Ebook Neurocritical care A guide to practical management Part 1

(BQ) Part 1 book Neurocritical care A guide to practical management presentation of content: Brain injury and dysfunction The critical role of primary management, monitoring the injured brain, the secondary management of traumatic brain injury, critical care management of subarachnoid hemorrhage, central nervous system infections, cervical spine injuries,...

Trang 1

www.ebook3000.com

Trang 2

Competency-Based Critical Care

www.ebook3000.com

Trang 3

Honorary Consultant Surgeon

Great Ormond Street Hospital for Children NHS Trust (GOSH)

London

UK

Other titles in this series

Renal Failure and Replacement Therapies

edited by Sara Blakeley

www.ebook3000.com

Trang 4

John P Adams • Dominic Bell • Justin McKinlay (eds.)

Neurocritical Care

A Guide to Practical

Management

www.ebook3000.com

Trang 5

John P Adams

The General Infirmary at Leeds

Great George Street

Leeds LS1 3EX

United Kingdom

John.Adams@leedsth.nhs.uk

Justin McKinlay

The General Infirmary at Leeds

Great George Street

Leeds LS1 3EX

United Kingdom

justin.mckinlay@leedsth.nhs.uk

Dominic Bell The General Infirmary at Leeds Great George Street

Leeds LS1 3EX United Kingdom dominic.bell@leedsth.nhs.uk

ISSN 1864-9998 e-ISSN 1865-3383

ISBN 978-1-84882-069-2 e-ISBN 978-1-84882-070-8

DOI 10.1007/978-1-84882-070-8

Springer London Dordrecht Heidelberg New York

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Control Number: 2009931330

Apart from any fair dealing for the purposes of research or private study, or criticism or review,

as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms of licenses issued by the Copyright Licensing Agency Enquiries concerning reproduction outside those terms should be sent to the publishers.

The use of registered names, trademarks, etc., in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore free for general use.

The publisher makes no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility or liability for any errors or omissions that may be made.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

www.ebook3000.com

Trang 6

John Adams dedicates this book to his wife Kate to compensate for neglect of his responsibilities as husband and father The families of his fellow editors did not specifically notice or comment and for this we are grateful.

www.ebook3000.com

Trang 7

Brain injury is a worldwide leading cause of mortality and morbidity and requires early and appropriate management to minimize these adverse sequelae Despite such needs, access to specialist centers is limited, forcing both immediate and secondary care of these patients onto generalist staff These responsibilities are made more problematical by differences in patient management between and even within specialist centers, due in part to an insufficient evidence-base for many interventions directed at brain injury.This book is borne out of the above observations and is targeted at emer-gency and acute medicine, anesthetic and general intensive care staff caring for brain injury of diverse etiology, or surgical teams responsible for the inpatient care of minor to moderate head trauma.

Although explaining the various facets of specialist care, the book is not intended to compete with texts directed at neurosciences staff, but aims to advise on optimal care in general hospitals, including criteria for transfer, by

a combination of narrative on pathophysiology, principles of care, templates for documentation, and highly specific algorithms for particular problems

It is intended that the content and structure can form the basis of guidelines and protocols that reflect the needs of individual units and that can be constantly refined Our ultimate goal is to promote informed, consistent, auditable, multidisciplinary care for this cohort of patients and we hope that this text contributes to that process

Preface

vii

www.ebook3000.com

Trang 8

We are indebted to our fellow authors who have not only made this book possible, but have approached the task with enthusiasm All understand and endorse the importance of clear, comprehensive, evidence-based, and con-sistent advice in the support of colleagues caring for these patients outside the regional center.

We are also grateful for the observations of colleagues responsible for the eventual rehabilitation of these patients, mainly that even minor reductions in neurological deficit by early and appropriate care, can have a significant impact

on quality of life, with proportional benefit not only for the patient, but family, health and social care institutions, and society These observations justify the book and warrant implementation of the contained principles

Finally, we thank Melissa Morton in the UK and Robin Lyon in New York for all their help and support in bringing this book to publication

Acknowledgments

ix

www.ebook3000.com

Trang 9

Chapter 1 Brain Injury and Dysfunction: The Critical

Role of Primary Management 1

M.D Dominic Bell

Chapter 2 Monitoring the Injured Brain 9

Simon Davies and Andrew Lindley

Chapter 3 The Secondary Management of Traumatic

Brain Injury 19

Dominic Bell and John P Adams

Chapter 4 Critical Care Management of Subarachnoid

Hemorrhage 33

Audrey C Quinn and Simon P Holbrook

Chapter 5 Central Nervous System Infections 43

Abigail Walker and Miles Denton

Chapter 6 Cervical Spine Injuries 51

John P Adams, Jake Timothy, and Justin McKinlay

Chapter 7 Recent Advances in the Management of Acute

Ischemic Stroke 61

Ahamad Hassan

Chapter 8 Seizures on the Adult Intensive Care Unit 69

Morgan Feely and Nicola Cooper

Chapter 9 Non-Neurological Complications of Brain Injury 77

John P Adams

Chapter 10 Acute Weakness in Intensive Care 89

Louise Barnes and Michael Vucevic

Chapter 11 Coma, Confusion, and Agitation in Intensive Care 97

Matthew Clark and Justin McKinlay

Contents

xi

www.ebook3000.com

Trang 10

Chapter 12 Death and Donation in Critical Care:

The Diagnosis of Brainstem Death 105

Paul G Murphy

Chapter 13 Death and Donation in Critical Care:

Management of Deceased Organ Donation 113

Paul G Murphy

Chapter 14 Imaging the Brain-Injured Patient 121

Tony Goddard and Kshitij Mankad

Chapter 15 Ethical Dilemmas Within Intensive Care 137

Trang 11

John P Adams

Leeds General Infirmary

Leeds Teaching Hospitals NHS Trust

Leeds

West Yorkshire LS1 3EX

UK

Louise Barnes

Hull Royal Infirmary

Hull and East Yorkshire Hospitals NHS Trust

Hull HU3 2JZ

UK

Dominic Bell

Leeds

UK

Matthew Clark

Department of Anesthetics and Intensive Care

Leeds

UK

Nicola Cooper

YorkNorth Yorkshire YO31 8HEUK

Miles Denton

Leeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds

West Yorkshire LS1 3EXUK

Morgan Feely

Department of NeurologyLeeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds General Infirmary

LeedsWest Yorkshire LS1 3EXUK

Tony Goddard

Department of NeuroradiologyLeeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds

Contributors

xiii

Trang 12

Ahamad Hassan

Department of Neurology

Leeds

UK

Simon Holbrook

Academic Unit of Anesthesia

St James’s University Hospital

Leeds

West Yorkshire LS9 7TF

UK

Andrew Lindley

Leeds

UK

Justin McKinlay

Department of Anaesthetics and Neurocritical Care

Audrey C Quinn

Leeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds

Jake Timothy

Department of NeurosurgeryLeeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds

Michael Vucevic

Department of AnestheticsLeeds General InfirmaryLeeds Teaching Hospitals NHS TrustLeeds

Abigail Walker

Department of AnesthesiaChristchurch HospitalChristchurch

CanterburyNZ

xiv

Trang 13

A/B/C Airway, breathing, circulation

APTT Activated partial thromboplastin timeBAL Bronchoalveolar lavage

CNS Central nervous systemCOAG Coagulation screenCPP Cerebral perfusion pressure (MAP-ICP)CRP C-reactive protein

FiO2 Fraction of inspired oxygen

ICU Intensive care unitINR International normalized ratio

xv

Trang 14

Glossary of Terms and Abbreviations

MRSA Methicillin-resistant Staphylococcus aureus

NEAD Non-epileptic Attack Disorder

NICE National Institute for health and Clinical Excellence

NSAID Non-steroidal anti-inflammatory drug

PaCO2 Partial pressure of carbondioxide (arterial blood)

PaO2 Partial pressure of oxygen (arterial blood)

PCR Polymerase chain reaction

PCWP Pulmonary capillary wedge pressure

PEEP Positive end-expiratory pressure

PbtO2 Partial pressure of brain tissue oxygen

PVS Persistent vegetative state

SaO2 Arterial oxygen saturation

WFNS World Federation of Neurosurgical Socities

xvi

Trang 15

Key Points

1 In traumatic brain injury, maintain mean

arte-rial (MAP) blood pressure >80 mmHg

2 Avoid hypoxia at all costs; keep PaO2 >13 kPa,

using PEEP if necessary

3 Keep PaCO2 4.5–5.0 kPa; hyperventilate only if

there are signs of impending brainstem

hernia-tion

4 Keep the neck in neutral position; always

con-sider the possibility of cervical spine injury

5 Maintain 15° head up position (as long as MAP

adequate)

6 Do not give mannitol if patient is hypotensive

Speak to a Regional Neurosurgical Center

be-fore giving additional doses

Introduction

The human brain, in structure and function,

rep-resents the pinnacle of biological evolution Even

the most rudimentary non-volitional role of

matching ventilation to demand or maintaining

homeostasis is phenomenally complex for an

organism vulnerable to disease or dysfunction of

the component tissues and organs, and more

par-ticularly when exposed to mechanical, chemical,

and thermal hazard as every environmental

extreme is challenged The coordination of

physi-cal movement, played out at the highest level in

sport and the performing arts, rightly warrants

recognition as a marker of complex neuronal

activity, but conventionally, as a form of gence, bows to the cognitive capacity of the human brain Numerical and literary skills, communica-tion, memory, and knowledge are entry-level cog-nitive skills, with man’s advances through understanding of both science and nature repre-senting a higher plane Reasoning and judgment, coupled with awareness of the needs of others and social skills arguably constitute the highest form

intelli-of human intelligence Interlinked with this tion are those characteristics of personality and emotional status which generate individual uniqueness These may be reflected in our achieve-ments, as in career choice, or functional and artis-tic creativity, or our behavior relating to those achievements, as in innovation, ambition, and leadership These higher functions also have an emotional dimension covering conscience, charity and self-sacrifice, enthusiasm, and the ability to love, rejoice and grieve

func-This refinement and complexity of normal cerebral function is, however, associated with certain inherent vulnerabilities carrying signifi-cant implications for the management of either primary or secondary brain pathology or dysfunc-tion Tissues such as bone are able to regain normal architecture after injury, complex organs such as the liver and kidney are able to regenerate with resto-ration of original levels of function, and heart, lung, and pancreas are able to withstand devascularization and subsequent transplantation The specializa-tion of cellular structure and function within the central nervous system, however, appears to exclude a capacity for repair and renewal after anything other than the most trivial insult Brain

