- Email: [email protected]
Hypotensive anaesthesia
Hypotensive Anaesthesia
P. J. S i m p s o n
An understanding of the normal physiological mechanisms controlling arterial blood pressure is essential for the use of elective intraoperative hypotension. Mechanisms of blood pressure reduction include the mechanical effects of intermittent positive pressure ventilation, posture and the application of PEEP. Pharmacological methods involve interruption of the normal sympathetic control of vascular tone, either centrally, neurologically, by receptor inhibition, or by a direct effect on peripheral vascular diameter. General anaesthesia will also produce hypotension by a direct myocardial effect, which provides a good background against which a more specific technique can be employed. There is no single ideal hypotensive agent since the demands of surgery vary from the simple need to reduce blood loss to operations which would otherwise be impossible. Elective hypotension is perfectly safe in experienced hands and its use may often be specifically indicated during surgery for many reasons.
Introduction
excessively low pressures, for example during clipping of a cerebral aneurysm. Further indications include resection of aortic coarctation where rapid fluctuations in blood pressure necessitating immediate control are required. In general, there is a need for two basic methods of hypotension; elective, slow onset, slow recovery hypotension on the one hand and 'dial a pressure' hypotension on the other. Operative bleeding can be arterial, in which case it is directly related to the mean arterial pressure (MABP), capillary, when it is dependent upon local flow in the capillary bed, or venous, when it is related to venous return, venous tone and therefore posturally dependent. Although arterial bleeding can only be abolished by the use of a tourniquet, it can be considerably reduced by a fall in M A B P or heart rate (HR). Capillary flow is also reduced by elective hypotension and by localised vasoconstriction due, for example, to adrenaline infiltration or metabolic factors such as acid-base disturbance or hypoxia.
When induced hypotension is superimposed upon a general anaesthetic technique, no single agent is capable of providing the ideal conditions for all operations, the requirements of surgery falling into three broad groups. The first demands relatively slow onset and sustained moderate hypotension with a slow return to normal pressures, and is ideal for most plastic, maxillofacial and ear, nose and throat surgery when rapid return to normal pressures may cause reactionary haemorrhage. In the second, where massive blood loss is anticipated, moderate sustained hypotension together with a reduction in heart rate is probably all that is required. In the third group, some operations are not only impossible without profound hypotension but also require short periods of
P. J. Simpson, Department of Anaesthetics, Frenchay Hospital, Bristol BS16 1LE Current Anaesthesia and Critical Care
(~ 1992LongrnanGroupUK Ltd
(1992)3, 91)-97 90
HYPOTENSIVE ANAESTHESIA 91 Venous tone can be completely abolished by spinal or epidural anaesthesia and by direct acting vasodilators such as sodium nitroprusside.
Cardiovascular physiology Mean arterial blood pressure is directly related both to cardiac output (CO) and to peripheral resistance (TPR). While cardiac output is dependent upon myocardial contractility, determining stroke volume (SV), and heart rate, peripheral resistance is a function of vascular dilation. Peripheral vasodilation is primarily controlled by sympathetic activity, vasodilation being produced by interruption of sympathetic pathways rather than by any parasympathetic effect. Some cholinergic vasodilator fibres do exist, for example in the smooth muscle of the gut. Hypotension due to a reduction in total peripheral resistance can therefore be achieved either centrally by drugs such as volatile agents acting on the vasomotor centre or peripherally at the level of the sympathetic ganglia, postganglionic noradrenergic (alpha) terminals, or directly on the blood vessels themselves. In order to produce a measurable reduction in bleeding, local blood pressure must be in the region of 30-40 mmHg and venous drainage of the area must be unimpaired.
Posture Posture influences intraoperative bleeding both by producing regional ischaemia if the operation site is elevated above the level of the heart, and also by augmenting the effect of agents, such as sympathetic ganglion blocking drugs, by pooling of blood in the dilated venous vascular bed. The effects of head-up tilt on regional cerebral perfusion pressure in relation to mean arterial blood pressure at heart level are considerable. For each 2.5 cm of vertical height above the heart, blood pressure falls by 2 mmHg. The use of posture to improve the operating field and to reduce bleeding in head and neck surgery is well known. Whenever possible, posture should be used to augment pharmacological methods of induced hypotension, particularly those which depend upon venous pooling.
Intermittent positive pressure ventilation (IPPV) Under normal circumstances, venous return to the heart occurs during inspiration when a negative intrathoracic pressure enhances blood flow to the heart, even against the force of gravity. During IPPV inspiration is associated with positive intrathoracic pressure, inevitably resulting in a reduction in venous return. If such an effect is augmented by posture, the resultant fall in venous return and therefore cardiac output can be considerable. In normotensive anaesthetised subjects, IPPV has little effect on
cardiac output due primarily to the reflex vasoconstriction produced in response to what is effectively a limited Valsalva manoeuvre a. In addition, the baroreceptors respond to hypotension by inducing a reflex tachycardia. There are, however, two methods of inducing hypotension in which cardiac output may fall considerably in response to IPPV, due to temporary autonomic paralysis. Both ganglion blocking drugs and beta adrenoceptor antagonists may produce a partial or completely blocked Valsalva manoeuvre,2 thus limiting the production of a normal compensatory response. IPPV is a useful adjunct to any hypotensive technique, largely because it augments pharmacological methods of blood pressure reduction, limiting the dose necessary to produce the desired effect and also the postoperative duration of hypotension. In addition, artificial ventilation allows the application of positive end expiratory pressure (PEEP) to the airway, producing a further reduction in venous return and cardiac output.
Respiratory physiology Carbon dioxide (C02) is itself a vasodilator and hyperventilation leading to hypocapnoea will induce vasoconstriction. C02 control can be achieved either by IPPV using moderate hyperventilation or by C02 absorption in a circle system using soda lime. Care, however, must be taken during hyperventilation with patients in the head-up position, since vasoconstriction may reduce cerebral blood flow to critical levels.
Effects of hypotension upon pulmonary gas exchange Since pulmonary blood flow is dependent upon gravity, the head-up position produces a reduction in flow to the apical parts of the lung. Alveolar ventilation, however, will occur throughout the lung, including the upper segments, resulting in considerable ventilation/perfusion (V/Q) mis-match and an increase in physiological deadspace. This may be as great as 80% of the tidal volume3 and is of particular importance in spontaneously breathing patients. The reduction in alveolar ventilation and increase in physiological shunt is an additional reason for the use of IPPV during elective hypotension. Furthermore, many anaesthetists increase the inspired oxygen concentration to minimise the effects of this ventilation perfusion imbalance.
Pharmacological methods of vasodilation Volatile anaesthetic agents Halothane. Although halothane produces a moderate degree of vasodilation, the overall fall in total peripheral resistance is of the order of 15-18%. Vasodilation in the skin and splanchnic vascular beds is
92
CURRENTANAESTHESIA AND CRITICALCARE
balanced by vasoconstriction in skeletal muscle, any additional hypotension produced by halothane being a direct result of myocardial depression. In addition, bradycardia induced by the stimulant effect of halothane on the vagus nerve will further reduce cardiac output. While halothane is often used successfully in low concentration as a background to hypotensive anaesthesia, its use as a sole hypotensive agent in larger doses should be discouraged. This is of particular importance in neurosurgery if the increase in intracranial pressure due to vasodilation is to be avoided. Enflurane. The mechanisms and effects of hypotension induced by enflurane are similar to those of halothane. Myocardial depression and vagal stimulation are still significant factors if excessive doses of the drug are used and for this reason it should also only be employed in moderate doses simply as a background agent. Isoflurane. Unlike both halothane and enflurane, isoflurane has no effect on myocardial contractility. The peripheral vasodilatory effect is readily reversible by alterations in inspired concentration of the drug. For this reason, it is becoming increasingly used as a hypotensive agent, particularly when only moderate reduction in arterial pressure is required. It has the additional benefit that increasing doses not only produce vasodilation and hypotension but also produce central nervous depression minimising any reflex vasoconstriction or tachycardia, which may occur as a result of baroreceptor stimulation under relatively light anaesthesia. Isoflurane also appears to have less effect than either halothane or enfturane upon intracranial pressure in patients in whom normal values are present pre-operatively. 4
Sympathetic ganglionic blockade Trimetaphan, (Pentolinium). These drugs produce autonomic ganglion blockade by competitive inhibition of acetylcholine. Their effects are not confined to the sympathetic system since acetylcholine transmission also occurs in parasympathetic ganglia. Interruption of sympathetic outflow produces vasodilation which tends to be relatively slow in onset and recovery. The duration of hypotension produced by trimetaphan (Arfonad) is relatively short (10-15 min) and for this reason the drug is often administered by intravenous infusion (3-4mg.min-1). By contrast, a single injection of 5-15mg pentolinium (Ansolysen) produces hypotension for about 45 min and allows a slow return of blood pressure to normal values. Pentolinium is no longer available in the UK, but considerable demand may result in production being resumed. Although a number of gastrointestinal and urinary symptoms may result from concomitant parasympathetic blockade, the two of clinical importance during induced hypotension are mydriasis and tachycardia. The increase in heart rate which
often accompanies hypotension produced by ganglion blockade may severely impair the effectiveness of these drugs in reducing bleeding. Tachyphylaxis, that is the need for increasing doses of the drug to produce the same effect, is particularly marked with trimetaphan and may make a stable blood pressure difficult to achieve. Continuous infusion is considerably superior to intermittent bolus dose administration in this respect.
