154

Review Article

Bruno Bissonnette MD FRCPC, Neuroleptanesthesia: Hilton Swan MBBS FANZCA, Patrick Ravussin MD,* current status Victor Un MD FRCPC

Purpose: To review the current status and possible future of neuroleptanalgesia/, techniques that may be nearly extinct. Souzce: Articles from 1966 to present were obtained from the Current Science and Medline databases. Search terms include neurolepananalgesia/anesthesia, conscious sedation, droperidol, benzodiazepines, propofol, keta- mine, and opioids. Information and abstracts obtained from meetings on this topic helped complete the collec- tion of information. Principal findings: Droperidol/fentanyl may still be clinically indicated in the management of surgical seizure therapy for electrocorticography. However, the high incidence of post-operative sedation and restlessness dis- courage its use for other surgical or diagnostic procedures. Many surgical interventions, once thought ideally suit- ed for neuroleptic agents, now meet better success with newer medications. The use of and/or propofol, in association with newer opioids, provides ideal anesthetic combinations. Conclusion: The advantages of newer anesthetic agents have redefined the clinical indications for neurolep- tanesthesia. In routine modem anesthesia, anxiolysis, sedation, and/or analgesia is better provided, with quicker recovery, by the new pharmacokinetic and pharmacodynamic characteristics of recent medications than by the neuroleptic component of neuroleptanesthesia.

Objectif : Fake une revue de I'&at actuel et de I'Evolution possible de la neuroleptanalgEsie qui semble main- tenant presque abandonnEe. Souree..s : Des articles de 1966 A aujourd'hui ont ~t6 obtenus A partir d'une consultation de Current Science et de Medline. Les mots-clEs comprenaient : neuroleptanalg&ie, sedation du patient EveillE, dropEridol, benzodi- azEpines, propofol, k&amine et opiac&. Les informations et les rEsumEs provenant de sEminaires sur le sujet ont permis de completer la cueillette de donnEes. Constatadons principales : La combinaison de drop&idol et d'alfentanil peut &re indiquEe pour I'Electrocor- ticographie utilisEe dans le traitement chirurgical de I'Epilepsie. Cependant, I'importante incidence de sedation et d'agitation postop&atoires d&ourage son utilisation pour d'autres interventions chirurgicales ou diagnostiques. Nombre d'interventions chirurgicales o6 on a cru que les neuroleptiques &aient les agents idEaux sont maintenant mieux rEussies avec de nouveaux medicaments. I'emploi de midazolam et/ou de propofol, associ& aux nou- veaux opiac&, fournit les meilleures combinaisons. Conclusion : Les avantages des nouveaux anesthEsiques ont amenE A red~finir les indications cliniques de la neu- roleptanalg&ie. Dans la pratique de I'anesthEsie moderne, les medicaments r&ents assurent mieux la reduction de I'anxiEtE, la sedation et/ou I'analgEsie et permettent une r6cup&ation plus rapide que les composes utilis& en neuroleptanalg&ie, grace ~ leurs nouvelles caract&istiques pharmacocin&iques et pharmacodynamiques.

From the Department of Anaesthesia, The Hospital for Sick Children, and University of Toronto, Toronto, Ontario, Canada and the Service d'Anesthesiologie,* H6pital Rrgional de Sion Herens Conthey, Switzerland. Address correspondence to: Dr. Bruno Bissonnette, Department of Anaesthesia, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1X8 Canada. Phone: 416-813-7445; Fax: 416-813-7543; E-mail: [email protected] Accepted for publication November 14, 1998

CAN J ANESTH 1999 / 46:2 / pp 154-168 Bissonnette et al.: NEUKOLEPTANESTHESIA 155

Conte~ts metabolic, and autonomic responses to nociceptive Introduction stimulation. They compared the anesthetic state to History that of certain selected animal species during hiberna- A The Neuroleptic Component tion. This new anesthetic technique, called ganglio- DROPERIDOL plegia or neuroplegia, used a lyric cocktail of three ALTERNATIVE DRUGS components: meperidine, promethazine, and chlor- Midazolam promazine. 2 It remains in use today in some pediatric Flumazenil centres, referred to as DPT (for their brand names, Ketamine: Dissociative Anesthesia Demerol, Phenergan, and Thorazine). Propofol The work of Campan and Lazothes I was expanded B The Opioid Component further by de Castro and Mundaleer 3 in Belgium. FENTANYL Using haloperidol (butyropherone) as a major neu- ALTERNATIVE DRUGS roleptic and phenoperidine (an opioid), they pro- Alfentanil duced a state of indifference and immobilization Sufentanil (often referred to by psychiatrists as mineralization) Remifentanil that they called neuroleptanalgesia. It was restricted to Conclusion patients who, under the combined influence of a neu- roleptic and an opioid, become analgesic, sedated, Abbreviations mineralized, and amnestic, maintaining autonomic, CBF: Cerebral blood flow neurological, and cardiovascular stability while able to CMRO2: Cerebral metabolic rate for oxygen comprehend and obey simple commands during CNS: Central nervous system surgery. 2 When the patient was rendered unconscious CPP: Cerebral perfusion pressure with the addition of N20 or a hypnotic drug, the term CSF: Cerebrospinal fluid neuroleptanesthesia was used. EEG: Electroencephalography When droperidol and fentanyl citrate, both intro- ICP: Intracranial pressure duced by Petr Janssen, 4,s became available, they replaced MAP: Mean arterial pressure haloperidol and phenoperidine as the most widely used PaCO2: Partial pressure of carbon dioxide compounds for inducing and maintaining neuroleptanal- PaO2: Partial pressure of oxygen gesia. Anesthesia was induced with Innovar, each millil- itre of which contained 2.5 mg of droperidol and 50 lag of fentanyl citrate, and supplemented with N20 and 02 T HE paucity of recent literature on neu- to attain neuroleptanesthesia. The drug combination was roleptanesthesia, once a topic of consider- titrated into a fast-running intravenous (iv) infusion until able interest in anesthesia, attests to its the patient was sedated and demonstrated some difficul- current uncertain clinical status. Has this ty in phonation from respiratory depression. Intubation technique, which defined a new anesthetic frontier 40 under direct laryngoscopy after topical anesthesia was yr ago, gone the way of ether and cyclopropane, to then possible, with or without muscle relaxants. become a subject only of historical interest? Re-evalu- However, this technique was not without problems, ation of its virtues and reassessment of its relevance to which included unconsciousness to the unrousable level, current practice is needed. This article, which discuss- venfilatory difficulties from muscular rigidity, and post- es the virtues and future of neuroleptanesthesia, is operative extrapyramidal excitation. intended to help clinicians understand and reassess These problems were caused by a lack of investiga- indications for a technique that may be nearly extinct. tion and understanding of the pharmacological effects of the individual components. 6 The tkxed-ratio mLxture History of Innovar, with the slow onset-slow offset droperidol The first reference to a neuroleptic anesthetic tech- and fast onset-fast offset fentanyl, was the obvious rea- nique was presented in 1954 by Campan and son for overdose and complications. The relatively slow Lazothes I in France. This new anesthetic technique onset time for droperidol (6-8 min) meant more fen- was designed to modify the effects of conventional tanyl was initially required for achieving sedation, general anesthesia in eliminating the perception of resulting in a relative overdose with development of nociceptive stimuli at the level of the cerebral cortex muscle rigidity and/or apnea. If not, as droperidol without affecting cognitive function. Its specific char- started to act, excessive sedation and respiratory depres- acteristics included modulation of certain endocrine, sion became evident; the relative overdose ofdroperidol 156 CANADIAN JOURNAL OF ANESTHESIA also caused extrapyramidal excitation. These complica- the cerebral vessels, the cerebral metabolic rate for tions could be prevented by separate, judicious admin- oxygen (CMRO2) remains unchanged. This may lead istration of each drug. 6,7 to a potential discrepancy between the metabolic sup- In the past 20 yr, numerous new anesthetic agents ply/demand ratio, which could become important in have become available, such as short-acting sedatives patients with impaired CBF from cerebrovascular ath- and hypnotics (midazolam and propofol) and the newer erosclerosis or surgical intervention, or with increased opioids (alfentanil, sufentanil, and remifentanil). To CMRO 2 from seizure activity. One of the advantages review the current status and list of indications and of droperidol is its lack of EEG effects,9 allowing reli- techniques for neuroleptanalgesia/anesthesia, we will able intra-operative EEG . It is not an anti- compare the pharmacology of droperidol and fentanyl convulsive agent; l~ in fact, droperidol can lower the with that of the newer alternative drugs. seizure threshold n and should be used with caution in Before proceeding, it seems appropriate to clarify patients with untreated epilepsy. This intrinsic effect some terms. Neuroleptanalgesia and neuroleptanes- might be an advantage during intra-operative record- thesia refer to the omission or addition, respectively, ing of seizure foci for intractable epilepsy; however, of N20 to the combination of a