V2.1_12 April 2012

Anesthesia Considerations for Neurophysiologic Monitoring using the ProPep Nerve Monitoring System™ during da Vinci® Prostatectomy

Because the ProPep Nerve Monitoring System is measuring stimulated electromyographic (EMG) signals emanating from the muscles in which the nerves of interest terminate, it is important the muscles not be paralyzed during that portion of the surgery when neurophysiologic monitoring is being performed. As a result, there are a number of anesthesia considerations that need to be kept in mind to optimize the validity and quality of the neurophysiologic readings. Please note that all decisions regarding anesthesia are the responsibility of the attending licensed medical practitioner administering anesthesia. It is important that the surgeon discuss these issues preoperatively with the attending licensed medical practitioner administering the anesthesia.

Caution: The use of paralyzing anesthetic agents will significantly reduce, if not completely eliminate, EMG responses to direct or passive nerve stimulation. Whenever nerve paralysis is suspected, consult the attending licensed medical practitioner administering the anesthesia.

Before the Start of the Surgery:- A conversation between the Surgeon and the attending medical practitioner administering the anesthesia should take

place to discuss: o At what point during the surgery will the monitoring occur; o How will the physician alert the medical practitioner administering the anesthesia that the portion of the case

requiring monitoring is approaching and how much lead time would the medical practitioner administering the anesthesia like to be given.

� This is important information that will allow the medical practitioner administering the anesthesia to ensure the muscle relaxants have worn off adequately so that the surgeon can obtain the best opportunity for recording useful and valid responses during the monitoring process.

During The Surgery:- Only short acting muscle relaxants should be used. - Muscle relaxants should be dosed incrementally.

o The goal is to keep the patient at 3-2 well defined twitches during the neurophysiologic monitoring. - The surgeon will communicate with the medical practitioner administering the anesthesia when they are approximately

20 minutes (or the previously agreed upon time) away from performing the neurophysiologic monitoring. o This will allow adequate time for the neuromuscular blockade to wear off sufficiently giving the surgeon the

best opportunity for optimal responses during the monitoring process.

Additional Considerations:- The medical practitioner administering the anesthesia should be prepared to use pressure control ventilation as a

means to improve the ability to ventilate the patient when they are becoming “light” on muscle relaxants. - During the period of reduced neuromuscular blockade, the stability of the surgical view for the operating surgeon can

be improved by reducing the drive to breath using: o over-ventilation to reduce CO2 o narcotics.

ProPep Surgical wishes to thank Dr. Paul Playfair - Chief of Anesthesia at Westlake Medical Center – Austin, TX for his contributions to this protocol.

References: Attached you will find references that address anesthesia considerations during neurophysiologic monitoring in more depth. Please refer to the highlighted sections for considerations specific to the mode of neurophysiologic monitoring the ProPep Nerve Monitoring System employs.

IntraoperativeNeurophysiological Monitoring

Second Edition

Aage R. Møller, PhD

University of Texas at DallasDallas, TX

© 2006 Humana Press Inc.999 Riverview Drive, Suite 208Totowa, New Jersey 07512

www.humanapress.com

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by anymeans, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from thePublisher.

All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect theviews of the publisher.

This publication is printed on acid-free paper. ∞ANSI Z39.48-1984 (American Standards Institute)Permanence of Paper for Printed Library Materials.

Production Editor: Jennifer Hackworth

Cover design by Patricia F. Cleary

For additional copies, pricing for bulk purchases, and/or information about other Humana titles, contact Humana at the aboveaddress or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail: orders@humanapr.com; or visitour www.humanapress.com

Photocopy Authorization Policy:Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted byHumana Press, provided that the base fee of US $30.00 per copy is paid directly to the Copyright Clearance Center at 222Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the CCC, aseparate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the TransactionalReporting Service is: [1-58829-703-9/06 $30.00].

eISBN: 1-59745-018-9

Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Møller, Aage R.Intraoperative neurophysiological monitoring / Aage R. Møller. -- 2nd ed.

p. cm.Includes bibliographical references and index.ISBN 1-58829-703-9 (alk. paper) 1. Neurophysiologic monitoring. 2. Evoked potentials (Elecrophysiology) I. Title.

RD52.N48M65 2006617.4'8--dc22

2005050259

ix

Contents

Preface ..................................................................................................................................... vAcknowledgments ..................................................................................................................vii

1 Introduction ..................................................................................................................... 1

SECTION I: PRINCIPLES OF INTRAOPERATIVE NEUROPHYSIOLOGICAL MONITORING

2 Basis of Intraoperative Neurophysiological Monitoring .................................................. 93 Generation of Electrical Activity in the Nervous System and Muscles .......................... 214 Practical Aspects of Recording Evoked Activity From Nerves,

Fiber Tracts, and Nuclei ............................................................................................ 39References to Section I .......................................................................................................... 49

SECTION II: SENSORY SYSTEMS

5 Anatomy and Physiology of Sensory Systems ................................................................ 556 Monitoring Auditory Evoked Potentials ......................................................................... 857 Monitoring of Somatosensory Evoked Potentials ......................................................... 1258 Monitoring of Visual Evoked Potentials ....................................................................... 145References to Section II ....................................................................................................... 147

SECTION III: MOTOR SYSTEMS

9 Anatomy and Physiology of Motor Systems ............................................................... 15710 Practical Aspects of Monitoring Spinal Motor Systems ............................................... 17911 Practical Aspects of Monitoring Cranial Motor Nerves ............................................... 197References to Section III ...................................................................................................... 213

SECTION IV: PERIPHERAL NERVES

12 Anatomy and Physiology of Peripheral Nerves ........................................................... 22113 Practical Aspects of Monitoring Peripheral Nerves ..................................................... 229References to Section IV ..................................................................................................... 233

SECTION V: INTRAOPERATIVE RECORDINGS THAT CAN GUIDE THE SURGEON IN THE OPERATION

14 Identification of Specific Neural Tissue ....................................................................... 23715 Intraoperative Diagnosis and Guide in Operations ..................................................... 251References to Section V ...................................................................................................... 273

SECTION VI: PRACTICAL ASPECTS OF ELECTROPHYSIOLOGICAL RECORDING IN THE OPERATING ROOM

16 Anesthesia and Its Constraints in Monitoring Motor and Sensory Systems ................. 27917 General Considerations About Intraoperative Neurophysiological Monitoring .......... 28318 Equipment, Recording Techniques, Data Analysis, and Stimulation ........................... 29919 Evaluating the Benefits of Intraoperative Neurophysiological Monitoring .................. 329References to Section VI ..................................................................................................... 339

Appendix ............................................................................................................................. 343Abbreviations ...................................................................................................................... 347Index ................................................................................................................................... 349

x Contents

INTRODUCTION

Because anesthesia could affect the results ofintraoperative monitoring, it is important that theperson who is performing the intraoperativeneurophysiological monitoring understand thebasic principles of anesthesia. The person who isresponsible for monitoring should communicatewith the anesthesiologist to obtain informationregarding the type of anesthesia that is to be used,if there are changes made in the anesthesia dur-ing the operation, and, if so, what other drugsmight be administered during the operation.

Maintaining a stable level of anesthesia isimportant and administration of drugs should beby continuous infusion; bolus administrationshould be avoided. The effect of anesthesia onspecific kinds of monitoring has been discussedin the preceding chapters. In this chapter, wewill discuss the various types of anesthesia mostcommonly used in connection with operationswhere intraoperative neurophysiological moni-toring of motor and sensory systems are used(for details about anesthesia in neurosurgery, seeref. 1. The classical text is ref. 2).