1

Brain Injury and Dysfunction:

The Critical Role of Primary Management

M.D Dominic Bell

1

Trang 16

M.D.D Bell

tissue has a high requirement for oxygen and

energy substrates to maintain both structure and

function, leaving little reserve in the face of

impaired delivery Even with normal arterial

oxygen content, circulatory arrest will result in loss

of consciousness within 15 seconds, and given the

high oxygen requirements simply to maintain

cel-lular integrity, more than 5 minutes of circulatory

arrest at normothermia will result in neuronal

death and a significant multifaceted neurological

deficit These aspects demonstrate the exquisite

vulnerability of the brain to the so-called

secon-dary cerebral insults, with cellular hypoxia being

the commonest final pathway

There is a gradient of sensitivity of the different

neural tissues to a global insult such as hypoxia,

whereby the loss of higher function precedes loss

of motor activity, with ventilatory effort

main-tained until immediately prior to death This

pattern parallels the picture of recovery from such

an insult, the extreme end of the spectrum being

the persistent vegetative state, where the patient is

self-ventilating, but has no awareness of

environ-ment or self This demonstrates that survival alone

cannot be considered a satisfactory outcome from

brain injury, and that all effort must be directed

toward preventing, where possible, even the most

subtle changes to personality and cognitive

func-tion at the other end of the spectrum, that would

require the skills of a clinical psychologist to

objectively quantify Failure to address these

aspects results not only in significant disability for

patient and family, but phenomenal burden and

cost to society

This edition of the series, devoted to

neurocriti-cal care and the prevention or minimization of

such avoidable neurological deficit, examines the

theory and evidence-base behind the various

man-agement strategies expected of a regional unit The

secondary aim is to define and promote principles

of care that can be deployed by any discipline, at

any level of seniority, at any location, at any time,

for any patient, with any pathology, and at any

stage Such principles, both clinical and procedural,

are essential, given that most neuropathology

arises outside the setting of a specialist center, and

many patients will not access that center, either

because neurosurgical intervention is not required,

other injuries require immediate management, or

because of limited bed availability

Given the vulnerability of the brain as outlined

earlier, it is unacceptable if the patient accrues

additional avoidable morbidity in these

circum-stances, or indeed while awaiting or during

trans-fer to the regional unit, through ignorance Clinical experience also highlights how patient care can be compromised due to a lack of clarity and consis-tency in the referral process and acceptance by the regional unit, resulting in a hiatus in care with neither party taking full responsibility for these aspects Such a scenario is arguably more unac-ceptable than ignorance, and demands explicit policy from the center and audit of process to monitor compliance

Role of the Regional Neurosurgical Center (RNC)

Fundamental to optimal patient management and any relationship with the regional center is an understanding of the specific services provided there The greatest demand will be for care of trau-matic brain injury, followed by subarachnoid hemorrhage, but the centers also have an emerg-ing involvement in conventional “strokes.” Throm-bolysis or interventional radiology for an ischemic infarct are being increasingly adopted as appro-priate emergency care, mirroring the approach taken to occlusive coronary events The implica-tions of managing these patients as medical emer-gencies cannot be overestimated, but the care and cost implications of the current conservative eval-uative approach to strokes are significant, regard-less of the impact on the patient and family.The role of the regional center for this range of pathology can be summarized as intervention, neuro-specific monitoring, and advice for referring units Given that vascular pathology is addressed in subsequent chapters, the role is only outlined in greater depth, as below, for traumatic brain injury:

1 To expedite removal of a significant nial hematoma

intracra-2 To monitor for the potential expansion of a less significant hematoma

3 To provide specialized monitoring (e.g., ranial pressure, jugular venous oximetry) to direct the neuro-intensive care of the diffuse axonal injury

intrac-4 To undertake radical surgical maneuvers for refractory intracranial hypertension, e.g., de-compressive craniectomy or lobectomy for ex-tensive contusion

Although it could be argued that a patient should be transferred to a specialized unit for

2

Trang 17

1 Brain Injury and Dysfunction: The Critical Role of Primary Management

imaging and assessment of the patient to make the

above distinctions, CT scanning in the referring

hospital has reduced the necessity for this and

digital image transfer should improve the quality

of discussion and decision-making Furthermore,

it is clearly not in the interest of the majority of

patients to be transferred for the sole purpose

of diagnosis

Indications for Patient Transfer

· Group 1: Transfer delayed only for correction of

secondary cerebral insults or for life-saving

surgery (e.g., expanding extradural hematoma

with localizing signs)

· Group 2: Requires urgent transfer following

optimization and life and limb saving surgery

(e.g., subdural hematoma with no mass

effect)

· Group 3(a): Patients should only be transferred

after absolute stabilization given that the overall

principles of care are to avoid secondary

cere-bral insults, rather than to offer neuro-specific

therapies (e.g., contusional injury with no mass

effect)

· Group 3(b): Some non-neurosurgical intensive

care units (ICU) monitor ICP in cases of diffuse

axonal injury; transfer may become necessary

if the ICP subsequently becomes difficult to

control

Organizing the Response

Groups 1-3(a) above demonstrate the importance

of the primary decision-making which involves

diagnostic skills, confident liaison with the

regional center, and an appropriate level of care in

the event of retention of the patient This

respon-sibility usually falls to the attending anesthetist or

intensive care specialist following initial

stabiliza-tion in the emergency department This individual

has a pivotal role in coordinating this process and

therefore assumes both clinical and logistical

responsibilities (see Table 1.1)

Avoidance of Secondary Cerebral Insults

No treatment strategy can reverse neuronal death

caused by the primary brain injury, but much can

be done to avoid preventable secondary neuronal death and subsequent deficit These secondary insults share a final common pathway that takes areas of the brain compromised by the primary injury, or indeed the whole brain, toward irrevers-ible ischemia (see Fig 1.1)

Secondary cerebral insults can be triggered by intracranial or systemic factors, which either reduce cerebral oxygen delivery or increase cere-bral oxygen consumption (Table 1.2) In addition,

an increase in the volume of brain, blood, or CSF,

or an expanding space occupying lesion (e.g., hematoma) may increase the pressure within the rigid skull and trigger global ischemia Focal damage may be caused by local compression or shearing forces

Cerebral Oxygen DeliveryCerebral oxygen delivery depends upon:

(a) An adequate circulating volume at a perfusion pressure above the lower level of cerebral autoregulation

(b) An adequate amount of oxygenated globin that dissociates appropriately at tissue level

hemo-Cerebral Oxygen Consumption

To avoid excessive cerebral oxygen consumption

in the context of compromised cerebral oxygen

T able 1.1 Roles of the attending specialist during the primary management of patients with traumatic brain injury

1 Primary resuscitation

2 Neurological assessment

3 Deciding on the need for intubation, sedation and ventilatory support

4 Management of problems such as convulsions

5 Interpretation of CT scans adequate for prioritization of treatment options

6 Prioritizing and expediting essential general surgical and orthopedic interventions

7 Deciding on transfer or retention after such interventions

8 Maintaining neurological observations

9 Avoiding secondary cerebral insults or expansion of any intracranial pathology

10 Organizing further CT scans in the event of retaining a patient

11 Maintaining dialog with the neurosurgeons and the neurosurgical intensive care

12 Deciding, in the face of massive injury, that no overall benefit from transfer exists

3

Trang 18

M.D.D Bell

delivery, it is essential to recognize and actively

treat any seizure activity and to provide adequate

analgesia and sedation, once a patient is intubated

and ventilated Pyrexia should be treated with

active cooling measures once the patient is

stabi-lized on the ICU Hyperglycemia, which is believed

to increase cerebral oxygen consumption, should

be targeted during all epochs of care

Expansion of Intracranial Contents

(a) Space-Occupying Lesions, for example,

he-matomata or contusions

The key priority is to determine whether urgent

neurosurgery is required General supportive

care includes avoidance of aspects that allow a

hematoma to expand through loss or dilution of

platelets or coagulation factors Hypothermia,

hypocalcemia, and administration of large

volumes of colloid solutions should be avoided

These aspects assume greatest significance

in the context of a subdural or intracranial hematoma, where such attention may avoid the need for surgical intervention

(b) Brain Edema – Four Mechanisms:

1 Hydrostatic edema: occurs when arterial

pressure exceeds the upper limit of regulation or when there is venous congestion

auto-(head-down position, pressure on the jugular veins, high intrathoracic pressure)

2 Osmotic edema: non-ionic crystalloid solutions such as dextrose become, in effect,

free water once the sugar component is metabolized

3 Oncotic edema: due to low plasma proteins;

can become important when the blood–brain barrier (BBB) is damaged

4 Inflammatory edema: the inflammatory

response to insults such as trauma or poxia can lead to increased capillary per-meability and disruption of the BBB It is critically important to avoid preventable insults such as osmotic edema when this has arisen

hy-The management of cerebral edema and raised intracranial pressure traditionally involves admin-istration of mannitol This can only be effective if the BBB is intact, there is mass rapid movement of water from the tissues into the circulating com-partment, and finally rapid excretion via the kidneys of both mannitol and water The main role of mannitol is to temporarily reduce the

amount of brain water and thereby reduce overall intracranial pressures and relieve pressure on vital structures such as the brainstem This buys time before definitive neurosurgical intervention By reducing the size of normal brain, abnormal areas including hematomata can expand, generating a greater shearing effect If mannitol is used indis-criminately with a deranged BBB, the molecule can diffuse across and ultimately contribute to the development of osmotic edema This is more likely

to occur with hypotension and poor renal fusion such that the mannitol is not excreted

per-Increase in Cerebral Blood Volume

1 Arterial: ↑PaCO2 is the commonest avoidable cause of cerebral arterial vasodilatation

2 Venous: discussed earlier, for example, neck

po-sitioning, endotracheal tube ties

ACID PRODUCTION OSMOTIC

PRESSURE

MEMBRANE DYSFUNCTION

OXYGEN REQUIREMENTS

EXCITATORY NEURO-TRANSMITTERS

APOPTOSIS

F igure 1.1 Mechanism of ischemia in brain injury.