Non-depolarising neuromuscular blocking drugs Alcuronium, (d-Tubocurarine). The use of nondepolarising neuromuscular blocking drugs such as curare to facilitate IPPV as an adjunct to elective hypotension has been advocated for some time. D-tubocurarine, formerly the most widely used drug in this situation and also alcuronium (Alloferin) were initially thought to induce a degree of sympathetic ganglionic blockade due to a relative lack of specificity for the skeletal neuromuscular junction. More recently it has been shown that the main reason for their hypotensive effect is the histamine release associated with their administration which itself induces vasodilation. This far outweighs any effect due to mild ganglionic blockade.
Alpha adrenoceptor blockade Phentolamine, Phenoxybenzamine, Chloropromazine, Droperidol. Alpha adrenergic blocking agents produce vasodilation by competitive blockade of post-synaptic noradrenergic receptors within the sympathetic system. While the effects of phentolamine are relatively short (20-40 min) and easily reversible, those of phenoxybenzamine may last several days since this drug, a nitrogen mustard derivative, forms an irreversible receptor complex. Phentolamine also exerts a direct myocardial stimulant effect, increasing both oxygen consumption and heart rate. Phenoxybenzamine may produce considerable sedation. While phentolamine (5-10 mg) is used in the rapid production of intraoperative vasodilation, phenoxybenzamine (0.5-2.0 mg.kg -1 for 10 days) is more commonly employed for chronic vascular expansion prior to surgery to minimise the effects of circulating catecholamines, e.g. in the surgical removal of phaeochromocytoma. Both chlorpromazine and droperidol produce mild alpha adrenergic blockade which is often useful in the preoperative preparation of patients prior to hypotensive anaesthesia and/or hypothermia.
Beta adrenoceptor blockade Propranolol, Oxprenalol, Atenolol, Labetalol, (Practolol). The main advantages of the beta adrenoceptor antagonists in induced hypotension are in the reduction of heart rate and cardiac output. Many anaesthetists believe that the maintenance of a slow heart rate
HYPOTENSIVE ANAESTHESIA 93 without any additional hypotension considerably reduces operative bleeding. Propranolol is often been used to produce this 'rheostatic' hyptension. Although preoperative oral therapy (40mg t.d.s.) is probably best, 1-2mg intravenously can be used during anaesthesia. Beta adrenoceptor blockade with either propranolol or oxprenalol is also employed either preoperatively or intra-operatively to counteract the tachycardia produced as a side effect of induced hypotension with either ganglion blocking or direct acting vasodilator drugs. Again it is best to administer the drugs orally rather than intravenously since this produces a steady intraoperative blood level of the drug. Although the combined alpha and beta adrenoceptor blocking drug labetalol would seem ideal for use in induced hypotension, it is important to realise that the alpha blocking effects of the drug only last for 30 min compared with a 90 min duration of beta blockade. In addition, the beta blocking effects are five to seven times as potent as the alpha blockade. The perioperative use of beta blockade with either propranolol or labetalol may have considerable benefit in the prevention of wide fluctuations in blood pressure particularly in patients with subarachnoid haemorrhage and vasospasm.
Direct acting vasodilators Sodium nitroprusside. The main advantage of this drug is its extremely evanescent action, allowing rapid reduction of blood pressure and equally rapid restoration to normal levels. It is the only drug capable of predictably producing 'dial-a-pressure' hypotension over relatively short periods, e.g. in the prevention of bleeding in meningiomas and major vascular surgery, or to facilitate clipping of cerebral aneurysms. As a vasodilator SNP inevitably produces an increase in intracranial pressure and for this reason should not be used during neurosurgery before the skull is open in a patient with raised intracranial pressure. Nevertheless autoregulation under induced hypotension with nitroprusside is maintained at cerebral perfusion pressures considerably lower than with other drugs. 5
Metabolism and toxicity of sodium nitroprusside. Shortly after the introduction of nitroprusside into clinical practice reports of fatalities during its use were directly attributed to cyanide poisoning. Each molecule of sodium nitroprusside contains 5 cyanide radicals which are lib~erated on breakdown of the drug in either plasma or red blood cells. Several studies, however, have failed to demonstrate significant effects upon' red cell oxygen transport of cyanide liberated during routine clinical use of
nitroprusside.6'7
The normal metabolic pathway of SNP breakdown is nonenzymatic, occurring in both red cells and plasma. The intracellular reaction is catalysed by the conversion of haemoglobin to methaemoglobin. Ultimately, more than 98% of the cyanide produced from
SNP is contained within the red blood cells while a small proportion is combined with either methaemoglobin or Vitamin B12. The majority of cyanide is metabolised in the liver by the enzyme rhodanese to thiocyanate which is then excreted in the urine. The rate limiting factor in cyanide metabolism appears to be the availability of sulphydryl groups and the administration of sodium thiosulphate can considerably enhance thiocyanate production and therefore reduce blood cyanide levels.8 The use of thiosulphate does not appear to affect the hypotension produced by nitroprusside. At the maximum safe doses recommended for nitroprusside administration, 1.5 mg/ kg -19 or 10 txg.kg -I min-1, ~° small increases in plasma lactate occur which are mirrored by increases in arterial base deficit. These changes are only minor, the maximum base deficit being of the order - 6 to - 7 mmols per litre, and are spontaneously reversible upon discontinuation of nitroprusside therapy. The routine measurement of acid-base balance during nitroprusside therapy would appear to provide adequate clinical information of the development of cyanide toxicity during routine clinical use. The use of sodium nitroprusside in patients already anaesthetised with a background hypotensive anaesthetic technique would still appear to be the method of choice for the production of extreme hypotension for neurosurgery. No other drug at present provides the predictable and rapid hypotensive effect necessary for many aspects of intracranial surgery. Natural apprehension over the potential toxicity of SNP has largely centred round a few reported cases all of which were directly attributable to cyanide poisoning. H'12'13 Close examination of these reports confirms that in all cases doses of SNP vastly in excess of those required for routine clinical use were needed to produce toxic symptoms. If the dose of nitroprusside is limited to that already stated above, toxic symptoms will not occur in patients with normal renal and hepatic function. For longer term infusion and in the presence of adequate sulphydryl groups as the substrate for cyanide detoxification by rhodanese, a maximum dose rate of 8 ixg.kg-114 has been shown to be satisfactory. Trinitroglycerine (TNG). Hepatic metabolism of trinitrate produces di- and mono-nitrate and finally glycerol.14 The vasodilator activity of these smaller nitrate molecules is reduced as their size decreases. Nitroglycerine produces a steadier and less dramatic reduction in arterial blood pressure, having a greater effect on systolic than diastolic pressure and tending to maintain the blood flow. Recovery from nitroglycerine-induced hypotension is also less rapid, taking between 10 and 20 rain in contrast to the 2-4 min with SNP. It has been suggested that this slower effect of nitroglycerine produces less overshoot of blood pressure either at induction of hypotension or following restoration of normal blood pressure but as the drug appears less effective in some cases in the production of extreme hypotension, its use may not
94
CURRENT ANAESTHESIA AND CRITICAL CARE
be ideal in all situations. Although TNG has been advocated as a direct acting vasodilator for neurosurgery, in one study,~5 hypotension to 50 mmHg was not possible in three out of 22 patients. Unlike sodium nitroprusside which dilates both resistance and capacitance vessels equally, nitroglycerine exerts its effect principally upon the venous capacitance system. As a result, diastolic blood pressure is maintained at higher levels than with nitroprusside and for this reason TNG maintains coronary artery perfusion more effectively than SNP. While this is probably of little importance in fit patients, it may be of considerable advantage in patients with impaired myocardial or cerebral circulation. However, the increase in intracranial pressure produced by nitroglycerine may be even greater than with SNP.