BASIC PRINCIPLES OF ANESTHESIA

The two primary purposes of general anes-thesia are to make the patient unconscious and to

provide analgesia (freedom from pain). A thirdpurpose is to keep the patient muscle relaxed,thus keeping the patient from moving during theoperation. In the Western world, general anes-thesia is predominantly accomplished by admin-istering pharmacological agents using either aninhalation or intravenous delivery method. Twoor more agents are often used together for addi-tive or (synergistic) action to achieve one of theanesthesia goals, as well as to reduce the sideeffects from a particular agent.

Different Kinds of AnesthesiaAnesthesia agents used in connection with

common operations can be divided into inhala-tion and intravenous anesthesia types. Often acombination of these two types is used. Morerecently, total intravenous anesthesia (TIVA) haswon popularity.

Inhalation AnesthesiaInhalation anesthesia is the oldest form of

general anesthesia. In its modern forms, it usu-ally consists of at least two different agents, sucha nitrous oxide and a halogenated agent, admin-istered together with pure oxygen. The relativepotency of inhalation agents is described bytheir MAC1 value.

Halogenated agents such as halothane(which is used rarely now), enflurane, isoflurane,

IntroductionBasic Principles of AnesthesiaEffects of Anesthesia on Recording Neuroelectrical Potentials

16Anes the s i a and I t s Cons t ra i n t s i n Mon i to r i ngMoto r and Sen so ry Sy s t ems

279

From: Intraoperative Neurophysiological Monitoring: Second EditionBy A. R. Møller © Humana Press Inc., Totowa, NJ.

1One MAC (minimal end-alveolar concentration) isthe equivalent of the sum of the effect of the anesthet-ics administered that prevent a response to painfulstimuli in 50% of individuals.

and so forth will cause increased central conduc-tion time (CCT) for somatosensory evokedpotentials (SSEPs) and essentially make itimpossible to elicit motor evoked potentials bysingle-impulse stimulation of the motor cortex(transcranial magnetic or electrical stimula-tion). This unfortunate effect is present even atlow concentrations.

Intravenous AnesthesiaSome intravenous agents have almost always

been used together with inhalational agents,but, recently, the TIVA regimen has becomeincreasingly prevalent. One reason for that isthat the inhalational agents, including nitrousoxide, are obstacles when electromyographic(EMG) responses are to be monitored in con-nection with transcranial stimulation of themotor cortex. It is an advantage that the mech-anism of action of intravenous agents appearsto be different from that of inhalational agentsin such a way that benefits monitoring EMGand of MEPs (see Chap. 10).

Analgesia. Achieving analgesia (pain relief)is a primary component of anesthesia, and formany years, opioids have been used in theanesthesia regimen together with agents suchas inhalation agents for achieving unconscious-ness (3). One of the oldest synthetic opioids isfentanyl, but now several different agents withsimilar action are in use for that purpose, suchas alfentanil, sufentanil, and remifentanil. Mus-cle responses evoked by transcranial corticalstimulation (electrical and magnetic) are onlyslightly affected by opioids. The effects of opi-oids can be reversed by administering nalox-one, suggesting that the effect is related toµ-receptor activity. Intravenous sedative agentsare frequently used to induce or supplementgeneral anesthesia, particularly with opioidsor ketamine, when inhalational agents are notutilized.

Ketamine is a valuable component of anes-thetic techniques allowing recording responsesthat might be depressed by other anesthetics.Ketamine could heighten synaptic functionrather than depress it (probably through its

interaction with the NMDA receptor) and it couldprovoke seizure activity in individuals withepilepsy but not in normal individuals. Ketaminehas been reported to increase cortical somatosen-sory evoked potential (SSEP) amplitude and toincrease the amplitude of muscle and spinalrecorded responses following spinal stimulationand it could potentate the H reflex. Ketamine hasminimal effects on muscle responses evoked bytranscranial cortical stimulation. Because of that,ketamine combined with opioids has become avaluable adjunct during some TIVA techniquesfor recording muscle responses. The fact that ket-amine could cause severe hallucinations post-operatively and increase intracranial pressure hasreduced its use in anesthesia.

Opioids provide analgesia but do not pro-vide sufficient degrees of sedation, relief ofanxiety, and loss of memory during operations(amnesia). Hence, TIVA usually includes somesedative–hypnotic agents such as barbiturates(thiopental) and benzodiazepines such as mida-zolam. Propofol is an agent that is in increasinguse because it provides excellent anesthesiaand limited effect on MEPs.

Barbiturates that are often used for inductionof general anesthesia have effects similar tothat of inhalation agents on evoked potentials.For example, muscle responses to transcranialstimulation are unusually sensitive to barbitu-rates and the effect lasts a long time, makingbarbiturates a poor choice in connection withmonitoring MEPs.

Etomidate is another popular agent to beused in intravenous anesthesia. It enhancessynaptic activity at low doses; thus, opposite tothe action of barbiturates and benzodiazepines,it might produce seizures in patients withepilepsy when given in low doses (0.1 mg/kg)and it might produce myoclonic activity atinduction of anesthesia. The ability to enhanceneural activity or reduce the depressant effectsof other drugs has been used to enhance theamplitude of both sensory and motor evokedresponses. The enhancing of evoked activityoccurs at doses similar to those that produce thedesired degree of sedation and loss of recall ofmemory when used in TIVA.

280 Intraoperative Neurophysiological Monitoring

Benzodiazepines, notably midazolam, areoften used in connection with TIVA in manykinds of operations because they provide excel-lent sedation and they suppress memories(recall). Benzodiazepines can also reduce therisk of hallucinations caused by ketamine.

Muscle RelaxantsMuscle relaxants are usually not regarded as

anesthetics but often combined with agents(intravenous or inhalation) that produce uncon-sciousness and freedom of pain. Muscle relax-ants are part of a common anesthesia regimen––so-called “balanced anesthesia” (neuroleptanesthesia)––that includes a strong narcotic foranalgesia plus a muscle relaxant to keep thepatient from moving, together with a relativelyweak anesthetic such as nitrous oxide.

Muscle relaxants used in anesthesia are of twodifferent types, each affecting muscle responsesdifferently: one blocks transmission in the neuro-muscular junction (muscle endplate) and theother type depolarizes the muscle endplate,thereby preventing it from activating the muscle.The oldest neuromuscular blocking agent iscurare, but that has been replaced by a longseries of steroid-type endplate blockers withdifferent action durations. Pancuronium bro-mide (Pavulon®) was one of the earliest of thisseries and the effects of pancuronium bromidelast more than 1 h when a dose that causes totalparalysis is administered. Other and newer drugsof the same family have a shorter duration ofaction (about 0.5 h for vecuronium bromide,[Norcuron®] and atracurium [Tracurium®]).

The most often used muscle-relaxing agentthat paralyzes by depolarizing the muscle end-plate is succinylcholine. The muscle-relaxingeffect of succinylcholine lasts only a very shorttime.

EFFECTS OF ANESTHESIA ON RECORDING NEUROELECTRICAL

POTENTIALS

Successful neurophysiological monitoringoften depends on the avoidance of certain types

of anesthetic agent; for instance, it is not possi-ble to record EMG potentials if the patient isparalyzed, as is the case for many commonlyused anesthesia regimens. Recording of corti-cal evoked potentials is affected by most of theagents commonly used in surgical anesthesia.Monitoring motor evoked responses elicited bytranscranial magnetic or electrical stimulationof the motor cortex requires special attentionon anesthesia and the use of a special anesthe-sia regimen is necessary.

