Learning objectives

After reading this article, you should be able to:

C understand the definition of consciousness and coma

C use an organized approach to identify the cause of coma

C initiate effective and comprehensive early management

NEUROINTENSIVE CARE

Clinical approach tocomatose patientsJoanne Tan

Marco Fedi

C identify conditions that can mimic coma

C identify relevant prognostic factors in acute coma

AbstractThe nature of consciousness itself belongs within a group of ‘underdeter-

mined questions’ to which we might not be able to find an answer. Simi-

larly, we have a limited understanding of disorders of consciousness. In

this brief article, we discuss a practical approach to the comatose patient

and the importance of promptly identifying the cause to prevent perma-

nent neurologic damage.

Keywords Coma; GCS; FOUR score; consciousness; neurological

assessment; prognosis

Royal College of Anaesthetists CPD Matrix: 3F00

Introduction

Consciousness remains one of the major unsolved questions,

despite being the subject of intense study for many centuries. In

the 19th century, autopsy studies of patients with impaired

conscious states (i.e. encephalitis lethargica and Wernicke’s en-

cephalopathy) identified lesions in the upper brainstem and

diencephalon. The role of these structures in maintaining

wakefulness was later confirmed by a series of elegant experi-

mental studies by Moruzzi and Mogun. However, it was only

with the advent of mechanical ventilation and the consequent

possibility of closely observing comatose patients that two

fundamental components of consciousness were described:

arousal and awareness. Arousal is mediated by the reticular

system and projections to the thalamus, whereas awareness is

mediated by the thalamus and the cerebral cortex.

Coma (from the Greek ‘koma’ to fall asleep) is a deep state of

unconsciousness characterized by the absence of arousal and

awareness. Neuroimaging studies have shown that regardless of

the aetiology, coma is associated with a marked and diffuse

reduction in metabolism, suggesting that coma is a state of

transient ‘cerebral energy failure’. Coma is typically a transitional

state evolving towards recovery of consciousness, minimal

conscious state (fluctuating minimal level of consciousness),

unresponsive wakefulness syndrome (preservation of arousal,

but the absence of awareness) or brain death.

Joanne Tan MBBS Clinical School, Austin Health, The University of

Melbourne, Australia. Conflicts of interest: none declared.

Marco Fedi MBBS FRACP PhD Department of Neurology, Department of

Intensive Care Medicine, Alfred Hospital, Prahran, Victoria, Australia.

Conflicts of interest: none declared.

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:9 375

Clinical assessment of conscious state

A number of different coma scales have been devised for the

clinical assessment of consciousness (e.g. Innsbruck coma scale,

Edinburgh-2 coma scale and Reaction level scale). However, the

Glasgow Coma Scale (GCS) remains the most widely used. It

comprises three subscales: eye opening, motor function and

verbal abilities. Coma is defined as a post-resuscitation GCS score

of 8 or less lasting for more than an hour. The GCS score should

be determined, focussing on individual component scoring rather

than the overall score. The examiner determines the level of

responsiveness with stimuli of increasing intensity, starting with

verbal cues. If no response is elicited, noxious stimuli can be

delivered to the sternum, trapezius, supraorbital nerve, temporo-

mandibular joints or to the peripheries. This enables the exam-

iner to test for localization of pain and abnormal motor re-

sponses. Despite its widespread use, the GCS has limitations in

the assessment of patients who are intubated or have craniofacial

trauma. An alternative to the GCS is the Full Outline of UnRe-

sponsiveness (FOUR) scale which is comprised of four subscales

assessing: motor, ocular responses, brainstem reflexes and

breathing (Table 1). Compared to the GCS score, the FOUR score

has a lower inter-observer variability and provides a tool to

identify patients with locked-in syndrome.

Management of coma in the first hour

The initial step should be directed to ensuring adequacy of resus-

citation measures, and to rapidly treat reversible causes of coma.

Urgent and empiric therapy must be given to avoid additional ce-

rebral insult. Oxygenation must be assured by establishment of an

airway and ventilation of the lungs. Maintenance of a PaCO2 in the

lowenormal range (35e40 mmHg) may have a therapeutic effect

on decreasing intracranial pressure, while not being so low as to

compromise cerebral oxygenation. High-concentration supple-

mental oxygen therapy is to be given to prevent cerebral hypoxia.