neuroleptic and an this association has yet to be supported. opioid. In recent years, the term conscious sedation Respiratory effectr. Clinical doses of droperidol (5 has appeared. It describes the administration of a seda- mg) have been reported to cause a reduction in tidal tive or hypnotic drug either alone (as in pediatric seda- volume (13.3%)) 2 minute ventilation (8.4%), 12 airway tion for radiological procedures or as an adjunct to resistance (50%), is and functional residual capacity local/regional anesthetic techniques) or in conjunc- (25%), whereas larger doses (0.3 mg.kg -1) in healthy tion with an analgesic agent for mild to moderately volunteers did not modify the respiratory drive to painful procedures. It constitutes a state of minimally CO2 .14 Partial pressures of carbon dioxide (PaCO2) depressed level of consciousness without affecting the measured before and after administration of droperi- ability of the patient to maintain airway reflexes and dol remained the same. Innovar (a proprietary appropriate responses to physical stimulation or verbal droperidol/fentanyl combination) has been shown to command. Despite the unambiguous nature of these cause a reduction in functional residual capacity phrases, frequently the terms neuroleptanesthesia and (FRC) that was reversible with succinylcholine, sug- neuroleptanalgesia are erroneously used as synonyms gesting increased expiratory muscle activity; 12 howev- to describe conscious sedation. er, the individual effects of droperidol and fentanyl were not compared in the study. It has been suggest- A THE NEUROLEPTIC COMPONENT ed is that the anti-dopaminergic action of droperidol All neuroleptic drugs used in clinical practice are ter- on the carotid body could reduce the depressant effect tiary aromatic amines based on methyl-ethylamine. of hypoxemia on the respiratory centre. Various substitutions around this basic structure pro- Cardiovascular effectr. Although droperidol does duce a series of drugs with a wide spectrum of neu- not affect myocardial contractility or heart rate, it does roleptic activities. Two of the main categories are the decrease systemic blood pressure--possibly because of butyrophenones and the phenothiazines. Drugs such as its peripheral a-adrenergic blockade and central anti- droperidol (dehydrobenzperidol) and haloperidol dopaminergic activity. 16 The decrease is usually mod- belong to the butyrophenones, whereas chlorpro- est; in the presence of hypovolemia, however, severe mazine and prochlorperazine are phenothiazines. These arterial hypotension could result. In hypovolemic drugs are effective in alleviating the anxiety accompany- patients, it is therefore prudent to administer droperi- ing psychotic disorders. Butyrophenones act as dol with extreme caution, to avoid changes in cerebral allosteric inhibitors at post-synaptic receptor sites to perfusion pressure (CPP) and intracranial pressure decrease the neurotransmitter activity of dopamine, s (ICP), which could cause cerebral ischemia. The antipsychotic effects of these agents have been Conversely, droperidol has been shown to cause mostly attributed to their antagonism of the post- hypertension in patients with pheochromocytoma, 17 synaptic dopaminergic receptors of the central nervous possibly as a result of the efflux of catecholamines system (CNS); however, other neurotransmitter sys- from adrenal medullary cells and/or the inhibition of tems could be involved. catecholamine reuptake into neuronal chromaffin granules, resulting in increased systemic cate- DROPERIDOL cholamine levels, is Neurological effectr. Although droperidol reduces Droperidol has cardiac antidysrhythmic proper- cerebral blood flow (CBF) 40% by vasoconstricting ties. 19 At a dose of 0.2 mg.kg q, it doubles the arrhyth- Bissonnette et al.: NEUROLEPTANESTHESIA 157 mic threshold to infusions of epinephrine. In animal Benzodiazepines were introduced in the 1960s and studies, Bertolo and associates2~ demonstrated that have assumed an increasing role as anxiolytics and droperidol delayed the onset of ventricular fibrillation sedative hypnotics, replacing barbiturates in many induced by coronary occlusion. Without g-blocking clinical applications. Because of their safety, as demon- activities, possible explanations for its antidysrhythmic strated by a high therapeutic index (50% lethal dose + activity include an antagonistic effect on the myocar- 50% effective dose) and a relatively benign adverse- dial a-adrenergic receptors combined with a local effect profile, they have achieved wide acceptance anesthetic-like stabilization of myocardial membranes. among clinicians. They have been the main class of Droperidol may be contraindicated in patients with drugs responsible for the decline in use of neuroleptic pre-existing myocardial conduction defects. In doses agents in anesthesia practice. ranging from 0.1 to 0.25 mg.kg -~, it has been shown In 1977, the discovery of the benzodiazepine to prolong the QT interval in surgical patients. 2~ receptor improved the understanding of their mecha- Antiemesir. Droperidol has potent antiemetic prop- nism of action and led to the development of more erties; it is commonly used in small doses to prevent versatile drugs, such as the short-acting agonist mida- perioperative nausea and vomiting in anesthetic prac- zolam and the specific benzodiazepine antagonist, tice. The antiemetic effect of droperidol may be due to flumazenil. Since then, the molecular pharmacology inhibition of dopaminergic (notably D2) receptors in of the benzodiazepine receptors and y-aminobutyric the chemoreceptor trigger zone (the area postrema) in acid (GABA) receptors has been further elucidated the floor of the fourth ventricle. However, vomiting and characterized. 2s In short, it was proposed that the induced by labyrinthine instability (motion sickness) is benzodiazepine receptor (with two subtypes) was dif- not prevented or ameliorated by droperidol. The inci- ferent from the GABA receptor (with types A and B dence of post-operative nausea and vomiting can be subunits), but that these two receptors are one macro- decreased by intra-operative administration of 2540 molecular complex associated with the chloride mem- pg.kg q droperidol in obstetric cases. 21 Although a brane channel. Thus, stimulation of the receptor by an larger dose (75 pg.kg -1) was suggested as effective in agonist would, by linkage through the GABA^ recep- children undergoing strabismus surgery, 21% of this tor-chloride channel complex, induce an influx of surgical population vomited after discharge from the chloride anions into the neurones, causing hyperpo- hospital. 22 At this dose, there was a high incidence of larization and inhibiting nerve transmission. Not all sedation and restlessness (63%). zs GABA^ receptors, however, are coupled to benzodi- ALTERNATIVE DRUGS azepine receptor sites; it is possible that the hetero- Many surgical procedures once thought to be ideal- geneity of the four subunits (~, 1~, n, A) have ly suited for the administration of a neuroleptic tech- functional differences and may represent a group of nique have, over the past two decades, been better different receptors rather than a single entity. 2s performed with non-neuroleptic drugs and techniques. Alone, benzodiazepines lack GABA-mimetic Several new classes of anesthetics have all but replaced effects; however, they have been shown to enhance the neuroleptic drugs for procedures requiting con- the response. This speculation is based on the sugges- scious sedation, total intravenous anesthesia (TIVA), or tion that a GABAa receptor, a benzodiazepine recep- balanced general anesthesia. The ideal drug to produce tor, and the chloride ionophore form one anxiolysis, sedation, and/or hypnosis should be phar- macromolecule complex; benzodiazepines are conse- macologically predictable and easily titratable in the quently not GABA-mimetic but can potentiate the duration and magnitude of its effect. It should be short- effects of binding with GABA^ receptors) s This acting as well, with a rapid recovery to preoperative lev- would explain why the binding of benzodiazepines to els of consciousness. Further favourable attributes their receptor complex can be reversed by a simple include a limited metabolism to active or toxic metabo- competitive antagonist (flumazenil) at the conclusion lites, water solubility, compatibility with commonly of anesthesia .26 used crystalloids, and lack of severe drug interactions. Furthermore, it should not be irritant upon intravenous Midazolam injection or cause tissue damage if inadvertently admin- The substance that comes closest to fulfilling the cri- istered extravascularly or intra-arterially. Finally, it teria mentioned above for an ideal drug is midazolam. should have no effect on cardiovascular, respiratory, Its effects on the cardiovascular and respiratory sys- hepatic, or renal function, and no untoward side-effects tems are dose-related, high doses being depressant. such as nausea and vomiting, allergic reactions, or psy- With sedative doses (0.075 mg-kgq), no change in chomotor disturbances. 24 ventilatory response to carbon dioxide or in blood 158 CANADIAN JOURNAL OF ANESTHESIA pressure has been reported; 27 however, the protective EEG changes observed after the administration of upper airway reflexes may be reduced. Caution with 0.15 mg.kg -1 midazolam. Flumazenil alone has no ventilation is required when an opioid is added, cerebral effects, suggesting that there is no intrinsic because of their synergistic effects. The central neuro- reverse-agonist activity. These intracerebral changes, logical anxiolytic, anticonvulsant, amnestic, sedative, however, differ in the diseased brain. Studies in ani- and hypnotic effects are dose-related. 