Recording of Sensory Evoked PotentialsIt is advantageous to reduce the use of

halogenated agents and nitrous oxide in anes-thesia when cortical evoked potentials aremonitored. Monitoring of short-latency sen-sory evoked potentials is not noticeablyaffected by any type of inhalation anesthesia;therefore, short-latency sensory evoked poten-tials should be used whenever possible forintraoperative monitoring instead of corticalevoked potentials. Auditory brainstem responses(ABRs), which are short-latency evoked poten-tials, are practically unaffected by inhalationanesthetics and can be recorded regardless ofthe anesthesia used. Short-latency componentsof SSEPs are not affected by inhalation anes-thetics, but only upper limb SSEPs haveclearly recordable short-latency components.Short-latency SSEPs evoked by stimulation ofthe median nerve are suitable for monitoringthe brachial plexus and the cervical portion ofthe spinal cord, but they are not useful for mon-itoring the spinal cord below the C6 vertebra orfor monitoring central structures such as thesomatosensory cortex. Therefore, it is usuallythe long-latency components, which are gener-ated in the cortex, that are used for intraopera-tive monitoring of SSEP.

The general effect of anesthetics is a lower-ing of the amplitude and a prolongation of thelatency of an individual component of therecorded potentials (4) (see Chap. 7, Fig. 7.10).The effect is different for different componentsof the evoked potentials, as the potentials areaffected by inhalation anesthetics or barbitu-rates to varying degrees (5) and the effect varies

Chapter 16 Anesthesia 281

from patient to patient, with children being gen-erally more sensitive than adults (6).

Because these components are affected byinhalation anesthetics it is important to discusswith the anesthesiologists in order to select atype of anesthesia that allow such monitoring.

Recording of EMG PotentialsResponse from muscles (electromyographic

[EMG] potentials or mechanical response) can-not be recorded in the presence of musclerelaxants. It is usually necessary to use a mus-cle-relaxing agent for intubation. When EMGrecordings are to be done during an operation,it is suitable to use succinylcholine togetherwith 3 mg of d-tubocurarine (curare) or short-acting endplate blockers, such as atracurium(Tracurium) or vecuronium bromide (Norcuron)during intubation. This will allow monitoring ofmuscle potentials 30–45 min after the adminis-tration of the drug, providing that only the min-imal amount of the drug is given and that it isgiven only once for intubation.

If a short-acting endplate-blocking agent isused, it is important to be aware that the para-lyzing action disappears gradually and at a ratethat differs from patient to patient. The rate atwhich muscle function is regained depends onthe age, weight, and so forth of the patient, whatother diseases might be present, and what othermedications might have been administered.

During the time that the muscle-relaxingeffect is decreasing, stimulation of a motornerve with a train of electrical shocks (such asthe commonly used “train of four” test) willgive rise to a relatively normal muscle contrac-tion in response to the initial electrical stimu-lus, but the response to subsequent impulsesdecreases and will be less than normal.

The effect of muscle relaxants of the endplate-blocking type can be shortened (“reversed”) byadministering agents such as neostigmine, whichinhibits the breakdown of acetylcholine andthereby makes better use of the acetylcholinereceptor sites that are not blocked by the musclerelaxant that is used. However, a prerequisite for

the use of such “reversing” agents is that a fairamount of muscle response (10–20%) hasreturned before reversing is attempted. It is alsoimportant to note that such reversing does notimmediately return the muscle function to nor-mal, as the effect of the muscle relaxant will lastfor some time.

When muscle relaxation is not used duringan operation, the patient could have noticeablespontaneous muscle activity, which increasesthe background noise level in recordings of dif-ferent kinds of neuroelectrical potential. This isimportant when monitoring of evoked poten-tials of low amplitude, such as ABR, is to bedone. The resulting background noise will pro-long the time over which responses must beaveraged in order to obtain an interpretablerecording. The muscle activity often increasesas the level of anesthesia lessens. If the muscleactivity becomes strong, it might be a sign thatthe level of anesthesia is too low. Early infor-mation about such increases in muscle activityis naturally important to the anesthesiologist sothat he/she can adjust the level of anesthesiabefore the patient begins to move sponta-neously. In this way, electrophysiological mon-itoring can often provide valuable informationto the anesthesiologist, because if anesthesiabecomes light, spontaneous muscle activity fre-quently manifests in the recording of evokedpotentials from scalp electrodes a long timebefore any movement of the patient is noticed.To do that, the output of the physiological ampli-fier must be watched continuously to detect anymuscle activity.

Intraoperative monitoring that involvesrecording EMG potentials from muscles isbecoming more and more common in thecomplex neurosurgical operations that cannow be performed and demands on theselection of an appropriate anesthesia regimenhave, therefore, increased. A close collaborationbetween the anesthesia team and the neuro-physiologist in charge of intraoperativeneurophysiological monitoring can often solvesuch problems.

282 Intraoperative Neurophysiological Monitoring

A Practical Approach toNeurophysiologic

Intraoperative Monitoring

Edited by

Aatif M. Husain, MDDepartment of Medicine (Neurology)

Duke University Medical Center

Durham, North Carolina

New York

Husain 00 1/17/08 11:51 AM Page iii

Acquisitions Editor: R. Craig PercyCover Designer: Aimee DavisIndexer: Joann WoyCompositor: TypeWritingPrinter: Edwards Brothers Incorporated

Visit our website at www.demosmedpub.com

© 2008 Demos Medical Publishing, LLC. All rights reserved. This book is protected by copyright. No partof it may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, elec-tronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the pub-lisher.

Library of Congress Cataloging-in-Publication Data

A practical approach to neurophysiologic intraoperative monitoring / edited by Aatif M. Husain.p. ; cm.

Includes bibliographical references and index.ISBN-13: 978-1-933864-09-9 (pbk. : alk. paper)ISBN-10: 1-933864-09-5 (pbk. : alk. paper)1. Neurophysiologic monitoring. I. Husain, Aatif M.[DNLM: 1. Monitoring, Intraoperative—methods. 2. Evoked Potentials—physiology. 3.

Intraoperative Complications—prevention & control. 4. Trauma, Nervous System—prevention &control. WO 181 P895 2008]

RD52.N48P73 2008617.4'8—dc22

2008000450

Medicine is an ever-changing science undergoing continual development. Research and clinical experienceare continually expanding our knowledge, in particular our knowledge of proper treatment and drug ther-apy. The authors, editors, and publisher have made every effort to ensure that all information in this bookis in accordance with the state of knowledge at the time of production of the book.

Nevertheless, this does not imply or express any guarantee or responsibility on the part of the authors, edi-tors, or publisher with respect to any dosage instructions and forms of application stated in the book. Everyreader should examine carefully the package inserts accompanying each drug and check with a his physi-cian or specialist whether the dosage schedules mentioned therein or the contraindications stated by themanufacturer differ from the statements made in this book. Such examination is particularly importantwith drugs that are either rarely used or have been newly released on the market. Every dosage schedule orevery form of application used is entirely at the reader’s own risk and responsibility. The editors and pub-lisher welcome any reader to report to the publisher any discrepancies or inaccuracies noticed.

