Anaesthesia and surgery carry increased risks in patients suffering from cardiovascular disease. This problem is accentuated in the elderly. Ischaemic heart disease, chronic lower respiratory tract infection, and cardiac failure are the disorders which are most commonly associated with postoperative deaths. Over the last decade, better preoperative assessment and the introduction of more sophisticated monitoring have increased the safety of surgery in patients with cardiovascular disease, as it has become possible for the anaesthetist to detect changes in the circulation before life-threatening complications occur. Nevertheless, myocardial infarction, progressive myocardial ischaemia, dysrhythmias, congestive cardiac failure and cerebrovascular accidents continue to occur relatively frequently, reflecting the trend to undertake invasive surgical procedures even in the severely ill patient. Relatively silent cardiac diseases are often initially diagnosed when a patient is admitted for surgery. Medical history, clinical examination, interpretation of the electrocardiogram, chest radiographs, and renal function tests form the basis of the preoperative evaluation. The need for this assessment to be supplemented by echocardiography, radionuclide cardiac imaging, cardiac catheterization, and angiography is, however, increasing. Objective data are required to enable decisions to be made about the overall surgical and anaesthetic strategy, including the extent of cardiovascular monitoring and the need for postoperative management in a high dependency or intensive care unit, as the most conscientious clinical assessment often fails to detect serious physiological abnormalities which should be remedied before anaesthesia and surgery. This requires familiarity with cardiovascular drugs which may modify haemodynamic responses to stresses.

Ischaemic heart disease is the most common cause of death in developed countries, accounting for 30 per cent of all deaths in men, and 25 per cent of deaths in women. Its incidence increases markedly with age and its presence is a major threat to surgical patients.

The risk of perioperative myocardial infarction is increased 30- to 300-fold in patients who have suffered a previous infarction; when surgery takes place within 3 months of the infarction the risk of reinfarction may be as high as 25 per cent. However, the delay between myocardial infarction and surgery is only one factor which needs to be taken into consideration. The type of surgical procedure is also important: reinfarction is more likely to occur after abdominal, thoracic, and prolonged surgery than after body surface surgery. The quality of the patient's ventricular function is paramount, as an ejection fraction of less than 40 per cent increases the risk of perioperative cardiac morbidity and mortality dramatically. Indeed, the presence of left ventricular failure at any time after myocardial infarction has been shown to be the most significant factor associated with mortality after emergency surgery for appendicitis or hip fracture. Persistence of angina after infarction is also an ominous sign, as it indicates the presence of compromised myocardium.

The inevitability of high morbidity and mortality rates associated with anaesthesia and surgery after a recent myocardial infarction has recently been questioned. Extended monitoring of ECG, arterial, central venous, and pulmonary capillary wedge pressure, combined with aggressive management of cardiovascular abnormalities during the perioperative period, may reduce the risk of reinfarction to about 5 per cent. To keep variations of heart rate, arterial pressure and pulmonary wedge pressure within very narrow limits, however, patients must be admitted to an intensive care unit for a relatively long period. It remains safer, in most circumstances, to delay elective surgery for at least 3 months after myocardial infarction.

Angina is a well recognized risk factor for perioperative morbidity and mortality, but there are subsets of patients in whom the risks are particularly high. Unstable angina is associated with risks comparable to those of recent myocardial infarction. Unless surgery can be considered life-saving, patients should be made stable before any other procedure is undertaken. Disabling angina increases the risk of perioperative ischaemia, while mild angina is a threat mostly for patients with preoperative ST–T segment abnormalities.

Over the last decade it has become obvious that many patients suffer from silent myocardial ischaemia, which can be detected by exercise testing or ambulatory ECG monitoring.

There are three types of silent myocardial ischaemia. Type 1 is observed in patients without any symptoms and occurs in about 2.5 per cent of the male population aged between 39 and 59 years, about 40 per cent of whom will develop symptomatic coronary artery disease. Type 2 occurs in about 16 per cent of patients who have suffered myocardial infarction, and Type 3 occurs in patients suffering from angina. In these patients 75 per cent of all episodes of ischaemia are silent, and only 25 per cent are painful.

The absence of pain may be a reflection of less severe ischaemia, a smaller affected area, or a different distribution of impairment of coronary blood flow. However, it is more likely that patients with silent ischaemia have a higher pain threshold, related to higher levels of plasma endorphins.

Silent ischaemia has been reported in 20 to 90 per cent of patients presenting for coronary artery surgery or vascular surgery, and is known to increase the risk of postoperative complications. This may be due to the association between silent ischaemia and poor left ventricular function.

The risk of perioperative infarction after successful coronary artery revascularization is low, and both morbidity and mortality are close to those of patients without heart disease. Thus, coronary revascularization is often advocated before major surgery is undertaken in patients with coronary heart disease. The low incidence of cardiovascular complications may be somewhat misleading, however, since the morbidity and mortality associated with the coronary revascularization itself should be taken into account; this makes the place of coronary revascularization prior to major surgery controversial. When it is indicated in its own right, for example, in patients with unstable angina, correctable disabling angina, or triple vessel disease, coronary revascularization should be performed first. When the indications are less clear cut, it is probably advisable only before surgery which is likely to be associated with major haemodynamic instability, such as major vascular surgery. In some patients, coronary bypass surgery improves cardiac function, probably because some areas of the left ventricle are still viable, if not functional. The concept of ‘hibernating’ myocardium explains this situation: myocardium with very poor blood supply may cease to contract, yet receives enough blood to maintain cell viability. Reperfusion restores mechanical activity and cardiac function. In patients with pre-existing ventricular dysfunction this does not always occur after coronary graft procedures and left ventricular function may remain severely compromised.