24 Midazolam mals subjected to incomplete global cerebral reduces CMRO 2 and CBF in parallel, such that the ischemia 34 showed an acute increase in CBF and ICP CBF/CMRO 2 ratio remains normal. after flumazenil reversal of midazolam. When high The EEG changes after midazolam (10 mg iv) are doses were studied in healthy dogs, ss the administra- comparable with those after diazepam, 2s with disap- tion of 1 mg-kgq flumazenil to reverse midazolam (40 pearance of 0~-rhyttun and the onset of higher fre- ]ag-kg-l) resulted in marked increases in ICP and CBF quency f~-activity, initially at 22 Hz, followed by for 15 min. Studies in humans have produced similar additional t-activity at 15 Hz. Subjects in that study2s findings. In a group of 18 ASA III patients (American were clinically asleep; these records were not typical of Society of Anesthesia status III) undergoing cranioto- the tracings found in light sleep, and persisted for my for tumour or aneurysm surgery, 36 the use of another 40 min, even after consciousness and orienta- flumazenil (bolus 0.5 mg with 0.1 mg increments to a tion were regained. Mthough midazolam may have a maximum of 1 mg) to reverse the residual effect of variable effect on cerebrospinal fluid (CSF) produc- midazolam following an induction dose of 0.2-0.3 tion and reabsorption, 29 it does not seem to affect ICP mg.kg -1 and an infusion of 0.2 mg.kg-Lhr -i at the if a clinically relevant dose (mean, 0.27 mg.kg -l) is conclusion of surgery resulted in a transient increase (>20% from baseline) of mean arterial pressure in given to patients with reduced intracranial compli- seven patients (39%). Measurements of ICP were not ance. 30 reported in this study. In a study of 15 head-injured Like other benzodiazepines, midazolam is able to patients, s7 midazolam reversal by flumazenil resulted produce reliable anterograde, but not retrograde, in an increase in ICP and mean arterial pressure. Until amnesia. A dose of 5 mg midazolam iv produced anme- more clinical experience is available, reversal of benzo- sia within two minutes, followed by a rapid recovery diazepines with flumazenil, especially in the head- over the next 20 min, with 40% of patients still having injured patient, must be done with prudence and memory impairment after 1.5 hr. sx Despite a wide preferably with simultaneous ICP monitoring. range of doses (0.05-0.27 mg.kg -1) required to achieve the endpoint in a cross-over study, midazolam pro- Ketamine: dissociative anesthesia duced better sedation and anterograde amnesia than A hypnotic agent structurally related to phencyclidine, did diazepam. This desirable property could reduce the ketamine produces a dose-related clinical state of dis- incidence of unfavourable recall of intra-operative sociative anesthesia, but also possesses well defined events. Finally, recent evidences2 also suggests that analgesic properties. The analgesic effects of ketamine midazolam may possess anti-emetic properties. are thought to be mediated by binding of the drug to N-methyl-D-aspartate (NMDA) receptors, although Flumazenil A recently synthesized benzodiazepine, flumazenil is a the drug may also bind to ~r receptors. specific competitive antagonist at the benzodiazepine Advantages include minimal cardiorespiratory depres- receptor, which can reverse all the central effects ofben- sion, active airway reflexes at lower doses, and marked zodiazepine agonists in a dose-related manner. It is analgesia. The airway and respiration may be deprived of almost no intrinsic agonist activity except a depressed at higher doses. Disadvantages include sym- slight subjective sedative effect. Large antagonistic pathetic and cardiovascular stimulation, and increases doses could result in withdrawal symptoms; carefully in muscular tone and activity, intracranial and intra- tit_rated increments of 0.1 mg, up to 1 mg, in healthy ocular pressures, CMRO2, and oropharyngeal secre- adults should provide a smooth recovery without any tions. Patients may recall vivid dreams and/or major adverse reactions. Combinations of midazolam experience hallucinations post-operatively. Ketamine and flumazenil have been successful during spinal or may be administered by oral, intramuscular, or intra- neurological surgery requiring a "wake up" test to venous routes. allow intra-operative neurological assessment. ~4 It is useful to distinguish between the analgesic and In healthy male volunteers (mean age 36 yr, stan- dissociative anesthetic effects ofketamine. It exhibits a dard deviation 5 yr), 33 administration of 0.1 mg.kg -1 range of subanesthetic concentrations whereby anal- flumazenil antagonized the reduction in CBF or the gesia is obtained without loss of consciousness. Lower Bissonnette et aL: NEUROLEPTANESTHESIA 159 bolus doses (0.2-0.75 mg.kg -1 iv, 0.4-2 mg-kg-1 ira) tion of anesthesia by , maintained with a produce analgesia lasting 60-90 min. A dose of combination of enflurane and N20. The current bal- 1.0-2.0 mg-kg -1 iv is recommended for the induction ance of opinion would suggest, however, that the of anesthesia, with unconsciousness usually lasting choice of intravenous induction agents does influence 5-15 min. For this purpose, in children, intranmscular recovery when anesthesia is subsequently maintained doses are mostly used, in the range of 4-10 mg.kg -1, for relatively short periods with inhalational agents. with onset in 3-5 min and a duration of action of Propofol infusions are used for the maintenance of 10-30 min. anesthesia when TIVA is indicated or, in part, for Continuous intravenous infusion of ketamine has intravenous sedation. For procedures lasting less than been recommended for procedures requiring analgesia 60 min, anesthesia maintenance with continuous and without unconsciousness. Idvall et al., as in establishing variable infusions ofpropofol combined with N20 and a concentration-effect relationship for analgesia and O z resulted in recovery times more favourable than hypnosis, found that the plasma concentration provid- those of barbiturate intravenous infusions, 42 inhala- ing an analgesic threshold was 160 ng.m1-1, whereas tional isoflurane,4~ or enflurane. 4s For longer or the hypnotic threshold was 1.5-2.5 lag.m1-1 (with major surgical procedures, however, the speed of N20 ). An intravenous loading dose of 2 mg.kg -1 fol- recovery and the incidence of post-operative vomiting lowed by an infusion of 40 lag-kg-l.min -1 resulted in a after propofol maintenance are similar to those of corresponding steady-state ketamine concentration thiopental-isoflurane anesthesia. 4~ between 1.7 and 2.4 ~ag.ml-~. Ketamine doses of 5-20 PROPOFOL FOR INTRAVENOUS SEDATION: Propofol is a }ag-kg-L-min-1, preceded by an intravenous bolus of good sedative agent used alone or supplementing 0.2-0.75 mg.kg -1 (or 0.4 4 mg.kg -1 ira), are suffi- regional anesthesia, either by repeated bolus administra- cient for analgesia without unconsciousness. For hyp- tion or, preferably, by continuous infusion. It allows ease nosis, in the presence of N20 the recommended 3s of adjustment and titration to the level of sedation maintenance dose is 15-40 lag.kg-l-min -~. As the sole desired. Intravenous infusions ofpropofol to induce hyp- agent, ketamine infusion rates of 60-80 lag.kg-~.min -~ nosis for patients undergoing central neural blockade provide clinical anesthesia. provide a faster recovery than intravenous anesthesia Ketamine has not been widely used for the induc- with methohexital. 46 Using lighter levels of sedation, tion or maintenance of neuroanesthesia because of other investigators have confirmed the rapid recovery CNS side-effects such as central excitation, increased observed with propofol. Dubois et al. 47 administered a CBF and ICP, and the production of post-operative mean infusion rate of 4.3 mg.kg-l.hr -1 to adults after a nightmares. slow induction bolus of 1.7 mg.kgq for sedation during endoscopy. Recovery was rapid, and 99% of the patients Propofol had "adequate" sedation. In another series, 47 a mean First studied clinically in 1977 as an induction agent, infusion rate of 4.9 mg.kg-1-hr -1 was required for seda- propofol (2,6 di-isopropylphenol) subsequently tion for inguinal hernia repair under field proved useful as an anesthetic agent for maintenance block. Again, recovery was rapid. In these two stud- of TIVA or as a sedative/hypnotic agent for diagnos- ies, 46,47 it was hypnosis rather than sedation that was tic procedures and minor surgical procedures to sup- maintained; one would anticipate smaller dose require- plement local or regional anesthesia. It has also been ments for conscious sedation. Low-dose propofol infu- used for longer periods for the maintenance of seda- sions (25-75 ~ag.kgq.min -1 = 1.5-4.5 mg.kg-i.hr -1) c~ua tion or hypnosis in intensive-care patients, and for be used to produce intra-operative sedation during local patient-controlled sedation (PCS). s9 and regional anesthesia. Dertwinkel and Nolte4s main- PROPOFOL FOR THE INDUCTION AND MAINTENANCE tained anesthesia for patients having surgery under spinal OF ANESTHESIA: When used for the induction of anes- anesthesia with low-dose propofol infusions of 1, 1.5, thesia for shorter procedures, propofol results in a and 2 mg-kg-l.hr q following an induction bolus of 1 quicker recovery and earlier return of psychomotor mg.kg-1 and an opioid premedication. All these rates of function than do other induction agents (thiopental infusion provided excellent sedation. Generally, at or methohexital), irrespective of the agent used for the propofol infusion rates > 30 lag-kg-X-min-1 (1.8 maintenance of anesthesia. 