Made in the United States of America

08 09 10 5 4 3 2 1

Husain 00 1/22/08 11:17 AM Page iv

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ixPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiiContributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

I BASIC PRINCIPLES

1. Introduction to the Operating Room . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Kristine H. Ashton, Dharmen Shah, and Aatif M. Husain

2. Basic Neurophysiologic Intraoperative Monitoring Techniques . . . . . . . . . . 21Robert E. Minahan and Allen S. Mandir

3. Remote Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Ronald G. Emerson

4. Anesthetic Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55Michael L. James

5. Billing, Ethical, and Legal Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67Marc R. Nuwer

6. A Buyer’s Guide to Monitoring Equipment . . . . . . . . . . . . . . . . . . . . . . . . . 73Greg Niznik

II CLINICAL METHODS

7. Vertebral Column Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95David B. MacDonald, Mohammad Al-Enazi, and Zayed Al-Zayed

8. Spinal Cord Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117Thoru Yamada, Marjorie Tucker, and Aatif M. Husain

Contents

vii

Husain 00 1/17/08 11:51 AM Page vii

9. Lumbosacral Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139Neil R. Holland

10. Tethered Cord Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155Aatif M. Husain and Kristine H. Ashton

11. Selective Dorsal Rhizotomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169Daniel L. Menkes, Chi-Keung Kong, and D. Benjamin Kabakoff

12. Peripheral Nerve Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181Brian A. Crum, Jeffrey A. Strommen, and James A. Abbott

13. Cerebellopotine Angle Surgery: Microvascular Decompression . . . . . . . . . 195Cormac A. O’ Donovan and Scott Kuhn

14. Cerebellopotine Angle Surgery: Tumor . . . . . . . . . . . . . . . . . . . . . . . . . . . 213Dileep R. Nair and James R. Brooks

15. Thoracic Aortic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Aatif M. Husain, Kristine H. Ashton, and G. Chad Hughes

16. Carotid Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261Jehuda P. Sepkuty and Sergio Gutierrez

17. Epilepsy Surgery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283William O. Tatum, IV, Fernando L. Vale, and Kumar U. Anthony

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

v i i i • Contents

Husain 00 1/17/08 11:51 AM Page viii

The practice of anesthesia has histori-cally relied on the induction of a

reversible state of amnesia, analgesia, andmotionlessness. With the improvement of med-ical technology, advancement of knowledge,and practice of evidence-based medicine, mod-ern anesthesiology comprises a great deal more.It has become the role of the anesthesiologistduring surgical, obstetrical, and diagnostic pro-cedures to provide anesthesia, optimize proce-dural conditions, maintain homeostasis, and,should it be necessary, manage cardiopul-monary resuscitation. Additionally, anesthesi-ology has found itself branching out intochronic and acute pain treatment as well as theintensive care unit. Obviously there has been anexpansion of expectations for the practice ofanesthesia over the last few decades; however,ultimately, anesthesiology is the practice ofmanipulating a patient’s neurologic system andphysiology to effect some beneficial end.

PRINCIPLES OF ANESTHESIA

There are four basic types of “anesthesia”:general anesthesia, regional anesthesia, localanesthesia, and sedation. For the purposes ofneurophysiologic intraoperative monitoring(NIOM), general anesthesia (the creation ofreversible coma) is nearly always required and

consists of four basic stages: premedication,induction, maintenance, and emergence. Priorto entering the operating suite, “premedica-tions” may be administered to prepare thepatient for the perioperative period. Usuallythis takes the form of mild sedation for anxiol-ysis, analgesics for preprocedural pain, antihy-pertensives, antiemetics for patients with ahigh likelihood of postoperative nausea andvomiting, antisialagogues to facilitate intuba-tion, etc. In the operating room the historicprinciples of anesthesia are still the foundationof practice, and analgesia (i.e., painlessness),amnesia (i.e., memory loss), motionlessness,and hemodynamic stability can be obtainedand maintained by a variety of means.Commonly, general anesthesia is inducedthrough the administration of a large bolusdose of an intravenous sedative-hypnotic (e.g.,propofol). A dose of intravenous opioid (e.g.,fentanyl) and a paralytic agent (e.g., vecuro-nium) may be given at this time as well to facil-itate endotracheal intubation. After induction,anesthesia maintenance usually consists ofsome amount of inhaled volatile anestheticagent (e.g., isoflurane) in a mix of oxygen andeither air or nitrous oxide and some dose ofintravenous opioid. The amount of volatileagent is quantified in terms of mean alveolarconcentration (MAC). MAC is expressed as apercentage of inhaled gas and is defined as the

4 Anesthetic Considerations

Michael L. James

55

Husain 04 1/17/08 11:55 AM Page 55

alveolar partial pressure of a gas at which 50%of patients will not move with a 1-cm abdom-inal surgical incision. However, in practice thenecessary amount of volatile agent is deter-mined by effect. It is during anesthesia mainte-nance that NIOM occurs (as does the surgicalprocedure). After the procedure is finished, theexpectation is that the anesthetic coma will becompletely reversible, and the patient mustemerge from anesthesia without experiencinglasting effects from the agents. Emergence isusually accomplished by reversing any residualneuromuscular blockade and allowing thepatient to eliminate volatile agent via breath-ing. Volatile anesthetic agents are minimallymetabolized and largely removed from thebody in the same manner they were intro-duced: ventilation.

In terms of NIOM, special considerationsfor general anesthesia are discussed later; how-ever, it is important for neurophysiologists andtechnologists to have a clear expectation of thestep-by-step nature whereby anesthetic andsurgical procedures are undertaken, and it isimportant to remember that the operatingroom is generally a highly active environmentwith people, monitors, equipment, and electri-cal cords all moving about at once. Any changein the NIOM may be due to many factors, notthe least of which is the surgical procedure, andevery attempt should be made to regain fadingor lost waveforms, as permanent loss may indi-cate postoperative impairment (1). Thereforethe entire process becomes most efficient wheneach individual in the room understands all thesteps, including those of every other individual,required to prepare for, perform, and enableemergence from a procedure in an environmentof open communication and respect for eachother’s responsibilities.

NONPHARMACOLOGIC FACTORS:ANESTHETIC CONSIDERATIONS

Physiologic function of the human bodyplays a major role in neuronal functioning; it

is arguably the most important factor, and agreat deal of human physiology is influencedby actions of the anesthesiologist. The mannerin which these physiologic functions aremanipulated often directly determines meas-urable neurophysiologic function. Further, itis reasonable to assume that physiologic func-tion determines, in large part, the survivabilityof nerves and their supporting structures.

Temperature

It is well established that temperatureplays a significant role in nerve function, espe-cially in the axon. Changes of a fraction of adegree can drastically alter latencies andamplitudes of neuronal potentials with corti-cal structures being more affected thanperipheral nerves (2). Relative hypothermiaproduces changes that invariably present asslowed latencies from slower nerve conduc-tion. In addition there are predictable, charac-teristic effects of profound hypothermia that,at least initially, begin with slowing to a deltafrequency (3). The opposite is true with rela-tive hyperthermia for both evoked potentials(EPs) and electroencephalograms (EEGs). It isimportant to note that regional temperaturechanges are invariably difficult to predict, fora variety of reasons. General anesthesia causesan overall cooling effect in the body core dueto peripheral vasodilatation, which is usuallyopposed by active surface warming andwarmed intravenous fluids. Additionally, coldand/or warm irrigants are nearly alwaysapplied to the surgical field. As a result, theextremities, brain, and spinal cord are beingheated or cooled depending on where they liein relation to warmed air blankets, intra-venous fluid lines, the surgical field, etc.Therefore, unless it is individually measured,the actual temperature of a given region isimpossible to know, but the potential effectsshould be kept in mind during the course ofmonitoring. It is very common for patients toexperience a decrease in core body tempera-ture for the first 15 minutes after anesthetic

56 • SECTION I : Bas ic Pr inc ip les

Husain 04 1/17/08 11:55 AM Page 56

induction. With active warming during theadministration of most anesthetic agents—unless the surgical procedure requires analternative strategy—the patient’s temperaturewill then be kept greater than 36°C by theanesthesiologist.