T able 1.2 Intracranial and systemic causes of secondary brain

injury

Expanding contusion/hematoma Hypotension

Vascular injury/carotid dissection Hypo or Hypercapnia

Trang 19

Cerebrospinal Fluid

The ventricular system and contained CSF are

usually capable of reducing in size to accommodate

brain edema without causing a rise in intracranial

pressure Pathologies such as subarachnoid

hemorrhage and bacterial meningitis can cause

obstructive hydrocephalus This requires insertion

of a ventricular drain

Overall Management Strategy

Optimal patient care derives from an understanding

of the common pathologies that compromise brain

structure or function, and of the principles

under-pinning appropriate treatment options The key

goal of this edition is to demystify this area of

activ-ity and thereby empower clinicians caring for these

patients, particularly within the primary receiving

hospital, since it is in this setting that there is the

greatest opportunity for patient harm through act,

omission, or delay in accessing the regional center

The clinical aspects of care, both neuro-specific and

general, need to be formalized through protocols to

ensure consistency, regardless of grade or discipline

of attendant staff It is vital that the logistical aspects

of care be similarly formalized, namely

documenta-tion, particularly observation charts, investigations,

involvement of other disciplines, communication,

and any referral process to the regional

neurosurgi-cal center Only with such a structure will the right

things be done on the right patient, in the right order,

and at the right time The challenge for clinicians

working within a regional unit is to recognize the

fundamental importance of achieving these goals in

the referring hospital, and to actively promote and

support such a system The challenge for those

working in the referring hospital is to ensure that this

responsibility of the regional unit is discharged

Such goals and the system directed at these are

defined as “care bundles”: strategies to not only

optimize care based on the strongest available

evi-dence, but also to facilitate audit of process

Readers are referred to the appendices for

exam-ples of how the princiexam-ples are translated into explicit

recommendations for care within the author’s region,

with responsibility for dissemination and

imple-mentation resting with the local critical care network1

There is, however, still much to be done to eradicate

inconsistencies of care through ignorance and

limited formalization of process, as much as limited

availability in the regional centers It is hoped that

those readers who recognize the magnitude of the

problem will be stimulated by this edition to

confi-dently address those issues, which are so critical for patient care and professional satisfaction

Principles of Management

of Brain InjuryThe primary clinical management of any patient with

a brain injury, regardless of the diagnosis or severity, consists of routine resuscitation maneuvers and diag-nosing the nature and severity of both CNS and non-CNS pathology Consideration should always be given to the possibility of a lesion for which there is

a specific surgical or medical intervention, or interim supportive measures that can prevent that lesion gen-erating morbidity or mortality In the event of there being more than one pathology, clinical judgment has to determine the priorities of treatment.Running parallel to that clinical process is a logistical process, which incorporates aspects such

as teamwork, leadership, communication, zation, documentation, and timekeeping

prioriti-The Clinical Process

1 Resuscitation: as per ALS/ATLS guidelines.

2 Diagnosis: CNS pathology/non-CNS injury/

co-morbidity

Indications for a CT brain scan after head injury are outlined in Table 1.3 (see NICE Guideline 2007)

(a) CNS pathology: diagnosis, CT findings, severity

(GCS, pupils, focal neurology, seizures), trends, confounding variables (e.g., drugs, alcohol, hypotension, hypothermia).(b) Non-CNS pathology: remember the possi-

bility of spinal injury

3 Consideration of need for neurosurgical referral:

Use standardized form for transfer of information (see example, Appendix)

4 Neuro-specific observations/monitoring: Use a

standardized chart

5 Neuro-specific treatment: for example, mannitol

(see Table 1.4), hypertonic saline (HSL), anticonvulsants

6 Define priorities for treatment:

(a) Urgent transfer (b) Life or limb-saving surgery (c) General support and stabilization

1 http://www.wyccn.org.uk/CareBund.htm

5

Trang 20

The Logistical Process

1 Involve all relevant specialties

2 Determine team leadership

3 Establish documentation of observations

4 Ensure explicit communication:

(a) Internally within the team

(b) With key support specialties; radiology,

transfusion, pharmacy, etc

(c) With the regional neurosurgical center

5 Determine satisfactory timescale for:

(a) Diagnostic procedures

(b) Care/interventions

(c) Communication with neurosurgical center

(d) Transfer

(e) Re-evaluation of all aspects of care

6 Ensure documentation (using standardized templates where available) of:

(a) Observations (b) All above clinical undertakings (c) Criteria for transfer

(d) Results of discussion with regional center

7 Ensuring all appropriate support for any fer is available (functioning equipment, trained personnel, means of communication)

trans-8 Define criteria for stabilization prior to transfer

Avoidance of Secondary Cerebral Insults

1 Maintaining cerebral oxygen delivery (a) Adequate circulating volume: Aim for capil-

lary refill time <2 s and CVP>PEEP+5 with crystalloids (0.9%NaCl) up to 2 L followed

by a colloid (e.g., voluven, gelofusine) Give blood and clotting factors to maintain Hb

~10 g/dL or hematocrit 30, INR <1.2 and platelet count >100,000

(b) Adequate oxygenation: Maintain PaO2

>13 kPa with supplemental oxygen and PEEP if necessary Intubate and ventilate for GCS <8, primary ventilatory distur-bance or airway management problems Keep PaCO2 4.5–5.0 kPa Insert orogastric tube after intubation to decompress the stomach and prevent gastric dilatation (c) Adequate perfusion pressure: Maintain MAP

>80 mmHg or within 15% of normal values

if normally hypertensive After volume resuscitation, vasopressors or inotropes may be required to maintain an adequate blood pressure, the choice depending upon the cardiovascular profile (see Appendix) Advanced monitoring (e.g., esophageal doppler, pulmonary artery catheter) may

be required to guide this process, cially if there is uncertainty about volume status

espe-2 Controlling cerebral oxygen consumption (a) Control seizure activity: Seizure activity is

usually treated with a benzodiazepine (e.g.,

T able 1.4 Indications for Mannitol

Unilateral pupillary dilatation, or unilateral progressing to bilateral

dilatation (primary bilateral dilatation may represent fitting, drug

intoxication or overdose, or overwhelming brain injury).

Dose: 0.5 g/kg (approximately 200 mL of 20% solution) over

10–15 min Can be repeated at 1–2 hourly intervals to maximum

serum osmolarity of 320 mosmol/L or Na + of 160 mmol/L Speak to

the Regional Neurosurgical Center prior to giving additional doses.

Alternative: Hypertonic saline (HSL) is being increasingly used for the

same purpose with good effect We use 30 mL of 20% HSL over

20 min via a CVC, with a similar serum [Na + ] cut-off of 160 mmol/L.

Precautions: Mannitol should not be given to patients who are

hypotensive or have evidence of inadequate renal perfusion All

patients require bladder catheterization.

T able 1.3 Indications for a CT brain scan after head injury

Trang 21

lorazepam 2–4 mg IV bolus) in the first

instance, followed by a longer-acting agent

(e.g., phenytoin 15 mg/kg over 20 min)

See Chap 8 for detailed description, and

the Appendix for the status epilepticus

algorithm

(b) Ensure adequate analgesia and sedation

if intubated: Use fentanyl or alfentanil by

infusion with propofol (midazolam can be

used if there is cardiovascular instability)

Maintain paralysis with infusion of muscle

relaxant (e.g., cisatracurium or vecuronium)

and monitor with a nerve stimulator

All head-injured patients require bladder

catheterization

3 Avoiding increases in intracranial pressure

(a) Avoid expansion of intracranial hematoma/

contusion: Maintain normal clotting and

platelet counts Monitor calcium in face of

massive transfusion Consider Factor VIIa

if intracranial hematoma or contusion in

the face of nonsurgical major hemorrhage

despite administration of platelets and

clotting factors (see Chap 3 for detailed

Br J Anaesth 93(6):761–762 Modernisation Agency/Department of Health (2004) The Neurosciences Critical Care Report London

www.dh.gov.uk/publications

NICE (2007) Head Injury: Triage, assessment, tions and early management of head injury in infants, children and adults London http://www.nice.org.uk/ nicemedia/pdf/CG56NICEGuideline.pdf

investiga-The Neuro Anaesthesia Society of Great Britain and Ireland and The Association of Anaesthetists of Great Britain and Ireland (2006) Recommendations for the Safe Transfer of Patients with Brain Injury London www.nasgbi.org.uk

7

Trang 22

Key Points

1 Repeated clinical assessment through the

Glas-gow Coma Scale (GCS) is the cornerstone of

neurological evaluation

2 Ventilated head-injured patients with

intracra-nial pathology on CT require ICP monitoring

3 Invasive or noninvasive neuro-specific

moni-toring requires careful interpretation when

as-sisting goal-directed therapies

4 Multimodal monitoring using a combination

of techniques can overcome some of the

limita-tions of individual methods

Neuro-Specific Monitoring

Accurate neurological assessment is fundamental

for the management of patients with intracranial

pathology This consists of repeated clinical

exam-ination (particularly GCS and pupillary response)

and the use of specific monitoring techniques,

including serial CT scans of the brain This chapter

provides an overview of the more common

moni-toring modalities found within the neuro-critical

care environment

In general terms, a combination of assessments

is more likely to detect change than one specific

modality Real-time continuous monitoring (e.g

ICP) will provide more timely warning about

adverse events (e.g., an expanding hematoma) as

compared to static assessments such as sedation

holds or serial CT brain scans

Clinical Assessment The Glasgow Coma ScaleThe Glasgow Coma Scale (GCS) provides a stand-ardized and internationally recognized method for evaluating a patient’s CNS function by record-ing their best response to verbal and physical stimuli A drop of two or more GCS points (or one

or more motor points) should prompt urgent re-evaluation and a repeat CT scan The GCS is described in detail in Chap 10

NB Eye opening is not synonymous with awareness, and can be seen in both coma and Per-sistent vegetative state(PVS) The important detail

is that the patients either open their eyes to command or fixes or follows a visual stimulus

Pupillary ResponseChanges in pupil size and reaction may provide useful additional information:

Sudden unilateral fixed pupil: Compression of

·the third nerve, e.g., ipsilateral uncal her niation

or posterior communicating artery aneurysmUnilateral miosis: Horner’s syndrome (consider

·vascular injury)Bilateral miosis: Narcotics, pontine hemor-

·rhageBilateral fixed, dilated pupils: Brainstem death,

·massive overdose (e.g tricyclic antidepressants)

