The aims of perioperative cardiovascular monitoring are to detect and quantify potentially dangerous haemodynamic changes, and to direct treatment and assess its effects.

Routine cardiovascular monitoring during and after cardiac surgery includes clinical monitoring and the continuous measurement and display of electrocardiogram, arterial pressure, central venous pressure, and pulse oximetry. Intermittent measurements of urine output, blood gases, and the core-peripheral temperature gradient also provide useful indirect information about the state of the cardiovascular system. Other variables that may be monitored in high risk patients include left atrial pressure, pulmonary artery pressures, and cardiac output.

Basic clinical monitoring consists of the direct observation of the patient by anaesthetist, surgeon, or nurse. It includes the assessment of skin colour and capillary refill time, palpation of peripheral pulses, skin temperature and sweating, and observation of chest movement and the noises of breathing. This monitoring requires no mechanical or electrical instrumentation and its importance must not be underestimated. Information obtained from mechanical monitors is subject to many errors, and should be interpreted in relation to the clinical picture.

The electrocardiogram (ECG) provides continuous information about heart rate and rhythm, and analysis of changes in the ST segments may help to detect myocardial ischaemia. A multiple lead electrode system consisting of four limb leads and a fifth at the V5 position allows selection of 7 out of the 12 standard ECG leads; V5 and II being the most helpful for detecting anterolateral and inferior ischaemia, respectively. ECG monitoring is subject to many types of interference in the operating theatre, and artefacts are common.

Direct monitoring of arterial blood pressure is essential for all but the simplest of cardiac surgical procedures so that rapid changes in blood pressure can be followed accurately. In addition, an arterial line permits repeated blood gas estimations. Indirect pressure measurements (using a blood pressure cuff) may be inaccurate, particularly in obese, hypertensive, hypothermic, or shocked patients.

Direct arterial pressure monitoring consists of an indwelling arterial cannula connected to a pressure transducer by a length of rigid fluid-filled manometer tubing. Amplification of the signal from the transducer is used to create a pressure trace on an oscilloscope. A digital readout of systolic, diastolic, and mean pressure is produced by averaging several beats. The arterial line is kept open either by intermittent flushing or by continuous infusion of a small volume of heparinized saline.

The radial artery is most commonly used for pressure monitoring. It is accessible, easy to cannulate and short-term cannulation carries little morbidity. Other possible sites include the brachial, axillary, femoral, posterior tibial, and dorsalis pedis arteries. The femoral artery is avoided by those who believe it to be a potential source of infection with organisms from the perineum. Prolonged femoral arterial cannulation has, however, been reported without major complications.

Transduced blood pressure is subject to inaccuracies due to transducer position, calibration, and the mechanical characteristics of the recording system. To obtain reliable readings the calibration of the transducer system should be checked regularly against a mercury column. Before use the system is zeroed against a reference point at right atrial level.

The pressure measured from a peripheral part of the arterial system does not always accurately reflect the intra-aortic pressure. Progressive alteration of the arterial pressure wave occurs as it is transmitted through the vascular system; measured systolic and diastolic pressures may be either higher or lower than aortic pressures depending on the degree of vasomotor tone and the rigidity of the vascular tree. Aortic to radial pressure gradients are particularly common immediately after cardiopulmonary bypass.

Central venous pressure is measured as an indication of right ventricular filling pressure using a catheter in a central (intrathoracic) vein. Venous pressure is affected by a number of interrelated factors: circulating blood volume, right and left heart function, peripheral and pulmonary vascular resistance, and intrathoracic pressure. In patients with normal ventricular and pulmonary function changes in central venous pressure occur roughly in parallel with left heart filling pressures. This relationship does not persist in pathological conditions, however.

Central venous pressure is measured with reference to a fixed zero point, at right atrial level. The position of the zero point will affect the absolute value of the central venous pressure; in a supine patient the pressure measured with reference to the mid-axillary line will be higher than that measured with reference to the sternal angle. Any change in central venous pressure will be of the same magnitude irrespective of the zero point.

Single measurements of central venous pressure are rarely helpful unless the measurement lies well outside the normal range (1–10 mmHg in a patient breathing spontaneously). Serial measurements and the response of central venous and arterial pressures to a fluid challenge (a rapid infusion of 100–200 ml) are, however, useful indicators of circulatory status and right ventricular function.

The units of central venous pressure depend on the method of measurement: if a simple saline manometer is used the measurement will be in centimetres of water but with an electronic pressure transducer the measurement is in mmHg (1 cmH&sub2;O 1.36 mmHg).