Circulatory hypotension requires intravenous resuscitation with

fluid volume infusion. Appropriate fluid resuscitation, aiming for a

mean arterial pressure at about 100 mmHg, is an adequate target

for most patients. As hypoglycaemia is a frequent cause of altered

consciousness, give glucose 25 g as a 50% solution intravenous

accompanied with thiamine 100 mg to prevent precipitation of

Wernicke’s encephalopathy. If opiate toxicity is suspected then

administer naloxone 0.4e2 mg intravenously repeated as needed

up to 4 mg. Benzodiazepine intoxication can be reversed with

flumazenil, with caution, so as to not induce awithdrawal seizure.

A rapid initial survey should follow, looking for injuries or other

notable physical signs (ecchymoses, jaundice, rashes, and needle

track marks). If trauma is suspected the cervical spine should be

immobilized. Once the patient is stabilized a search for the causes

� 2013 Elsevier Ltd. All rights reserved.

Scales used for the assessment of conscious state

Full Outline of UnResponsiveness (FOUR) scale

Eye response

Eyelids open or opened, tracking, or blinking to command 4

Eyelids open but not tracking 3

Eyelids closed but open to loud voice 2

Eyelids closed but open to pain 1

Eyelids remain closed with pain 0

Motor response

Thumbs-up, fist or peace sign 4

Localizing to pain 3

Flexion response to pain 2

Extension response to pain 1

No response to pain or generalized myoclonic status 0

Brainstem reflexes

Pupil and corneal reflexes present 4

One pupil wide and fixed 3

Pupil OR corneal reflexes absent 2

Pupil AND corneal reflexes absent 1

Absent pupil, corneal and cough reflex 0

Respiration

Own airways, regular breathing pattern 4

Own airways, CheyneeStokes breathing pattern 3

Own airways, irregular breathing 2

Breathes above ventilator rate 1

Breathes at ventilator rate or apnoea 0

Glasgow Coma Scale

Eye opening

Spontaneous 4

To voice 3

To pain 2

No response 1

Verbal

Oriented 5

Confused 4

Inappropriate words 3

Incomprehensible sounds 2

No response 1

Motor

Obeys commands 6

Localizes to pain 5

Withdrawal to pain 4

Flexion posturing 3

Extension posturing 2

No response 1

Table 1

Half-lives of medications that affect neurologicalassessment of coma

Medication Half-life (hours)

Phenobarbital 100

Diazepam 40

Amitriptyline 24

Lorazepam 15

Thiopental 10

Midazolam 6

Fentanyl 6

Morphine 3

NEUROINTENSIVE CARE

of the coma begins with serologic investigations and arterial blood

gases to identify potentially reversible metabolic or toxicological

causes.

Rocuronium 1

Atracurium 0.5

History

Table 2

Collateral history should focus on: (a) the time course of the

alteration in consciousness (abrupt onset suggests a stroke,

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:9 376

seizure, drug poisoning or trauma, whereas a gradual onset is

more suggestive of a metabolic process); (b) the presence of

premonitory signs (headaches or focal signs prior to coma); (c)

the setting (objects in the vicinity of the patient such as empty

medicine bottles); and (d) active and past medical and surgical

relevant conditions.

Neurological assessment

The initial assessment should focus on the detection of lateral-

izing signs that suggest a focal lesion and require rapid imaging

evaluation. In examining a comatose patient, confounding fac-

tors such as sedation, muscle paralysis, hypothermia, shock and

drug intoxication should be taken into consideration (Table 2).

For instance, sedation can not only suppress brainstem reflexes

(opioids and midazolam are more suppressive than propofol and

dexmedetomidine), but also can generate false localizing signs

(i.e. phenytoin suppression of oculocephalic and vestibulo-

ocular reflexes).

Fundoscopy and pupillary responses

Fundoscopy may reveal important signs such as papilloedema

(i.e. raised intracranial pressure or acute asphyxia) and sub-

hyaloid haemorrhage (subarachnoid haemorrhage and shaken

baby syndrome). Visual fields and neglect can be grossly

assessed with the blink threat response. The size, symmetry,

shape and reactivity of the pupils should be noted. A unilateral

dilated pupil suggests pressure on the oculomotor nerve (i.e.

uncal herniation or a posterior communicating artery aneurysm).

Unilateral miosis is suggestive of Horner’s syndrome. Bilateral

fixed midposition pupils are a sign of extensive midbrain injury.

Bilateral pinpoint pupils are often due to narcotic overdose, a

ponto-tegmental lesion, or cholinergic toxicity. Pupillary reflexes

and ciliospinal reflexes are preserved in metabolic coma.

Ocular motor function

The primary position of the eyes should be documented. Eyes

found to be heterotropic (non-parallel) must be classified ac-

cording to their relative positional defect: esotropia (one eye

deviated towards nose), exotropia (one eye deviated towards

temple) and hypertropia or skew deviation (one eye deviated

upwards). Two patterns are commonly encountered: oculomotor

� 2013 Elsevier Ltd. All rights reserved.