Spinal and epidural anaesthesia The sympathetic blockade produced by spinal and epidural anaesthesia is a very effective way of inducing hypotension. Lumbar epidural anaesthesia will produce arteriolar dilation and hypotension together with a reduction in venous tone. This is enhanced by posturally dependent pooling of venous blood leading to a fall in venous return and therefore cardiac output. If the block is extended to the mid-thoracic region, the cardiac sympathetic fibres passing in segments T1 to T4 will also be blocked, preventing the compensatory tachycardia which will otherwise occur. This also limits any baroreceptor response, and prevents tachycardia occurring as a result of other pharmacological methods of induced hypotension. Regional anaesthesia is most commonly employed in lower abdominal or pelvic surgery to minimise blood loss, particularly that occurring from the pelvic venous plexuses. The complete abolition of venous tone is extremely effective in minimising blood loss without the need for profound arteriolar hypotension. If adrenaline is added to the local anaesthetics used, its systemic absorption may partially counteract the hypotensive effect of the regional block. Adrenaline has the advantage of prolonging the effects of the local anaesthetics used, although this can be satisfactorily achieved more simply by the use of intermittent or continuous epidural injections via a catheter.
Monitoring during induced hypotension Routine monitoring should include ECG, oxygen saturation, end-tidal carbon dioxide concentration and either direct or indirect arterial blood pressure measurement. Although instruments such as the cerebral function monitor are potentially available for use during hypotensive anaesthesia, their use and, indeed, their possible benefits are relatively limited. 16
Electrocardiographic monitoring ECG monitoring is essential to demonstrate two vital signs of inadequate myocardial perfusion, the development of ectopic beats and ST segment depression. Although a single channel ECG is incapable of demonstrating the exact site of any ischaemia, ST segment changes do occur and are usually readily reversible by increasing the blood pressure. The myocardial response to relative hypoxia and hypoperfusion is a sensitive monitor of hypotension being excessively exploited.
Measurement of blood pressure Indirect m e a s u r e m e n t
Hutton and Prys-Roberts 17 have shown that errors in interpretation of oscillotonemetry have tended to err on the side of safety since what was originally thought to be systolic blood pressure has now been demonstrated to more closely represent the mean pressure. The semi-automated versions (Dinamap) are an improvement since they remove the observer bias and indeed display not only systolic but also mean and diastolic pressures. Direct m e a s u r e m e n t
Although indirect methods of blood pressure measurement are used during relatively mild hypotensive techniques, such as intermittent ventilation with isoflurane, epidural anaesthesia, or even in some cases, the use of ganglion blocking drugs, direct monitoring is an essential part of the technique when rapidly acting direct vasodilators such as sodium nitroprusside are employed. The position of the transducer is the level at which blood pressure is measured and if the mean arterial blood pressure at head level is desired, the transducer should be placed at this height. Assuming a normal intracranial pressure and with the appropriate postural adjustments, the measurement is then directly related to cerebral perfusion pressure.
Oxygen saturation and end-tidal carbon dioxide monitoring Induced systemic hypotension will inevitably result in a reduction in pulmonary artery pressure, and since, pulmonary perfusion is gravity dependent, an increase in V/Q mismatch within the lung. The ability to non-invasively monitor respiratory gas exchange allows alterations in inspired oxygen and minute volume to be made in specific relation to the patient's needs, rather than on an empirical basis. Inspired oxygen may need to be increased during a period of hypotension to compensate for changes in lung perfusion. In addition end-tidal CO2 analysis is a sensitive method of detecting air embolism, which
HYPOTENSIVE ANAESTHESIA 95 may also occur during some cranial or head and neck operations.
Practical technique of induced hypotension In all cases of elective hypotension, a background anaesthetic against which hypotension can be induced is essential, the principles of balanced anaesthesia dictating that it is better to employ individual agents to achieve specific effects rather than to pursue the toxic properties of an agent like halothane in the production of hypotension by myocardial depression. An ideal background anaesthetic consists of omission of atropine premedication, since this induces tachycardia, but the use of generous sedation or analgesia. It is essential to avoid preoperative anxiety and the release of adrenaline, since the effects take some time to abate under anaesthesia. Induction of anaesthesia with thiopentone, fentanyl and a long-acting, non-depolarising neuromuscular blocking drug, for example alcuronium, should be employed. Following topical anaesthesia of the larynx, intubation is performed with a non-kinking endotracheal tube to remove any possible risk of partial airway obstruction and CO2 retention. Moderate hyperventilation with nitrous oxide and oxygen together with 0.5-1.0% isoflurane is then the background technique against which hypotensive agents can be used. Under these stable conditions, specific hypotensive drugs can then be employed with the minimum of side effects, for example tachycardia or excessive hypotension. During hypotension, as mentioned earlier, the inspired oxygen concentration may be increased to 40 or even 50% if excessively low pressures are being employed over a short period, for example during neurosurgery.
Postoperative management Recovery staff must be made aware of patients in whom elective hypotension has been employed. Accurate and regularly blood pressure monitoring together with meticulous airway care to avoid CO2 retention and partial obstruction is essential. The patient's position should be determined by the blood pressure measured, and postural changes may be necessary for several hours to ensure adequate cerebral perfusion. Supplementary oxygen should be administered in all cases until the patient is adequately awake and may be required for longer where oxygenation is thought to be critical. In cases where pharmacological modification of sympathetic responses has been undertaken, such as with the use of ganglion blocking drugs, patients should remain in bed for 12-18h postoperatively and, if necessary, lying virtually flat until they are able to sit up without feeling dizzy.
Relative contra-indications to induced hypotension Although many anaesthetists are reluctant to employ induced hypotension, there are very few patients in whom it cannot be used safely. Most would refrain from utilising the technique in patients with evidence of severe cardiovascular or cerebro-vascular disease, although both these are relative contraindications if, for example, cerebral aneurysm surgery is proposed.
Myocardial ischaemia This is made worse by an increase in the rate pressure product or a reduction in myocardial oxygen delivery. Since hypotensive techniques are designed to reduce both heart rate and arterial blood pressure, the amount of cardiac work is reduced considerably. Many such patients are already on beta-adrenoceptor blockade and this should be continued intraoperatively. In cases where a reduction in afterload is produced by direct acting vasodilators such as sodium nitroprusside, myocardial work may be further reduced, but an adequate perfusion pressure must be maintained.
Hypertension Although patients with treated hypertension may be abnormally sensitive to hypotensive drugs, such techniques can still be employed with care. Untreated hypertension, however, is a contraindication, since the blood pressure may be extremely labile and profound hypotension results. It is also important to remember that volatile anaesthetic agents enhance the hypotensive effects of drugs which the patient is already receiving for routine control. Monitoring the ECG is essential in patients with cardiovascular disease. Rollason and Hough is demonstrated ST depression in hypertensive patients and Simpson et a119 noted similar changes during anaesthesia and nitroprusside induced hypotension. While the importance of such changes is probably doubtful, they do demonstrate the need for care when utilising hypotension in such patients.