Blood Flow

Logic dictates that ischemic nerves do notfunction normally; therefore measurable neu-ral potentials would become abnormal. In factit has been demonstrated that somatosensoryevoked potentials (SEPs) can be lost whencerebral blood flow falls below 15mL/min/100 g (2). This can be assumed to betrue for the spinal cord and peripheral nervesas well. Unfortunately, it is difficult to actuallymeasure blood flow to any given structure, sosystemic blood pressure is often used as a sur-rogate. Furthermore, systemic blood flowdoes not necessarily dictate regional bloodflow, especially in the brain, which makes iteven more difficult to predict. Monitors arebecoming available that purport to quantifyregional blood flow (e.g., cerebral oximetry,microdialysis), but a discussion of these isbeyond the scope of this chapter. Essentiallythen, there are two main considerations forthe neurophysiologist: systemic hypotensionand decreased regional blood flow. When pro-found, systemic hypotension results in glob-ally reduced blood flow, which translates intotissue ischemia of varying degrees basedlargely on autoregulation. For example, dur-ing spinal surgery, controlled deliberatehypotension is often requested of the anesthe-siologist so as to assist in controlling bloodloss; however, surgical traction and hypoten-sion can aggravate each other with deleteriouseffects to the patient, and NIOM can assist indetermining the acceptable limit of systemichypotension (4). There are many examples ofcauses of decreased regional blood flow, andalmost all are due to some interruption inblood supply either due to compression fromsurgical instruments (intentionally or uninten-

tionally), patient positioning, tourniquets,vasospasm, vascular ligation, etc. Anecdotally,some have reported discovering incidentalulnar nerve ischemia secondary to compres-sion during routine monitoring for spinalfusion. When the compression was released,the nerve potentials returned to normal.

Ventilation

Optimal neural functioning depends onmaintenance of a homeostatic extracellularenvironment. Hypo- or hypercapnea can altercellular metabolism by changing the acid-basestatus of the individual. In general individualstolerate relatively profound acid-basederangements, especially upward trends inpH. Unless the pH of a patient drops below7.2, neuronal mechanisms are maintained.Additionally, there is a suggestion thatextremes in hypocarbia (< 20 mmHg partialpressure) can alter SEP monitoring (5).Alternatively, profound hypoxia is poorly tol-erated, especially in the surgical setting ofongoing blood loss and potential hypotension.

Hematology

Like hypoxia, profound anemia can con-tribute to neural dysfunction. Normally, ane-mia is well tolerated to levels of hemoglobinless than 7 g/dL. However, in the surgical set-ting of possible large volume blood loss,hypotension, and hypoxia, it is generallyaccepted that hemoglobin levels should bekept above 8 g/dL and may require optimizingat 10 g/dL. At approximately 10 g/dL ofhemoglobin, oxygen delivery appears to bemaximized and transfusion above this thresh-old does not appear to improve augmenta-tion. There are animal data that support thissupposition in SEP monitoring (6).

Intracranial Pressure

Increase in intracranial pressure is a rela-tively well documented cause of shifts in cor-

CHAPTER 4: Anesthet ic Considerat ions • 57

Husain 04 1/17/08 11:55 AM Page 57

tical responses of EPs and prolongation ofmotor evoked potentials (MEPs), presumablydue to compression of cortical structures.There is a pressure-related increase in latencyand decrease in amplitude of cortical SEPsand as intracranial pressure becomes patho-logic, uncal herniation occurs with subsequentloss of subcortical SEP responses and brain-stem auditory evoked potentials (BAEPs) (7).Alleviation of this pressure can return EPs tonormal.

Other Factors

Neuronal function depends on mainte-nance of a homeostatic intra- and extracellu-lar environment determined by potassium,calcium, and sodium concentrations. It is log-ical to assume that alteration in these concen-trations would result in dysfunction andpossible changes in measurable neuronalpotentials. The concentration of these ions islargely in the control of the anesthesiologist,and maintenance within ranges of normal val-ues is necessary. In addition, profound hyper-or hypoglycemia should be avoided, as eitherextreme can result in cellular dysfunction;although there is no evidence that they resultin intraoperative changes in NIOM, there aredata to suggest that both can lead to poor out-comes (8).

EFFECTS OF SPECIFIC ANESTHETIC AGENTS

In general the anesthesiologist and neuro-physiologist are constantly at odds in thatnearly all anesthetic agents, given in highenough doses, cause depression of NIOMpotentials. However, with open communica-tion and mutual understanding of each other’sactivities, NIOM can be successful with nearlyany anesthetic technique. The crucial conceptis that any change in either anesthetic orNIOM must be communicated to the team, sothat every person in the operating room is act-ing under appropriate assumptions.

Inhalation Agents

Despite being the oldest form of anesthe-sia, the exact mechanism of action of inhala-tion agents remains unclear. Inhalationanesthetics consist of two basic gases avail-able in the United States: halogenated agents(halothane, isoflurane, sevoflurane, desflu-rane) and nitrous oxide. Doses of gas aregiven as percentage of inhaled mixture, andeffective doses are expressed as some amountof MAC. As discussed before, one MAC of anagent is sufficient to prevent 50% of patientsfrom moving to the stimulation of surgicalincision (Tables 4.1 and 4.2).

58 • SECTION I : Bas ic Pr inc ip les

TABLE 4.1 Effects of Inhaled Agents on Evoked Potentials

BAEP SEP MEP

Agents Latency Amplitude Latency Amplitude Latency Amplitude

Desflurane Inc 0 Inc Dec Inc Dec

Enflurane Inc 0 Inc Inc Inc Dec

Halothane Inc 0 Inc Dec Inc Dec

Isoflurane Inc 0 Inc Dec Inc Dec

Sevoflurane Inc 0 Inc Dec Inc Dec

Nitrous oxide 0 Dec 0 Dec Inc Dec

Inc = increased; Dec = decreased; 0 = no change.

Husain 04 1/17/08 11:55 AM Page 58

Halogenated AgentsThe halogenated agents consist of the his-

toric agent halothane, which is still used inmost countries outside the United States, andthe modern agents consisting of isoflurane,sevoflurane, and desflurane. Each has its ownMAC, onset and offset times, and metabolismbased on the inherent properties of the gas.Their use results in a dose-related decrease inamplitude and slowing of latency of SEPs,with the least effect seen in peripheral andsubcortical responses (2). BAEPs are mini-mally affected by halogenated anesthetics atusual doses but can be ablated at high doses.

MEPs are enormously affected by the useof halogenated agents and can be entirelyablated even with doses of 0.5 MAC. Itappears that this effect occurs proximal to theanterior horn cell due to evidence that wavesrecorded distal to the anterior horn cell andproximal to the neuromuscular junctionremain recordable even at high doses of anes-thetic (9). MEP monitoring may also occurthrough spinal or epidural stimulation withminimal effect on recorded responses; how-

ever, cord stimulation results in stimulationof the sensory and motor pathways, andhalogenated gases preferentially block themotor responses (10). Therefore it is impor-tant to remember that NIOM utilizing spinalcord stimulation may not reliably monitormotor function in the presence of halo-genated gases. For this and reasons men-tioned above—namely, easy ablation whenMEP monitoring is essential—halogenatedgases should usually not be part of the anes-thetic regimen when using this modality.

The EEG is affected but usually withouthindrance to monitoring. All halogenatedanesthetics produce a frontal shift of therhythm predominance when used at inductiondoses (two to three times MAC doses). Thegases then produce a dose-dependent reduc-tion in frequency and amplitude. It is impor-tant to note that both isoflurane anddesflurane can produce burst suppression andelectrocerebral silence at clinical doses. Forpractical purposes, however, all halogenatedagents can be used for maintenance anesthesiawhen NIOM requires EEG monitoring.