4~ However, some investi- mg-kg-l-hr-1), patients are anmestic of the procedure. 49 gators have found that this difference could be of less- The effects ofpropofol sedation on the incidence of er clinical significance when the agent is restricted to intra-operative seizures and the adequacy of electro- use as an induction agent. Valanne and Korttilla 41 corticographic (ECoG) recordings during awake cran- found that it offered little advantage over the induc- iotomy for the management of refractory epilepsy 160 CANADIAN JOURNAL OF ANESTHESIA have been reported, s~ When the authors studied 30 fully after two hours, whereas 96% of patients after patients undergoing temporal or frontal lobectomy for propofol had recovered by 15 rain. Amnesia findings epilepsy under bupivacaine scalp block with a basal were similar between groups. In patients undergoing infusion ofpropofol vs neuroleptanalgesia with a com- minor general or urological procedures with local or bination of fentanyl and droperidol, they found a regional anesthesia, White and Negus s* compared higher incidence ofintra-operative seizures among the maintenance infusions ofpropofol and midazolam for a neurolept patients (0 vs 6, P = 0.008). However, evi- sedation level of 3 ("sleepy/easily roused") as their end- dence of low spike activity on ECoG did not correlate point. Patients receiving sedation during local anesthe- with the type of sedation administered. It is also sia required larger maintenance doses than during important that higher-frequency background ECoG regional anesthesia (propofol 6.3 vs 4.3 mg.min -1, activity occurred among the patients who received midazolam 0.19 vs 0.15 mg.min-1). Recovery of cogni- propofol, which is suggested not to interfere with the tive function occurred more rapidly with propofol, ECoG interpretation. The use of propofol sedation whereas midazolam was associated with greater intra- does not appear to affect the recording of ECoG dur- operative and post-operative amnesia. Patients, accord- hag to follow-up questionnaires, were highly satisfied ing epilepsy surgery, as long as its administration is with both sedative techniques. suspended at least 15 min before recording. In another study, the administration of 2 mg mida- In another study on the safety and efficacy of PCS zolam iv prior to a propofol maintenance infusion was using propofol during awake seizure surgery per- found to be highly effective in enhancing sedation, formed under bupivacaine scalp blocks) l 37 patients amnesia, and anxiolysis, without prolonging recovery- were randomized to receive either propofol PCS with room stay (compared with propofol alone), s2 Propofol a basal infusion of propofol or neuroleptanalgesia was compared with diazepam in an infusion titrated using an initial bolus dose of fentanyl and droperidol until speech was slurred (the endpoint) and then followed by a fentanyl infusion. The authors com- adjusted to maintain clinical requirements based on pared sedation, memory, cognitive function, patient the desired level of sedation. It resulted in equally satisfaction, and incidence of complications between good sedation in both groups but recovery and amne- the two groups. Levels ofintra-operative sedation and sia were better in the propofol group. A mean infusion patient satisfaction in both groups were similar; mem- rate of 4.2 mg-kg-l.hr -1 ofpropofol equated to a mean ory and cognitive function were well preserved in diazepam dose of 0.28 mg.kg-1, ss both. However, transient episodes of ventilatory rate PROPOFOL FOR PATIENT-CONTROLLED SEDATION: depression (<8 beats-min) were observed more fre- Propofol has been used successfully in PCS. s9 Propofol quently among patients given propofol (5 vs O, P = and midazolam were compared for intra-operative PCS 0.04), especially after supplemental doses of opioid. in patients undergoing dental extractions under local Although intra-operative seizures were more common anesthesia. After 0.7 ~ag.kg-l fentanyl iv, patients self- among neurolept patients (0 vs 7, P = 0.002), it was administered either 20 mg propofol or 0.5 mg midazo- not stated if the increase in EEG seizure activity limit- lam with a lockout time of one minute in both groups. ed the ability of the neurologist and neurosurgeon to Propofol had recovery characteristics superior to those perform the procedure. Furthermore, there was no ofmidazolam and was judged by the investigators to be mention of the effect ofpropofol's reported respirato- a more suitable agent because of its rapid response to ry depression on cerebral venous congestion and fluctuating intra-operative requirements, s~ bleeding during the surgical procedure. However, PROPOFOL FOR SEDATION IN INTENSIVE CARE: PCS with propofol was concluded to represent an Propofol has been used for sedation in patients receiv- effective alternative to neuroleptanalgesia during ing mechanical ventilation in the intensive care unit awake seizure surgery, as long as it was performed in a (ICU). s4 It is easy to titrate to obtain a proper level of monitored environment. 1 sedation and a rapid recovery, irrespective of the dura- Comparative studies with midazolam 49 have shown tion of the propofol infusion. In a study of agitat- the superiority ofpropofol for ease of control and rapid- ed/restless ICU patients, propofol was administered ity of recovery. Fanard and colleagues49 compared sub- intravenously for sedation for four days (mean dose, anesthetic inflations of 1.75 mg-kg-l.hr -l of propofol 2.8 mg.kg-l.hr-1). When the infusion was discontin- after 1.5 mg.kg-1 induction with intermittent intra- ued, adequate recovery with response to verbal com- venous boluses of midazolam to produce light sleep mands was obtained in most patients by ten minutes. during epidural anesthesia. Recovery was much slower Recovery times and decreases in blood propofol con- after midazolam: 25% of patients had not recovered centrations were similar after stopping the infusion 24, Bissonnette et al.: NEUROLEPTANESTHESIA 161

48, 72, or 96 hr after initiation. Untoward cumulative who had never taken anticonvulsants. A history of effects were not seen. The plasma concentrations smoking, alcohol, or caffeine consumption also required for sedation and awakening were similar at increased the requirement for fentanyl. The theory of 24 and 96 hr, implying that tolerance did not devel- induction of microsomal liver enzymes resulting in op. s4 Propofol can also be administered for analgesia accelerated fentanyl metabolism is attractive, but and sedation on the basis of "target-controlled infu- changes in the state of opioid receptors induced by sion", in which the plasma concentration of propofol chronic anticonvulsant exposure may also be operative. is maintained between 0.5 and 2.0 lag.m1-1 (for seda- Neurological propertier. Fentanyl has a variety of tion) or betweem 3 and 8 ~ag.m1-1 (for hypnosis). effects on cerebral dynamics. These "specific" effects Another report ss implicated propofol in the deaths attributable solely to fentanyl have been difficult, if (due to secondary respiratory infections) of five chil- not impossible, to single out, because of the complex dren treated with propofol infusions in ICU. The like- relationship between CBF, CPP, and ICE Moreover, lihood ofpropofol being responsible has subsequently changes in PaO 2 from the respiratory depressant been challenged. Although propofol is not yet effects of fentanyl can exert their own effects. approved for sedation of children, it will continue to The effect of opioids on CBF and CMRO 2 is not be used with success for specific surgical and invasive completely known, mainly because of "contamination" medical procedures. from other non-opioid anesthetics. It has been suggest- ed6~ that fentanyl and sufentanil can reduce CBF B THE OPIOID COMI'ONENT and/or CMRO 2 while maintaining cerebral autoregula- tion: 15 rain after administration, 6 ~ag.kg-~ of fentanyl FENTANYL has been able to reduce CBF by as much as 47% and Fentanyl is related to, but not derived from, meperi- CMRO 2 by as much as 18%.61 In another study in adult dine. Although often quoted to be 60-200 times as patients undergoing cardiac surgery,62 100 ]ag.kg-1 of potent as morphine, clinically 0.2 mg fentanyl (200 fentanyl and 0.4 mg.kg -1 of diazepam caused a 25% lag) is equi-analgesic to 10 mg morphine, with a short- reduction in CBF, with no changes in CMRO 2, er duration of action (15-30 min vs 120 min, respec- Moss and colleagues,4s studying patients undergoing tively), s6 Fentanyl is a pure agonist and acts craniotomy and receiving controlled ventilation and predominantly on p-receptors. There is greater reten- hypocapnia, found that 200 lag fentanyl alone did not tion of fentanyl in poorly perfused compartments, result in changes in ICE The authors concluded that leading to subsequent release that could result in res- CPP was the most important determinant of CBF when piratory depression, up to 15 hr after injection. they observed that CBF was reduced in parallel with Neonates are more sensitive to fentanyl because of MAP. This contradicted an earlier study (using normo- their lower brain myelin content, higher CBF, altered capnia) 4a that showed isolated increases in ICP after protein binding, and immature respiratory control, s7 droperidol or fentanyl, without details of statistical analy- Attempts to relate plasma fentanyl concentration to sis. In a recent double-blind study, Jamali etal. 63 demon- its effect on the function of the target organ have pro- strated a reduction in MAP when sufentanil (0.8 vided a better picture of the distribution of fentanyl in lag.kg-1) or fentanyl (4.5 lag.kg-1) was given for elective the brain, ss With a high-dose infusion of fentanyl, there supratentorial craniotomy. When phenylephrine was is slowing of EEG progressing to the formation of delta administered to maintain MAP within 15% of initial val- waves (4 Hz, amplitude 50 }aV). The steady-state serum ues, there were no changes in lumbar CSF pressure. 