Autoregulation of the coronary circulation is impaired in the face of coronary artery disease and the balance between oxygen demand and supply is easily compromised, leading to regional myocardial ischaemia. The major causes of ischaemia are: tachycardia, during which oxygen demand is increased while oxygen supply is impaired because of the shorter duration of diastole; hypotension, during which coronary perfusion pressure is reduced (as coronary blood flow through narrowed vessels is directly proportional to the perfusion pressure) so that the reduction in perfusion may be greater than the reduction in oxygen demands; hypertension, which increases oxygen demand, in the face of an insufficient coronary flow reserve; and left ventricular overfilling, which increases wall tension and oxygen demand, while compromising coronary blood flow because of the augmented tissue pressure. These cardiovascular abnormalities are responsible for perioperative myocardial ischaemia and infarction.

Abnormalities of left ventricular wall motion are frequently detected in patients suffering from coronary artery disease. Non-invasive diagnostic methods relying on radionuclides (radionuclide cineangiography, perfusion scans), and ultrasound (echocardiography) are of proven value and should be used systematically in the preoperative assessment of patients with significant coronary artery disease. Where substantial abnormalities of wall function are present, or if left ventricular aneurysms are discovered, the negative inotropy of most anaesthetic agents is likely to cause severe depression of global cardiac function.

It is becoming increasingly obvious that clinical assessment of left ventricular function is difficult, especially in patients whose activity is limited by peripheral vascular disease or other disabilities. Radionuclide studies show that severe dysfunction is frequently present in patients who appear essentially asymptomatic: about one-third of patients presenting for major vascular surgery have ejection fractions below 50 per cent.

It is convenient to consider that systolic pressure above 160 mmHg, and diastolic pressure above 90 mmHg define arterial hypertension. Hypertension is present in between 10 and 20 per cent of the adult population and carries the risk of serious cardiovascular complications such as stroke, ischaemic heart disease, peripheral vascular disease, and renal dysfunction.

Clinical examination may reveal an enlarged heart and signs of left ventricular failure. The chest radiograph and electrocardiogram may confirm left ventricular hypertrophy. Signs of previous myocardial infarction, myocardial ischaemia, and disorders of intraventricular conduction are often noted. The cerebral complications of hypertension include transient ischaemic attacks and major cerebrovascular accidents. Renal complications may be shown by proteinuria, and elevated serum creatinine and urea. Inspection of the retina may show diminished vessel calibre, nipping, and flame-shaped haemorrhages. Approximately 90 per cent of hypertensive patients suffer from essential hypertension; only 10 per cent have secondary hypertension. Some causes of secondary hypertension are listed in Table 1 87, and it is a possibility that must be kept in mind.

Hypertension is associated with a high incidence of perioperative hypertensive crises, dysrhythmias and myocardial ischaemia, and increases the risk of reinfarction.

While maintenance of the antihypertensive drug regimen is essential in order to ensure cardiovascular stability and prevent rebound hypertension, the need to initiate antihypertensive therapy before anaesthesia and surgery in untreated patients is more controversial. As postponement of surgery may cause considerable inconvenience to patients and disrupt operative programmes, it is essential to identify patients in whom the risks of anaesthesia and surgery will be significantly improved by treatment of their hypertension. This may be done in considering the severity of the hypertensive heart disease (Table 2) 88.

The presence of mild hypertension increases the risks of anaesthesia and surgery but there is no evidence that treatment reduces this risk. Moderate hypertension poses a more substantial threat, especially in the presence of target organ disease, such as significant coronary, cerebrovascular, or renal disease; treatment of hypertension prior to elective surgery is then recommended. In severe arterial hypertension optimal treatment before elective surgery is essential as the incidence of dysrhythmias and/or myocardial ischaemia is far higher in untreated than in treated patients. A major cause for concern is the possibility of very large increases in arterial pressure accompanied by tachycardia and increases in the pulmonary wedge pressure at the time of endotracheal intubation, extubation, and during the recovery period. The increase in pulmonary wedge pressure may represent both failure of the left ventricle to eject and reduced ventricular compliance. Hypertensive crises may result in strokes, myocardial infarction, pulmonary oedema, or life-threatening dysrhythmias. Treatment of hypertension blunts these responses, especially when it includes &bgr;-adrenoceptor blockers.

If treatment of hypertension is indicated, it should be continued for several weeks before surgery since ‘cosmetic’ reductions in blood pressure will be obtained very quickly, but improvement of the underlying vascular hyper-reactivity will occur gradually over weeks and not days.

Regional anaesthesia is often considered a safe alternative to general anaesthesia in untreated hypertensive patients. This is not true: lumbar epidural anaesthesia has been shown to cause greater reductions in systolic and diastolic arterial pressure in untreated than in treated hypertensive patients.

As left ventricular hypertrophy reduces ventricular compliance, excessive volume loading may cause pulmonary oedema, while reduced filling may result in a dramatic reduction in cardiac output. Moreover, in patients with left ventricular hypertrophy marked discrepancies may exist between right and left ventricular filling pressures. Monitoring the pulmonary capillary wedge pressure may therefore be essential for safe perioperative management after major surgery.

Isolated systolic hypertension is due predominantly to the loss of elasticity of the aorta and its major branches. The high pressure increases myocardial oxygen consumption and may cause myocardial ischaemia. However, the treatment of isolated systolic hypertension may be associated with subjective complaints and objective deterioration of cardiac, cerebral, or renal function; in addition titration of blood pressure is often difficult. Treatment of purely systolic hypertension in the elderly is probably not justified before surgery, but adequate cardiovascular monitoring is essential to enable excessive hypo- or hypertension to be detected and treated immediately.

When both systolic and diastolic pressures are elevated, the risk of cardiovascular complications are increased in the elderly as well as in younger patients. Antihypertensive treatment reduces the risk of complications but it is important to achieve the reduction gradually in order for cerebral autoregulation to return to normal limits.