Respiratory disease Contraindications to the use of hypotensive anaesthesia in patients with chronic respiratory disease are related to the disturbance of normal pulmonary physiology. The increase in physiological dead space due to ventilation/perfusion imbalance is more important in patients in whom preoperative gas exchange is limited. Under normal circumstances, hypoxic pulmonary vasoconstriction occurring in poorly ventilated segments of the lung prevents gross disorders of ventilation and perfusion. Vasodilation induced by direct acting drugs such as sodium
96
CURRENT ANAESTHESIA AND CRITICAL CARE
nitroprusside, abolishes this response and will therefore make shunting worse. Reversible airways obstruction and bronchospasm may be made worse by the use of either ganglion blocking drugs or beta adrenoceptor antagonists which are not cardiospecific and such drugs are contraindicated in asthmatics. High inflation pressures may lead to CO2 retention and severe hypotension due to impairment of venous return.
Diabetes mellitus This relative contraindication to induced hypotension concerns the drugs used rather than the actual induction of arterial hypotension. Ganglion blocking drugs, by producing sympathetic blockade, impair stress induced gluconeogenesis mediated by adrenaline. Beta blockers may also potentiate hypoglycaemia in insulin dependent diabetics and it is the combination of hypoglycaemia plus hypotension which may produce severe consequences, particularly on cerebral metabolism. Under normal circumstances, however, it is safe to employ volatile agents, direct acting vasodilators or local anaesthetic techniques without impairment of blood sugar levels.
Complications of hypotensive anaesthesia Failed hypotension Although not regarded by many as a complication of hypotensive anaesthsia, failure to appreciate this may lead to the excessive use of certain agents and consequent toxic effects. Some patients, for example, with a sensitive renin-angiotensin system have been shown to be 'resistant' to sodium nitroprusside. Before this was appreciated, excessive doses of the drug were employed in the mistaken belief that eventually hypotension would be achieved. This led in some cases to sodium nitroprusside toxicity, cyanide poisoning, and even death. If hypotension using one drug is not sufficient, then a second agent acting at a different site in the sympathetic system should be employed. This has been put to good effect, for example by combinations of ganglion blocking drugs such as trimetaphan with sodium nitroprusside or, more recently, the combination of isoflurane and nitroprusside.
Excessive hypotension Moderate hypotension is achieved during many anaesthetics and many patients are subjected to systolic arterial pressures of 60 mmHg during routine surgery. It is unlikely that hypotension to this degree, particularly when induced by vasodilation and accompanied by additional oxygen administration in fit patients, would lead to any permanent adverse effect. Nevertheless, situations have arisen where
permanent damage has resulted from the use of elective hypotension. This is related either to excessive hypotension, impaired oxygenation or inadequate blood pressure monitoring. Lindop, 2° surveying the evidence concerning the effects of hypotension on vital organs, stressed the importance of flow rather than blood pressure in the development of complications. This is particularly important since normal oxygen extraction by most organs with the exception of the heart is only about 25% of its potential. In considering the effects of hypotension on the brain, autoregulation becomes important. It has been shown that the lower limit of autoregulation is better preserved with nitroprusside induced hypotension than with trimetaphan, flow remaining constant down to mean pressures of 4 0 - 5 0 mmHg. 5 These figures need to be interpreted with some caution when extrapolated to humans because of the effects of anaesthesia and other metabolic derangements upon cerebral blood flow. More detailed studies, such as continuous E E G monitoring 16 and jugular venous oxygen measurements, 21 have produced inconclusive results. Probably the most important rule is not to reduce the intraoperative systolic pressure to below the preoperative diastolic, and to avoid where possible severe head-up tilt unless blood pressure is being measured at the head level. The monitoring of ST segment depression to detect myocardial ischaemia would appear to be the most reliable method of demonstrating adverse cardiac effects due to hypotension. ~9 Provided myocardial work is reduced and a compensatory tachycardia due to hypotension does not occur, severe problems appear to be very rare.
Outcome Although those antagonistic to hypotensive anaesthesia will always cite cases of permanent damage or even death related directly to the lowering of blood pressure, large series, particularly that collected by Enderby 22 do not bear this out. At East Grinstead, the mortality rate was 1 in 4128 cases of induced hypotension and Kerr 23 in a separate series reported no mortality or morbidity in a series of 700 patients.
References 1. Prys-Roberts C, Kelman GR, Greenbaum R, Robinson RH. Circulatory influencesof artificial ventilation during nitrous oxide anaesthesia in man. II. Results: the relative influence of mean intrathoracic pressure and arterial carbon dioxide tension. Br J Anaesth 1967; 39:533-548 2. Blackburn JP, Conway CM, Davies RM, Enderby GEH, Eldridge A, Leigh JM, Lindop MJ, Strickland DAP. Valsalva responses, and systolictime intervals during anaesthesia and induced hypotension. Br J Anaesth 1973;45: 704-710 3. Eckenhoff JE, Enderby GEH, Larson A, Eldridge A, Judevine DE. Pulmonary gas exchange during deliberate hypotension. Br J Anaesth 1963; 35:750-758
HYPOTENSIVE ANAESTHESIA 4. Murphy FL Jnr, Kennell EM, Johnstone RE et al. The effects of enflurane, isoflurane and halothane on cerebral blood flow and metabolism in man. Abstracts of Scientific Papers. Annual Meeting of the American Society of Anesthesiologists 1974; 61-62 5. Stoyka WW, Schutz H. The cerebral response to sodium nitroprusside and trimetaphan controlled hypotension. Canadian Anaesthetists Society Journal 1975; 22:275-282 6. du Cailar J, Mathieu-Dande JC, Duschade J, Lamarche Y, Castel J. Nitroprusside, its metabolites and red cell function. Canadian Anaesthetists Society Journal 1978; 25:92-105 7. Vesey C J, Krapez JR, Cole PV. The effects of sodium nitroprusside and cyanide on haemoglobin function. J Pharm Pharmacol 1980; 32:256-261 8. Krapez JR, Vesey CJ, Adams L, Cole PV. Effects of cyanide antidotes used with sodium nitroprusside infusion: sodium thiosulphate and hydroxycobalamin given prophylactically to dogs. Br J Anaesth 1981; 53:793-804 9. Vesey CJ, Cole PV, Simpson PJ. Sodium nitroprusside in anaesthesia. BMJ 1975; 3:229 10. Tinker JH, Michenfelder JD. Sodium nitroprusside; pharmacology, toxicology and therapeutics. Anesthesiology 1976; 45:340-354 11. Davies DW, Kadar D, Steward D J, Munroe IR. A sudden death associated with the use of sodium nitroprusside for the induction of hypotension during anaesthesia. Canadian Anaesthetists Society Journal 1975; 22:547-552 12. Jack R. The toxicity of sodium nitroprusside. Br J Anaesth 1974; 46:952
97
13. Merrifield A, BlundeU M. Toxicity of sodium nitroprusside. Br J Anaesth 1974; 46:324 14. Michenfelder JD, Tinker JH. Cyanide toxicity and thiosulphate protection during chronic administration of sodium nitroprusside in the dog. Anesthesiology 1977; 47: 441-448 15. Chestnut JS, Albin MS, Gonzalez-Abola E, Newfield P, Maroon JC. Clinical evaluation of intravenous nitroglycerin for neurosurgery. J Neurosurg 1978; 48:704-711 16. Patel H. Experience with the cerebral function monitor during deliberate hypotension. Br J Anaesth 1981; 53: 639645 17. Hutton P, Prys-Roberts C. The oscillotonometer in theory and practice. Br J Anaesth 1982; 54:581-593 18. Rollason WN, Hough JM. A study of hypotensive anaesthesia in the elderly. Br J Anaesth 1960; 32:276-285 19. Simpson PJ, Bellamy D, Cole PV. Electrocardiographic studies during hypotensive anaesthesia using sodium nitroprusside. Anaesthesia 1976; 31:1172-1178 20. Lindop MJ. Complications and morbidity of controlled hypotension. Br J Anaesth 1975; 47:799-803 21. Larson CP Jnr, Ehrenfeld WK, Wade JG, Wylie EJ. Jugular venous oxygen saturation as an index of adequacy of cerebral oxygenation. Surgery 1967; 62:31-39 22. Enderby GEH. Hypotensive Anaesthesia. In: Gray TC, Nunn JF, Utting JE, eds. General Anaesthesia 3rd end. London: Butterworths 1980:1149-1168 23. Kerr A. Anaesthesia with profound hypotension for middle ear surgery. Br J Anaesth 1977; 49:447-452
P. J. S i m p s o n
An understanding of the normal physiological mechanisms controlling arterial blood pressure is essential for the use of elective intraoperative hypotension. Mechanisms of blood pressure reduction include the mechanical effects of intermittent positive pressure ventilation, posture and the application of PEEP. Pharmacological methods involve interruption of the normal sympathetic control of vascular tone, either centrally, neurologically, by receptor inhibition, or by a direct effect on peripheral vascular diameter. General anaesthesia will also produce hypotension by a direct myocardial effect, which provides a good background against which a more specific technique can be employed. There is no single ideal hypotensive agent since the demands of surgery vary from the simple need to reduce blood loss to operations which would otherwise be impossible. Elective hypotension is perfectly safe in experienced hands and its use may often be specifically indicated during surgery for many reasons.