6s concentration of fentanyl that caused one-half of the Another unblinded outcome study in patients undergo- maximal EEG slowing is 6.9 • 1.5 ng.m1-1. There is a ing elective supratentorial craniotomy with three differ- time lag between the peak concentration of fentanyl ent anesthetic techniques (propofol/fentanyl vs and the reduction in EEG spectral edge frequency - the thiopental/isoflurane/N20 vs thiopental/fentanyl frequency below which 95% of the area under the sig- /NzO)64 also failed to show any important intergroup nal power vs frequency histogram was contained. The differences in mean ICP and in short-term outcomes. lag may be due to the high lipid solubility of fentanyl Although probably of minimal clinical importance, fen- (serum-brain ratio of 1:5), resulting in the solution of tanyl has also been shown to reduce the resistance to fentanyl in CNS lipid rather than at receptor sites. reabsorption of CSF, while CSF production was In neurosurgical patients, it was found that the intra- unchanged .44 operative maintenance dose of fentanyl was increased in When fentanyl combined with droperidol was used, those receiving anticonvulsant therapy, s9 although the Innovar decreased CBF by 50% and CMRO z by 23%, initial loading dose did not differ from that in patients and also reduced CBF response to changes in PaCO 2. 162 CANADIAN JOURNAL OF ANESTHESIA

Unlike volatile agents, which could increase CSF pres- cannot be reversed by naloxone, which suggests a sure, 6s the combination of 5 mg droperidol and 100 non-opioid-related mechanism of action. ~ag fentanyl decreases CSF pressure and CBF (as a Respiratory effects. I3ke all opioids, fentanyl causes a result of decreased MAP), leading to a reduction in dose-related respiratory depression (2 pg.kg-1 can cerebrovascular volume. 66 In a study by Fitch and co- decrease respiratory rate), which is, however, associated workers, 66 this combination resulted in decreased CSF with a compensatory increase in tidal volume. With pressure in all six patients with normal CSF pathways higher doses, tidal volume is decreased, leading to irreg- and in eight of nine patients with intracranial ular breathing and finally apnea. 74 The central response space-occupying lesions during normocapnia. This to carbon dioxide is shifted to the tight. Plasma con- result, however, was not confirmed in a later study centrations of fentanyl of about 3 ng.ru1-1 cause a 50% with Innovar (0.1 ml.kg-1), 67 which found no effect depression of the CO 2 response curve; however, an on CBF and CMRO 2 during normocapnia. On the apparent plasma concentration of i ng-m1-1 is needed to other hand, in the presence of hypocapnia (PaCO 2 < cause ventilatory depression in the presence of enflu- 5 kPa), a moderate decrease in both CBF and CMRO 2 rane. 74 An iv fentanyl dose of 10 lag.kg-1 in dogs raised has been observed. 6s Innovar-induced respiratory cisternal CSF concentration within the first two to three depression in patients breathing spontaneously pro- minutes, even though respiratory depression was evi- duced an increase in ICP. 69 dent within the first minute. 74 The CSF concentration The EEG changes caused by opioids mimic those of of fentanyl was 46% of that in plasma because of the other general anesthetic agents, but to a lesser extent. lower protein concentration in CSF (fentanyl assays With a high infusion dose of fentanyl, EEG alpha activ- were measured for total fentanyl, not for the unbound ity slows, and beta activity ceases within one to two portion of the drug). It is likely that this threshold is minutes. Diffuse theta and some delta activity follow higher clinically in the absence of any other sedative or rapidly. Within five minutes of induction, high-ampli- hypnotic drug. tude delta activity is dominant, and is synchronized in a The respiratory effects of the fentanyl and dropetidol high percentage of patients. All these changes reverted combination have been examined. Ventilatory response to normal upon discontinuation of the fentanyl infu- to carbon dioxide was studied post-operatively after 21 sion. 4s Small doses of fentanyl (0.5-1.0 ~ag.kg-1) and patients received a mean dose of 5 pg.kg-1 fentanyl and droperidol can induce slow-wave activity that is consis- 0.2 mg.kg-1 droperidol. 7s An impaired response was tent and does not interfere with EEG monitoring. 7~ found up to 3.5 hr after the induction of anesthesia, Other neurological effects of fentanyl include the which could extend to six hours in those who also induction of nausea and vomiting by stimulation of received meperidine premedication. Despite animal the chemoreceptor trigger zone in the area postrema experiments suggesting that droperidol had potentiat- of the floor of the fourth ventricle, and muscle rigidi- ing analgesic and respiratory depressant effects on fen- ty. Opioid-induced nausea and vomiting can occur in tanyl, 76 subsequent studies in humans with 77 or without up to 40% of patients post-operatively. Unfortunately, surgery7s have failed to confirm it. Both studies77,78 this is not entirely preventable with antiemetic pro- found a four hour respiratory depression, with or with- phylaxis. Nausea and vomiting have been reported out the addition of droperidol to fentanyl. intra-operatively in patients undergoing neurolep- Neuroleptanalgesia, using droperidol (0.1 mg.kg-1), tanesthesia for awake crartiotomy, 7x with an incidence fentanyl, and diazepam, could also reduce laryngeal as high as 50% with fentanyl, 7~ despite prophylaxis competence. 79 with droperidol and dimenhydrinate. Thoracic and Delayed respiratory depression post-operatively has abdominal muscle rigidity of a tonic or clonic nature been reported after small intravenous doses of fentanyl have been reported after fentanyl, sufentanil, alfen- given intra-operatively. 79 Although an enterohepatic cir- tanil, and morphine, without any EEG evidence of culation of fentanyl had been proposed as a means to seizure activities. Its onset can occur within minutes of cause a secondary peak in plasma fentanyl concentra- administration, or even manifest itself for the first time tion, 8~ the near-complete pre-systemic metabolism that post-operatively. The mechanism naight involve opioid exists (hepatic extraction ratio = 1) renders this unlike- receptors in the brainstem and basal ganglia, since ly. A more plausible explanation is the release of fentanyl treatment with naloxone is rapidly effective in revers- from body stores, such as muscle, associated with an ing the rigidity. The threshold of fentanyl plasma con- increase in body movement during the recovery phase. centration below which muscular rigidity ceased was Cardiovascular effects. Fentanyl is a drug with 6.9 • 1.5 ng.ml -~ in human volunteers. 73 It is inter- marked cardiovascular stability, even at high doses (150 esting to note that normeperidine-induced seizures lag.kg-l). Its lack of histamine release and any myocar- Bissonnette et aL: NEUROLEPTANESTHESIA 163 dial depressant effect make it a popular drug for use in that the 95% effective dose (ED9s) for superficial and cardiac surgery. The main cardiovascular effect is brady- intra-abdominal operations exceeded 300 ng.ml -l and cardia, which is readily reversible by atropine in most 400 ng.m1-1, respectively. The clearance rate of alfen- cases. At doses up to 20 lag.kg-1 in animals, left-ventric- tanil can be lower in the presence of liver disease, result- ular performance was not affected, sl ing in the wide variation observed in clinical There have been many reports on the clinical use of responses. 91 When the EEG spectral edge frequency the fentanyl-droperidol combination in the context of was plotted against the alfentanil plasma concentration, neuroleptanalgesia/anesthesia. In healthy volunteers no time lag was found, indicating its rapidly targeted not undergoing surgery, 5 lag-kg -1 fentanyl and 0.22 association with intracerebral opioid receptors, ss With mg.kg -1 droperidol, given either alone or in combina- 89% of alfentanil being non-ionized at physiological pH tion, were compared, s2 Changes in cardiovascular para- (pKa 6.5), more drug is available to diffuse out of solu- meters were minimal, except for a decrease in systemic tion, resulting in a faster clinical effect than with fen- vascular resistance. In general, both medications cause tanyl. Indeed, alfentanil and droperidol have been used minimal myocardial depression, even in poor-risk successfully to provide neuroleptanalgesia for awake patients undergoing major surgery, provided good fluid craniotomy. 