failure worsens cardiac volumes and pressures continue to increase leading to pulmonary venous congestion. If this process is allowed to continue, increased pulmonary venous pressure leads to right ventricular hypertension, failure and increased central venous pressures.

Figure 3: Any increase in preload in congestive heart failure due to systolic failure yields a smaller increase in stroke volume than in normal ventricles if inotropy and afterload are kept constant. A decrease in afterload or increase in contraction may improve ventricular performance, but will not affect filling pressures. A reduction in preload with either or both an increase in inotropic force and afterload yields reductions in ventricular filling pressures and improvement in ventricular performance.

Since Starling's curves are all demonstrated at constant afterload and inotropy, then these are the only two ways to improve the relationship between preload and increased systolic function. As seen in Figure 3, only either a direct improvement in the inotropic state of the heart or a reduction in afterload will move the ventricle onto a more favourable part of the curve. Reducing preload merely reduces stroke volume by continuing along the descending part of the curve. However, combing a reduction in the preload with either a direct inotropic stimulus or reduction in afterload will provide both better systolic ventricular function and a reduction in ventricular filling pressures while improving pulmonary congestion. Treatment of systolic heart failure is fundamentally based on an attempt to meet two goals, stabilising or improving systolic function and reducing ventricular filling pressure.

Assuming adequate preload, the normal ventricle is relatively insensitive to small changes in afterload. However, in the heart with systolic failure, very small increases in afterload may lead to marked reductions in ventricular performance and conversely a small decrease in afterload may improve left ventricular function (Figure 4). This relationship between left ventricular systolic function and afterload forms the basis for afterload reduction therapy used in the treatment of systolic failure and congestive cardiac failure.

Figure 4: Afterload has very little effect on the normal ventricle. However, as systolic failure develops even small increases in afterload have significant effects on compromised ventricular systolic function. Conversely, small reductions in afterload in a failing ventricle can have significant beneficial effects on impaired contractility.

Afterload is dependent on a variety of physiological factors relating to the body's vasculature and how it inter-relates with the left ventricle. The most important factors are systemic vascular resistance and arterial blood pressure but left ventricular volume and large artery impedance also contribute. The law of Laplace illustrates an important principle relating to the regulation of afterload, which again has implications in governing heart function in the failing heart. End systolic wall stress is proportional to both end systolic pressure and end systolic volume. Therefore, if ejection fraction is reduced and ventricular size increases, end systolic volume also increases. Reduction of ventricular systolic function coupled with increased preload immediately leads to an increase in afterload.

Several factors contribute to changes in ventricular filling in systolic heart failure:

In summary, systolic ventricular dysfunction is caused by decreased left ventricular contractility, increased cardiac volume and pressure, decreased responsiveness to preload, and increased sensitivity to increased afterload. This leads to the situation where the heart can no longer meet the metabolic demands of the body, firstly at exercise then progressively at rest. Dyspnoea is due to increased pulmonary venous pressure although what the precise cause of the feeling of breathlessness is unknown. Volume expansion occurs due to salt and water retention due to neuroendocrine imbalances, which exacerbates both abnormalities in preload and afterload. The end result is a patient who has cold peripheries due to vasoconstriction and hypoperfusion with dyspnoea due to congested lungs and a heart that cannot function adequately. The postoperative patient is at risk of these changes because of the stress response, analgesic therapy, fluid management (mismanagement), hypoxia and previous cardiac compromise.

In stark contrast to patients with systolic dysfunction, patients with diastolic dysfunction have normal or often enhanced contractile function of the left ventricle and there is a normal increase in stroke volume with an increase in preload. Since the ventricle in these patients is capable of a normal increase in cardiac output, there should be no mismatch between metabolic demands and the ability of the left ventricle to supply sufficient blood to meet the needs. Yet these patients are still symptomatic, complaining of dyspnoea and fatigue similar to those with impaired contractile function. The main problem in diastolic failure is that ventricular stiffness or reduced ventricular compliance leads to the impairment of ventricular function. This results in the heart being unable to use preload reserve since high cardiac filling pressures occur at normal or decreased intraventricular volumes. Figure 5 shows the relationship between ventricular pressure and ventricular volume. Increased stiffness (decreased compliance) shifts the pressure-volume curve to the left, producing a higher pressure for a given diastolic volume. Conversely, decreased stiffness or compliance leads to a lower pressure for any given volume. Changes in compliance can be due to several factors influencing diastolic function such as fibrosis, ischaemia, valvular and hypertensive heart disease.

Figure 5: Ventricular chamber compliance is shown by the above relationship of ventricular pressure to volume during diastole. Curve (A) demonstrates the change in pressure that occurs with a given change in volume throughout diastole. Curve (B) shows an increase in chamber compliance. With a decrease in chamber compliance at curve (C), at any given ventricular volume, there is an increased ventricular pressure. Decreased compliance also results in a more rapid increase in ventricular pressure for a given change in volume and a steeper slope of the pressure volume curve.