2

Monitoring the Injured Brain

Simon Davies and Andrew Lindley

9

Trang 23

S Davies and A Lindley

In the non-specialist center, neurological

assess-ment of the ventilated patient consists of serial CT

brain scans, pupillary response, and assessment of

GCS during sedation holds A reduction in

seda-tion level will usually be at the suggesseda-tion of the

Regional Neurosurgical Center (RNC) and its

timing will depend upon a number of factors

Responses such as unilateral pupillary dilatation,

extensor posturing, seizures, or severe

hyper-tension should prompt rapid re-sedation, repeat CT

scan, and contact with the RNC In the patient

with multiple injuries, consideration must be

given to their analgesic requirements prior to any

decrease in sedation levels

Invasive Monitoring

Intracranial Pressure Monitoring

Cerebral perfusion pressure (CPP) reflects the

pressure gradient that drives cerebral blood flow

(CBF), and hence cerebral oxygen delivery

Meas-urement of intracranial pressure (ICP) allows

estimation of CPP

CPP = Mean Arterial Pressure − ICP

Sufficient CPP is needed to allow CBF to meet the

metabolic requirements of the brain An

inade-quate CPP may result in the failure of

autore-gulation of flow to meet metabolic needs whilst

an artificially induced high CPP may result in

hyperemia and vasogenic edema, thereby

wors-ening ICP The CPP needs to be assessed for each

individual and other monito ring modalities (e.g.,

jugular venous oximetry, brain tissue

oxygen-ation) may be required to assess its adequacy

Despite its almost universal acceptance, there

are no properly controlled trials demonstrating

improved outcome from either ICP- or CPP-targeted

therapy However, in the early 1990s Marmarou

et al showed that patients with ICP values

consi-stently greater than 20 mmHg suffered worse

outcomes than matched controls, and poorer

outcomes have been described in patients whose

CPP dropped below 60 mmHg (Juul 2000; Young

et al 2003) As such, ICP- and CPP-targeted therapy

have now become an accepted standard of care in

head injury management

The 2007 Brain Trauma Foundation Guidelines

(Brain Trauma Foundation 2007) recommend

treating ICP values above 20 mmHg and to target CPP in the range of 50–70 mmHg Patients with intact pressure autoregulation will tolerate higher CPP values Aggressive attempts to maintain CPP

>70 mmHg should be avoided because of the risk

of ARDS

Measuring ICP

· Intraventricular devices consist of a drain

inserted into the lateral ventricle via a burr hole, and connected to a pressure transducer, manometer, or fiber optic catheter This remains the gold standard but is associated with a higher incidence of infection and greater potential for brain injury during placement It has the added benefit of allowing CSF drainage Historically, saline could be injected to assess brain compliance

· Extraventricular systems are placed in

paren-chymal tissue, the subarachnoid space, or in the epidural space via a burr hole This can be inserted at the bedside in the ICU These systems are tipped with a transducer requiring calibra-tion, and are subject to drift (particularly after long-term placement) Examples of extraven-ticular systems are the Codman and Camino devices These devices have a tendency to underestimate ICP

In general, both types of device are left in situ for

as short a time as possible to minimize the risk of introducing infection Prophylactic antibiotics are not generally used

Indications for ICP monitoring

More specific indications:

Traumatic brain injury, in particular:

·Severe head injury (GCS <8)

·

ICP values Normal ICP <15 mmHg.

Focal ischemia occurs at ICP >20 mmHg Global ischemia occurs at ICP >50 mmHg Usual treatment threshold is 20 mmHg

Head injury + ventilator + abnormal CT brain scan = ICP monitor

10

Trang 24

2 Monitoring the Injured Brain

Focal pathology on CT brain scan

necessitate the use of sedation or anesthesia

Subarachnoid hemorrhage with associated

Coagulopathy is the primary contraindication to

insertion The ICP device will generally be removed

as soon as the patient is awake with satisfactory

neu-rology (GCS motor score M5 or M6) or when

physi-ological challenges (removal of sedation, normalizing

PaCO2) no longer produce a sustained rise in ICP

Intracranial Pressure Waveforms

and Analysis

The normal ICP waveform is a modified arterial

trace and consists of three characteristic peaks

The “percussive” P1 wave results from arterial

pressure being transmitted from the choroid

plexi, the “tidal” P2 wave varies with brain

compli-ance (fig 2.1), whilst P3 represents the dicrotic

notch and closure of the aortic valve It is

impor-tant to establish the accuracy of the ICP trace and

value before initiating therapy based upon the numbers generated Transient sequential occlu-sion of the internal jugular veins or removing the head-up tilt should produce an increase in ICP

In addition to simple pressure measurement, if ICP is recorded against time, a number of charac-teristic wave forms (Lundberg waves) can be seen

A-waves: Pathological sustained plateau waves of

50–80 mmHg lasting between 5 and 10 min, possibly representing cerebral vasodilatation and an increase

in CBF in response to a low CPP (Fig 2.2)

B-waves: Small, transient waves of limited

amplitude every 1–2 min representing fluctuations

in cerebral blood volume These may be seen in normal subjects, but are indicative of intracranial pathology when the amplitude increases above

10 mmHg (Fig 2.3)

Time

Compliant Brain P3

P2

P1

P3 Non-compliant Brain

P2

P1

F igure 2.1 ICP traces showing the three distinct peaks In the

non-compliant brain, the amplitude of P2 increases Reproduced

with kind permission from Anaesthesia UK (www.frca.co.uk).

F igure 2.2. Lundberg A waves Reproduced with kind permission

from Anaesthesia UK (www.frca.co.uk).

0

10 20 30 40

Time (minutes)

F igure 2.3. Lundberg B waves Reproduced with kind permission

from Anaesthesia UK (www.frca.co.uk).

11

Trang 25

C-waves: Small oscillations in ICP that reflect

changes in systemic arterial pressure

With cerebral autoregulation intact, a rise in

MAP produces vasoconstriction and a fall in ICP

However, when autoregulation fails, the

circula-tion becomes pressure passive and changes in

MAP are reflected in changes in the ICP

Continu-ous analysis of MAP and ICP allows a correlation

coefficient called the pressure reactivity index to

be derived (PRx) Positive values indicate

dis-turbed cerebral vascular reactivity, whilst negative

values indicate that reactivity remains intact

(Gupta 2002)

Despite the fact that trial results have not always

been compelling, most clinicians regard the ICP

monitor as an essential tool that allows estimation

of CPP (Czosnyka and Pickard 2004; Czosnyka

et al 1996), gives early warning of developing

pathology, allows the response to therapy to be

objectively measured, and also has value as a

prognostic indicator (Joseph 2005)

Jugular Venous Oximetry (SjvO 2 )

SjvO2 is an indicator of global oxygen extraction

of the brain Jugular venous desaturation

sug-gests an increase in cerebral oxygen extraction

which indirectly implies that there has been a

decrease in cerebral oxygen delivery, and hence

perfusion

The internal jugular vein drains the majority of

blood from the brain, and in most patients the

right lateral sinus is larger Despite the fact that

flow is different on the two sides, oxygen

satura-tions are normally very similar This also appears

to be the case in diffuse brain injury, whilst in

focal injuries there tends to be a greater difference

in the saturations of the two veins

Jugular venous saturations can be measured

using the principle of infrared refractometry via

a specially designed catheter (Gopinath et al

with increased cerebral lactate production

Insertion of Jugular Venous Saturation Catheter

Insertion involves retrograde cannulation of the internal jugular vein A pediatric pulmonary artery catheter introducer can be used through which the fiber optic SjvO2 catheter is advanced beyond the outlet of the common facial vein to the level of the jugular bulb at the base of the skull Ultrasound is often used for accurate identifica-tion of vein position to avoid arterial puncture, and to minimize the risk of hematoma formation which can in turn impede venous drainage Correct positioning is confirmed on a lateral neck X-ray with the catheter tip lying at the level of the mastoid air cells

Indications for SjvO2 Monitoring

· Acute brain injury An association between

SjvO2 desaturation and poor neurological out come has been observed Fandino showed that in traumatic head injury SjvO2 was the only factor associated with outcome, whilst Gopinath showed that multiple SjvO2 desaturations were associated with an increased incidence of poor neurological outcome compared to those who showed no desaturations (Moppett and Mahajan

2004)

· Monitoring of therapy response If ICP and SjvO2

are both raised, hyperemia is implied and ventilation is appropriate SjvO2 should be moni-tored and kept above 55% in these circumstances,

hyper-as excessive hyperventilation may cause found cerebral vasoconstriction and cerebral ischemia More recent work using PET scanning, however, has cast some doubt on the value of SjvO2, with hyperventilation appearing to increase ischemic brain volume without necessarily pro-ducing a fall in jugular venous saturation

pro-· To guide optimal blood pressure and P a CO 2 agement during operative treatment of aneu- rysms following SAH During the operative

man-treatment of an aneurysm, hypertension must be avoided because of the risk of rupture and bleeding However, excessive reductions in

Optimal CPP would appear to be at the point when further increases

in MAP do not lead to a rise in SjvO2.