92 pre-loading is given. In patients with renal failure under- The effect of alfentanil on ICP has also been studied going major surgery using neuroleptanesthesia, fentanyl in children requiring insertion of ventriculoperitoneal (5.5 lag.kg-1) and droperidol (0.275 lag.kg -x) with N20 shunts. 93 Alfentanil in doses up to 40 lag-kg -~ did not provided good cardiovascular stability with no delayed result in an increase in ICP transduced electronically via wakening 83 but there was a 20% incidence ofintra-oper- the shunt. There was, however, a decrease in CPP ative awareness. In another group, undergoing renal (although CBF in dogs with acute intracranial hyper- transplantation, s4 a similar technique resulted ha very tension was maintained at doses of 300 pg.kg-l). 9s labile blood pressure perioperatively; arrhythmias were Low-dose alfentanil for sedation in the ventilated new- very uncommon; and there was no mortality. born could cause moderate to severe muscle rigidity. 94 In 106 children undergoing cardiac catheteriza- tion, fentanyl (1.25 lag-kg-1, maximum dose 50 Iag) and droperidol (62.5 lag&g-i, maximum dose 2.5 rag) Sufentanil im did not cause changes in heart rate, respiratory Sufentanil is five times more potent than fentaawl , while arrest, or cyanotic spells (no patients had a tetralogy of its pharmacokinetic properties are intermediate Fallot). 8s When larger doses of droperidol (0.3 between fentanyl and alfentanil. Ninety-two percent is mg.kg -1) and fentanyl (10 lag&g-i) were used for protein-bound (mainly to 0~l-acid glycoprotein), with major surgery in 15 neonates (<10 days old) and 25 19.7% non-ionized at physiological pH (pKa = 8.01). infants (<18 months old), remarkably good cardiovas- The minimal effective blood level is approximately 0.2 cular stability was reported. 86 The decision to admin- ng.mFl. 9s Although it differs from fentanyl in potency ister an intravascular pre-load with 20 ml.kg -1 Ringer's and lipid solubility, the pharmacodynamic effects of lactate and atropine before induction and the subse- sufentanil are generally similar to those of fentanyl. 96 In quent maintenance with N20 contributed to the suc- a study of patients undergoing cardiac surgery, sufen- cess of this technique. tanil (10 lag-kg -1 infused at 2 lag-kg-l.min -l) caused a 25% reduction in CBF and a 29% reduction in CMRO2, with a corresponding increase in cerebrovascular resis- ALTERNATIVE DRUGS tance due to its primary metabolic depressant effects. 97 Alfentanil Low-dose sufentanil (0.5 lag&g-i) given as a single A more predictable drug, alfentanil has one-quarter the agent to healthy volunteers did not result in any change potency of fentanyl. 87 Its protein binding approaches in CBF, 9s which suggests that neuroleptic doses should 90% (mainly to ~xffacid glycoprotein), but this figure is not affect CBF. However, in the presence of intracere- reduced at high doses, which could result in non-linear bral space-occupying lesions, sufentanil (1 lag.kg-1) kinetics and some enhancement of the concentra- resulted in an 87% increase in CSF pressure, 99 which tion-effect relationship. In children 8s and infants, s9 it could be due to cerebral vasodilation. has a faster clearance than in adults. With a target plas- There have been numerous studies comparing the ma alfentanil concentration of 400 ng.ml -l to provide various clinical effects of fentanyl, alfentanil, and sufen- adequate analgesia in the presence of N20 67%, a bolus tanil. On the whole, it is difficult to declare that either dose of 176 lag.kg-1 and a maintenance dose of 1.3 of the newer agents is superior to fentanyl, on the basis lag-kg-Lmin -I was necessary. 9~ Clinically, it was found of clinical effective outcome, side-effects, or cost. 164 CANADIAN JOURNAL OF ANESTHESIA

When their use in neurosurgery for tumour excision theoretically, there should therefore be no prolongation was compared, fentanyl (10 pg-kg-1 loading dose, fol- of drug elimination. Although N-dealkylation of lowed by 2 lag.kg-l.hr-1 infusion), sufentanil, and alfen- remifentanil occurs, its major route of biotransforma- tanil (each at doses equipotent to fentanyl) were found tion is by de-esterification to a carboxylic acid metabo- to be equally safe and effective. 1~176In a separate study,x~ rite (G190291) which, in animal models, has only high-dose sufentanil (20 pg.kg-1) for both intra- and 1/300th to 1/1000th the potency of the parent com- extracranial neurosurgery provided more adequate pound. The mean terminal half-life of G190291 has anesthesia than that offentanyl (100 lag.kg-1 ). In anoth- been measured at 88-137 min, 1~ longer than that of er prospective study of patients undergoing awake cran- remifentanil; it is excreted by the kidneys. Given the fast iotomy for epilepsy surgery,1~ fentanyl (bolus 0.75 onset and offset of action with inactive metabolites, no lag.kg-1 with infusion of 0.01 pg-kg-l.min-1) was as major adjustment dosage appears to be required. effective as sufentanil (bolus 0.075 pg-kg-~, infusion However, most studies to date have been performed in 0.0015 pg.kg-l.min-1) or alfentanil (bolus 7.5 pg-kg-1, healthy young adults, so extrapolation of these results infusion 0.5 pg-kg-l.rnin-1). However, the incidence of to include the elderly and children might be inappro- nausea and vomiting, despite prophylactic treatment, pilate. Furthermore, studies of its effects in neurosurgi- was 50% with fentanyl, 30% with sufentanil, and 70% cal and cardiac anesthesia will be needed as part of with alfentanil. The drug cost ofalfentanil was 13 times remifentanil's clinical evaluation. and sufentanil 30 times that of fentanyl. When their Possible advantages of remifentanil include (1) rapid effects in patients with brain tumours were measured, titration of effect, (2) reduced post-operative opioid- fentanyl was found to have negligible effect on CSF induced side-effects, (3) lack of cumulative effects, and pressure, whereas sufentanil had the greatest CBF (4) no requirement for dosage adjustment in cases of increase. 1~ In a separate study,99 the cerebral vasodila- hepatic or renal disease. Theoretical disadvantages tory effect (with a 70% increase in CSF pressure) of include (1) the cost of extra equipment and expertise in alfentanil also became apparent when systemic blood infusion techniques, and (2) a lack of prolonged opioid pressure was held constant with phenylephrine. With effect (e.g., analgesia into the post-operative period) the evidence to date, fentanyl remains the opioid of when such effects are desirable. Remifentanil may choice for use in neuroanesthesia. become the ideal opioid for neuroanesthesia, allowing When sufentanil and fentanyl were compared with rapid recovery and facilitating early clinical neurological morphine and meperidine in patients undergoing gen- assessment. eral, orthopedic, or gynecological surgical proce- dures) ~ these newer synthetic drugs were found to Conclusion provide much better intra-operative hemodynamic Neuroleptanesthesia was developed to maintain nor- conditions, earlier return of mental function, and a mal cerebral cognitive function while eliminating the lower incidence of respiratory depression post-opera- perception of nociceptive stimuli at the level of the tively. Overall, it seemed that sufentanil provided the cerebral cortex. The procedure was designed to mod- most satisfactory conditions. Patients having short ulate neuroendocrines and metabolic and autonomic surgical procedures recovered faster with alfentanil responses in response to the nociceptive stimulation. than with fentanyl. However, when used in sedative The development of an injectable combination of two doses in infants and children (1-23 mo), fentanyl was new drugs, a neuroleptic agent (droperidol) and an found to provide more stable conditions and fewer opioid (fentanyl), extensively improved the original side-effects than alfentartil. technique and quickly extended its use to many surgi- cal procedures. The advantages included an excellent Remifentanil cardiovascular stability, rapid post-operative recovery, Remifentanil (G187084B) is the latest opioid to be and the availability of an antidote for the adverse introduced into clinical use. Chemically related to fen- effects of the opioid component. tanyl, its analgesic potency is similar to that of fentanyl, However, this technique was not without prob- but it is 20-30 times more potent than alfentanil. ~~ lems. Unconsciousness to the level of being unrous- Remifentanil is a p-agortist, analgesic, sedative, and res- able and post-operative extrapyramidal excitation, piratory depressant. The ester linkage of the molecule restlessness, and confusion were among several side- makes it susceptible to hydrolysis by circulating and tis- effects associated with its use. The relatively slow sue-non-specific esterases, resulting in a short elimina- onset of action of droperidol and the fast onset of fen- tion half-life. Unlike plasma cholinesterase, these tanyl renders them difficult to manage clinically, with non-specific enzymes are not inhibited by neostigmine; associated overdose of fentanyl and consequential Bissonnette et al.: NEUROLEPTANESTHESIA 165 muscle rigidity and/or respiratory depression necessi- 2 Bailey PIo Stanley TH. Intravenous Opioid anesthet- tating airway support. ics. In: Miller RD (Ed.). Anesthesia, 4th ed. New New pharmacological agents developed over the York: Churchill Livingstone Inc.,. 