CHAPTER 4: Anesthet ic Considerat ions • 59

TABLE 4.2 Effects of Anesthetics Agents on Electroencephalogram

INCREASED FREQUENCY SUPPRESSED

Barbiturate (low dose) Barbiturates (high dose)

Benzodiazepine Propofol (high dose)

Etomidate Benzodiazepine (high dose)

Propofol

Ketamine

Halogenated agents

(< 1 MAC)

INCREASED AMPLITUDE ELECTROCEREBRAL SILENCE

Barbiturate (moderate dose) Barbiturates

Etomidate Propofol

Opioid Etomidate

Halogenated agents Halogenated agents

(1–2 MAC) (> 2 MAC except halothane)

Inc = increased; Dec = decreased; 0 = no change.

Husain 04 1/17/08 11:55 AM Page 59

Nitrous OxideNitrous oxide is similar to halogenated

anesthetic agents and causes a dose-relateddecrease in amplitude and prolongation oflatency of cortical SEPs and ablation of MEPs.This effect seems somewhat limited in subcor-tical and peripheral potentials of the SEPs. Atequipotent doses to halogenated agents,nitrous oxide may, in fact, cause greater EPdepression (2). Additionally, nitrous oxide hassomewhat indeterminate effects on the EEGthat is highly dependent on other agents anddoses being used simultaneously. The effectson the EEG are not wholly predictable, butgenerally, there is frontally dominant high-fre-quency activity and posterior slowing. Despitethis, a frequent anesthetic technique used dur-ing NIOM is a “nitrous-narcotic” technique.The modern version of this technique consistsof a high-dose remifentanil infusion (0.2 to0.5 µg/kg/min) with 60% to 70% inhaledfraction of nitrous oxide. A high, but con-stant, amount of nitrous oxide is deliveredwith varying amounts of remifentanil basedon surgical stimulation. As long as the per-centage of inhaled nitrous oxide is held con-stant, this practice generally allows recordableresponses for most NIOM except transcranialMEPs, although even then 50% to 60%nitrous oxide may be used. The benefit ofusing nitrous oxide is that brain concentra-

tions vary rapidly with inhaled concentra-tions, so that if NIOM is problematic andneeds maximizing intraoperatively, discontin-uance of nitrous oxide will quickly result inthe its elimination from the brain and body.

Intravenous Agents

Intravenous anesthetic agents are gener-ally used to induce anesthesia and afterwardsto supplement inhalation maintenance anes-thesia. Most modern anesthetic techniquesconsist of a variety of agents, intravenous andinhaled; nearly always an intravenous opioidis administered to augment other agents foreither tracheal intubation at induction orintense surgical stimulation exceeding a stablemaintenance anesthesia. If halogenated agentsare contraindicated or NIOM becomes prob-lematic with their use, a complete anestheticcan consist of intravenous drugs, or totalintravenous anesthesia (TIVA). TIVA exists inmany forms. The most common regimen isbased on continuous propofol infusion andsupplementation with intravenous opioid.However, all manner of TIVAs have beendescribed, including the use of ketamine, bar-biturate, midazolam, dexmedetomidine, etc.,with drug selection depending on utilizingspecific attributes of an agent to effect a spe-cific outcome (Tables 4.2 and 4.3).

60 • SECTION I : Bas ic Pr inc ip les

TABLE 4.3 Effects of Intravenous Agents on Evoked Potentials

BAEP SEP MEP

Agents Latency Amplitude Latency Amplitude Latency Amplitude

Barbiturate

Low dose 0 0 0 0 Inc Dec

High dose Inc Dec Inc Dec Inc Dec

Benzodiazepine 0 0 Inc Dec Inc Dec

Opioid 0 0 Inc Dec 0 0

Etomidate 0 0 Inc Inc 0 0

Propofol Inc 0 Inc Dec Inc Dec

Ketamine Inc 0 Inc Inc 0 0

Inc = increased; Dec = decreased; 0 = no change.

Husain 04 1/17/08 11:55 AM Page 60

BarbituratesSome of the oldest intravenous anesthet-

ics include barbiturates (e.g., thiopental, pen-tobarbital, phenobarbital, methohexital).These drugs exist in alkaline salt solution andexert their mechanism of action at the GABAAreceptor. Of these, thiopental remains in com-mon use, in certain surgical cases, as an induc-tion agent and as a means of achievingneuroprotection through “burst suppression.”Additionally, methohexital is frequently usedto facilitate electroconvulsive therapy (ECT).However, much like halogenated agents, bar-biturates will produce EEG slowing and, athigher doses, burst suppression and electro-cerebral silence. There appears to be littleclass effect of barbiturates on SEPs, with eachagent producing somewhat different results.Thiopental produces transient decreases inamplitude and increases in latency with bolusdosing for induction, but phenobarbital pro-duces little effect until doses causing cardio-vascular collapse are reached (11). As withinhaled agents, SEP cortical potentials seemedto be most affected, with relative sparing ofsubcortical and peripheral responses. In con-trast, whether with low-dose continuous infu-sion or single-bolus dosing, MEP responsescan be entirely abolished with the use of bar-biturates. Any anesthetic given for a surgicalprocedure requiring MEP monitoring shouldexclude the use of barbiturates in any formunless their use (i.e., neuroprotection) super-sedes the benefit from MEP monitoring.

BenzodiazepinesMidazolam is a common intravenous

benzodiazepine used in preoperative areasprior to transfer to the operating suite.Benzodiazepines also have their site of actionat the GABA receptor and have the desirableeffects of amnesia, sedation, and anxiolysis. Ingeneral, single one-time doses of midazolamgiven prior to induction have little effect onNIOM during critical portions of the proce-dure. However, induction doses of midazolam(0.2 mg/kg) or continuous infusions of the

drug can slow SEP latencies and decreaseamplitudes (12). Furthermore, as with mostother anesthetics, even small doses of benzo-diazepines (1 to 2 mg) can lead to a markedreduction in MEP responses. However, owingto relatively rapid metabolism of singleadminstration, if small doses of midazolamare given preoperatively, their effects onNIOM are usually minimal. Of note, benzodi-azepines are anticonvulsants and will all pro-duce slowing of the EEG into the theta range;however, at small doses they create beta-rhythm predominance in frontal leads, whichis also seen with chronic oral administration.

PropofolPropofol remains one of the most com-

mon agents used for the induction of anesthe-sia and is the most common agent used formaintenance anesthesia during TIVA. It ispackaged in a lipid-soluble solution and itssite of action is also at the GABA receptor.Owing to rapid redistribution after dosing,propofol is easily titratable to the desiredeffect, which makes it very useful for TIVAtechniques. Induction doses of propofol (2 to5 mg/kg) cause amplitude depression of EEG,SEP, and MEP responses, as does high-dosecontinuous infusion (80 to 100 µg/kg/min).However, there is generally rapid recoveryafter termination if long infusion times (>8hours) are avoided (13). In recording SEPs orMEPs from the epidural space, there seems tobe limited effect of the drug on the EPs; thisseems to hold true for recordings from thescalp or peripheral muscle as well (14).Propofol is also notable as an agent that canproduce burst suppression and electrocerebralsilence on the EEG. Despite profound EEGsuppression at high dose, propofol retains itsrelatively quick termination, allowing for anawake, alert, and neurologically testablepatient at the end of a surgical procedure.