NEUROINTENSIVE CARE

nerve palsy (affected eye is deviated outwards and downwards

associated with ipsilateral ptosis and dilated pupil), and abdu-

cens nerve palsy (eye ipsilateral to the lesion is deviated inward)

as a result of focal lesions or increased intracranial pressure.

Conjugate lateral eye deviation could indicate an ipsilateral

hemispheric lesion, a contralateral pontine lesion or a contra-

lateral seizure focus. Forced vertical gaze deviation is suggestive

of a hemispheric lesion. Spontaneous eye movements (periodic

alternating gaze, ocular dipping, and refractory nystagmus) are

often seen but with no localizing features. In contrast, ocular

bobbing, a rapid downward jerk followed by a slow return to

midposition, is typical of a pontine lesion. Oculocephalic should

be tested by turning the head side to side if not contraindicated

(i.e. cervical spine trauma). Vestibulo-ocular responses are tested

by irrigating the right external auditory canal with 50 ml of ice

water, followed by warm water into the left if no response to

cold. An asymmetric caloric response is consistent with a struc-

tural abnormality.

Corneal responses

Corneal responses should be tested by applying a cotton wisp

across the cornea or by dripping sterile saline onto the cornea.

Bulbar function and respiratory patterns

Classification and major causes of acute coma

In intubated patients, the gag reflex may be difficult to elicit and is

probably unreliable. Lack of a cough response to bronchial suction-

ing canbedemonstratedbypassing a suctioning catheter through the

endotracheal tube. Absence of cough is suggestive of a medullary

injury. Abnormal respiratory patterns are commonly seen in coma-

tose patients and include central neurogenic hyperventilation from

pontine or mesencephalic lesions, or cluster (Biot’s) breathing from

pontine lesions. Ataxic breathing, irregular in both rate and tidal

volume, is suggestive of damage to the medulla or lower pons.

Structural

Traumatic brain injury

Motor function

Vascular Infarction, haemorrhage, vasculitis,

subarachnoid haemorrhage, sinus venous

thrombosis

Brain tumour Brainstem glioma, lymphoma, multiple

metastasis

Infection Abscess, meningitis and encephalitis

Demyelination Central pontine or extra-pontine myelinolysis,

acute disseminated encephalomyelitis,

acute hydrocephalus

Non-structural

Metabolic-endocrine derangement

Metabolic Hypoxia, hypercarbia, hypoglycaemia,

hyperglycaemia, acute uraemia, hepatic failure,

Wernicke’s encephalopathy, hyponatraemia,

hypernatraemia

Endocrine Acute hypothyroidism, Addisonian crisis,

hypopituitarism

Motor function is assessed by observing any spontaneous

movements or posturing. Decerebrate posturing (lesion below

the midbrain) is characterized by adduction, extension, and

pronation of the upper limbs with extension of the lower limbs.

Conversely, decorticate posturing (lesion above the midbrain) is

marked by upper limb flexion and adduction, again with lower

limb extension. An active assessment of tone may be performed

for spasticity, rigidity or flaccidity. Spasticity is most commonly

observed in the anti-gravitational musculature (i.e. knee exten-

sors and elbow flexors) and it is rate dependent. Examination of

deep tendon and cutaneous reflexes, including Babinski and

Hoffman signs, should be performed with particular attention

to symmetry. Paratonic rigidity characterized by irregular resis-

tance to passive movements which increases in intensity as the

speed of the movements increases, is typically seen in metabolic

coma. Unconscious patients may display involuntary or reflexive

movements such as tremor, myoclonus and asterixis.

Genetic Inborn errors of metabolism, mitochondrial

diseases

Coma mimics

Epilepsy Status epilepticus

Toxins Illicit drugs, carbon monoxide, lead, gas inhalation

Others Hypothermia, hyperthermia, acute catatonia,

malignant neuroleptic syndrome

Table 3

Despite being rare, coma mimics need to be systematically

excluded. The locked-in syndrome is a ‘de-efferented state’ in

which the patient is fully conscious but can make no spontaneous

movements except for lid and vertical eye movements. Akinetic

mutism due to bifrontal lesions is characterised by preserved

consciousness but with virtually no response to external stimuli.

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:9 377

Eyelid-apraxia in bilateral thalamic infarcts is characterized but

partial consciousness but the inability to open eyes to command.

Psychogenic unresponsiveness is rare and difficult to diagnose.