1994: 291-389. past two decades have helped simplify this anesthetic 3 de Castro G, Mundeleer P. Anesth&ie sans sommeii: technique. Neuroleptanalgesia/anesthesia has evolved "neuroleptanalg&ie'. Acta Clair Belg 1959; 58: into "conscious sedation", in which control of the 689-93. patient's vital functions with maintenance of normal 4 Janssen PAJ, Niemegeers CJE, SeheUekens KHI~ cognitive function has been made easier. New benzo- Verbruggen FJ, Van Neuten JM. The pharmacology of diazepines such as midazolam and its reversal agent, dehydrobenzperidol, a new potent and short-acting flumazenil, have revolutionized this anesthetic tech- neuroleptic agent chemically related to haloperidol. nique. Propofol, an anesthetic agent capable of induc- Arzniemittelforschung 1963; 13: 205-8. ing and maintaining general anesthesia, can also be 5 Janssen PAJ. A review of the chemical features associ- used successfully and predictably to provide sedation ated with strong morphine-like activity. Br J Anaesth and hypnosis for diagnostic procedures and minor and 1962; 34: 260-8. even more complex surgical procedures. The recovery 6 FoldesFF. Neuroleptanesthesia for general surgery. Int characteristics of these drugs have made them suitable Anesthesiol Clin 1973; 11: 1-35. for ambulatory anesthesia and short surgical or diag- 7 FoldesFF, Kepes ER, Kronfeld PP, Shiffman HP. A nostic procedures. The introduction of newer opioids rational approach to neuroleptanesthesia. Anesth have added more flexibility to these techniques com- Analg 1966; 45: 642-54. pared to fentanyl, especially in the context of con- 8 RichterJJ. Current theories about the mechanisms of scious sedation, where analgesia must be excellent. benzodiazepines and neuroleptic drugs. Alfentanil, sufentanil, and remifentanil have con- 1981; 54: 66-72. tributed to improve intra-operative analgesia while 9 Marshall BM, Gordon RA. Electroencephalographic offering better control of post-operative recovery, monitoring in anaesthesia with droperidol and fen- acute pain management, and discharge time. tanyl. Can Anaesth Soc J 1968; 15: 357-61. Unfortunately, the cost of many of these new medica- 10 Hashem A, Frey HH. The effect ofneuroleptics and tions can be considerable---one of the few major con- neuroleptic/analgesic combinations on the sensitivity cerns about their use. to seizures in mice. Anaesthetist 1988; 37: 631-5. The need to provide high-quality anesthesia associated 11 Herrick IA, Craen RA, Gelb AW, et al. Propofol seda- with faster recovery, better comfort, and reduced cost are tion during awake craniotomy for seizures: patient- all factors in favour of the concept of neuroleptanesthesia. controlled administration versus neurolept analgesia. These new medications have now been used extensively Anesth Analg 1997; 84: 1285-91. in children, adults, and elderly patients to facilitate the 12 SorokerD, Barzilay E, Konichezky S, Bruderman I. expansion of programs such as ambulatory anesthesia, Respiratory function following premedication with "satellite" anesthesia (radiology, interventional cardiolo- droperidol or diazeparn. Anesth Analg 1978; 57: gy, oncology, endoscopy, rheumatology, etc.), acute and 695-9. chronic pain therapy, and programs for many other indi- 13 CortrellJE, Wolfson B, Siker ES. Changes in airway cations. Finally, intra-operative electrocorticography and resistance following droperidol, hydroxyzine, and seizure foci removal may remain the only indication left diazepam in normal volunteers. Anesth Analg 1976; for the use of "conventional" neuroleptanalgesia/anes- 55: 18-21. thesia in modern anesthesia practice. 14 Prokocimer P, Delavault E, Rey F, Lefevre P, Mazze RI, DesmontsJM~ Effects of droperidol on respiratory Acknowledgment drive in humans. Anesthesiology 1983; 59: 113-6. This paper was prepared with the assistance of 15 Ward DS. Stimulation ofhypoxic ventilatory drive by Editorial Services, The Hospital for Sick Children, droperidol. Anesth Analg 1984; 63: 106-10. Toronto, Ontario, Canada. 16 WhitwamJG, Russell WJ. The acute cardiovascular changes and adrenergic blockade by droperidol in References man. Br J Anaesth 1971; 43: 581-91. 1 Campan Io Lazothes G. L'hibernoth&apie en chirurgie 17 Sumikawa K, Amakata Y. The pressor effect of ctrtbrale, sa justification clinique et pathologt3nique. droperidol on a patient with pheochromocytoma. In: Laborit H, Huguenard P (Eds.). Practique de Anesthesiology 1977; 46: 359-61. l'Hibernothtrapie en Chirurgie et en Mtdicine. Paris: 18 Sumikawa K, Hirano H, Amakata Y, Kashimoto T, Masson et Cie, 1954: 137-80. Wada A, Izumi F. Mechanism of the effect of 166 CANADIAN JOURNAL OF ANESTHESIA

droperidol to induce catecholamine efflux from the cerebral blood flow. Anesth Analg 1987; 66: 309-13. adrenal medulla. Anesthesiology 1985; 62: 17-22. 34 Schulte am EschJ, Kochs E. Midazolam and flumazenil 19 Long G, Dripps RD, Price HL. Measurement ofanti- in neuroanaesthesia. Acta Anaesthesiol Scand 1990; arrhythmic potency of drugs in man: effects of dehy- 92(Suppl): 96-102. drobenzperidol. Anesthesiology 1967; 28: 318-23. 35 FleischerJE, Milde JH, Moyer TP, Michenfelder JD. 20 Bertol6 Lo Novakovi L, Penna M. Antiarrhythmic effects Cerebral effects of high-dose midazolam and subse- ofdroperidol. Anesthesiology 1972; 37: 529-35. quent reversal with RO 15-1788 in dogs. 21 SantosA, Datta S. Prophylactic use ofdroperidol for Anesthesiology 1988; 68: 234--42. control of nausea and vomiting during spinal anesthe- 36 Chiolero ILL, Ravussin PA, Freeman J. Using flumaze- sia for cesarean section. Anesth Analg 1984; 63: 85-7. nil (Ro 15-1788) after prolonged midazolam infusion 22 Lerman J, Eustis S, Smith DR. Effect ofdroperidol for anaesthesia for intracranial surgery. Ann Fr Anesth pretreatment on postanesthetic vomiting in children Reanim 1988; 7: 17-21. undergoing strabismus surgery. Anesthesiology 1986; 37 Chiolero R-L, Ravussin P, Anderes J-P, Ledermann P, 65: 322-5. de Tribolet N. The effects of midazolam reversal by 23 Yentis SM, Bissonnette B. Ineffectiveness of acupunc- RO 15-1788 on cerebral perfusion pressure in ture and droperidol in preventing vomiting following patients with severe head injury. Intensive Care Med strabismus repair in children. Can J Anaesth 1992; 1988; 14: 196-200. 39: 151-4. 38 Idvall J, Ahlgren I, Aronsen KR~ Stenberg P Ketamine 24 Thomson D, Geller E, Lauven P, Whitwam J. infusions: pharmacokinetics and clinical effects. Br J Midazolam and flumazenil: the agonist-antagonist Anaesth 1979; 51: 1167-73. concept for sedation and anaesthesia. Acta 39 Rudkin GE, Osborne GA, Finn BP, Jarvis DA, Vickers Anaesthesiol Scand 1990; 92(Suppl): 108-9. D. Intra-operative patient-controlled sedation. 25 Goodchild CS. GABA receptors and benzodiazepines. Comparison of patient-controlled propofol with patient- Br J Anaesth 1993; 71: 127-33. controlled midazolam. Anaesthesia 1992; 47: 376-81. 26 Lauven PM, Kulka PJ. Anaesthesia techniques for 40 Doze VA, Sharer A, White PF. Propofol- midazolam and flumazenil - an overview. Acta versus thiopental-isoflurane-nitrous oxide for general Anaesthesiol Scand 1990; 92(Suppl): 84-9. anesthesia. Anesthesiology 1988; 69: 63-71. 27 Murphy PJ, Erskine Re Langton JA. The effect of 41 ValanneJ, Korttila K. Effect of a small dose of intravenously administered diazepam, midazolam and droperidol on nausea, vomiting and recovery after flumazenil on the sensitivity of upper airway reflexes. outpatient enflurane anaesthesia. Acta Anaesthesiol Anaesthesia 1994; 49: 105-10. Scand 1985; 29: 359-62. 28 Brown CR, Sarnquist FH, Canup CA, Pedley TA. 42 Doze VA, Westphal LM, White PF. Comparison of Clinical, electroencephalographic, and pharmacoki- propofol with methohexital for outpatient anesthesia. netic studies of a water-soluble benzodiazepine, mida- Anesth Analg 1986; 65: 1189-95. zolam maleate. Anesthesiology 1979; 50: 467-70. 43 Mi~VeldtBB, J~rgensen PB, Spotoft H, Ronde F. The 29 Arrru AA. Dose-related changes in the rate ofcere- effects of droperidol and fentanyl on intracranial pres- brospinal fluid formation and resistance to reabsorp- sure and cerebral perfusion pressure in neurosurgical tion of cerebrospinal fluid following administration of patients. Br J Anaesth 1976; 48: 963-8. thiopental, midazolam, and etomidate in dogs. 44 Artru AA. Effects of halothane and fentanyl anesthe- Anesthesiology 1988; 69: 541-6. sia on resistance to reabsorption of CSF. J Neurosurg 30 Giffin JP, CorrrellJE, Shwiry B, Hartung J, Epstein J, 1984; 60: 252-6. Lira K. Intracranial pressure, mean arterial pressure, 45 MossE, Powel119, Gibson RM, McDowall DG. Effects and heart rate following midazolam or thiopental in of fentanyl on intracranial pressure and cerebral per- humans with brain tumors. Anesthesiology 1984; 60: fusion pressure during hypocapnia. Br J Anaesth 491-4. 1978; 50: 779-84. 31 DundeeJW,,, Wilson DB. Amnesic action of midazo- 46 Jessop E, Grounds RM, Morgan M, Lumley J. lam. Anaesthesia 1980; 35: 459-61. Comparison of infusions of propofol and methohexi- 32 Splinter WM, MacNeill HB, Menard EA, Rhine EJ, tone to provide light during Roberts DJ, Gould MH. Midazolam reduces vomiting surgery with regional blockade. Br J Anaesth 1985; after tonsillectomy in children. Can J Anaesth 1995; 57: 1173-7. 42: 201-3. 47 Dubois A, Balatoni E, PeetersJP, Baudoux M. Use of 33 Forster A, Juge O, Louis M, Nahory A. Effects of a propofol for sedation during gastrointestinal endo- specific benzodiazepine antagonist (RO 15-1788) on scopies. Anaesthesia 1988; 43(Suppl): 75-80. Bissonnette et al.: NEUROLEPTANESTHESIA 167