12

Trang 26

blood pressure may risk cerebral ischemia,

especially in those patients with preoperative

hypertension SjvO2 monitoring allows the

anesthetist to assess the degree to which blood

pressure can be safely lowered during the

oper-ative period Similarly, a low PaCO2 will cause

SjvO2 desaturation

Problems with SjvO2 Monitoring

The major criticism of SjvO2 is that it is a measure

of global oxygen delivery and does not reflect

metabolic inadequacies in focal areas of injury,

and hence may miss regional areas of ischemia

Inaccuracies can occur with catheter

misplace-ment, contamination with extra cerebral blood,

when the catheter abuts the vessel wall, or if

thrombosis occurs around the catheter tip

Contraindications and complications are similar

to those of an IJV central line

Interpretation of Changes in SjvO2

If cerebral oxygen delivery is impaired, oxygen

extraction increases and SjvO2 decreases If

autoregulation is intact, CBF increases to meet

metabolic demand and SjvO2 is restored However,

in the injured brain autoregulation is often

impaired and cerebral ischemia ensues

· ↓SjvO 2 : This implies inadequate cerebral oxygen

delivery that may be due to decreased oxygen

delivery (systemic hypoxia, anemia), decreased

CBF (hypotension, raised ICP, excessive

hypoc-apnia or vasospasm), or increased cerebral oxygen

consumption (seizures, hyperthermia, pain)

· ↑SjvO 2: This is somewhat more difficult to

inter-pret, and may represent either hyperemia (e.g.,

when the autoregulation mechanisms are lost)

or reduced oxygen consumption (e.g., hypothermia,

deep sedation, or cerebral infarction)

· Lactate Oxygen Index: During cerebral

hypop-erfusion the brain can become a net producer

of lactate, with the jugular venous lactate rising

above arterial values The lactate oxygen index

is discussed in more detail in Chap 3

Brain Tissue Oximetry

Interest in measuring brain tissue oxygenation via

implantable sensors has grown in recent years

The Licox sensor is an implantable polarographic electrode that measures tissue oxygen tensions It

is inserted through a compatible bolt and ideally should be placed into the penumbral area of the injury Oxygen diffuses from the tissue through the catheter into an electrolyte chamber where an electrical current is generated Brain tissue oxygen tension is normally lower than arterial oxygen tension (15–50 mmHg), whilst tissue CO2 is normally higher (range 40–70 mmHg) The sensors are useful in monitoring local changes and trends

in tissue oxygenation that might be missed by SjvO2 measurements

At present it is primarily used in severe head injury and poor-grade subarachnoid hemor-rhage, and in conjunction with other monitoring moda lities The technique allows a continuous method of monitoring of regional tissue oxygen-ation and in particular, monitoring areas of high ischemic risk, and is a promising and reliable clinical tool

Noninvasive Monitoring

Transcranial Doppler Ultrasound

Transcranial Doppler is a noninvasive technique that calculates blood flow velocity in the cerebral vasculature An ultrasound beam is reflected back

by the moving blood stream at a different frequency than it was transmitted (Doppler shift), and from the Doppler equation the velocity of blood flow (FV) can be calculated Changes in FV correlate well with changes in CBF, as long as the orienta-tion of the transducer and the vessel diameter remain constant It is used clinically to diagnose vasospasm, to test cerebral autoregulation, and to detect emboli during cardiac surgery and carotid endarterectomy (Moppett and Mahajan 2004)

is also increased in hemodilutional states

13

Trang 27

Technique for Insonating the Middle Cerebral

Artery (MCA)

The M1 branch of the MCA is the commonest

vessel to be insonated, and is visualized through a

transtemporal window (Fig 2.4) with a 2 MHz

pulsed Doppler signal The anterior and posterior

cerebral arteries can also be accessed through this

window, whilst a transorbital approach allows

access to the carotid siphon and the suboccipital

route to the basilar and vertebral arteries

Analysis of Doppler waveform

Analysis of the Doppler waveform gives rise to

useful derived variables as well as blood velocity

information

Pulsatility Index (PI): FV

value: 0.6–1.1)

This reflects distal cerebrovascular resistance

and correlates with CPP

Change in CBF with arterial CO

(cerebral vascular reactivity)

Uses of TCD in Intensive Care

Head Injury

Three distinct phases have been shown in severe

head injury with regard to CBF and MCA FV

Phase 1 occurs on the day of injury and has a

·

normal CBF, normal MCA FV, and normal AVDO2

Phase 2 occurring 1–2 days post-injury, a

hyper-·

emic state is encountered with an increased

CBF, MCA FV and decreased AVDO2

The final phase seen at days 4–15 is the

vasos-·pastic phase and is associated with a signifi-cantly decreased CBF and increased MCA FV The use of TCD allows interpretation of the dynamic physiological changes seen in severe head injury, and in combination with other modalities allows perfusion and oxygenation

to be optimized for the individual patient.The highest MCA FV recorded at any stage had been shown to be an independent predictor

of outcome from head injury, and the loss of autoregulation (calculated by regression of CPP

on MCA FV) has also been shown to be a predictor

of poor outcome from head injury

Subarachnoid Hemorrhage

Vasospasm occurs in approximately 50% of people with subarachnoid hemorrhage between 2–17 days post-event, and is associated with significant morbidity and mortality TCD may be used to detect vasospasm by the increase in MCA FV associated with vessel narrowing Spasm is also assumed to be occurring when blood velocity is

>120 cm/s (see Fig 2.5a, b) High MCA FV is ciated with worse-grade SAH, larger blood loads

asso-on CT (assessed by Fischer Grade) and hence worse outcome (Steiger et al 1994)

Electroencephalography

An electroencephalogram (EEG) is obtained using the standardized system of electrode placement Practically, this is not often readily available and requires expert interpretation The EEG is affected

by anesthetic agents and physiological abnormalities such as hypoxia, hypoperfusion, and hypercarbia

A number of methods have been developed to simplify and summarize the EEG data

Cerebral Function Monitor (CFM): This is a

modified device from a conventional EEG It uses

a single biparietal or bitemporal lead, and is cessed to give an overall representation of average cortical activity

pro-Cerebral function analyzing monitor: Developed

from the CFM but displays information about both amplitude and frequency separately

Bispectral Analysis: This modification of the

EEG analyzes the phase and power between any two EEG frequencies The bispectral index (BIS) is

a dimensionless number statistically derived

F igure 2.4 Insonation of the middle cerebral artery through a

trans-temporal window.

14

Trang 28

from these phased and power frequencies and

ranges from 0 to 100 (100-awake, 60-unconscious,

0-isoelectric EEG) This technology was derived

with normal subjects and is not readily transferable

to the injured brain, but may have a use in guiding

sedation and analgesia

Spectral Edge Frequency:

Compressed Spectral Array: Raw EEG data is

processed into a number of sine waves (Fourier

analysis) Power spectral Analysis then

investi-gates the relationship between power and

fre-quency of the sine waves over a short time period

(Epoch) Compressed spectral array is obtained

by superimposing linear plots of successive epochs to produce a three-dimensional “hill and valley” plot (Fig 2.6) The spectral edge frequency looks at the frequency below which a determined power of the total power spectrum occurs SEF90 indicates a spectral edge frequency of 90% and is the frequency below which 90% of activity is occurring

Application of the EEG in the ICUSeizure management: Confirms the diagnosis of

·seizures and identifies a focal or lateralized source of activity It also helps to distinguish between involuntary movements, posturing,

F igure 2.5 (a) and (b) TCD examination of a patient following a subarachnoid hemorrhage and endovascular coiling of an anterior communicating artery aneurysm The patient’s GCS had dropped and they had developed a right-sided hemiparesis The velocities on the right were normal whereas those on the left were high and indicative of vasospasm.

F igure 2.6 Hill and valley plot of the Compressed Spectral Array.

15

Trang 29

and eye signs that are common in the intensive

care, and true seizure activity

Nonconvulsive status epilepticus: This

repre-·

sents a state that lasts more than 30 min with

clinical evidence in alteration in mental state

from normal, and seizure activity on the EEG

Between 4 and 20% of patients with status

epi-lepticus have nonconvulsive episodes

Metabolic suppression: Burst suppression

·

(isoelectric EEG) is a definable end point when

pharmacological reduction of the cerebral

metabolic rate of the injured brain is required

for either neuroprotection or intractable

intrac-ranial hypertension

Ensuring adequate sedation in patients who

·

require prolonged neuromuscular paralysis

Prognosis: The EEG can be of prognostic value

·

following brain injury, with absence of

spon-taneous variability being associated with poor

outcome

Near Infrared Spectroscopy

While the criticism of jugular venous oximetry is that

it is representative of global oxygen delivery, near

infrared spectroscopy (NIRS) is a noninvasive

tech-nique that measures regional cerebral oxygenation

Light in the near infrared wavelength (700–

1,000 nm) can pass through bone, skin, and other

tissues with minimal absorption, but is partly

scat-tered and partly absorbed by brain tissue The

amount of light absorbed is proportional to the

con-centration of chromophobes (iron in hemoglobin,

and copper in cytochromes), and measurement of

absorption at a number of wavelengths provides an

estimate of oxygenation (Owen-Reece et al 1999)

The probes illuminate a volume of about

8–10 mL of tissue and are ideally suited for use in

neonates because of their thin skull, but have been

used with success in adults

Advantages of this technique are that it is

non-invasive, and provides a regional indicator of

cere-bral oxygenation Its major limitation is its inability

to distinguish between intra- and extra-cranial

changes in blood flow

Multimodal Monitoring

In any type of brain injury, the available monitoring

modalities are prone to artifact and misinterpretation

By utilizing more than one monitoring technique, the observer is more likely to determine whether

a genuine change in cerebral physiology has occurred and what the most appropriate interven-tion should be For instance, in traumatic brain injured patients we routinely monitor ICP, proc-essed EEG, SjvO2 and brain-tissue oxygen tension (PbtO2), allowing us to observe both local and regional changes in cerebral hemodynamics General rules cannot always be applied to indi-vidual patients, and multimodal monitoring can allow more informed decision making such

as determining CPP thresholds or the ability of the cerebral vasculature to autoregulate (Matta

et al 2000).Conclusions

A wide range of monitoring techniques is available, each with their different strengths and limitations Multimodal monitoring using a combination of techniques can overcome some of the limitations

of the individual methods discussed The choice

of monitoring is often guided by clinical familiarity and local policy

inter-Czosnyka M, Guazzo E, Whitehouse H, Smielewski P, Czosnyka Z, Kirkpatrick P et al (1996) Significance

of intracranial pressure waveform analysis after head injury Acta Neurochir (Wien) 138(5):531–542 Gopinath SP, Robertson CS, Contant CF et al (1994) Jugular venous desaturation and outcome after head injury J Neurol Neurosurg Psychiatry 57:717–723 Gupta AK (2002) Monitoring the injured brain in the intensive care unit J Postgrad Med 48(3):218–225 Joseph M (2005) Intracranial pressure monitoring: vital information ignored Indian J Crit Care Med 9(1): 35–41

Juul N, Morris GF, Marshall SB, Marshall LF (2000) Intracranial hypertension and cerebral perfusion pressure: influence on neurological deterioration and outcome in severe head injury the executive committee of the international selfotel trial J Neuro- surg 92:1–6

16

Trang 30

Marmarou A, Anderson RL, Ward JD et al (1991) Impact

of ICP instability and hypotension on outcome in

patients with severe head trauma J Neurosurg 75:

S59–S66.