The lack of arousability can be expression of a psychiatric disorder

(i.e. conversion disorder, severe depression, catatonic stupor) or a

deliberate fabrication. Examination reveals inconsistent volitional

responses, preserved pupillary and vestibulo-ocular reflexes, and

an awake electroencephalography (EEG) pattern.

Causes of coma

Several schemes have been proposed to classify the differential

diagnosis of the cause of coma. In patients admitted to intensive

care, approximately 23% of cases were due to anoxo-ischaemic

encephalopathy, 20% to drug overdose, 10% to stroke, 10%

metabolic brain dysfunction, 7% status epilepticus and the

remainder to general medical disorders. For practical purposes,

the causes should be divided into two categories: structural and

non-structural (metabolic and toxic causes) (Table 3).

Structural (compressive or destructive) causes of coma include:

(a) unilateral lesions which often have an acute onset (i.e.

ischaemia or haemorrhage); (b) bi-hemispheric structural injuries

(i.e. hypoxic-ischaemic encephalopathy); and (c) intrinsic brain-

stem injuries. Non-structural causes of coma include toxic and

metabolic coma, which often have a gradual onset and are associ-

ated with preservation of pupillary reflexes. Non-convulsive status

epilepticus (NCSE) is a prolonged and self-sustained seizure that

may have subtle or almost no behavioural manifestations.

Although NCSE can occur in patients with a known diagnosis of

� 2013 Elsevier Ltd. All rights reserved.

NEUROINTENSIVE CARE

epilepsy often due to non-compliance to anti-epilepticmedications,

it is not uncommon to encounter patients with a de-novo NCSE in

the settingof anacuteneurological (i.e. traumatic brain injury (TBI)

or stroke) or metabolic-toxic condition (i.e. hyponatraemia, acute

withdrawal from psychotropic medications).

Initial investigations

Immediate tests include glucose and electrolytes. Blood gases

should be drawn as they can provide important clues for the

diagnosis. Draw blood cultures on all febrile patients and those

who are hypothermic without a clear cause. Urine can be

collected and screened for toxic substances and drugs. The vast

majority of the patients with acute onset of coma require urgent

computed tomography (CT). CT imaging will detect intracranial

haemorrhage, hydrocephalus and structural abnormalities. MRI

is superior in early detection of ischaemic stroke. CT venogram

and angiogram may also be necessary to exclude conditions such

as a basilar artery thrombus or sinus venous thrombosis. An EEG

should be considered to rule out NCSE. Lumbar puncture should

be performed if infection or inflammation is suspected.

Outcome after coma

The prognosis of comatose patients is heavily dependent on the

underlying cause. However there are factors such as age, coma

duration and the presence of neurological signs (longitudinal

assessment) that should be taken into consideration regardless of

the aetiology. Comas caused by toxic-metabolic derangements,

with the exception of hypoxic brain injury, have better outcomes

compared to comas due to structural causes. Clinical information

should be integrated with biochemical (i.e. high serum S100B

and hyperglycaemia), imaging and electrophysiological data.

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:9 378

MRI remains the imaging investigation of choice in terms of

determining damage extent and the involvement of eloquent

areas, hence functional prognosis can be inferred. There is a

growing role for diffusion tensor imaging (DTI), functional MRI

and proton magnetic resonance spectroscopy (MRS) in indicating

early neurological outcome. Traditional EEG measures have

shown their efficacy in predicting outcome after hypoxic or TBI.

However, recent studies have shown that somatosensory and

event related evoked potentials (N20 and mismatch negativity)

have predictive value superior to EEG.

Conclusions

Determining the cause of an acutely depressed level of con-

sciousness is a difficult clinical challenge. An organized approach

to comatose patients is likely to lead to a better outcome. A

FURTHER READING

1. Laureys S, Schiff ND. Coma and consciousness: paradigms (re)

framed by neuroimaging. Neuroimage 2012; 61: 478e91.

2. Young GB. Coma. Ann N Y Acad Sci 2009; 1157: 32e47.

3. Stevens RD, Bhardwaj A. Approach to the comatose patient. Crit Care

Med 2006; 34: 31e41.

4. Huff JS, Stevens RD, Weingart SD, Smith WS. Emergency neurological

life support: approach to the patient with coma. Neurocrit Care

2012; 17(suppl 1): S54e9.

5. Nelson DW, Rudehill A, MacCallum RM, et al. Multivariate outcome

prediction in traumatic brain injury with focus on laboratory values.

J Neurotrauma 2012; 29: 2613e24.

6. Trinka E, Hofler J, Zerbs A. Causes of status epilepticus. Epilepsia

2012; 53(suppl 4): 127e38.

� 2013 Elsevier Ltd. All rights reserved.