Introduction
excessively low pressures, for example during clipping of a cerebral aneurysm. Further indications include resection of aortic coarctation where rapid fluctuations in blood pressure necessitating immediate control are required. In general, there is a need for two basic methods of hypotension; elective, slow onset, slow recovery hypotension on the one hand and 'dial a pressure' hypotension on the other. Operative bleeding can be arterial, in which case it is directly related to the mean arterial pressure (MABP), capillary, when it is dependent upon local flow in the capillary bed, or venous, when it is related to venous return, venous tone and therefore posturally dependent. Although arterial bleeding can only be abolished by the use of a tourniquet, it can be considerably reduced by a fall in M A B P or heart rate (HR). Capillary flow is also reduced by elective hypotension and by localised vasoconstriction due, for example, to adrenaline infiltration or metabolic factors such as acid-base disturbance or hypoxia.
When induced hypotension is superimposed upon a general anaesthetic technique, no single agent is capable of providing the ideal conditions for all operations, the requirements of surgery falling into three broad groups. The first demands relatively slow onset and sustained moderate hypotension with a slow return to normal pressures, and is ideal for most plastic, maxillofacial and ear, nose and throat surgery when rapid return to normal pressures may cause reactionary haemorrhage. In the second, where massive blood loss is anticipated, moderate sustained hypotension together with a reduction in heart rate is probably all that is required. In the third group, some operations are not only impossible without profound hypotension but also require short periods of
P. J. Simpson, Department of Anaesthetics, Frenchay Hospital, Bristol BS16 1LE Current Anaesthesia and Critical Care
(~ 1992LongrnanGroupUK Ltd
(1992)3, 91)-97 90
HYPOTENSIVE ANAESTHESIA 91 Venous tone can be completely abolished by spinal or epidural anaesthesia and by direct acting vasodilators such as sodium nitroprusside.
Cardiovascular physiology Mean arterial blood pressure is directly related both to cardiac output (CO) and to peripheral resistance (TPR). While cardiac output is dependent upon myocardial contractility, determining stroke volume (SV), and heart rate, peripheral resistance is a function of vascular dilation. Peripheral vasodilation is primarily controlled by sympathetic activity, vasodilation being produced by interruption of sympathetic pathways rather than by any parasympathetic effect. Some cholinergic vasodilator fibres do exist, for example in the smooth muscle of the gut. Hypotension due to a reduction in total peripheral resistance can therefore be achieved either centrally by drugs such as volatile agents acting on the vasomotor centre or peripherally at the level of the sympathetic ganglia, postganglionic noradrenergic (alpha) terminals, or directly on the blood vessels themselves. In order to produce a measurable reduction in bleeding, local blood pressure must be in the region of 30-40 mmHg and venous drainage of the area must be unimpaired.
Posture Posture influences intraoperative bleeding both by producing regional ischaemia if the operation site is elevated above the level of the heart, and also by augmenting the effect of agents, such as sympathetic ganglion blocking drugs, by pooling of blood in the dilated venous vascular bed. The effects of head-up tilt on regional cerebral perfusion pressure in relation to mean arterial blood pressure at heart level are considerable. For each 2.5 cm of vertical height above the heart, blood pressure falls by 2 mmHg. The use of posture to improve the operating field and to reduce bleeding in head and neck surgery is well known. Whenever possible, posture should be used to augment pharmacological methods of induced hypotension, particularly those which depend upon venous pooling.
Intermittent positive pressure ventilation (IPPV) Under normal circumstances, venous return to the heart occurs during inspiration when a negative intrathoracic pressure enhances blood flow to the heart, even against the force of gravity. During IPPV inspiration is associated with positive intrathoracic pressure, inevitably resulting in a reduction in venous return. If such an effect is augmented by posture, the resultant fall in venous return and therefore cardiac output can be considerable. In normotensive anaesthetised subjects, IPPV has little effect on
cardiac output due primarily to the reflex vasoconstriction produced in response to what is effectively a limited Valsalva manoeuvre a. In addition, the baroreceptors respond to hypotension by inducing a reflex tachycardia. There are, however, two methods of inducing hypotension in which cardiac output may fall considerably in response to IPPV, due to temporary autonomic paralysis. Both ganglion blocking drugs and beta adrenoceptor antagonists may produce a partial or completely blocked Valsalva manoeuvre,2 thus limiting the production of a normal compensatory response. IPPV is a useful adjunct to any hypotensive technique, largely because it augments pharmacological methods of blood pressure reduction, limiting the dose necessary to produce the desired effect and also the postoperative duration of hypotension. In addition, artificial ventilation allows the application of positive end expiratory pressure (PEEP) to the airway, producing a further reduction in venous return and cardiac output.
Respiratory physiology Carbon dioxide (C02) is itself a vasodilator and hyperventilation leading to hypocapnoea will induce vasoconstriction. C02 control can be achieved either by IPPV using moderate hyperventilation or by C02 absorption in a circle system using soda lime. Care, however, must be taken during hyperventilation with patients in the head-up position, since vasoconstriction may reduce cerebral blood flow to critical levels.
Effects of hypotension upon pulmonary gas exchange Since pulmonary blood flow is dependent upon gravity, the head-up position produces a reduction in flow to the apical parts of the lung. Alveolar ventilation, however, will occur throughout the lung, including the upper segments, resulting in considerable ventilation/perfusion (V/Q) mis-match and an increase in physiological deadspace. This may be as great as 80% of the tidal volume3 and is of particular importance in spontaneously breathing patients. The reduction in alveolar ventilation and increase in physiological shunt is an additional reason for the use of IPPV during elective hypotension. Furthermore, many anaesthetists increase the inspired oxygen concentration to minimise the effects of this ventilation perfusion imbalance.
Pharmacological methods of vasodilation Volatile anaesthetic agents Halothane. Although halothane produces a moderate degree of vasodilation, the overall fall in total peripheral resistance is of the order of 15-18%. Vasodilation in the skin and splanchnic vascular beds is
92
CURRENTANAESTHESIA AND CRITICALCARE
balanced by vasoconstriction in skeletal muscle, any additional hypotension produced by halothane being a direct result of myocardial depression. In addition, bradycardia induced by the stimulant effect of halothane on the vagus nerve will further reduce cardiac output. While halothane is often used successfully in low concentration as a background to hypotensive anaesthesia, its use as a sole hypotensive agent in larger doses should be discouraged. This is of particular importance in neurosurgery if the increase in intracranial pressure due to vasodilation is to be avoided. Enflurane. The mechanisms and effects of hypotension induced by enflurane are similar to those of halothane. Myocardial depression and vagal stimulation are still significant factors if excessive doses of the drug are used and for this reason it should also only be employed in moderate doses simply as a background agent. Isoflurane. Unlike both halothane and enflurane, isoflurane has no effect on myocardial contractility. The peripheral vasodilatory effect is readily reversible by alterations in inspired concentration of the drug. For this reason, it is becoming increasingly used as a hypotensive agent, particularly when only moderate reduction in arterial pressure is required. It has the additional benefit that increasing doses not only produce vasodilation and hypotension but also produce central nervous depression minimising any reflex vasoconstriction or tachycardia, which may occur as a result of baroreceptor stimulation under relatively light anaesthesia. Isoflurane also appears to have less effect than either halothane or enfturane upon intracranial pressure in patients in whom normal values are present pre-operatively. 4
Sympathetic ganglionic blockade Trimetaphan, (Pentolinium). These drugs produce autonomic ganglion blockade by competitive inhibition of acetylcholine. Their effects are not confined to the sympathetic system since acetylcholine transmission also occurs in parasympathetic ganglia. Interruption of sympathetic outflow produces vasodilation which tends to be relatively slow in onset and recovery. The duration of hypotension produced by trimetaphan (Arfonad) is relatively short (10-15 min) and for this reason the drug is often administered by intravenous infusion (3-4mg.min-1). By contrast, a single injection of 5-15mg pentolinium (Ansolysen) produces hypotension for about 45 min and allows a slow return of blood pressure to normal values. Pentolinium is no longer available in the UK, but considerable demand may result in production being resumed. Although a number of gastrointestinal and urinary symptoms may result from concomitant parasympathetic blockade, the two of clinical importance during induced hypotension are mydriasis and tachycardia. The increase in heart rate which
often accompanies hypotension produced by ganglion blockade may severely impair the effectiveness of these drugs in reducing bleeding. Tachyphylaxis, that is the need for increasing doses of the drug to produce the same effect, is particularly marked with trimetaphan and may make a stable blood pressure difficult to achieve. Continuous infusion is considerably superior to intermittent bolus dose administration in this respect.