OpioidsIntravenous opioids represent a critical

mainstay in the practice of modern “balanced”

CHAPTER 4: Anesthet ic Considerat ions • 61

Husain 04 1/17/08 11:55 AM Page 61

anesthesia to control perioperative pain.Nearly all general anesthetics will have someform and dose of intravenous opioid as a cen-tral component. Intravenous opioids in currentuse during the perioperative period includemorphine, hydromorphone, fentanyl, alfen-tanil, sufentanil, and remifentanil; they areadministered for various indications and at awide variation in dosing regimens. All intra-venous opioids have almost no effect onNIOM even at very high doses, making themof essential importance during anesthesia forprocedures requiring NIOM. Even when givenin the epidural or intrathecal space, they haveminimal effect on EPs (2). It has been notedthat generous application of opioids can resultin improved MEP monitoring owing to thereduction of spontaneous muscle contractionand lowering of the MAC for other anestheticagents. With regards to the EEG, opioids pro-duce a mild slowing into the delta range with-out effect on amplitude. Opioids will notproduce burst suppression or an isoelectricEEG even at the highest doses. Of particularimportance, the development of remifentanilhas revolutionized opioid use in TIVA.Remifentanil is an ultra-short-acting opioidwith a half-life on the order of 5 minutesregardless of dose. This allows for very rapidtitration of analgesia with little or no effect onemergence times, thus permitting high-doseopioid TIVA to minimize the dose of an asso-ciated sedative-hypnotic.

KetamineKetamine is one of the older anesthetic

agents and has undergone a recent resurgenceof use owing to the finding that it helps toalleviate postoperative pain and chronic painstates. Ketamine influences a variety of recep-tors and has the unique characteristic amonganesthetic agents of enhancing EP responses,especially in the cortex and spinal cord (15).Whether given as single bolus at induction oras continuous infusion, ketamine can increaseEP amplitude in SEP, MEP, and BAEP record-ing, making it an attractive agent for use dur-

ing NIOM. Additionally, the use of ketaminecan produce larger amplitudes, with mildslowing into the theta range on the EEG, andthere is anecdotal evidence that ketaminemay be proconvulsant. The downside to ket-amine use (and the reason ketamine fell outof favor prior to the last 5 years) is the occur-rence of emergence delirium and dissociativehallucinations. Additionally, increase inintracranial pressure from enhanced cerebralblood flow due to ketamine makes it of lim-ited use in neurosurgical patients withintracranial hypertension as well as in someother patient populations. Ketamine has beenfound particularly useful as a low-dose infu-sion (10 to 20 µg/kg/min) to supplement apropofol/opioid TIVA technique in proce-dures that require anesthetic-sensitive NIOM(e.g., MEP). The addition of low-dose keta-mine to a propofol-based TIVA allows for asubstantial reduction in propofol infusiondoses and enhancement of EP responses whileminimizing the undesirable side effects of ket-amine. For procedures requiring NIOM tech-niques that are highly sensitive to the effectsof anesthetics (e.g., transcranial MEP), theuse of ketamine in the anesthetic armamen-tarium should be considered.

EtomidateEtomidate represents another contradic-

tion to the general rule that anesthetic agentscause EP depression. Induction doses and con-tinuous intravenous infusion enhance bothMEP and SEP recordings (16). Etomidate hasbeen used in the past as a component of TIVAduring procedures that require anesthetic-sen-sitive NIOM (e.g., transcranial MEPs).Etomidate is also somewhat contradictory inits EEG effects; at low doses it may be some-what proconvulsant, and it is occasionallyused for ECT or epilepsy surgery; although athigher doses it may produce burst suppres-sion. However, among its many unpleasantside effects, concerns have been raised regard-ing etomidate-induced adrenal suppression,which can occur with even single-bolus induc-

62 • SECTION I : Bas ic Pr inc ip les

Husain 04 1/17/08 11:55 AM Page 62

tion doses (0.2 to 0.5 mg/kg). Increased mor-tality has been seen with prolonged infusionof etomidate, mainly in the intensive care set-ting (17). Nevertheless, etomidate remainsvaluable in cases where NIOM responses aredifficult to obtain and otherwise may not berecordable.

DexmedetomidineDexmedetomidine is a relatively new

agent used in human anesthesia. This selectivealpha-2 agonist has seen widespread use inveterinary medicine and has found its wayinto intensive care units and operating roomsbecause of its desirable effects of sedation,analgesia, and sympatholysis without respira-tory depression. Though increasing ancedotalreports are emerging, there are limited data onthe effects of dexmedetomidine on NIOM;however, animal data suggest that there is lit-tle effect (18). It may be used as a low-doseinfusion (0.2 to 0.5 µg/kg/hr) to augment anyanesthetic technique, and it allows for the useof considerably less volatile or intravenousanesthesthesia or opioid. Its definitive role inanesthetic techniques for highly sensitiveNIOM remains to be determined.

Paralytics

Neuromuscular blockers exert their effectby blocking acetylcholine at the nicotinicreceptor in the neuromuscular junction. Theyhave no effect on monitoring modalities thatare not derived from muscle activity (e.g.,EEG, BAEPs, and SEPs). They will com-pletely negate MEP monitoring if intense neu-romuscular blockade is utilized. However,employing partial blockade will allow sub-stantial reduction in patient movement withtesting, improved surgical retraction, andfavorable MEP monitoring. There are manyways to monitor the amount of neuromuscu-lar blockade; the most common is the “trainof four” (TOF) technique. It consists of meas-uring muscle responses, or compound muscleaction potentials, after four 2-Hz peripheral

nerve stimuli. MEP monitoring is acceptablewhen neuromuscular blockade is maintainedat a TOF of two responses. In using MEPmonitoring, it is important for the neuro-physiologist and surgeon to know whetherthe patient is paralyzed. If the patient is notparalyzed, MEP stimulation must be done attimes when patient movement is acceptable.If the patient is paralyzed, there are likely tobe brief periods when MEP responses are notrecordable owing to intense paralysis; it isthen imperative to communicate when a neu-romuscular blocking agent is redosed.However, either practice, paralysis or not, isacceptable; the main principle is, again, effec-tive and open communication with all partiesin the surgical suite.

ANESTHETIC TECHNIQUES

A variety of anesthetic techniques areacceptable for use during NIOM; the type ofanesthetic should be tailored to the type ofNIOM and the requirements of the surgicalprocedure. There are, however, a few generalprinciples. First, the least amount of anes-thetic agent necessary should be utilized aslong as there is little possibility of awarenessor discomfort on the part of the patient. Theliberal use of opioids can allow for a signifi-cant decrement in MAC. Second, the morestable an anesthetic dose can remain for theduration of the case, the less likely that theanesthetic agent might be contributing tointraoperative changes in NIOM waveforms.Supplementation of baseline anesthetic drugswith opioids or less NIOM-offending agentscan be made at times of more intense surgicalstimulation. Overall, there are essentially fourclasses of NIOM based on how easily themonitoring technique is ablated by anestheticagents. As the relative sensitivity of NIOM toanesthesia increases, the anesthetic techniqueshould be adjusted to maximize the leastoffending agents. Each group and its anes-thetic implications are discussed below.

CHAPTER 4: Anesthet ic Considerat ions • 63

Husain 04 1/17/08 11:55 AM Page 63

Relative Insensitivity

NIOM that is relatively insensitive toanesthetic agents in general includes BAEPsand SEPs recorded from the epidural space.With these monitoring methods, nearly allanesthetic practices can be used with theunderstanding that the general objective is tomaintain a constant level of anesthesia supple-mented with intermittent opioid dosing tocontrol increased surgical stimulus. Of coursethe least amount of anesthetic necessary toensure amnesia and analgesia should be used.Generally all patients have baseline EPs, sothat once in the operating room, deviationfrom that baseline can be assessed. If needed,anesthetic level or technique can then beadjusted to refine NIOM recordings.