Many techniques of central venous cannulation have been described. The most reliable is via the right internal jugular vein, since its short straight course almost assures that the catheter tip will pass into the superior vena cava or right atrium. The internal jugular veins are preferred to the subclavian veins since there is a lower incidence of major complications such as pneumothorax and arterial puncture.

In many patients with left-sided heart disease or pulmonary disorders the central venous pressure is not a reliable indicator of left heart filling pressures. In these patients left atrial pressure can be measured directly via a surgically inserted catheter. The use of left atrial catheters carries a number of unique risks, including the possibility of air or clot emboli to the systemic circulation and the potential for bleeding at the time of removal.

Pulmonary artery pressures can be measured using a flow directed pulmonary artery catheter, also known as a Swan-Ganz catheter. The balloon-tipped catheter is inserted into the central venous system and ‘floated’ through the right heart into the pulmonary artery using the changes in pressure waveform as the catheter passes through the chambers of the heart to guide insertion (Fig. 1) 1664. The catheter can be used to measure central venous pressure, pulmonary artery systolic, diastolic, and mean pressures, and pulmonary capillary wedge pressure (also known as pulmonary artery occlusion pressure). The latter is the pressure measured when the catheter balloon is inflated so that the catheter is impacted into a peripheral pulmonary artery (Fig. 2) 1665. The catheter then registers the pressure transmitted retrogradely from the left atrium thus providing an estimate of left heart filling pressure.

The pulmonary artery catheter, when fitted with a thermistor, can also be used to measure cardiac output by thermodilution. Once the cardiac output is known, stroke volume, systemic and pulmonary vascular resistances, and left and right ventricular stroke work indices can all be calculated. Serial measurements of cardiac output and vascular resistances are used to assess the state of the circulation and, if necessary, to determine appropriate therapy.

The introduction of pulmonary artery catheters has been hailed as a major advance in management of patients during and after cardiac surgery, and their use has contributed greatly to knowledge and understanding of cardiovascular pathophysiology and pharmacology. To date, however, no study has shown that morbidity and mortality rates are reduced by their routine use. In view of their expense and potential complications, it seems reasonable to restrict their use to patients with poor left ventricular function, pulmonary hypertension, or complex lesions. Much routine cardiac surgery can be carried out without the use of pulmonary artery pressure monitoring.

Oxygen saturation can be monitored non-invasively and continuously by pulse oximetry. The oximeter probe measures the pulsatile absorption of light from two light emitting diodes and calculates the ratio of oxygenated to deoxygenated haemoglobin. The result is a waveform whose magnitude depends on the pulse pressure and the arterial oxygen saturation. It is thus a monitor of both the cardiovascular and respiratory systems. Unfortunately the accuracy, and therefore the usefulness, of the instrument is greatly reduced in conditions in which skin blood flow is reduced, such as hypothermia or low cardiac output.

Useful information about cardiovascular function can be gained by monitoring its effects on other organ systems.

Urine output is critically dependent on adequate renal perfusion pressure and intravascular volume. In an adult maintenance of a urine output of more than 0.5 ml/kg.h without the use of diuretics is a reassuring sign of satisfactory cardiac output and renal blood flow.

Inadequate cardiac output and consequent tissue hypoxia produces metabolic acidosis and a decrease in mixed venous oxygen saturation. Appropriate treatment will reverse these changes.

The core–peripheral temperature gradient has been used as a guide to cardiac output, since peripheral vasoconstriction is one of the compensatory mechanisms for a fall in output. It is, however, an unreliable indicator of cardiovascular function since it is affected by a number of variables including ambient temperature and the presence or absence of peripheral vascular disease.

The principles of postoperative care in the cardiac surgical patient are maintenance and optimization of tissue perfusion and oxygenation and prevention and, if necessary, prompt treatment of complications.

Patients are nursed in a high dependency area with the same monitoring facilities as those in the operating theatre and with access to blood gas and electrolyte analysis. A small minority of patients who have multisystem disorders will require admission to an intensive care unit, but the majority can be safely managed in a recovery area. This area should be sited close to the operating theatre and there should be adequate space for equipment and the performance of emergency surgical procedures. A one-to-one nurse to patient ratio is essential in the immediate postoperative period, until the patient is haemodynamically stable and extubated.