Incipient right or left ventricular failure, indicated by the presence of a third heart sound or distended jugular veins, increases the relative risk of complications by 2 to 10 times. Overt ventricular failure must be treated before any surgical procedure is undertaken in order to reduce the risk of postoperative pulmonary oedema, and invasive monitoring may be necessary to maintain optimal ventricular filling throughout the perioperative period.

Low ejection fractions are associated with increased risks of anaesthesia and surgery. The author's personal views regarding major surgical procedures are as follows: when the ejection fraction is greater than 50 per cent, cardiac function is likely to be adequate and the risks are acceptable in the absence of other risk factors, such as reversible myocardial ischaemia and impaired pulmonary or renal functions. When the ejection fraction is between 40 and 50 per cent, the risks are increased, and the presence of additional risk factors should prompt reconsideration of the type of surgery, with a view to performing the most limited operation possible. Postoperative care in an intensive care or high dependency unit is necessary. When the ejection fraction is between 30 and 40 per cent, whether or not additional risk factors are present, management must be carefully considered. Surgery may be postponed to allow optimal medical treatment; alternatively, less extensive procedures may have to be selected; surgery may be delayed until the balance of risk is increasingly in favour of surgery rather than abstention, or surgery may be cancelled altogether. Again, intensive care management after major surgery is essential. Ejection fractions less than 30 per cent are usually observed in patients who have intermittent left ventricular failure. Recent episodes of left ventricular failure may require admission of the patient to an intensive care unit prior to surgery so that full monitoring may be implemented and treatment optimized with intravenous drugs ahead of major procedures if, and only if, surgery is deemed essential.

The incidence of rheumatic fever has fallen in the Western world, but many elderly patients still suffer from rheumatic heart disease; others suffer from degenerative valvular disease and the prevalence of valvular disease in people over the age of 65 may be as high as 4 per cent.

Aortic stenosis is now the most common valvular lesion; it leads to the development of a pressure gradient between the left ventricle and the aorta. Left ventricular hypertrophy develops as a result of increased stroke work; hypertrophy is accompanied by an increase in diastolic stiffness and an increase in end-diastolic pressure. Raised ventricular pressure and increased muscle mass augment myocardial oxygen consumption, but myocardial blood flow is impaired due to the reduction in aortic diastolic pressure. During the perioperative period, close attention must be paid to changes in heart rate and arterial pressure. Increases in heart rate are poorly tolerated because they reduce the duration of diastole and thus decrease coronary blood flow. Similarly, interventions which reduce systemic vascular resistance decrease both the coronary perfusion pressure and coronary blood flow and may precipitate myocardial ischaemia (Table 3) 89.

Aortic regurgitation causes an increase in left ventricular stroke volume with a corresponding increase in left ventricular cavity size. In moderately severe aortic regurgitation, the stroke volume may be double the normal value. End-diastolic pressure in the aorta is low. Forward flow is facilitated by a relatively low vascular resistance, while regurgitant flow is increased by peripheral vasoconstriction, or by the prolonged diastole associated with bradycardia.

Mitral stenosis impedes left ventricular filling. Flow through the narrowed mitral valve depends on the left atrial pressure, which determines early diastolic flow, and the contribution of atrial systole, which determines late diastolic flow. As duration of diastole is an important determinant of ventricular filling, tachycardia is poorly tolerated. Atrial fibrillation further compromises ventricular filling, especially when the ventricular rate is fast. In patients with mitral stenosis, the main benefit of digitalis therapy is to maintain heart rate below 80 per minute. In many patients, mean left atrial pressure is greater than 15 mmHg. Factors which facilitate overfilling of the atria will rapidly lead to pulmonary oedema.

Mitral regurgitation results in dilatation of both left atrium and left ventricle. The regurgitant flow causes high pressure in the atrium during systole, but there is no obstruction to diastolic forward flow except in combined stenosis and regurgitation. Any increase in systemic vascular resistance will limit left ventricular forward ejection and exaggerate retrograde flow into the atrium.

Hypertrophic obstructive cardiomyopathy is characterized by massive ventricular hypertrophy associated with failure of the ventricles to relax adequately. Systolic function is maintained and rapid ejection often causes pressure gradients within the cavity of the ventricle. Interventions that increase the inotropic state of the myocardium worsen left ventricular function, while this is improved by drugs that reduce contractility, hence the beneficial effects of &bgr;-adrenoceptor blockers and calcium antagonists.

Dilated cardiomyopathy implies marked dilatation and impaired contractile performance. The ejection fraction is usually extremely low. Previous viral myocarditis and excessive alcohol consumption are the main predisposing factors.

Cerebrovascular disease is responsible for death in 9 per cent of men and 15 per cent of women in the United Kingdom. The symptoms and signs of previous cerebrovascular accidents are easily detected when they have resulted in a significant functional deficit. However, transient disturbances of cerebral function should not be overlooked as they may indicate the presence of atherosclerotic lesions of the cerebral vasculature. Particular attention should be paid to reduced carotid artery pulsation and to carotid artery bruits, as carotid artery stenosis may be responsible for transient ischaemic attacks. The question of carotid angiography and carotid endarterectomy may have to be considered, as hypotensive episodes during any type of surgery may precipitate a stroke in patients with significant carotid artery stenosis. The prognosis of postoperative stroke is poor, with a mortality of about 50 per cent. If carotid disease is asymptomatic, prophylactic carotid endarterectomy is probably not indicated.

Carotid artery disease may be considered a marker for coronary disease, as the mortality associated with carotid artery surgery is predominantly due to myocardial infarction.