Matta B, Menon D, Turner J (2000) Multimodal

monitoring in neurointensive care Textbook of

Neuroanaesthesia and Critical Care Greenwich

Medical Media, Cambridge

Moppett IK, Mahajan RP (2004) Transcranial Doppler

ultrasonography in anaesthesia and intensive care

Br J Anaesth 93:710–724

Owen-Reece H, Smith M, Elwell CE, Goldstone JC (1999) Near infrared spectroscopy Br J Anaesth 82:418–26 Steiger HJ, Aaslid R, Stooss R, Seiler RW (1994) Transcra- nial Doppler monitoring in head injury: relations between type of injury, flow velocities, vasoreactivity, and outcome Neurosurgery 34:79–85

Young JS, Blow O, Turrentine F, Claridge JA, Schulman A (2003) Is there an upper limit of intracranial pressure

in patients with severe head injury if cerebral perfusion pressure is maintained? Neurosurg Focus 15(6):E2 Anaesthesia UK www.frca.co.uk

17

Trang 31

Key Points

1 The management of traumatic brain injury

(TBI) has increasingly become more tailored to

the individual patient; measuring adequacy of

cerebral oxygenation may allow lower cerebral

perfusion pressures to be targeted and more

ra-tional adjustments of PaCO2 levels

2 Patients with TBI who are hypothermic at

pres-entation should not be rapidly rewarmed

3 Hypertonic saline can be a useful alternative

to mannitol in the management of intracranial

hypertension

4 Steroids are not currently recommended in the

management of TBI

5 Recombinant Factor VIIa may be useful in cases

where correction of acidosis and hypothermia

and administration of appropriate blood

prod-ucts has failed to control continued

nonsurgi-cal bleeding

6 Decompressive craniectomy is a useful

thera-peutic maneuver in selected cases of refractory

intracranial hypertension

The Secondary Management

of Traumatic Brain Injury

The management of Traumatic Brain Injury (TBI)

is challenging, right from the point of injury

through to rehabilitation Although this chapter

sets out the evidence-base behind certain

treat-ment strategies, it will be clear to the reader that there is still no consensus position on many aspects of care Traumatic Brain Injury constitutes the key cause of death in trauma, with trauma itself the principal cause of death and disability up

to the age of 50 Given both the magnitude of the problem and the significant negative impact, there

is an urgent need to promote both seamless and consistent care

Pathogenesis of Brain InjuryImpact to the cranium may be wholly absorbed by fragmentation of the skull with no direct brain injury Fractures in the temporo-parietal region may be associated with tears to the middle menin-geal artery and a resultant extradural hematoma, which if identified and evacuated quickly, is not usually associated with any significant longer-term implications The underlying brain is, however, vulnerable to injury even without penetration of the skull Internal movement results in compression, stretch, and shearing of neurons and supporting tissue, causing direct neuronal damage, hemorrhage,

or contusion

The Monro–Kellie doctrine describes the ciple whereby skull contents of brain, blood, and CSF are normally in equipoise with a pressure of

prin-<15 mmHg At global pressures above 20 mmHg, microvascular flow will be compromised, leading

to ischemia Osmotically active metabolites mulate and intracellular membranes are disrupted,

Trang 32

D Bell and J.P Adams

thereby aggravating edema Cellular dysfunction

is associated with shifts in ionic concentrations,

potassium leaving to be taken up by the glial cells,

resulting in cytotoxic edema and astrocyte

swell-ing The concurrent influx of calcium and sodium

into the neurons promotes release of excitatoxic

neurotransmitters (e.g., glutamate), which further

exacerbate calcium influx, and eventually results

in irreversible change (see Chap 1, Fig 1.1)

Destru ctive enzyme systems such as lipases and

proteases are activated, aggravating cellular

destruction with cell death ultimately triggered by

the release of mitochondrial apoptotic proteins

There appears to be a higher order of

inflam-matory response than that seen after injury to

other tissues or organs, compounded by the

pres-ence of the rigid skull This explains why certain

patients who initially appear to have a relatively

trivial injury progress to intractable intracranial

hypertension (ICH) despite all appropriate care,

culminating in death or profound deficit Even

those patients who eventually make a reasonable

recovery may demonstrate a progressive rise in

intracranial pressure (ICP) beyond a week after

injury, in contrast to an inflammatory response

after peripheral trauma, which typically peaks at

24–48 h, and contrary to the perceived wisdom

that high ICP will begin to abate after 48 h

The absence of any treatment strategy that

mod-ifies this complex and progressive inflam matory

response illustrates the limitations of a

neurosur-gical center to continuing the principle of avoiding

secondary cerebral insults (Chap 1, Table 1.2)

The Regional Neurosurgical Center (RNC)

carries these principles further by taking active

measures to increase oxygen delivery and reduce

oxygen consumption and by controlling the volume

of brain, blood, or CSF This process requires

spe-cialized monitoring to ensure that the strategies

are effective (i.e., ICP is reduced) and that

addi-tional cerebral insults are avoided (e.g.,

hyperven-tilation compromising cerebral oxygen delivery)

Timely neurosurgical input allows rapid removal

of any significant intracranial hematoma,

monitor-ing for the potential expansion of a less significant

hematoma and radical surgical maneuvers for

refractory ICH (e.g., decompressive craniectomy)

Although flow diagrams are provided in the text

and appendices, this chapter is not directed toward

an empirical and prescriptive approach to care,

but to an analysis of the various treatment options,

such that any practitioner can exercise sional judgment when faced with different pathol-ogy, at a different time after injury, with a different clinical presentation

profes-Apart from obvious examples such as ate evacuation of an extradural hematoma, there are very few scenarios where there is a universally accepted unequivocal treatment strategy With a subdural hematoma, bleeding arises from bridg-ing veins and the surface of the brain itself, and there is no single identifiable source to target Any mass effect is as likely to be attributable to swell-ing of the underlying brain from the associated injury, as much as the hematoma Further brain swelling is likely to take up any space created by removal of a hematoma Embarking on surgery for removal of an intracerebral hematoma or con-tusion is even more problematical because of dif-ficulties in identifying, accessing, and controlling bleeding points, with the distinct possibility of collateral damage to surrounding vulnerable neural tissue The decision in these scenarios is therefore based not just on evidence of a hema-toma radiologically, but also on location, size, pro-gression, associated intra and extra-cranial injury, co-morbidity, impact on neurological function, and the level of ICP

immedi-The medical management of ICH is also vexed, with a wish to reduce cerebral oxygen demand through sedation making it impossible to under-take a functional neurological assessment Further-more, sedative strategies have a negative impact on cardiovascular, respiratory, gastrointestinal, and immune status, all of which may at times generate significant secondary insults Maneuvers under-taken to improve cerebral oxygen delivery such as increasing cerebral perfusion pressure (CPP) may constitute a secondary cerebral insult by other mechanisms Strategies directed at control of a rising ICP such as hyperventilation or administra-tion of mannitol may similarly constitute second-ary insults This chapter therefore aims to explore the benefits, hazards, and the weight of evidence to support the use of these various interventions

Control of Cerebral Oxygen Demand

Seizure Control

Chapter 8 has a detailed description of the nition and management of seizure activity; an

recog-20

Trang 33

3 The Secondary Management of Traumatic Brain Injury

algorithm for the treatment of status epilepticus

is also included in the appendices

Sedation

Sedation not only reduces cerebral oxygen

demand, but also contributes to the reduction of

other secondary insults by facilitating airway

control, optimizing ventilatory support, and

reducing global oxygen demands The negative

aspects of sedation, however, affect most systems

of the body and may ultimately contribute to both

morbidity and mortality The combination of

sedation and positive pressure ventilation usually

leads to a requirement for inotropic and/or

vaso-pressor support, and not infrequently, one

wit-nesses a rising requirement for these despite

excluding failure of the pituitary–adrenal axis

This in turn is often associated with signs of

coro-nary ischemia, particularly in previously fit young

males.(Cremer et al 2001) Adverse effect on

gas-trointestinal function may be instrumental in the

additional complication of VAP

(ventilator-associ-ated pneumonia), as well as compromising

nutri-tional status Immune impairment may contribute

to the development of infection, generating a

common scenario whereby sedation has to

con-tinue to manage gas-exchange problems caused

by the acquired pneumonia Sedation and

immo-bility also predispose to thrombotic complications

and skincare problems One of the greatest

diffi-culties with sedation, however, is the inability to

make a functional neurological assessment, and

clearly the longer sedation is continued, the greater

the subsequent period in which the clinical picture

may be compromised by drug accumulation or

withdrawal phenomena, the latter possibly

requir-ing the use of further sedative regimens

Thiopentone creates singular difficulties in this

regard, with an exceptionally long elimination

half-life, no antagonist, and the ability after higher-level

administration to mimic the signs of brainstem

death with irregular, dilated, and unreactive pupils

Sedation should be provided:

1 When the patient’s level of consciousness is

ob-tunded (GCS 8 or less) such that they cannot

maintain or protect the airway or adequately

self-ventilate and oxygenate

2 When intubation and ventilation is required to

address other aspects of injury or disturbance

of respiratory function, or

3 When ICP remains high, despite avoidance of all other cerebral insults and in the absence of any functional neurological activity

Such “primary sedation” should ideally be mulative in the interests of early clinical assessment, with propofol/alfentanil a reasonable combination, but midazolam an acceptable addition or alternative

noncu-to propofol, if high dosage causes cardiovascular problems or lipid accumulation Remifentanil is gaining popularity as a single agent or in combi-nation with a sedative

When using secondary sedation strategies for refractory ICH, it is essential to have a measurable endpoint (Winer et al 1991) This includes the use

of processed EEG and the achievement of burst suppression using the minimum amount of seda-tion necessary In addition, the reduction in elec-trical activity should be accompanied by a fall in ICP or an increase in jugular venous oxygen saturation (SjvO2) The usual dose of thiopentone required to achieve the electrical end-point of burst suppression is a loading dose of 5–10 mg/kg, with a subsequent infusion rate of 5–10 mg/kg/h

If only used where high ICP is not responsive to all other strategies and following these principles, the hazards of barbiturate coma (pneumonia, sepsis syndrome, and hepatic dysfunction(Schwab et al

1997) can be offset, not only against the control of ICH but also potential longer-term recovery ben-efits (Lee et al 1994; The Brain Trauma Foundation The American Association of Neurological Sur-geons 2000a; Dereeper et al 2002) Recent local experience with brain-tissue oxygen measurement, however, has raised some concerns about the use

of thiopentone Despite seeing a fall in ICP and maintenance of a satisfactory SjvO2, thiopentone can lead to a reduction in brain-tissue oxygen levels (presumably risking a further ischemic insult), and therefore its use should probably be reserved for specialist centers

Additional agents such as lidocaine (1 mg/kg 4–6 hourly) or ketamine may have a role in modi-fying surges in ICP in response to interventions such as suctioning, but again their use should probably be discussed with the RNC

Hypothermia/Temperature Control

Barbiturate coma has the capacity to reduce brain oxygen and energy consumption by between 50–60% Induced hypothermia can reduce this further by

21

Trang 34

slowing cellular constitutive process Control of

pyrexia is universally accepted with symptomatic

therapy (paracetamol, nonsteroidal

antiinflamma-tory drugs, surface cooling) and by treating sources

of infection, but the step beyond this to induction of

hypothermia does not constitute routine practice

Clinical studies have given conflicting results,

(Marion et al 1997; Clifton et al 2001) but a more

recent meta-analysis concluded that patients with

high ICP refractory to all other maneuvers may

benefit (Henderson et al 2003) Given the positive

conclusions of studies evaluating neurological

outcome after induced hypothermia following

cardiac arrest (The Hypothermia after Cardiac

Arrest Study Group 2002), it is likely that this

strategy will continue to be evaluated

However, even mild–moderate (33–35°C)

hypo-thermia can produce cardiovascular instability,

disturbance of coagulation, and immune impairment

In addition, cerebral oxygen delivery may be

com-promised through the reduction of cardiac output,

cerebral vasoconstriction, increased plasma

viscos-ity, and a shift in the oxygen dissociation curve to

the left It is our current practice to control pyrexia

and promote normothermia rather than to induce

hypothermia, unless the ICP is persistently elevated

and unresponsive to other therapeutic maneuvers

Patients with traumatic brain injury who are

hypo-thermic on admission should not be rapidly re-warmed

as this may be associated with a poorer outcome.