Circulatory management during the postoperative period is an extension of the management begun during weaning from cardiopulmonary bypass. The aim is to achieve and maintain an adequate cardiac output and ensure oxygenation at cellular level. This goal requires optimization of the determinants of cardiac output and of oxygen carrying capacity. Some degree of decreased myocardial function is almost invariable in the immediate postoperative period as the heart recovers from the insult of surgery.

In common with all surgical patients, the postoperative cardiac patient requires fluids both for maintenance of normal homeostasis and to replace abnormal losses. Hypovolaemia is frequently seen in the early postoperative period due to excessive bleeding or diuresis, increased capillary permeability, and the effects of vasodilators or rapid rewarming. Blood is given to maintain a haematocrit of 30 to 35 per cent; autotransfusion of red cells salvaged from mediastinal drainage decreases the need for homologous blood transfusion.

Controversy exists over the employment of colloid as opposed to crystalloid solutions to maintain the circulation. In the United Kingdom the usual practice is to restrict crystalloid administration on the first postoperative day to 0.5 to 1.0 ml/kg. h of 5 per cent glucose because of the obligatory retention of sodium and water after major surgery, and to use colloid solutions (plasma or plasma expanders) to increase the circulating volume. Potassium is added to intravenous fluids to maintain extracellular potassium at or above 4.5 mmol/l and avoid hypokalaemia which may be associated with dysrhythmias.

The use of controlled ventilation after cardiac surgery became widespread in the 1960s, with the belief that it allowed a smoother and more stable recovery from surgery. As surgical and anaesthetic techniques have improved, a number of studies have shown that earlier return to spontaneous respiration and early extubation are reasonable options in selected patients. Early extubation improves patient comfort, decreases the need for sedative drugs, reduces nursing dependency, and may decrease morbidity. Factors that should be considered before ventilatory support is withdrawn include haemodynamic stability, resolution of the effects of narcotics and muscle relaxants, adequate arterial blood gases with inspired oxygen concentrations of 50 per cent or less, and normal core temperature.

Extubation can be performed when, in addition to the above, the patient is awake and alert and able to maintain adequate gas exchange without distress. Using these criteria many patients can be extubated within a few hours of surgery. However, each patient must be assessed on an individual basis and some patients will undoubtedly benefit from a longer period of controlled ventilation after surgery.

Effective analgesia helps to relieve anxiety and tachycardia. In the early postoperative period intravenous opiates should be given either by continuous infusion or by a patient-controlled system.

Hypertension is common following coronary artery surgery. High blood pressure increases cardiac work and oxygen consumption and may lead to myocardial ischaemia and increased bleeding. Recovery of consciousness with inadequate analgesia and sedation is a common but easily treated cause. Other management is directed at reducing afterload by infusion of vasodilators such as glyceryl trinitrate or sodium nitroprusside. Both drugs have a short duration of action, and the dose can be titrated against response. Reflex tachycardia can be treated by a &bgr;-blocking drug, although these should be used with caution in patients with left ventricular dysfunction.

Excessive postoperative blood loss is due to inadequate haemostasis, a deficiency in coagulation, or both. Continued heavy bleeding in a patient with normal or near normal clotting is an indication for surgical re-exploration: when coagulation is abnormal the decision to re-explore is more difficult. Coagulopathy after cardiopulmonary bypass is frequently multifactorial. Platelet function is affected by preoperative treatment with aspirin, haemodilution, and mechanical destruction during cardiopulmonary bypass. Levels of circulating clotting factors may be greatly reduced, and excessive fibrinolysis can occur. Residual unneutralized heparin may also contribute to bleeding. Treatment should include replacement of platelets and clotting factors as well as red cells, and administration of protamine if indicated.

Tamponade occurs if clotted blood accumulates in the mediastinum, producing mechanical obstruction to cardiac filling. This results in hypotension due to decreased stroke volume, and an increase in filling pressures, together with tachycardia, oliguria, and peripheral cooling. Hypotension from tamponade may improve initially with administration of fluids and inotropic support, but circulatory collapse eventually occurs. The diagnosis rests on a high index of suspicion although chest radiographs may show an enlarging heart shadow, and echocardiography maybe helpful in demonstrating pericardial clot and a compressed ventricle. Pulsus paradoxus (an exaggeration of the normal fall in arterial pressure with inspiration) is often cited as a sign of cardiac tamponade, but is rarely seen. The treatment is mediastinal exploration with removal of clot. Opening the chest usually produces a marked haemodynamic improvement.