A high proportion of patients suffer from both cerebrovascular and hypertensive disease. Their management is difficult because their blood pressure can be very labile during anaesthesia and postoperatively. The incidence of neurological deficit is greater in patients who develop hypertension after surgery and neurological deficits are much more likely to occur in untreated or poorly controlled hypertensive patients. Hypertensive crises can cause intracerebral haemorrhages. Gradual control of blood pressure is necessary in order for cerebral autoregulation to return to near normal.

Between one-third and two-thirds of patients undergoing peripheral vascular surgery suffer from coronary heart disease, and the morbidity and mortality associated with anaesthesia and surgery are higher in such patients than in those undergoing non-vascular surgery. Clinical evaluation is often inadequate because of physical inactivity, and objective assessment of cardiac function and myocardial perfusion is invaluable, especially in patients presenting for surgery of the abdominal aorta and other intra-abdominal vessels. About one-third of these patients have low cardiac output and ejection fraction less than 50 per cent.

The high incidence of complications following surgery of the abdominal aorta is a reflection of major haemodynamic disturbances. Cross-clamping of the aorta causes a sudden, large increase in left ventricular afterload. In patients with compromised left ventricular function, the pulmonary wedge pressure increases, reflecting acute left ventricular dilatation associated with acute myocardial ischaemia. Myocardial ischaemia is particularly common when cross-clamping is applied above the renal arteries or the coeliac axis. Administration of vasodilators may be required to prevent or treat myocardial ischaemia during aortic cross-clamping. Removal of the clamps has three effects: firstly, vascular resistance is abruptly reduced; secondly, the sudden return of acidic blood to the heart causes cardiac depression; thirdly, the circulating volume is reduced because vasodilatation has developed in the lower limbs. Adequate volume loading before the gradual removal of cross-clamps is essential.

It is often assumed that surgery of limb vessels is relatively well tolerated because it does not cause major haemodynamic instability. While peripheral vascular surgery is better tolerated than aortic surgery, cardiovascular complications and mortality are still more frequent than in non-vascular surgery, reflecting the association with hypertensive and coronary disease. Indeed, in elderly patients vascular surgery for limb salvage carries a mortality rate of up to 16 per cent.

Any rhythm other than sinus rhythm is associated with a significant increase in the risk of cardiac complications because preoperative dysrhythmias reflect the underlying heart disease.

Atrial dysrhythmias may precede the development of atrial tachycardia and atrial fibrillation. The loss of correctly timed atrial contraction reduces the ventricular filling, particularly when the ventricular rate is fast. In patients with atrial fibrillation or atrial flutter sinus rhythm should be restored if possible, or the ventricular rate controlled, prior to surgery. Fluid load may be poorly tolerated and, for major surgery, patients with atrial fibrillation may benefit from perioperative monitoring of the pulmonary capillary wedge pressure.

Ventricular ectopic activity is often associated with organic heart disease or digitalis toxicity. Premature ventricular beats frequently occur in patients with ventricular wall dysfunction and poor ejection fraction. Cardiac function may worsen: many anaesthetic agents depress cardiac performance, and alter the electrophysiology of the myocardium either by a direct effect on the conduction tissue or by enhancing the arrhythmogenic activity of adrenaline.

Heart blocks (Fig. 1) 70 usually have an organic cause and may be accentuated by vagal stimulation, digitalis, &bgr;-adrenoceptor antagonists, and calcium antagonists. Anaesthesia may precipitate the development of complete heart block in patients with a wide range of atrioventricular or intraventricular conduction disorders by several mechanisms. Some anaesthetic agents, particularly the halogenated inhalational anaesthetics, decrease atrioventricular conduction. Secondly, anaesthesia often causes dysrhythmias, and premature beats are known to facilitate the development of post-ectopic heart block. Thirdly, alterations of serum potassium concentration may occur: acute hypokalaemia may be caused by respiratory alkalosis (hypocapnic intermittent positive pressure ventilation), while acute hyperkalaemia may follow the rapid infusion of large quantities of stored blood. These changes exacerbate dysrhythmias and compromise conduction, and insertion of a temporary pacemaker is often necessary before elective or emergency surgery, even though permanent pacing may not be necessary (Table 4) 90.

Insertion of a temporary pacemaker is not usually recommended in asymptomatic patients with uncomplicated first degree heart block. Such patients do, however, occasionally develop profound bradycardia accompanied by hypotension while under general anaesthesia. Temporary pacing should be considered when this type of block is unresponsive to atropine, and must also be considered when first degree heart block is accompanied by bundle branch block. In Mobitz type I (Wenckebach) block, temporary pacing is necessary only when the patient is symptomatic. However, temporary (or permanent) pacing is necessary in Mobitz type II second degree heart block and in third degree heart block.

The question of whether temporary pacing is required in patients with a bifascicular block is unsettled. Bifascicular blocks can be identified relatively easily: right bundle branch block with left axial deviation of greater than −75° is associated with left anterior hemiblock; right bundle branch block with right axial deviation greater than 110° is associated with left posterior hemiblock. Whenever patients with bifasicular blocks have experienced symptoms which can be attributed to transient complete heart block, temporary pacing is necessary. In totally asymptomatic patients, the risk of anaesthesia facilitating the development of complete heart block is probably very small. Complete left bundle branch block accompanied by first degree heart block is also an indication for temporary pacing.

The sick sinus syndrome represents a group of disturbances of impulse formation in the sinoatrial node which can easily lead to severe arrhythmias, including cardiac arrest during anaesthesia and surgery. The diagnosis should be suspected if there are periodic episodes of slow sinus rate alternating with tachycardia. Establishment of temporary pacing before anaesthesia and surgery is recommended since bradycardia may result from treatment of tachycardia, and this is usually refractory to drug therapy.

Regional anaesthesia is not safe in patients with conduction disorders: absorption of lignocaine injected into the extradural space may cause sinus bradycardia and induce heart blocks, especially in patients with bundle branch block. The same criteria for insertion of a temporary pacemaker should be applied before regional anaesthesia and before general anaesthesia.