Sensitivity to Paralytics

Forms of NIOM that are sensitive to neu-romuscular blockade include all monitoringthat requires elicitation of muscle actionpotentials (i.e., electromyography, MEP,spinal reflex testing, etc.). For these cases, ifvery fine control of the amount of neuromus-cular blockade can be maintained throughvigilant monitoring and drug dosing, neuro-muscular blocking agents can be employed.Otherwise they should be entirely avoidedonce the patient has been intubated. In fact,there are some practices that utilize intraoper-ative neuromuscular blockade reversal whencritical monitoring periods approach. In gen-eral, with the exception of MEP recording,which is exquisitely sensitive to anesthetictechnique, other forms of anesthetic agentsare acceptable. For cases that rely on anunparalyzed patient, relatively “deep” anes-thesia (e.g., high doses of anesthetic agents)can be used to offset lack of patient paralysis,allowing optimal surgical conditions of immo-bility and relaxation while maintaining theintegrity of NIOM. However, the generalprinciple of stable, though relatively high,anesthetic dose should be maintained.

Sensitivity to Anesthetics withoutSensitivity to Paralysis

NIOM that is not negated by neuromuscu-lar blockade but is sensitive to anesthetic agentsincludes SEP monitoring. Care must be takento minimize offending anesthetic agents andoptimize non-anesthetic variables (i.e., temper-ature). Generally, volatile or intravenous anes-thesia is acceptable if relatively low doses aremaintained (0.5 MAC for anesthetic gases orless than 80 µg/kg/min of propofol). The use ofneuromuscular blockade in this situationallows for a modest decrement in anestheticdose, as patient movement and relaxation thenbecome improbable. However, care must betaken that anesthetic dose is not so low as topermit patient recall or discomfort.

Relative Sensitivity

The need for MEP monitoring can initiatesome of the more challenging anesthetic issues.Designing an anesthetic technique to optimizeMEP monitoring adds to an already complexsurgical procedure. A TIVA technique that lim-its the amount of sedative-hypnotic agent (i.e.,propofol, barbiturate) is usually required.Limiting the dose of sedative-hypnotic toallow for optimal response recording ofNIOM requires the use of a second agent, usu-ally opioid, to supplement and augment theanesthetic properties. For instance, using apropofol-based anesthetic requires the addi-tion of opioid, ketamine, or dexmedetomidineinfusion to allow a much smaller dose ofpropofol to be administered. Additionally, ifneuromuscular blockade is used, it must betightly controlled so that profound paralysisdoes not preclude MEP responses from themuscles. It is not uncommon for the patient tobe unparalyzed during critical monitoring por-tions of the procedure. Therefore the anesthe-siologist is often faced with an unparalyzedpatient, whose monitoring requires relativelylow doses of an anesthetic, and whose airwayand accessibility is often remote. One current

64 • SECTION I : Bas ic Pr inc ip les

Husain 04 1/17/08 11:55 AM Page 64

practice is to utilize high-dose remifentanilinfusion to supplement a low-dose propofol-ketamine based anesthetic. This allows verylow dose propofol (20 to 30 µg/kg/min), whichhas minimal effects on MEP responses, to beoffset by low-dose ketamine (10 to 20µg/kg/min), which enhances MEP responses,and an amount of remifentanil that keeps thepatient motionless and relaxed.

CONCLUSIONS

In developing an anesthetic plan, the typeof NIOM is often as important a considerationas the type of surgical procedure. The crucialfactor for a successful procedure is open andcandid communication between the operatingroom staff, neurophysiologist, anesthesiologist,and surgeon. The majority of problems withintraoperative monitoring arise when operatingroom communication does not allow for eachindividual to have a clear understanding of theactions of each of the other members. Wheneveryone involved in the procedure is knowl-edgeable about reasonable expectations andaware of the current situation, the patient bene-fits from an operating team that is poised andfluid in its execution. With that understanding,it is imperative for the neurophysiologist tounderstand the limitations produced by ananesthetic and for the anesthesiologist to under-stand the effects of certain medications on mon-itoring. Without that fundamental knowledge,there can be little coordinated activity betweenthe two parties, resulting in ineffective monitor-ing. However, with the knowledge of the basiceffect of a given anesthetic agent on monitoringmodalities, nearly any anesthetic technique canbe administered safely and effectively with alltypes of monitoring.

REFERENCES

1. James ML, Husain AM. Brainstem auditoryevoked potential monitoring: when is change

in wave V significant? Neurology 2005;65:1551–1555.

2. Sloan TB. Evoked potentials In: Albin MS, ed.A Textbook of Neuroanesthesia withNeurosurgical and Neuroscience Perspectives.New York: McGraw-Hill, 1997:221–276.

3. Stekker MM, Escherich A, Patterson T, et al.Effects of acute hypoxemia/ischemia on EEGand evoked responses at normothermia andhypothermia in humans. Med Sci Monit2002;8:CR223–CR228.

4. Dolan EJ, Transfeldt EE, Tator CH, et al. Theeffect of spinal distraction on regional spinalcord blood flow in cats. J Neurosurg 1980;53:756–764.

5. Grundy BL, Heros RC, Tung AS, et al.Intraoperative hypoxia detected by evokedpotential monitoring. Anesth Analg 1981;60:437–439.

6. Nagoa S, Roccaforte P, Moody RA. Theeffects of isovolemic hemodilution and reinfu-sion of packed erythrocytes on somatosensoryand visual evoked potentials. J Surg Res 1978;25:530–537.

7. Mackey JR, Hall JW III. Sensory evokedresponses in head injury. Central Nerv SystTrauma 1985;2:187–206.

8. McGirt MJ, Woodworth GF, Brooke BS, et al.Hyperglycemia independently increases therisk of perioperative stroke, myocardial infarc-tion, and death after carotid endarterectomy.Neurosurgery 2006;58:1066–1073.

9. Gugino LD, Aglio LS, Segal NE, et al. Use oftranscranial magnetic stimulation for monitor-ing spinal cord motor paths. Semin Spine Surg1997;9:315–336.

10. Deletis V. Intraoperative monitoring of thefunctional integrity of the motor pathways.Adv Neurol 1993;63:201–214.

11. Drummond JC, Todd MM, U HS. The effectsof high dose sodium thiopental on brainstemauditory and median nerve somatosensoryevoked responses in humans. Anesthesiology1985;63:249–254.

12. Sloan TB, Fugina ML, Toleikis JR. Effects ofmidazolam on median nerve somatosensoryevoked potentials. Br J Anaesth 1990;64:590–593.

13. Kalkman CJ, Drummond JC, Ribberrink AA.Effects of propofol, etomidate, midazolam,

CHAPTER 4: Anesthet ic Considerat ions • 65

Husain 04 1/17/08 11:55 AM Page 65

and fentanyl on motor evoked responsestotranscranial electrical or magnetic stimulationin humans. Anesthesiology 1992;76:502–509.

14. Angel BA, LeBeau F. A comparison of theeffects of propofol with other anestheticagents on the centripetal transmission of sen-sory information. Gen Pharmacol 1992;23:945–063.

15. Schubert A, Licina MG, Lineberry PJ. Theeffect of ketamine on human somatosensoryevoked potentials after high frequency repeti-tive electrical stimulation. ElectroencephalogrClin Neurophysiol 1998;108:175–181.

16. Kochs E, Treede RD, Schulte AM, Esch J.Increase in somatosensory evoked potentialsduring induction of anaesthesia with etomi-date. Anaesthetist 1986;35:359–364.

17. Ledingham IM, Watt I. Influence of sedationon mortality in critically ill multiple traumapatients. Lancet 1983;1:1270.

18. Li BH, Lohmann JS, Schuler HG, Cronin AJ.Preservation of the cortical somatosensory-evoked potential during dexmedetomidineinfusion in rates. Anesth Analg 2003;96:1150–1160.

66 • SECTION I : Bas ic Pr inc ip les

Husain 04 1/17/08 11:55 AM Page 66