The ventricular diastolic pressure - volume relationship may be abnormal because of changes in active relaxation, passive compliance factors or both. Whatever the specific abnormality may be, in left ventricular diastolic failure the result is impaired ventricular filling with high left atrial and pulmonary venous pressures. Clinically, as with systolic failure, the patient presents with increasing dyspnoea and fatigue, particularly on exercise or during stress. The most likely reason for this is elevation of pulmonary venous pressures as a result of elevated ventricular pressure, reflected on the left atrium due to decreased ventricular compliance. Stroke volume is limited due to reduced ventricular preload. It is possible for a patient to have an inability to increase cardiac output in response to stress despite having normal systolic function and if systolic function is also abnormal then this effect is exaggerated. This factor is important to consider when a patient's symptoms change, e.g. when a previous hypertrophied ventricle begins to decompensate in progressive valvular disease or uncontrolled hypertension.

Different disease states seem to influence diastolic function in different ways. When the ventricle hypertrophies in response to uncontrolled hypertension or volume overload, the ventricle has slow relaxation period and increased chamber stiffness. This decreased compliance is probably due to the increase in myocardial cell diameter and altered collagen of the left ventricle. Myocardial ischaemia can cause transient changes in ventricular relaxation. Restrictive cardiomyopathies caused by infiltrative diseases such as sarcoidosis, haemochromatosis and amyloidosis are less common causes of diastolic failure but when they do present cause increased passive chamber stiffness due to the increased deposition of the foreign substance in the myocardium. Aging also has a detrimental effect on the ventricular relaxation which is prolonged the older we become. All of these diseases result in the end pathological process of impaired ventricular filling, leading to left atrial and pulmonary venous hypertension.

To be able to differentiate between systolic and diastolic function properly requires imaging of the heart with echocardiography and radionuclide ventriculography. However, a good history, thorough clinical examination, chest radiograph and ECG will offer some clues. A patient presenting with a history of myocardial infarction, Q waves on the ECG, an S3 gallop rhythm and an enlarged heart is obviously suggestive of systolic failure. Hypertension and ischaemia both cause systolic and diastolic dysfunction. This leads to the diagnosis being unclear and imaging of the heart becomes essential. If systolic function is found to be normal, then the diagnosis of diastolic dysfunction is one of exclusion. Clearly, it is also important to exclude lung disease as a cause of dyspnoea.

The basic therapeutic principles of treating systolic failure are based on reversing or preventing further deterioration in left ventricular function.

Diuretics remove excess salt and water from the circulation -ventricular preload decreases and pulmonary venous pressure declines - resulting in lessening of dyspnoea and fatigue and improved ventricular performance. This may be due to diminished wall stress as the ventricle decreases in size and also better subendocardial blood flow at any given diastolic pressure. Both of which may move the ventricle onto a more favourable part of the Starling curve. This may actually result in improved left ventricular function despite decreased preload. A vigorous diuresis, therefore, is essential in the management of the volume-expanded patient with systolic failure, particularly if they develop acute heart failure. However, diuretic use leads to electrolyte depletion and activation of the rennin-angiotensin system with negative consequences and these may be alleviated somewhat by the use of ACE inhibitors.

The most obvious way to improve left ventricular function would be to give an inotropic agent e.g. dobutamine. Inotropic drugs improve myocardial contractility and increase cardiac output in low cardiac output states - particularly useful for acutely decompensated congestive heart failure. However, the long-term use of this group of drugs has been less successful, perhaps due to the general effect of all these drugs, which increase myocardial oxygen consumption, diminish myocardial energy stores and increase the incidence of arrythmias.

Direct vasodilating drugs currently available include hydralazine and isosorbide mononitrate. These agents respectively reduce systemic vascular resistance and pulmonary venous pressure and so in combination improve overall ventricular loading conditions with decreased pulmonary congestion and increased cardiac output. As with diuretics, however, use of these agents can lead to neuroendorcine activation.

Angiotensin converting enzyme inhibitors produce essentially the same haemodynamic profile as the combination of hydralazine and isosorbide mononitrate since they reduce cardiac filling pressure and improve cardiac output. These effects are due to decreased levels of angiotensin II but probably also due to the release of prostaglandins, which cause vasodilatation.

Calcium channel blockers are extremely potent vasodilators, especially the dihdropyridine class. Nifidipine is probably the most potent peripheral vasodilator currently available but has significant negatively inotropic effects and may activate the sympathetic nervous system and the rennin-angiotensin system. Diltiazem is both a less potent vasodilator and less potent negative inotrope. Verapamil is simply too potent a negative inotrope to be recommended. Newer agents such as amlodipine have proven to be more useful especially in combination with an ACE inhibitor.

Diastolic dysfunction can be present for several years before any symptoms occur. It is important to detect diastolic dysfunction early and to start treatment before irreversible structural alterations and systolic dysfunction occurs. However, no single drug has exclusive lusitropic properties, i.e. increase relaxation of the heart in diastole, which enhance myocardial relaxation without negative effects on ventricular contractility. Therefore, there are four treatment therapies advocated for the treatment of diastolic dysfunction:

Diuretics are effective in reducing pulmonary congestion by shifting the pressure-volume curve downwards. This positive effect on chamber stiffness causes a reduction in systemic blood volume and lowering of right atrial blood volume. However, they must be used with care because of the sensitivity of some patients to cardiac volume and excessive diuresis leading to a sudden drop in stoke volume and cardiac output.