Most pacemakers are implanted prophylactically for atrioventricular blocks or sick sinus syndrome, and they have recently been implanted for the control of recurrent tachydysrhythmias. Over the past 30 years the complexity of pacemakers has increased and they are now classified by the chamber paced, activity sensed, mode of response, and type of programmability. Prior to anaesthesia and surgery it is essential to establish the type of pacemaker, the risk of programmes being lost during diathermy, and the means of inducing a fixed rate, which may be necessary during the procedure, particularly when electrocautery is used.

The risk of interference by electrocautery is minimized by placing the indifferent diathermy plate well away from the pacemaker, and limiting the duration of electrocautery to 1-s bursts at intervals of at least 10 s.

Several authors have tried to determine which clinical preoperative factors contribute to the development of cardiac complications in patients undergoing major non-cardiac surgery. The index of cardiac risk in non-cardiac surgery developed by Goldman and his colleagues was based on the relationship between postoperative complications and preoperative assessment data in just over 1000 patients. Using discriminant analysis to identify the factors that contributed significantly to the cardiac risks of anaesthesia and surgery, they produced an index which was able to predict the prognosis in over 80 per cent of cases (Table 5) 91. When applied to a selected population of patients undergoing surgery for abdominal aortic aneurysm this index tended to underestimate the risk of complications. This is not surprising in view of the marked cardiovascular disturbances which occur during major vascular surgery.

Another index, developed by Detsky and colleagues takes coronary artery disease into account more fully, considering not only recent myocardial infarction (within 6 months) but also older infarcts and severe angina (Table 6) 92. Patients with a low score (0–5) have below average risk, while higher scores predict an above average risk of cardiovascular complications.

Global indices of risk are useful but have their limitations, as the severity of heart disease is often underestimated by pure clinical assessment. More detailed preoperative assessment, including functional evaluation of ventricular function, is becoming an integral part of indices of risk in patients undergoing major surgery.

Although the cost-effectiveness of the routine chest radiograph has been questioned, it provides important information in patients with heart disease. Diagnostic alterations of the cardiac shadow, such as prominent left atrium or marked left ventricular hypertrophy, may be observed in valvular heart disease. Cardiomegaly (heart shadow >50 per cent of the diameter of the thorax) in patients with coronary artery disease is usually associated with a low ejection fraction, and substantial enlargement of the left ventricle in hypertensive heart disease is associated with poor diastolic ventricular function. Pulmonary congestion and presence of pleural effusion may also indicate impaired cardiac function.

Although the resting electrocardiogram is normal in 25 to 50 per cent of patients with coronary heart disease, the ECG remains a very important preoperative test in these patients. Characteristic patterns associated with ischaemia, injury, and infarction are easy to recognize (Fig. 2) 71. When Q waves extend to a large area on the precordial leads, left ventricular function is likely to be substantially reduced. Abnormalities of the ST segment or T wave suggestive of myocardial ischaemia are associated with a three-fold increase in perioperative ischaemia. Patients with coronary artery disease who are admitted for elective surgery following previous investigations should undergo repeat ECG: totally silent myocardial infarction often occurs within the days preceding admission. Overlooking a very recent infarction is a most serious hazard.

Characteristic patterns of ventricular or atrial hypertrophy are seen in valvular heart disease, and in severe arterial hypertension (Fig. 2) 71.

Continuous ambulatory ECG recording is of considerable value in the diagnosis of dysrhythmias. Short episodes of tachyarrhythmias, including ventricular tachycardia, may indicate that the dysrhythmia is potentially life-threatening and may warrant drug therapy prior to anaesthesia and surgery. Similarly, in the presence of conduction disorders, continuous ECG monitoring may reveal episodes of complete heart block demonstrating the severity of what may otherwise be considered a relatively benign conduction disorder.

Over the past 5 years the use of continuous ambulatory monitoring has revealed that many episodes of myocardial ischaemia are completely silent. Such silent ischaemia accounts for 75 per cent of ischaemic episodes in patients with stable angina and 90 per cent of episodes of ischaemia in those presenting for coronary artery bypass grafting. As silent ischaemia is an adverse factor in both stable and unstable angina, 24-h ECG monitoring is likely to become an important preoperative test in patients with coronary heart disease presenting for surgery.

Exercise electrocardiography has a high specificity in the prediction of coronary disease and gives an indication of the coronary reserve. Exercise-induced ischaemia usually occurs in territory supplied by coronary arteries that are moderately or severely obstructed, or which develop vasospasm during exercise. There is a good correlation between poor exercise tolerance and perioperative cardiac complications in patients undergoing vascular surgery.

M-mode echocardiography represents a one-dimensional view of the heart plotted against time, while two-dimensional echocardiography enables evaluation of an entire sector of single-dimensional beams. The latter makes it possible to obtain estimates of muscle mass, ejection fraction, velocity of fibre shortening, and end-diastolic and end-systolic volumes. Segmental wall motion abnormalities due to myocardial ischaemia can be described qualitatively and quantitatively. Images in multiple planes are necessary to obtain accurate estimations of ejection fraction and volumes.

Both the extent of left ventricular hypertrophy and the degree of diastolic dysfunction can be determined in patients with arterial hypertension. After myocardial infarction the size of ventricular aneurysms can be assessed, as can papillary muscle abnormalities and the presence of mural thrombi. In valvular heart disease the anatomy of the valves can be examined. Other conditions such as pericardial effusion, presence of clots after cardiac surgery, and atrial myxoma can be detected.

Advances in Doppler technology have made it possible to measure flow across the heart valves; forward and regurgitant flows can be estimated, and pressure gradients across stenoses may be calculated. Such measurements correlate well with direct measurements obtained at cardiac catheterization.