Non-depolarising neuromuscular blocking drugs Alcuronium, (d-Tubocurarine). The use of nondepolarising neuromuscular blocking drugs such as curare to facilitate IPPV as an adjunct to elective hypotension has been advocated for some time. D-tubocurarine, formerly the most widely used drug in this situation and also alcuronium (Alloferin) were initially thought to induce a degree of sympathetic ganglionic blockade due to a relative lack of specificity for the skeletal neuromuscular junction. More recently it has been shown that the main reason for their hypotensive effect is the histamine release associated with their administration which itself induces vasodilation. This far outweighs any effect due to mild ganglionic blockade.
Alpha adrenoceptor blockade Phentolamine, Phenoxybenzamine, Chloropromazine, Droperidol. Alpha adrenergic blocking agents produce vasodilation by competitive blockade of post-synaptic noradrenergic receptors within the sympathetic system. While the effects of phentolamine are relatively short (20-40 min) and easily reversible, those of phenoxybenzamine may last several days since this drug, a nitrogen mustard derivative, forms an irreversible receptor complex. Phentolamine also exerts a direct myocardial stimulant effect, increasing both oxygen consumption and heart rate. Phenoxybenzamine may produce considerable sedation. While phentolamine (5-10 mg) is used in the rapid production of intraoperative vasodilation, phenoxybenzamine (0.5-2.0 mg.kg -1 for 10 days) is more commonly employed for chronic vascular expansion prior to surgery to minimise the effects of circulating catecholamines, e.g. in the surgical removal of phaeochromocytoma. Both chlorpromazine and droperidol produce mild alpha adrenergic blockade which is often useful in the preoperative preparation of patients prior to hypotensive anaesthesia and/or hypothermia.
Beta adrenoceptor blockade Propranolol, Oxprenalol, Atenolol, Labetalol, (Practolol). The main advantages of the beta adrenoceptor antagonists in induced hypotension are in the reduction of heart rate and cardiac output. Many anaesthetists believe that the maintenance of a slow heart rate
HYPOTENSIVE ANAESTHESIA 93 without any additional hypotension considerably reduces operative bleeding. Propranolol is often been used to produce this 'rheostatic' hyptension. Although preoperative oral therapy (40mg t.d.s.) is probably best, 1-2mg intravenously can be used during anaesthesia. Beta adrenoceptor blockade with either propranolol or oxprenalol is also employed either preoperatively or intra-operatively to counteract the tachycardia produced as a side effect of induced hypotension with either ganglion blocking or direct acting vasodilator drugs. Again it is best to administer the drugs orally rather than intravenously since this produces a steady intraoperative blood level of the drug. Although the combined alpha and beta adrenoceptor blocking drug labetalol would seem ideal for use in induced hypotension, it is important to realise that the alpha blocking effects of the drug only last for 30 min compared with a 90 min duration of beta blockade. In addition, the beta blocking effects are five to seven times as potent as the alpha blockade. The perioperative use of beta blockade with either propranolol or labetalol may have considerable benefit in the prevention of wide fluctuations in blood pressure particularly in patients with subarachnoid haemorrhage and vasospasm.
Direct acting vasodilators Sodium nitroprusside. The main advantage of this drug is its extremely evanescent action, allowing rapid reduction of blood pressure and equally rapid restoration to normal levels. It is the only drug capable of predictably producing 'dial-a-pressure' hypotension over relatively short periods, e.g. in the prevention of bleeding in meningiomas and major vascular surgery, or to facilitate clipping of cerebral aneurysms. As a vasodilator SNP inevitably produces an increase in intracranial pressure and for this reason should not be used during neurosurgery before the skull is open in a patient with raised intracranial pressure. Nevertheless autoregulation under induced hypotension with nitroprusside is maintained at cerebral perfusion pressures considerably lower than with other drugs. 5
Metabolism and toxicity of sodium nitroprusside. Shortly after the introduction of nitroprusside into clinical practice reports of fatalities during its use were directly attributed to cyanide poisoning. Each molecule of sodium nitroprusside contains 5 cyanide radicals which are lib~erated on breakdown of the drug in either plasma or red blood cells. Several studies, however, have failed to demonstrate significant effects upon' red cell oxygen transport of cyanide liberated during routine clinical use of
nitroprusside.6'7
The normal metabolic pathway of SNP breakdown is nonenzymatic, occurring in both red cells and plasma. The intracellular reaction is catalysed by the conversion of haemoglobin to methaemoglobin. Ultimately, more than 98% of the cyanide produced from
SNP is contained within the red blood cells while a small proportion is combined with either methaemoglobin or Vitamin B12. The majority of cyanide is metabolised in the liver by the enzyme rhodanese to thiocyanate which is then excreted in the urine. The rate limiting factor in cyanide metabolism appears to be the availability of sulphydryl groups and the administration of sodium thiosulphate can considerably enhance thiocyanate production and therefore reduce blood cyanide levels.8 The use of thiosulphate does not appear to affect the hypotension produced by nitroprusside. At the maximum safe doses recommended for nitroprusside administration, 1.5 mg/ kg -19 or 10 txg.kg -I min-1, ~° small increases in plasma lactate occur which are mirrored by increases in arterial base deficit. These changes are only minor, the maximum base deficit being of the order - 6 to - 7 mmols per litre, and are spontaneously reversible upon discontinuation of nitroprusside therapy. The routine measurement of acid-base balance during nitroprusside therapy would appear to provide adequate clinical information of the development of cyanide toxicity during routine clinical use. The use of sodium nitroprusside in patients already anaesthetised with a background hypotensive anaesthetic technique would still appear to be the method of choice for the production of extreme hypotension for neurosurgery. No other drug at present provides the predictable and rapid hypotensive effect necessary for many aspects of intracranial surgery. Natural apprehension over the potential toxicity of SNP has largely centred round a few reported cases all of which were directly attributable to cyanide poisoning. H'12'13 Close examination of these reports confirms that in all cases doses of SNP vastly in excess of those required for routine clinical use were needed to produce toxic symptoms. If the dose of nitroprusside is limited to that already stated above, toxic symptoms will not occur in patients with normal renal and hepatic function. For longer term infusion and in the presence of adequate sulphydryl groups as the substrate for cyanide detoxification by rhodanese, a maximum dose rate of 8 ixg.kg-114 has been shown to be satisfactory. Trinitroglycerine (TNG). Hepatic metabolism of trinitrate produces di- and mono-nitrate and finally glycerol.14 The vasodilator activity of these smaller nitrate molecules is reduced as their size decreases. Nitroglycerine produces a steadier and less dramatic reduction in arterial blood pressure, having a greater effect on systolic than diastolic pressure and tending to maintain the blood flow. Recovery from nitroglycerine-induced hypotension is also less rapid, taking between 10 and 20 rain in contrast to the 2-4 min with SNP. It has been suggested that this slower effect of nitroglycerine produces less overshoot of blood pressure either at induction of hypotension or following restoration of normal blood pressure but as the drug appears less effective in some cases in the production of extreme hypotension, its use may not
94
CURRENT ANAESTHESIA AND CRITICAL CARE
be ideal in all situations. Although TNG has been advocated as a direct acting vasodilator for neurosurgery, in one study,~5 hypotension to 50 mmHg was not possible in three out of 22 patients. Unlike sodium nitroprusside which dilates both resistance and capacitance vessels equally, nitroglycerine exerts its effect principally upon the venous capacitance system. As a result, diastolic blood pressure is maintained at higher levels than with nitroprusside and for this reason TNG maintains coronary artery perfusion more effectively than SNP. While this is probably of little importance in fit patients, it may be of considerable advantage in patients with impaired myocardial or cerebral circulation. However, the increase in intracranial pressure produced by nitroglycerine may be even greater than with SNP.