Beta-blockers have been used for many years to control blood pressure and, thus, myocardial hypertrophy. Slowing of the heart seems to be the primary benefit and not improvement of isovolumetric relation as once thought. Beta-blockers also cause regression of left ventricular hypertrophy, which will also improve diastolic filling.

Calcium channel blockers have been shown to improve myocardial relaxation and enhance diastolic filling. These drugs seem to be best matched to the pathophysiology of relaxation disturbances due to their ability to decrease calcium concentration within cells and also reduce afterload. Calcium channel blockers with negative dromotropic (heart rate) action such as verapamil or diltiazem may improve diastolic filling by a reduction in heart rate. They have also been shown to reduce muscle mass in patients with hypertension, thus, improving the elastic properties of the myocardium. Calcium blockers are first line drugs in patients with hypertrophic cardiomyopathy due to their beneficial effect on relaxation and diastolic filling which are often severe in these patients.

The patient with sepsis has severely altered physiology in a number of ways, which can influence cardiac function. Firstly, there is a loss of intravascular volume due to excessive third space loss that results in a decrease in preload. Systemic vascular resistance is decreased which results in a fall in afterload. In addition, end diastolic volumes often increase and ejection fraction falls. However, many of these changes are overcome by an increase in heart rate that may result in an increase in cardiac output. However, it should be remembered that even in the presence of high cardiac outputs it is usually always possible to demonstrate ventricular dysfunction in patients with sepsis. Echocardiographic studies consistently confirm that there is decreased left ventricular systolic function in humans with sepsis.

In addition, there have been many studies in animals and a few in humans which have confirmed the presence of diastolic dysfunction - particularly in those patients that go on to die from sepsis. In the presence of adequate fluid resuscitation there is an increase in end diastolic volume and this is probably a normal response to a decrease in contractility. However, in the non-survivors of sepsis there is a normal or low end diastolic volume that is the result of a decrease in ventricular diastolic compliance. Thus, there is a decreased end diastolic volume at the same filling pressure.

During sepsis, a decrease in contractility results in a shift to the right of the end-systolic pressure / volume curve (Figure 1) and if this is not compensated for results in a decrease in stroke volume and cardiac output. When patients with sepsis are appropriately fluid resuscitated there is an increase in enddiastolic pressure that increases stroke volume. In addition, the decrease in afterload will also increase stroke volume and will prevent a decrease in ejection fraction. Alas, because there is a decrease in systolic contractility it would be expected that there would also be a decrease in diastolic stiffness which would allow cardiac output to be maintained despite the relatively low filling pressures. However, if this diastolic compliance change does not occur (as in the nonsurvivors of sepsis) then it is apparent from Figure 1 that the ability of the ventricle to generate a stroke volume is impaired at both ends of the curve.

The cause of the altered cardiac function in sepsis remains unknown although there are many theoretical explanations. Clearly, one of the most important mechanisms which can be readily corrected is hypovolaemia. Myocardial oedema may contribute to a decrease in contractility. Increased circulating catecholamines can result in a decrease in diastolic compliance, particularly important since these agents are often used to improve myocardial contractility. Increased intrathoracic pressure caused by positive pressure ventilation can also result in decreased diastolic compliance. In addition, many of the mediators of the inflammatory response, including products of activated endothelial cells and polymorphonuclear leucocytes (e.g. nitric oxide, tumour necrosis factor and interleukins 1 and 2) have all been postulated as negative inotropes and negative lusitropes. Another, as yet, unidentified agent which is believed to be released from the splanchnic bed - myocardial depressant factor - is postulated to play a role.

Treatments aimed at correcting the effects of these various inflammatory mediators may be eventually found but until these approaches have been proven to be beneficial the septic patient will continue to be managed according to the physiological principles outlined by Starling.

Pre-existing cardiac disease and myocardial dysfunction is common in the surgical patient. Moreover, the stress response of surgery and the alterations in various biological processes seen in the post-operative period may further aggrevate any cardiac compromise. The postoperative patient is at risk of these changes because of the stress response, analgesic therapy, fluid management (mismanagement), hypoxia and previous cardiac compromise.

The patient with sepsis is at further risk because of alterations in both systolic and diastolic function, which may be the result of inadequate fluid resuscitation and also release of a variety of inflammatory mediators. Until treatments aimed at correcting the effects of these mediators are proven to be beneficial the septic patient will be managed according to the physiological principles outlined by Starling.

Correspondence: Professor N.R. Webster, Anaesthesia and Intensive Care, Institute of Medical Sciences, Foresterhill Aberdeen AB25 2ZD, U.K.

EDUCATIONAL REVIEW

Altered cardiac function