The introduction of transoesophageal echocardiography (the transducer being attached to the distal end of a gastroscope) has made it possible to obtain high quality images during anaesthesia and surgery. In awake patients, high quality images of dissection of the thoracic aorta, prosthetic valve dysfunction, intracardiac thrombi, cardiac and paracardiac masses, and acute mitral regurgitation can be obtained. This technique is particularly useful when standard echocardiography is inconclusive because of chest trauma.

Scintillation cameras detect &ggr;-rays emitted by radiopharmaceuticals. Multicrystal cameras (dynamic studies) and single probe detectors (ejection fraction, cardiac output, pulmonary transit time) extend the range of studies.

Myocardial imaging is used extensively to determine the presence and size of myocardial infarction and the presence of areas of poor perfusion. Hot-spot imaging relies on the avidity of infarcted segments for technetium-99m pyrophosphate. Uptake is detectable as early as 12 to 16 h following infarction, while the maximum abnormality is seen at 48 to 72 h. Cold-spot imaging relies on the uptake of thallium (²&sup0;¹Tl) being proportional to regional myocardial blood flow: areas of ischaemia, infarction, or decreased perfusion appear as cold spots. Perfusion scans may be entirely normal at rest, even in the presence of significant coronary stenoses; heterogeneity of thallium uptake may become obvious during metabolic stress or infusion of a coronary vasodilator such as dipyridamole. Repeat imaging at rest 3 to 4 h later may show that some perfusion defects have disappeared. Reversible defects indicate transient myocardial ischaemia without infarction, and their presence is associated with perioperative ischaemia in patients undergoing vascular surgery.

Gated blood pool imaging relies on technetium-99m to label serum albumin or red cells. Following equilibrium of the radionuclide in the intravascular space, activity is counted over 300 to 500 cardiac cycles. This allows gating to a physiological marker (usually the R wave of the ECG) in order to define end-diastole. The cardiac cycle is then divided into 20 to 30 periods, so that dimensions can be estimated throughout contraction and relaxation. Global and regional ejection fractions may be calculated as well as extent and synchrony of contraction. Gated blood pool imaging is regarded as the gold standard for measuring the ejection fraction. The severity of coronary artery disease can be gauged by the changes in function observed during exercise or in response to dipyridamole. A reduction of the ejection fraction during exercise indicates that an area of the myocardium that was functioning normally at rest has become ischaemic and has stopped contributing to ventricular ejection. Radionuclide testing often reveals ventricular dysfunction that was essentially silent because the patients could not exercise or had simply become used to their disability.

Cardiac catheterization allows the extent of heart disease to be delineated anatomically by use of contrast media and measurement of intracardiac pressure, cardiac output, and shunt fraction. The exact site of coronary artery lesions can be identified. Coronary angiography is not without risks and should be carried out prior to non-cardiac surgery only when coronary revascularization is being seriously considered as the first step in a staged management. Recent studies have shown that approximately 25 per cent of patients with coronary artery disease have a ‘steal prone’ coronary circulation, that is, the presence of an occluded artery, a stenosed artery, and demonstrable collaterals between the two. In such patients arterial vasodilators may cause a steal phenomenon and induce ischaemia.

The insertion of a balloon-tipped pulmonary artery catheter allows left and right ventricular filling pressures, cardiac output, and mixed venous oxygen saturation to be measured and oxygen consumption and physiological shunt to be calculated. There is a high incidence of unrecognized cardiorespiratory abnormalities, particularly in the elderly (Table 7) 93, and identification and accurate estimation of functional defects makes it possible to postpone surgery in order to initiate further treatment, to modify the operation itself, or to cancel surgery altogether if the risks are unacceptable.

The preoperative assessment of patients with cardiovascular disease cannot be complete without considering their long-term cardiovascular medication. Adrenergic &bgr;-adrenoceptor blockers, calcium antagonists, angiotensin converting enzyme inhibitors, nitrates, and cardiac glycosides are commonly used and may interact with drugs given during anaesthesia or during recovery from surgery.

&bgr;-Blockers are useful in both acute and long-term therapy of patients with ischaemic heart disease, arterial hypertension, hypertrophic cardiomyopathies, and certain forms of dysrhythmias. They prevent the cardiac effects of &bgr;-adrenoceptor stimulation, thus protecting the heart against the adverse effects of tachycardia and enhanced contractility. Heart rate and cardiac output are decreased by &bgr;-blockers in proportion to the patient's pre-existing sympathoadrenal activity.

For many years the prevailing opinion was that &bgr;-receptor antagonists should be discontinued before elective surgery. However, detailed haemodynamic studies have shown that arterial pressure and cardiac output are well maintained in hypertensive patients receiving these drugs. Direct benefits of &bgr;-adrenoceptor blockade include the blunting of hypertensive responses to laryngoscopy and intubation, and a reduced incidence of dysrhythmias and myocardial ischaemia.

Prevention of tachycardia is particularly important in patients with coronary disease. &bgr;-Blockers protect the ischaemic myocardium, limit the degree of myocardial infarction, and improve the functional recovery of myocardium subjected to short periods of ischaemia. Maintenance of such therapy throughout the perioperative period is therefore strongly recommended. However, &bgr;-blockade also modifies some of the responses of the circulation and if hypovolaemia occurs the adrenergic stimulation which is normally responsible for tachycardia will be substantially blunted. Reliance on tachycardia as an indicator of the severity of hypovolaemia may therefore cause unnecessary delays in fluid replacement in these patients.

Calcium antagonists are used extensively in the management of supraventricular dysrhythmias, angina, and arterial hypertension. As calcium ions play a crucial role in the electrical activity of the heart and the excitation–contraction coupling of cardiac and vascular smooth muscle, calcium antagonists can be expected to cause marked alterations in the cardiovascular system. The selective inhibition of transmembrane and intracellular Ca²⫀ fluxes is responsible for the depression of sinus node activity and atrioventricular node conduction, the negative inotropy, and the vasodilatation that attend the administration of calcium antagonists.