Spinal and epidural anaesthesia The sympathetic blockade produced by spinal and epidural anaesthesia is a very effective way of inducing hypotension. Lumbar epidural anaesthesia will produce arteriolar dilation and hypotension together with a reduction in venous tone. This is enhanced by posturally dependent pooling of venous blood leading to a fall in venous return and therefore cardiac output. If the block is extended to the mid-thoracic region, the cardiac sympathetic fibres passing in segments T1 to T4 will also be blocked, preventing the compensatory tachycardia which will otherwise occur. This also limits any baroreceptor response, and prevents tachycardia occurring as a result of other pharmacological methods of induced hypotension. Regional anaesthesia is most commonly employed in lower abdominal or pelvic surgery to minimise blood loss, particularly that occurring from the pelvic venous plexuses. The complete abolition of venous tone is extremely effective in minimising blood loss without the need for profound arteriolar hypotension. If adrenaline is added to the local anaesthetics used, its systemic absorption may partially counteract the hypotensive effect of the regional block. Adrenaline has the advantage of prolonging the effects of the local anaesthetics used, although this can be satisfactorily achieved more simply by the use of intermittent or continuous epidural injections via a catheter.
Monitoring during induced hypotension Routine monitoring should include ECG, oxygen saturation, end-tidal carbon dioxide concentration and either direct or indirect arterial blood pressure measurement. Although instruments such as the cerebral function monitor are potentially available for use during hypotensive anaesthesia, their use and, indeed, their possible benefits are relatively limited. 16
Electrocardiographic monitoring ECG monitoring is essential to demonstrate two vital signs of inadequate myocardial perfusion, the development of ectopic beats and ST segment depression. Although a single channel ECG is incapable of demonstrating the exact site of any ischaemia, ST segment changes do occur and are usually readily reversible by increasing the blood pressure. The myocardial response to relative hypoxia and hypoperfusion is a sensitive monitor of hypotension being excessively exploited.
Measurement of blood pressure Indirect m e a s u r e m e n t
Hutton and Prys-Roberts 17 have shown that errors in interpretation of oscillotonemetry have tended to err on the side of safety since what was originally thought to be systolic blood pressure has now been demonstrated to more closely represent the mean pressure. The semi-automated versions (Dinamap) are an improvement since they remove the observer bias and indeed display not only systolic but also mean and diastolic pressures. Direct m e a s u r e m e n t
Although indirect methods of blood pressure measurement are used during relatively mild hypotensive techniques, such as intermittent ventilation with isoflurane, epidural anaesthesia, or even in some cases, the use of ganglion blocking drugs, direct monitoring is an essential part of the technique when rapidly acting direct vasodilators such as sodium nitroprusside are employed. The position of the transducer is the level at which blood pressure is measured and if the mean arterial blood pressure at head level is desired, the transducer should be placed at this height. Assuming a normal intracranial pressure and with the appropriate postural adjustments, the measurement is then directly related to cerebral perfusion pressure.
Oxygen saturation and end-tidal carbon dioxide monitoring Induced systemic hypotension will inevitably result in a reduction in pulmonary artery pressure, and since, pulmonary perfusion is gravity dependent, an increase in V/Q mismatch within the lung. The ability to non-invasively monitor respiratory gas exchange allows alterations in inspired oxygen and minute volume to be made in specific relation to the patient's needs, rather than on an empirical basis. Inspired oxygen may need to be increased during a period of hypotension to compensate for changes in lung perfusion. In addition end-tidal CO2 analysis is a sensitive method of detecting air embolism, which
HYPOTENSIVE ANAESTHESIA 95 may also occur during some cranial or head and neck operations.
Practical technique of induced hypotension In all cases of elective hypotension, a background anaesthetic against which hypotension can be induced is essential, the principles of balanced anaesthesia dictating that it is better to employ individual agents to achieve specific effects rather than to pursue the toxic properties of an agent like halothane in the production of hypotension by myocardial depression. An ideal background anaesthetic consists of omission of atropine premedication, since this induces tachycardia, but the use of generous sedation or analgesia. It is essential to avoid preoperative anxiety and the release of adrenaline, since the effects take some time to abate under anaesthesia. Induction of anaesthesia with thiopentone, fentanyl and a long-acting, non-depolarising neuromuscular blocking drug, for example alcuronium, should be employed. Following topical anaesthesia of the larynx, intubation is performed with a non-kinking endotracheal tube to remove any possible risk of partial airway obstruction and CO2 retention. Moderate hyperventilation with nitrous oxide and oxygen together with 0.5-1.0% isoflurane is then the background technique against which hypotensive agents can be used. Under these stable conditions, specific hypotensive drugs can then be employed with the minimum of side effects, for example tachycardia or excessive hypotension. During hypotension, as mentioned earlier, the inspired oxygen concentration may be increased to 40 or even 50% if excessively low pressures are being employed over a short period, for example during neurosurgery.
Postoperative management Recovery staff must be made aware of patients in whom elective hypotension has been employed. Accurate and regularly blood pressure monitoring together with meticulous airway care to avoid CO2 retention and partial obstruction is essential. The patient's position should be determined by the blood pressure measured, and postural changes may be necessary for several hours to ensure adequate cerebral perfusion. Supplementary oxygen should be administered in all cases until the patient is adequately awake and may be required for longer where oxygenation is thought to be critical. In cases where pharmacological modification of sympathetic responses has been undertaken, such as with the use of ganglion blocking drugs, patients should remain in bed for 12-18h postoperatively and, if necessary, lying virtually flat until they are able to sit up without feeling dizzy.
Relative contra-indications to induced hypotension Although many anaesthetists are reluctant to employ induced hypotension, there are very few patients in whom it cannot be used safely. Most would refrain from utilising the technique in patients with evidence of severe cardiovascular or cerebro-vascular disease, although both these are relative contraindications if, for example, cerebral aneurysm surgery is proposed.
Myocardial ischaemia This is made worse by an increase in the rate pressure product or a reduction in myocardial oxygen delivery. Since hypotensive techniques are designed to reduce both heart rate and arterial blood pressure, the amount of cardiac work is reduced considerably. Many such patients are already on beta-adrenoceptor blockade and this should be continued intraoperatively. In cases where a reduction in afterload is produced by direct acting vasodilators such as sodium nitroprusside, myocardial work may be further reduced, but an adequate perfusion pressure must be maintained.
Hypertension Although patients with treated hypertension may be abnormally sensitive to hypotensive drugs, such techniques can still be employed with care. Untreated hypertension, however, is a contraindication, since the blood pressure may be extremely labile and profound hypotension results. It is also important to remember that volatile anaesthetic agents enhance the hypotensive effects of drugs which the patient is already receiving for routine control. Monitoring the ECG is essential in patients with cardiovascular disease. Rollason and Hough is demonstrated ST depression in hypertensive patients and Simpson et a119 noted similar changes during anaesthesia and nitroprusside induced hypotension. While the importance of such changes is probably doubtful, they do demonstrate the need for care when utilising hypotension in such patients.