48 Dertwinkel R, Nolte H. Continuous sedation for bypass: the influence of PaCO 2. Anesth Analg 1987; regional anesthesia with propofol (Disoprivan| and 66: 825-32. midazolam (Dormicum| A comparative study. Reg 63 Jamali S, Ravussin P, Archer D, Goutallier D, Parker F, Anaesth 1988; 11: 84-91. Ecoffey C. The effects of bolus administration of opi- 49 Fanard L, Van Steenberge A, Demeire X, van der Puyl oids on cerebrospinal fluid pressure in patients with F. Comparison between propofol and midazolam as supratentorial lesions. Anesth Analg 1996; 82: 600-6. sedative agents for surgery under regional anaesthesia. 64 Todd MM, Warner DS, Sokoll MD, et al. A prospec- Anaesthesia 1988; 43(Suppl): 87-9. tive, comparative trial of three anesthetics for elective 50 Herrick IA, Craen RA, Gelb AW, et al. Propofol seda- supratentorial craniotomy. Propofol/fentanyl, isoflu- tion during awake craniotomy for seizures: electro- rane/nitrous oxide, and fentanyl/nitrous oxide. corticographic and epileptogenic effects. Anesth Anesthesiology 1993; 78: 1005-20. Analg 1997; 84: 1280-4. 65 Fitch W, Barker J, McDowall DG, Jennett WB. The 51 White IF, NegusJB. Sedative infusions during local effect of on cerebrospinal fluid pres- and regional anesthesia: a comparison of midazolam sure in patients with and without intracranial space- and propofol. J Clin Anesth 1991; 3: 32-9. occupying lesions. Br J Anaesth 1969; 41: 564-73. 52 TaylorE, Ghouri AF, White PF. Midazolam in combi- 66 Fitch W, Barker J, Jennett WB, McDowall DG. The nation with propofol for sedation during local anes- influence of neuroleptanalgesic drugs on cerebrospinal thesia. J Clin Anesth 1992; 4: 213-6. fluid pressure. Br J Anaesth 1969; 41: 800-6. 53 Valtonen M, Salonen M, ForssellH, Scheinin M, 67 Sari A, Okuda Y, Takeshita H. The effects of thalam- Viinamiiki O. Propofol infusion for sedation in out- onal on cerebral circulation and oxygen consumption patient oral surgery. A comparison with diazepam. in man. Br J Anaesth 1972; 44: 330-4. Anaesthesia 1989; 44: 730-4. 68 Cold GE, Christensen KJS, Nordentoft J, Engberg M, 54 Belier JP, Pottecher T, Lugnier A, Mangin P, Otteni Pedersen MB. Cerebral blood flow, cerebral metabolic JC. Prolonged sedation with propofol in ICU rate of oxygen and relative CO 2 reactivity during neu- patients: recovery and blood concentration changes rolept anaesthesia in patients subjected to craniotomy during periodic interruptions in infusion. Br J for supratentorial cerebral tumours. Acta Anaesthesiol Anaesth 1988; 61: 583-8. Scand 1988; 32: 310-5. 55 Parke TJ, StevensJE, Rice AS, et al. Metabolic acido- 69 Miller R, Tausk HC, Stark DCC. Effect of Innovar, sis and fatal myocardial failure after propofol infusion fentanyl and droperidol on the cerebrospinal fluid in children: five case reports. BMJ 1992; 305: 613-6. pressure in neurosurgical patients. Can Anaesth Soc J 56 Morrison JD, Loan WB, Dundee JW.. Controlled com- 1975; 22: 502-8. parison of the efficacy of fourteen preparations in the 70 Manninen P, Contreras J. Anesthetic considerations relief of postoperative pain. BMJ 1971; 3: 287-90. for craniotomy in awake patients. Int Anesthesiol Clin 57 Koehntop DE, Rodman JH, Brundage DM, Hegland 1986; 24: 157-74. MG, BuckleyJJ. Pharmacokinetics of fentanyl in 71 Archer DP, McKenna JMA, Morin L, Ravussin P. neonates. Anesth Analg 1986; 65: 227-32. Conscious-sedation analgesia during craniotomy for 58 ScottJC, Ponganis KV, Stanski DR. EEG quantitation intractable epilepsy: a review of 354 consecutive of narcotic effect: the comparative pharmacodynamics cases. Can J Anaesth 1988; 35:338 41. of fentanyl and alfentanil. Anesthesiology 1985; 62: 72 Gignac E, Manninen PIff, Gelb AW. Comparison of 234-41. fentanyl, sufentanil and alfentanil during awake cran- 59 Tempelhoff R, Modica PA, Spitznagel EL Jr. iotomy for epilepsy. Can J Anaesth 1993; 40: 421-4. Anticonvulsant therapy increases fentanyl require- 73 Bowdle TA, Rooke GA. Postoperative myoclonus and ments during anaesthesia for craniotomy. Can J rigidity after anesthesia with opioids. Anesth Analg Anaesth 1990; 37: 327-32. 1994; 78: 783-6. 60 Schell RM, Kern FH, Greeley WJ, et al. Cerebral 74 BovilJG. Opioids. In: Dundee JW, Clarke RSJ, blood flow and metabolism during cardiopulmonary McCaughey W (Eds.). Clinical Anaesthetic bypass. Anesth Analg 1993; 76: 849-65. Pharmacology. Philadelphia: Churchill Livingstone 61 Michenfelder JD, Theye RA. Effects offentanyl, Inc., 1991: 203-30. droperidol, and Innovar on canine cerebral metabo- 75 VejlstedH, Hansen M, Jacobsen E. Postoperative ventila- lism and blood flow. Br J Anaesth 1971; 43: 630-6. tory response to carbon dioxide following neurotept 62 Murkin JM, Farrar JK, Tweed WA, McKenzie FN, anaesthesia. Acta Anaesthesiol Scand 1977; 21: 529-33. Guiraudon G. Cerebral autoregulation and 76 Greene MJ. Some aspects of the pharmacology of flow/metabolism coupling during cardiopuimonary droperidol. Br l Anaesth 1972; 44: 1272-6. 168 CANADIAN JOURNAL OF ANESTHESIA

77 Smedstad KG, Rigg JRA. Control of breathing after 93 Markovitz BP, Duhaime A-C, Sutton Lo Schreiner MS, fentanyl and Innovar anaesthesia. Br J Anaesth 1982; Cohen DE. Effects of alfentanil on intracranial pres- 54: 599-604. sure in children undergoing ventriculoperitoneal 78 HarperMH, HickeyRF, Cromwell TH, Linwood S. The shunt revision. Anesthesiology 1992; 76: 71-6. magnitude and duration of respiratory depression pro- 94 PokelaM-I o Ryh~nen PT, Koivisto ME, Olkkola KT, duced by fentanyl and fentanyl plus droperidol in man. Saukkonen A-L. Alfentanil-induced rigidity in new- J Pharmacol Exp Ther 1976; 199: 464-8. born infants. Anesth Analg 1992; 75: 252-7. 79 BeckerLD, Paulson BA, Miller RD, Severinghaus 95 BovilIJG, Sebel PS, Blackburn CL, Oei-Lim V, Eger E1 II. Biphasic respiratory depression after fen- HeykantsJJ. The pharmacokinetics of sufentanil in tanyl-droperidol or fentanyl alone used to supple- surgical patients. Anesthesiology 1984; 61: 502-6. ment nitrous oxide anesthesia. Anesthesiology 1976; 96 Sanford TJJr. A review of sufentanil. Semin Anesth 44: 291-6. 1988; VII: 127-36. 80 StoeckelH, Hengstmann JH, SchartlerJ. 97 Murkin JM, Farrar JK, McNeill B, GelbAW, Lok P. Pharmacokinetics of fentanyl as a possible explanation Cerebral oxygen consumption during sufentanil anes- for recurrence of respiratory depression. Br J Anaesth thesia: measurement by N20 uptake vs 133xenon 1979; 51: 741-5. clearance. Can J Anaesth 1990; 37: $32. 81 OstheimerGW, Shanahan EA, Guyton RA, Daggett 98 Mayer N, Weins~abl C, Podreka I, Spiss CK. Sufentanil WM, Lowenstein E. Effects of fentanyl and droperidol does not increase cerebral blood flow in healthy on canine left-ventricular performance. Anesthesiology human volunteers. Anesthesiology 1990; 73: 240-3. 1975; 42: 288-91. 99 Jung R, Shah N, Reinsel R, et al. Cerebrospinal fluid 82 Graves CL, Downs NH, Browne AB. Cardiovascular pressure in patients with brain tumors: impact of fen- effects of minimal analgesic quantities of Innovar@, tanyl versus alfentanil during nitrous oxide-oxygen fentanyl, and droperidol in man. Anesth Analg 1975; anesthesia. Anesth Analg 1990; 71: 419-22. 54: 15-23. 100 From RP,, Warner DS, Todd MM, Sokoll MD. 83 TrudnowskiR J, MostertJW,, Hobika GH, Rico R. Anesthesia for craniotomy: a double-blind compari- Neuroleptanalgesia for patients with kidney malfunc- son of alfentanil, fentanyl, and sufentanil. tion. Anesth Analg 1971; 50: 679-84. Anesthesiology 1990; 73: 896-904. 84 Lindahl-Nilsson Ch, Lundh R, Groth C-G. Neurolept 101 Shupak R C, Harp JR. Comparison between high- anaesthesia for the renal transplant operation. Acta dose sufentanil-oxygen and high-dose fentanyl-oxy- Anaesthesiol Scand 1980; 24: 451-7. gen for neuroanaesthesia. Br J Anaesth 1985; 57: 85 Graham TPJr, Arwood GF, Werner B. Use ofdroperi- 375-81. dol-fentanyl sedation for cardiac catheterization in 102 Meretoja OA, Rautiainen P. Alfentanil and fentanyl children. Am Heart J 1974; 87: 287-93. sedation in infants and small children during cardiac 86 Kay B. Neuroleptanesthesia for neonates and infants. catheterization. Can J Anaesth 1990; 37: 624-8. Anesth Analg 1973; 52: 970-3. 103 GlassPS, Hardman D, Kamiyama Y, et al. 87 Smnski DR, Hug CJ. Alfentanil--a kinetically pre- Preliminary pharmacokinetics and pharmacodynamics dictable narcotic analgesic (Editorial). Anesthesiology of an ultra-short-acting opioid: remifentanil 1982; 57: 435-8. (GI87084B). Anesth Analg 1993; 77: 1031-40. 88 Roure P, Jean N, LeclercA-C, Cabanel N, LevronJ-C, 104 DavisPJ, Lerman J, Suresh S, et al. A randomized Duvaldestin P. Pharmacokinetics of alfentanil in children multicenter study of remifentanil compared with undergoing surgery. Br J Anaesth 1987; 59: 1437-40. alfentanil, isoflurane, or propofol in anesthetized 89 GoreskyGV, Koren G, Sabourin MA, SaleJP, Strunin pediatric patients undergoing elective strabismus L. The pharmacokinetics of alfentanil in children. surgery. Anesth Analg 1997; 84: 982-9. Anesthesiology 1987; 67: 654-9. 90 Fragen RJ, Booij LHDJ, Braak GJJ, Vree TB, Heykants J, CrulJF. Pharmacokinetics of the infusion of alfen- tanil in man. Br J Anaesth 1983; 55: 1077-81. 91 Shafer A, Sung M-L, White PF. Pharmacokinetics and pharmacodynamics of alfentanil infusions during gen- eral anesthesia. Anesth Analg 1986; 65: 1021-8. 92 WellingEC, Donegan J. Neuroleptanalgesia using alfentanil for awake craniotomy. Anesth Analg 1989; 68: 57-60.