Verapamil causes myocardial depression and decreases sinus node and atrioventricular node activity. It is effective in slowing the atrial rate in sinus tachycardia and is most effective in terminating re-entrant paroxysmal supraventricular tachycardias. It slows the ventricular rate in atrial fibrillation and atrial flutter. Verapamil is also used in the management of coronary heart disease, as it decreases myocardial oxygen consumption and may cause coronary vasodilatation.

Nifedipine is predominantly a vasodilator and is ineffective in the management of supraventricular arrhythmias mediated by reflex sympathetic overactivity. Nifedipine has become a very important drug in the management of arterial hypertension, angina, and coronary spasm.

The effects of diltiazem are intermediate between those of verapamil and those of nifedipine. It is used mostly in the management of ischaemic heart disease, though it is also effective in the management of supraventricular dysrhythmias.

New calcium antagonists, such as nicardipine and nimodipine, have recently been introduced: like nifedipine, they are nitrendipine derivatives. Nicardipine is used in the treatment of angina and hypertension, while nimodipine is used mostly to prevent or treat cerebral vasospasm.

The reflex-mediated adrenergic responses to calcium antagonists are suppressed by the addition of &bgr;-blockers, and the possibility of adverse interactions between &bgr;-blockers and calcium antagonists must be recognized, particularly when they are administered intravenously.

Halogenated anaesthetics reduce calcium fluxes in cardiac cells and may potentiate the negative inotropy and chronotropy of the calcium antagonists. This in no way indicates that calcium antagonists should be discontinued before elective surgery, but monitoring must be sufficiently detailed for early warning signs of adverse effects to be detected immediately.

Calcium antagonists offer little or no protection against perioperative myocardial ischaemia. In this respect they appear to be much less effective than &bgr;-blockers, probably due to their inability to protect the heart against the effect of sympathetic overactivity. They do not prevent increases in heart rate.

Angiotensin converting enzyme inhibitors cause peripheral vasodilatation and, in patients with normal renal circulation, increase renal plasma flow, decrease renal vascular resistance, and have little effect on glomerular filtration. However, in patients with renal artery stenosis, these drugs may precipitate renal failure, as the reduction of vascular resistance in the efferent arteriole reduces the glomerular filtration pressure.

Angiotensin converting enzyme inhibitors are becoming increasingly important in the management of congestive heart failure because they increase cardiac output and exercise tolerance; they also prolong survival. They are also effective in controlling blood pressure in hypertensive patients. Unlike other vasodilators, they do not cause sodium retention because the renin–angiotensin–aldosterone system is blocked.

Converting enzyme inhibitors may minimize the pressor responses to intubation and surgical stimulation without causing exaggerated reductions in blood pressure in response to induction or maintenance of anaesthesia.

Nitrates cause venodilatation and, to a lesser extent, arteriolar vasodilatation: these combined effects decrease left ventricular wall tension and myocardial oxygen consumption. Glyceryl trinitrate causes a reflex tachycardia which may counteract this beneficial effect, unless a &bgr;&sub1;-adrenoceptor antagonist is also administered. Glyceryl trinitrate redistributes coronary blood flow to the subendocardium by increasing collateral coronary blood flow. It is effective in the treatment of unstable angina, coronary artery spasm (Prinzmetal's variant angina) and in acute myocardial infarction as a means of limiting infarct size.

The cardiac glycosides inhibit transport of sodium and potassium ions across cell membranes by inhibiting the Na⫀, K⫀-ATPase. One major effect is the release of sequestered calcium ions from the mitochondria of the failing heart, thus enhancing calcium availability in the sarcoplasmic reticulum and improving contractile force. This advantage is counteracted by a marked increase in systemic vascular resistance. Thus, in acute care units, the combination of vasodilators and inotropes such as dopamine or dobutamine is used routinely instead of cardiac glycosides.

Digitalis preparations increase central vagal activity, delaying the atrioventricular conduction time, an effect well suited to the treatment of atrial fibrillation, atrial flutter, and occasionally the termination of supraventricular tachycardia in the Wolff–Parkinson–White syndrome.

Cardiac glycosides may accumulate in patients with poor renal function, and toxic blood concentrations may be achieved following administration of conventional doses. This tendency may be enhanced following anaesthesia and surgery because of the alterations in renal function associated with the stress response to surgery. Digitalis toxicity may be precipitated during anaesthesia because respiratory alkalosis exacerbates the effects of hypokalaemia, especially in patients taking diuretics. Nevertheless, the use of digoxin to maintain normal heart rate (55–70 beats/min) in patients with atrial fibrillation has the advantage over all the other antiarrhythmics of having positive rather than negative inotropic effects. This may be important in patients with poor left ventricular function.

Preoperative administration of digitalis has been advocated to minimize the effect of potent anaesthetic agents on the heart, and to minimize the incidence of perioperative tachyarrhythmias. However, the risk of intraoperative dysrhythmias is increased. Laver and Lowenstein have advocated the following guidelines: acute digitalization is justified in the patient with preoperative heart failure; digitalis is best continued until the evening before operation in patients with chronic atrial fibrillation if the ventricular rate is in excess of 80 beats/min; preoperative digitalization may be considered for reducing the incidence of supraventricular dysrhythmias in patients undergoing abdominal or intrathoracic operations in whom a previous myocardial infarction is associated with abnormal left ventricular function; preoperative digitalization should be discouraged if it is intended solely to counteract the cardiac depression of anaesthesia.