Optimization of Cerebral Oxygen Delivery

An adequate circulating volume, with an adequate

level of functional hemoglobin, with no cerebral

vasoconstriction, and no factors compromising

blood rheology or oxygen dissociation are key

to ensuring cellular oxygen delivery The brain

requires a critical perfusion pressure above the

lower limit of autoregulation When ICP exceeds

20 mmHg flow through the microcirculation will

be compromised, with an associated higher mortality (Johnston et al 1970) However, it is simplistic to believe that increasing the CPP will compensate for this If the blood brain barrier is disrupted and microvascular flow is impaired, an increase in MAP may simply aggravate brain swell-ing through the formation of hydrostatic edema There is no evidence that increasing CPP improves the perfusion of pericontusional ischemic tissue (Steiner et al 2003) A polarized debate continues

on the correct approach in these circumstances, with proponents of the Lund philosophy (Grande

2004) (see Fig 3.1) targeting the causes of high ICP, rather than relentlessly pursuing a fixed differen-tial pressure Despite endorsement by national and international bodies, (The Brain Trauma Foundation The American Association of Neurological Surgeons

2000b; Maas et al 1997), there are no definitive controlled trials to conclusively prove that ICP-guided therapy is efficacious Recommendations are based

on the association between secondary insults and poor outcome (Jones et al 1994), but the evidence base to take this one step beyond avoiding such insults to actively manipulating these variables is not fully established, with conflicting results in the literature Although recent studies have demon-strated reduced mortality (Clayton et al 2004)

and improved outcome from protocolized ICP /CPP-directed care (Fakhry et al 2004), other retro-spective cohort studies from hospitals with different strategies for head-injured patients demonstrated

no evidence of benefit, as determined by the extended Glasgow Outcome Scale (Cremer et al

2005) The CPP-directed strategy was however noted

to be associated with prolonged mechanical lation and increased levels of therapy intensity.Our approach is to maintain a CPP of > 60 mmHg, with the target CPP being referenced to achie ving

venti-a SjvO2 value of greater than 60% and a lactate–

The Lund Approach to severe traumatic brain injury

Pharmacological principles:

Reduction of capillary hydrostatic pressure with a metoprolol and clonidine

Reduction of cerebral blood volume with thiopental and dihydroergotamine

Reduction of stress response with opioids, benzodiazepines and thiopental

Maintaining normal colloid oncotic pressure with albumin, blood and plasma transfusions

F igure 3.1 The pharmacological principles of the Lund Approach to traumatic brain injury.

22

Trang 35

oxygen index of <0.03 (see Fig 3.2), whilst avoiding

an escalating regimen of vasopressor (particularly

if there is any cardiac dysfunction) There is little

evidence to suggest that any one pressor agent is

superior to another in this situation We use

nora-drenaline (norepinephrine) as first line (up to

1 mcg/kg/min) but would favor phenylephrine

(1–10 mcg/kg/min) or dopamine (2–15 mcg/kg/

min) in the young patient with evidence of cardiac

dysfunction (see Appendix for suggested infusion

regimens) Increasing vasopressor requirements

should prompt reevaluation of volume status with

invasive monitoring, as well as consideration of a

short snynacthen test and administration of

“physi-ological dose” hydrocortisone (e.g., 1 mg/kg/day in

three divided doses) Vasopressin can be used in

refractory cases

If oxygen delivery is compromised as measured

by SjvO2, lactate-oxygen index or tissue oxygen

mon-itoring, then transfusing to Hb >10 gm/dl, incr e asing

inspired oxygen concentration, and norma lizing CO2

can be associated with improved tissue oxygenation

and potentially more favorable outcomes (Stiefel

et al 2005), whilst avoiding the hazards of pursuing

CPP in the face of a relentless rise in ICP

Control of ICP

Elevation of ICP after traumatic brain injury can

be attributed to an expansion of the primary

com-ponents (brain, blood, and CSF) or any focal

pathology (hematoma or contusion), or a

combi-nation of these factors

Management is directed toward treatment or

control of these contributing factors (see Fig 3.3),

with occasional symptomatic relief in the form of

decompressive craniectomy The warning signs

and subsequent management of acute brain

her-niation are outlined in Table 3.1

Management of Hematoma/Contusion

With the exception of an extradural hematoma, evacuation of a mass lesion is a significant under-taking that may in fact aggravate brain injury There is no guarantee that either a hematoma or further brain swelling will not fill any space created, and surrounding ischemic tissue may suffer further collateral damage Limiting progression involves scrupulous optimization of coagulation (platelets > 100,000 and INR and APTTR <1.2) using blood products, with additional manage-ment of other factors compromising coagulation such as hypothermia, hypocalcaemia, and hypophosphatemia

Administration of rFVIIa (recombinant Factor VIIa) should be considered when:

1 Provision of all blood components and mization of temperature and pH has failed to control microvascular ooze

opti-2 When there is difficulty or delay in accessing blood components

3 Where radical surgery is countenanced (e.g., sacrifice of an otherwise potentially salvage-able limb to prevent ongoing blood loss).rFVIIa generates a “thrombin burst” and has published benefits in cases of spontaneous intracerebral hemorrhage (Kaufmann and Cardoso 1992) and in trauma, where not only has the hemorrhage been controlled, but the reduced transfusion requirements has been associated with a lower incidence of acute lung injury and multiple organ failure (MOF) (Cruz

et al 2004) It is expensive and can only achieve maximum benefit if administered in association with all blood products, correction of acidosis, and hypothermia A hematologist should be involved

La ctate Ox ygen Index (LOI )

1 4 3 (

[lactate] jv− [lactate] art

SaO2 = arterial oxygen satura tion

SjvO2 = jugular ve nous oxygen satura tion

F igure 3.2 Calculation of the lactate oxygen index.

23

Trang 36

Case History

Problem: An alcoholic patient with a recurrent extradural hematoma had

ongoing “nonsurgical” bleeding from associated lower-limb injuries,

despite provision of all blood products The neurosurgeons were reluctant

to operate without optimization of coagulation status.

Treatment: Single dose of rFVIIa 120 mcg/kg over 3 min (Vialet et al

2003) following administration of platelets, FFP, and cryoprecipitate

and bicarbonate to adjust pH to 7.25.

Outcome: Uneventful surgery, blood loss from other injuries minimized.

Comment: No laboratory test for this treatment, the endpoint being

clinical control of bleeding Further dose to be given if no positive

response within 15 min.

CIRCULATION

MAP > 80mmHg CPP > 60, if ICP measured*

Hb ~10g/dl Maintain adequate circul ating volume (e g CVP, PC WP, ODM)

If hypotensive, check for bleeding Consider need for inotropes or vasopressors*

SEDATION

Propofol 1-6mg/kg/hr Alfentanil 1- 4mg/hr Midazolam if unstabl e Consider paralysis EEG for Thiopentone coma**

GENERAL MEAS URES

15° Head up tilt, neck neutral Check ETT ties, hard collar OGT/ NGT Early enteral feeding Metoclopramide if not absorbing

H 2 blocker / PPI Insulin: maintain glucose 4-8mmo l/l VTE prophylaxis*

Arterial blood gases Group & Save

FLUIDS *

Daily Fluid Balance 0.9% NaCl maintenance ( unless

grossly hypernatremic) 6% starch ( up to 1.5l /day) Blood, Hb ~ 10 g/dl Clotting products (INR, APTT<1.2, platelets >100)

PYREXIA

Culture blood, sputum, urine

CRP CXR, consider BAL Paracetamol 1g qd s +/- NSAIDs**

Active Cooling Consider line change Antibiotics*

* see accompan ying discussion in text

24

Trang 37

the management of brain edema, do however

include reduced CSF production, and in certain

cases this may be of marginal benefit

Manipulation of Intracranial Blood Volume

As an increase in blood volume may contribute to

ICH, secondary cerebral insults such as

hypercap-nia and venous obstruction must be avoided

Hyperventilation can reduce intra-cranial pressure

by inducing cerebral vasoconstriction, but may

worsen cerebral ischemia It should only be

used as a temporary measure to prevent imminent

brainstem herniation unless SjvO2 and the lactate–

oxygen index are being measured In a similar way,

hyperoxia will also lead to cerebral

vasoconstric-tion and a reducvasoconstric-tion in cerebral blood volume and

as such, may also be a useful temporary holding

measure when ICP is very high

Manipulation of Brain Swelling

Osmotherapy

Prevention includes avoidance of hypotonic

solu-tions such as 5% dextrose and by maintaining

nor-moglycemia With intact autoregulation, mannitol

increases cerebral blood flow by expanding the culating volume and by improving rheology This results in a reflex cerebral vasoconstriction and a rapid fall in ICP A subsequent reduction in brain water then occurs because of the osmotic differen-tial However, in contusional or diffuse axonal injury, the BBB is frequently disrupted This may result in mannitol redistributing in the brain interstitium and contributing to the edema, rather than improv-ing it (Kaufmann and Cardoso 1992) Mannitol also increases serum sodium and osmolality, the latter becoming nephrotoxic at >320mOsm/L The rise in serum sodium is also mirrored in the brain intersti-tium, this again generating osmotic edema Man-nitol remains, therefore, most effective in shrinking relatively normal brain as a temporizing measure prior to definitive surgical relief of a mass lesion It should be used as intermittent rather than continu-ous therapy and there is some evidence that it is more efficacious if used at higher dose (1.4 g/kg rather than conventional 0.5 g/kg) (Cruz et al 2004)

cir-It is our practice to restrict mannitol to a dose of 0.5–1 g/kg every 6 h and to stop if serum osmolality

exceeds 320 or serum sodium >160 mmol/l We

rarely use it beyond the first 24–36 h after injury, after which our preference is to switch to hypertonic saline (HSL) This is claimed to be more effective than mannitol in reducing ICH, without compro-mising the hemodynamic status of the patient (Vialet et al 2003; Munar et al 2000) Sodium chlo-ride is completely excluded from the intact blood–brain barrier (reflection coefficient = 1.0), and is theoretically a better osmotic agent than mannitol (reflection coefficient 0.9) Additional benefits may include antagonism of excitatory neurotransmitters