Rhythm disturbances are common after cardiac surgery. Many factors may contribute: hypokalaemia, acid-base abnormalities, hypothermia, myocardial ischaemia, circulating catecholamines, and surgical damage. Treatment includes correction of hypokalaemia and acidosis, administration of antidysrhythmic drugs, and occasionally overdrive pacing or DC cardioversion. About 30 per cent of patients undergoing coronary artery surgery will develop atrial fibrillation by the third postoperative day. Prophylatic administration of amiodarone may be effective in preventing this.

Postoperative myocardial dysfunction can be caused by inadequate myocardial protection during surgery, perioperative infarction, dysrhythmias, air embolism, or coronary spasm: pre-existing ventricular dysfunction is a major risk factor. Inadequate cardiac output is characterized by hypotension and hypoperfusion associated with vasoconstriction, oliguria, and acidosis. Successful management depends on optimization of cardiac output and reduction in cardiac workload by physiological, pharmacological, and mechanical means. General supportive measures include the treatment of acidosis and hypoxaemia. A pulmonary artery catheter may be required to direct treatment.

Cardiac output is the product of heart rate and stroke volume. Increasing the rate will therefore increase the cardiac output (if stroke volume were unchanged). At heart rates above 110 to 120 beats/min, however, shortening of the duration of diastole starts to impair ventricular filling and markedly reduces stroke volume. Myocardial ischaemia may also occur in patients with severe tachycardia, and the function of prosthetic valves may be significantly impaired.

Dysrhythmias will adversely affect cardiac output and should be promptly diagnosed and treated. The co-ordinated atrial contraction that occurs in sinus rhythm contributes up to 30 per cent of ventricular filling and may be vital for adequate cardiac performance in patients with poor ventricular function.

Measurements of both right and left heart filling pressures are required to optimize preload. Although increased filling pressures are needed to augment cardiac output, ventricular overdistension and pulmonary congestion must be avoided.

Myocardial contractility is increased by the use of positively inotropic drugs. Catecholamines such as dopamine, dobutamine, adrenaline, and isoprenaline will improve contractility, albeit at the cost of an increase in myocardial oxygen demand. The overall clinical effects depend on the relative degrees of stimulation of &agr;- and &bgr;-adrenergic receptors. The newer phosphodiesterase inhibitors such as amrinone, milrinone, and enoximone combine positive inotropy with vasodilatation; these drugs may find a place in the treatment of low cardiac output syndrome.

The use of vasodilator drugs to reduce afterload decreases myocardial wall tension and oxygen demand and allows the ventricles to empty more fully. Successful treatment is marked by improving cardiac output in the face of unchanged or only slightly decreased arterial pressure. A combination of vasodilators and inotropes is often required to maximize cardiac output.

When pharmacological support is insufficient to reverse the low cardiac output syndrome, mechanical assistance should be considered. The intra-aortic balloon pump is a volume displacement device in which inflation and deflation of an elongated balloon in the descending thoracic aorta is timed to coincide with the cardiac cycle. Balloon inflation increases diastolic blood pressure, thus improving coronary flow: active deflation just prior to systole unloads the aorta and reduces resistance to ventricular ejection. The balloon pump does not generate blood flow but augments existing cardiac output and increases coronary perfusion. In contrast, ventricular assist devices are capable of maintaining cardiac output without ventricular function. At present experience with these devices is limited, but it seems likely that their use will increase as knowledge and technology advance.

Kaplan JA, ed. Cardiac Anesthesia. 2nd edn. Orlando: Grune and Stratton Inc, 1987.

Kormos RL, Griffith BP. Ventricular assist devices. In: Shoemaker WC, Ayres S, Grenvik, A, Holbrook PR, Thompson WL, eds, Textbook of Critical Care, 2nd edn. Philadelphia: WB Saunders Co, 1989: 428–38.

Reemtsma K, Bregman D, Cohen SS, Kashel P. Mechanical circulatory support—advances in intra-aortic balloon pumping. In: Shoemaker WC, Ayres S, Grenvik A, Holbrook PR, Thompson WL, eds, Textbook of Critical Care, 2nd edn. Philadelphia: WB Saunders, Co, 1989: 420–8.

Swan HJC, Ganz W, Forrester JS, Marcus H, Diamond G, Chonette D. Catheterization of the heart in man with the use of a flow-directed balloon-tipped catheter. N Engl J Med 1970; 283: 447–51.

Tuman KJ, et al. Effect of pulmonary artery catheterization on outcome in patients undergoing coronary artery surgery. Anesthesiology 1989; 70: 199–206.