Respiratory disease Contraindications to the use of hypotensive anaesthesia in patients with chronic respiratory disease are related to the disturbance of normal pulmonary physiology. The increase in physiological dead space due to ventilation/perfusion imbalance is more important in patients in whom preoperative gas exchange is limited. Under normal circumstances, hypoxic pulmonary vasoconstriction occurring in poorly ventilated segments of the lung prevents gross disorders of ventilation and perfusion. Vasodilation induced by direct acting drugs such as sodium
96
CURRENT ANAESTHESIA AND CRITICAL CARE
nitroprusside, abolishes this response and will therefore make shunting worse. Reversible airways obstruction and bronchospasm may be made worse by the use of either ganglion blocking drugs or beta adrenoceptor antagonists which are not cardiospecific and such drugs are contraindicated in asthmatics. High inflation pressures may lead to CO2 retention and severe hypotension due to impairment of venous return.
Diabetes mellitus This relative contraindication to induced hypotension concerns the drugs used rather than the actual induction of arterial hypotension. Ganglion blocking drugs, by producing sympathetic blockade, impair stress induced gluconeogenesis mediated by adrenaline. Beta blockers may also potentiate hypoglycaemia in insulin dependent diabetics and it is the combination of hypoglycaemia plus hypotension which may produce severe consequences, particularly on cerebral metabolism. Under normal circumstances, however, it is safe to employ volatile agents, direct acting vasodilators or local anaesthetic techniques without impairment of blood sugar levels.
Complications of hypotensive anaesthesia Failed hypotension Although not regarded by many as a complication of hypotensive anaesthsia, failure to appreciate this may lead to the excessive use of certain agents and consequent toxic effects. Some patients, for example, with a sensitive renin-angiotensin system have been shown to be 'resistant' to sodium nitroprusside. Before this was appreciated, excessive doses of the drug were employed in the mistaken belief that eventually hypotension would be achieved. This led in some cases to sodium nitroprusside toxicity, cyanide poisoning, and even death. If hypotension using one drug is not sufficient, then a second agent acting at a different site in the sympathetic system should be employed. This has been put to good effect, for example by combinations of ganglion blocking drugs such as trimetaphan with sodium nitroprusside or, more recently, the combination of isoflurane and nitroprusside.
Excessive hypotension Moderate hypotension is achieved during many anaesthetics and many patients are subjected to systolic arterial pressures of 60 mmHg during routine surgery. It is unlikely that hypotension to this degree, particularly when induced by vasodilation and accompanied by additional oxygen administration in fit patients, would lead to any permanent adverse effect. Nevertheless, situations have arisen where
permanent damage has resulted from the use of elective hypotension. This is related either to excessive hypotension, impaired oxygenation or inadequate blood pressure monitoring. Lindop, 2° surveying the evidence concerning the effects of hypotension on vital organs, stressed the importance of flow rather than blood pressure in the development of complications. This is particularly important since normal oxygen extraction by most organs with the exception of the heart is only about 25% of its potential. In considering the effects of hypotension on the brain, autoregulation becomes important. It has been shown that the lower limit of autoregulation is better preserved with nitroprusside induced hypotension than with trimetaphan, flow remaining constant down to mean pressures of 4 0 - 5 0 mmHg. 5 These figures need to be interpreted with some caution when extrapolated to humans because of the effects of anaesthesia and other metabolic derangements upon cerebral blood flow. More detailed studies, such as continuous E E G monitoring 16 and jugular venous oxygen measurements, 21 have produced inconclusive results. Probably the most important rule is not to reduce the intraoperative systolic pressure to below the preoperative diastolic, and to avoid where possible severe head-up tilt unless blood pressure is being measured at the head level. The monitoring of ST segment depression to detect myocardial ischaemia would appear to be the most reliable method of demonstrating adverse cardiac effects due to hypotension. ~9 Provided myocardial work is reduced and a compensatory tachycardia due to hypotension does not occur, severe problems appear to be very rare.
Outcome Although those antagonistic to hypotensive anaesthesia will always cite cases of permanent damage or even death related directly to the lowering of blood pressure, large series, particularly that collected by Enderby 22 do not bear this out. At East Grinstead, the mortality rate was 1 in 4128 cases of induced hypotension and Kerr 23 in a separate series reported no mortality or morbidity in a series of 700 patients.
References 1. Prys-Roberts C, Kelman GR, Greenbaum R, Robinson RH. Circulatory influencesof artificial ventilation during nitrous oxide anaesthesia in man. II. Results: the relative influence of mean intrathoracic pressure and arterial carbon dioxide tension. Br J Anaesth 1967; 39:533-548 2. Blackburn JP, Conway CM, Davies RM, Enderby GEH, Eldridge A, Leigh JM, Lindop MJ, Strickland DAP. Valsalva responses, and systolictime intervals during anaesthesia and induced hypotension. Br J Anaesth 1973;45: 704-710 3. Eckenhoff JE, Enderby GEH, Larson A, Eldridge A, Judevine DE. Pulmonary gas exchange during deliberate hypotension. Br J Anaesth 1963; 35:750-758
HYPOTENSIVE ANAESTHESIA 4. Murphy FL Jnr, Kennell EM, Johnstone RE et al. The effects of enflurane, isoflurane and halothane on cerebral blood flow and metabolism in man. Abstracts of Scientific Papers. Annual Meeting of the American Society of Anesthesiologists 1974; 61-62 5. Stoyka WW, Schutz H. The cerebral response to sodium nitroprusside and trimetaphan controlled hypotension. Canadian Anaesthetists Society Journal 1975; 22:275-282 6. du Cailar J, Mathieu-Dande JC, Duschade J, Lamarche Y, Castel J. Nitroprusside, its metabolites and red cell function. Canadian Anaesthetists Society Journal 1978; 25:92-105 7. Vesey C J, Krapez JR, Cole PV. The effects of sodium nitroprusside and cyanide on haemoglobin function. J Pharm Pharmacol 1980; 32:256-261 8. Krapez JR, Vesey CJ, Adams L, Cole PV. Effects of cyanide antidotes used with sodium nitroprusside infusion: sodium thiosulphate and hydroxycobalamin given prophylactically to dogs. Br J Anaesth 1981; 53:793-804 9. Vesey CJ, Cole PV, Simpson PJ. Sodium nitroprusside in anaesthesia. BMJ 1975; 3:229 10. Tinker JH, Michenfelder JD. Sodium nitroprusside; pharmacology, toxicology and therapeutics. Anesthesiology 1976; 45:340-354 11. Davies DW, Kadar D, Steward D J, Munroe IR. A sudden death associated with the use of sodium nitroprusside for the induction of hypotension during anaesthesia. Canadian Anaesthetists Society Journal 1975; 22:547-552 12. Jack R. The toxicity of sodium nitroprusside. Br J Anaesth 1974; 46:952
97
13. Merrifield A, BlundeU M. Toxicity of sodium nitroprusside. Br J Anaesth 1974; 46:324 14. Michenfelder JD, Tinker JH. Cyanide toxicity and thiosulphate protection during chronic administration of sodium nitroprusside in the dog. Anesthesiology 1977; 47: 441-448 15. Chestnut JS, Albin MS, Gonzalez-Abola E, Newfield P, Maroon JC. Clinical evaluation of intravenous nitroglycerin for neurosurgery. J Neurosurg 1978; 48:704-711 16. Patel H. Experience with the cerebral function monitor during deliberate hypotension. Br J Anaesth 1981; 53: 639645 17. Hutton P, Prys-Roberts C. The oscillotonometer in theory and practice. Br J Anaesth 1982; 54:581-593 18. Rollason WN, Hough JM. A study of hypotensive anaesthesia in the elderly. Br J Anaesth 1960; 32:276-285 19. Simpson PJ, Bellamy D, Cole PV. Electrocardiographic studies during hypotensive anaesthesia using sodium nitroprusside. Anaesthesia 1976; 31:1172-1178 20. Lindop MJ. Complications and morbidity of controlled hypotension. Br J Anaesth 1975; 47:799-803 21. Larson CP Jnr, Ehrenfeld WK, Wade JG, Wylie EJ. Jugular venous oxygen saturation as an index of adequacy of cerebral oxygenation. Surgery 1967; 62:31-39 22. Enderby GEH. Hypotensive Anaesthesia. In: Gray TC, Nunn JF, Utting JE, eds. General Anaesthesia 3rd end. London: Butterworths 1980:1149-1168 23. Kerr A. Anaesthesia with profound hypotension for middle ear surgery. Br J Anaesth 1977; 49:447-452