We use a regime of 30 ml of 20% HSL over 20 min through a central venous catheter Repeated doses can be given after approximately 6 h as long as plasma [Na+] has not exceeded 160 mmol/l

Loop Diuretics

Furosemide is effective in reducing brain water and is synergistic when used with mannitol It reduces CSF production and increases sodium and water transfer through the arachnoid granu-lations It also eliminates sodium and water through the kidneys, thereby avoiding the higher sodium levels seen with recurrent mannitol administration Unlike mannitol, it does not con-tribute to brain edema Our policy is to commence

T able 3.1 Recognition and management of acute brain herniation

Warning signs:

Reduction in conscious level

Unilateral third Nerve palsy

Lateralising motor signs e.g., hemiparesis, extensor posturing

Hypertension, bradycardia or respiratory irregularity (Cushing’s Triad)

Management

• Rapid intubation and ventilation (great care needed to avoid

exaggerated pressor response to laryngoscopy and intubation –

experienced anaesthetist essential Invasive blood pressure

monitoring ideal, but do not delay establishing ventilation)

• Hyperventilate to Pa CO 2 3.5–4.0 kPa as a temporary measure

• Mannitol 20% 0.5 g/kg over 10 min

• Sedation to reduce cerebral metabolic rate (e.g., propofol,

thiopentone) supplemented with opioid analgesic (e.g., fentanyl,

alfentanil)

• Head up position and good neck position to encourage venous

drainage

• Maintain adequate MAP (ideally 90–100 mmHg) with pressor Do

not treat hypertension (may reduce cerebral perfusion)

• 100% O 2 (hyperoxia) may reduce cerebral blood volume and ICP

(especially in younger patients) and can be utilised whilst more

definitive treatment is sought

Trang 38

an infusion at 0.3mg/kg/day and adjust to achieve

neutral water balance over a 24-h period

In instances where rapid control of ICH is

required (i.e., herniation syndromes) and mannitol

has not caused a significant diuresis, furosemide

0.25–0.5 mg/kg can be administered

Steroids

The efficacy of steroids on modification of

trau-matic edema or outcome has not been validated

(Alderson and Roberts 1997) A recent large

multi-center, prospective, randomized, placebo-controlled

trial (CRASH study) demonstrated no reduction

in mortality (Roberts et al 2004),and their use is

not currently recommended

Management of ICH by Craniectomy

Following encouraging trial results (Guerra et al

1999; Albanese et al 2003), surgical intervention

for ICH is currently being reevaluated It should

be considered in cases of intractable ICH

unrespon-sive to all medical maneuvers, or when medical

maneuvers are generating such significant side

effects (most commonly myocardial ischemia)

that morbidity or mortality are likely to arise from

complications of treatment The magnitude of this

intervention should not be underestimated, with

the possibility of uncontrolled bleeding and

further brain injury As our experience grows, it

may be possible to identify those patients who will

benefit most from early decompression Although

a randomized ICP rescue trial of thiopentone

versus craniectomy is currently being carried out

in the United Kingdom, (www.rescueicp.com), it

is the authors’ current practice to consider the

procedure for diffuse axonal injury in young

patients with escalating vasopressor requirements

and a LOI approaching 0.08 despite optimization

Other Aspects of the Management

of Traumatic Brain Injury

The general principles of intensive care such as early

enteral nutritional support apply equally to the

patient with traumatic brain injury (Fig 3.3) However,

certain aspects of care such as thrombo-prophylaxis,

antibiotic therapy, and optimal timing of surgical

intervention for other injuries are more contentious

Neurosurgical patients are at a significant risk from thromboembolic complications Each case has to be judged on individual merits depending upon CT findings, coagulation status, and associated injuries Low molecular weight heparin (LMWH) is usually withheld until at least 24–48 h after injury and possibly longer if surgical intervention is required, or if there is any radiological evidence

of extension of focal pathology Patients with matic brain injury have at least a moderate risk of venous thromboembolism LMWH for example, tinzaparin 3500–4500iu s.c daily is now routinely started 24–48 h after admission (obese patients may require larger doses based on body weight) Exceptions to this would include coagulopathy, low platelet count, hemorrhagic contusions, or imminent surgical intervention, but not the pres-ence of an ICP catheter as such With any contrain-dication to LWMH, mechanical compression devices should be started within 24 h of admission.Antibiotic therapy is problematical, with an inevitable balancing act between vulnerability to secondary infection with resistant organisms if antibiotics are started without proven infection, and the secondary cerebral insults triggered by established infection if there is a delay in initiating therapy This decision is made particularly difficult

trau-by the brain injury itself driving a central pyrexia, with other markers of infection such as white cell count and C-Reactive Protein (CRP) not having diagnostic specificity in these circumstances Pro-phylactic antibiotics are not routinely prescribed for base of skull fracture and CSF leak, but antibi-otics to cover Staph aureus and Hemophilus influ- enzae (e.g., ampicillin and flucloxacillin) should be

started empirically for the young head-injured patient with rising oxygen requirements and a sus-pected aspiration lung injury, but no proven infec-tion Input from a dedicated microbiologist is invaluable as local resistance patterns will vary.The optimal timing of surgical intervention for other injuries depends upon their nature and severity, the likely outcome of the head injury, and the impact on patient management of either undertaking or deferring any procedure The most common surgical undertakings are maxillofacial reconstruction and stabilization of long-bone frac-tures In any patient with ICH, the impact of trans-ferring and embarking on any surgical procedure should not be underestimated, particularly if it is

26

Trang 39

lengthy and associated with significant blood loss,

cardiovascular disturbance, or secondary

derange-ment of coagulation It would appear reasonable to

ensure adequate surgical toilet and closure of open

wounds, and to undertake procedures such as

fas-ciectomy to maintain viability and future function

However, there is little justification for embarking

on extensive reconstruction procedures if it

remains unlikely that in view of the severity of the

brain injury, there is little chance of ever requiring

a completely stable knee, for example The benefits

of early fixation of long-bone fractures in

modify-ing the systemic inflammatory response or

pre-venting complications (e.g., fat embolism) will

continue to be debated, but it is the authors’ current

practice to rely on external fixation in the first

instance until the direction of the brain injury has

been determined

Conclusions

The secondary management of traumatic brain

injury is a continuation of the principles of

primary management, namely promoting cerebral

oxygen delivery, controlling cerebral oxygen

demand, and modifying, where possible, those

factors contributing to a rise in intracranial

pres-sure The RNC can offer specialized monitoring

to direct a more aggressive approach to these

factors, as well as surgical intervention for focal

hematomata or decompressive craniectomy In

addition, specialist neurosurgical intensive-care

units improve both the quality and efficiency of

care for neurologically injured patients (Mirski

et al 2001) The challenge in secondary

manage-ment rests not in the actual undertaking, but

in knowing when to embark upon, repeat, or

continue these medical or surgical strategies, and

when to withhold or discontinue Given the

hetero-geneity of brain injury, the lack of unequivocal

evidence to support many of the interventions

undertaken and potential hazards associated with

certain strategies, optimal care is a constant

multi-disciplinary exercise of professional judgment,

rather than rigid adherence to a simplistic

proto-col Certain principles can however be derived,

and it is hoped that consistency of care for a certain

pathology, or combination of pathologies, may

ultimately generate a robust evidence base and

modify the impact of an injury which is ing in terms of incidence, mortality, and long-term serious disability

devastat-Secondary Head-Injury ManagementThe algorithm given here describes how we would initially manage a brain-injured patient on our Neurosurgical ICU This is also illustrated in the flow diagram (Fig 3.3) Apart from the neuro-specific monitoring, the physiological goals and parameters would be exactly the same for a patient managed outside the RNC

Initial Stabilization on Neuro-ICU

Ventilation: FiO2 1.0, tidal volume 7–10 ml/kg,

12 breaths/min, PEEP 2.5–5 cmH2O until first blood gas done

Sedation: Propofol 1–6 mg/kg/h + Alfentanil 1-4mg/h

titrated to effect

Paralysis: Cisatracurium bolus/infusion if indicated Circulation: MAP >80 mmHg or CPP >60 mmHg Monitoring: ECG, IABP, CVP, EtCO2, Temperature Neuro-specific monitoring: Processed EEG, jugular

venous oximetry (consider brain-tissue oxygenation measurement if available)

Investigations: Arterial blood gases, U&Es, Glucose,

FBC, INR/APTT, G + S

Reassessment

1 Blood gases: Adjust ventilation – PaO2 >13 kPa,

PaCO2 4.5–5.0 kPa until SjvO2 available

2 Circulation:

(a) Maintain MAP >80 mmHg or CPP >60 if ICP is measured

(b) Keep Hb ~10 g/dl or haematocrit ~30.(c) If hypotensive, look for sources of bleeding.(d) If hypotension persists after volume resus-citation, start vasopressor support (see CPP guidelines, Fig 3.4)

(e) Use advanced cardiovascular monitoring (e.g., PAFC, ODM) if cause of hypotension

is unclear, or if vasopressor requirements are rapidly escalating

3 ICP monitoring: if ICP >20 mmHg see ICP

guidelines (Fig 3.5)

27

Định dạng
Số trang	79
Dung lượng	2,51 MB