Rigorous preoperative assessment is essential in order to identify patients in whom the cardiac risks of anaesthesia and surgery are particularly high. Cardiologists and radiologists are frequently called upon to establish the nature and severity of cardiac disorders and to perform objective tests of cardiac function, as unrecognized poor cardiac function is associated with excessive morbidity and mortality. Sound management rests on communication between surgical and anaesthetic teams in order to define the safest strategy for the perioperative period. Collaboration with cardiologists is essential in order to ensure the best therapy of cardiovascular abnormalities, both pre- and postoperatively.

Al Wathiqui M H, Farber N, Pelc L, Gross GJ, Brooks HL, Warltier DC. Improvement in functional recovery of stunned myocardium by long-term pretreatment with oral propranolol. Am Heart J, 1989; 117: 791–8.

Barnes RW. Asymptomatic carotid disease in patients undergoing major cardiovascular operations: can prophylactic endarterectomy be justified? Ann Thoracic Surg 1986; 42: S36–S40.

Brunner HR, Gavras H, Waeber B, Textor SC, Turini GA, Wauters JP. Clinical use of an orally acting converting enzyme inhibitor, captopril. Hypertension 1980; 2: 558–66.

Carliner N, et al. Routine preoperative exercise testing in patients undergoing major non-cardiac surgery. Am J Cardiol 1985; 56: 51–8.

Chung F, et al. Calcium channel blockade does not offer adequate protection from perioperative myocardial ischemia. Anesthesiology 1988; 69: 343–7.

Clements FM, de Bruijn NP. Perioperative evaluation of regional wall motion by transoesophageal two-dimensional echocardiography. Anesth Analg 1987; 66: 249–61.

Cohn PF. Silent myocardial ischemia: classification, prevalence, and prognosis. Am J Med 1985; 79 (Suppl 3A): 2–6.

Coriat P, Harari A, Ducardonet A, Targot JP, Viars P. Risk of advanced heart block during extradural anaesthesia in patients with right bundle branch block and left anterior hemiblock. Br J Anaesth 1981; 53: 545–7.

Cutler BS, Leppo JA. Dipyridamole thallium 201 scintigraphy to detect coronary artery disease before abdominal aortic surgery. J Vasc Surg 1987; 5: 91–100.

Dagnino J, Prys-Roberts C. Strategy for patients with hypertensive heart disease Clin Anaesthesiol 1989; 3: 261–89.

Deanfield JE, Selwyn AP, Chierchia S, Maseri A, Ribiero P. Myocardial ischaemia during daily life in patients with stable angina: its relation to symptoms and heart rate changes. Lancet 1983; ii 753–8.

Del Guercio LRM, Cohn JD. Monitoring operative risk in the elderly. JAMA 1980; 243: 1350–5.

Detsky AS, Abrams HB, Forbath N, Scott JG, Hilliard JR. Cardiac assessment for patients undergoing non-cardiac surgery. Arch Intern Med 1986; 146: 2131–4.

Dirksen A, Kjoller E. Cardiac predictors of death after non-cardiac surgery evaluated by intention to treat, Br Med J 1988; 297: 1011–3.

Foëx P. Pharmacology of cardiovascular drugs. Clin Anaesthesiol 1989; 3: 131–61.

Foster ED, Davis KB, Carpenter JA, Abele S, Fray D. Risk of noncardiac operation in patients with defined coronary disease: The coronary artery surgery study (CASS) registry experience. Ann Thoracic Surg 1986; 41: 42–50.

Gifford RW Jr. Myths about hypertension in the elderly. Med Clin N Am 1987; 71: 1003–11.

Goldman L. Multifactorial index of cardiac risk in non-cardiac surgery: ten-year status report. J Cardiothorac Anesth 1987; 1: 237–44.

Goldman L. Assessment of the patient with known or suspected ischaemic heart disease for non-cardiac surgery. Br J Anaesth 1988; 61: 38–43.

Goldman L, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977; 297: 845–50.

Haagensen R, Steen PA. Perioperative myocardial infarction. Br J Anaesth 1988; 61: 24–37.

Hertzer NR, et al. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199: 223–33.

Jeffrey CC, Kunsman, J Cullen, DJ, Brewster, DC. A prospective evaluation of cardiac risk index. Anesthesiology 1983; 58: 462–4.

Knight AA, et al. Perioperative myocardial ischemia: importance of the preoperative ischemic pattern. Anesthesiology 1988; 68: 681–8.

Laver MB, Lowenstein E. Anesthesia and the patient with heart disease. In: Johnson RA, Haber E, Austen WG, eds. The practice of cardiology. Boston; Little Brown and Company, 1980: 1090–1109.

Leppo J Plaja J, Gionet M, Tumolo J, Pasakos JA, Cutler BS. Non-invasive evaluation of cardiac risk before elective vascular surgery. J Am Coll Cardiol 1987; 9: 269–76.

Morise AP, et al. The prediction of cardiac risk in patients undergoing vascular surgery. Am J Med Sci 1987; 293: 150–8.

Prys-Roberts C, Meloche R, Foëx P. Studies of anaesthesia in relation to hypertension. I. Cardiovascular responses of treated and untreated patients. Br J Anaesth 1971; 43: 122–137.

Raby KE, et al. Correlation between preoperative ischemia and major cardiac events after peripheral vascular surgery. N Engl J Med 1989; 321: 1296–300.

Rao TLK, Jacobs KH, El-Etr AA. Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 1983; 59: 499–505.

Sorensen MB, Engell HC. Preoperative haemodynamic evaluation of patients submitted for major surgery. Acta Anaesth Scand 1978; 22: 391–9.

Steen PA, Tinker JH, Tarhan S. Myocardial reinfarction after anesthesia and surgery. JAMA 1978; 239: 2566–70.

Zaidan JR. Pacemakers. Anesthesiology 1984; 60: 319–34.