Initial Diagnosis and

Management of Coma
a, b
Stephen J. Traub, MD *, Eelco F. Wijdicks, MD

KEYWORDS
 Coma  Coma mimics  Pathophysiology

KEY POINTS
 Coma is a life-threatening process that requires immediate stabilization and a structured
approach to diagnosis and management.
 The differential diagnosis for coma is long, but is often divided into structural vs. diffuse
neuronal dysfunction; the latter is subdivided into toxic vs. metabolic.
 When available, historical information may be of great use in determining the etiology of
coma; in all cases, a focused physical examination can help greatly refine the differential
diagnosis.
 The definitive treatment of patients with coma is ultimately disease-specific.

INTRODUCTION

Many patients present to the emergency department with an alteration in mental sta-
tus simply as a complication of many serious illnesses. A subset of these patients will
present comatose, a clinical state that is a true medical emergency. Although coma is
a relatively rare presenting condition in the emergency department, patients who pre-
sent with coma are often in extremis and necessitate immediate evaluation and
stabilization.
The approach to coma by the emergency physician is described, beginning with a
discussion of pathophysiology and cause. Then, the practical clinical aspects of coma
are addressed, including initial stabilization, obtaining the correct historical informa-
tion, performing a thorough physical examination, ordering appropriate testing and
imaging studies, and providing appropriate treatment.

PATHOPHYSIOLOGY

A neuronal network in the dorsal pons and midbrain give rise to the ascending reticular
activing system (ARAS), which is responsible for arousal.1 Neurons from these centers

a
Department of Emergency Medicine, Mayo Clinic Arizona, 1-738, 5777 East Mayo Boulevard,
Phoenix, AZ 85054, USA; b Department of Neurology, Mayo Clinic, Rochester, MN, USA
* Corresponding author.
E-mail address: [email protected]

Emerg Med Clin N Am - (2016) -–-

http://dx.doi.org/10.1016/j.emc.2016.06.017 emed.theclinics.com
0733-8627/16/ª 2016 Elsevier Inc. All rights reserved.
2 Traub & Wijdicks

run together through the thalamus and then to the bilateral cerebral cortex; the cortex
controls sensory processing and understanding, which generates awareness.2,3
Coma results from an impairment of this axis by a process that affects the brain’s
arousal center, consciousness center, the tracts that connect them, or some combi-
nation thereof. Patients are, therefore, not aware and not awake. Importantly, coma
from cortical impairment can only result from a bilateral insult; unilateral cortical def-
icits do not cause coma. Prolonged coma may result in awakening cycles (eyes
open coma) without awareness. Because the comatose state is difficult to quantify,
some patients diagnosed as comatose may be minimally aware (minimally conscious
state) and others may be more aware than can be assumed or tested.
Although the final common physiologic pathway of coma is neuronal dysfunc-
tion in the ARAS-thalamic-cortical pathway, it is useful to subdivide the patho-
physiology into structural versus diffuse neuronal dysfunction. Structural causes
of coma are defined as those that precipitate cellular dysfunction through a me-
chanical force, such as pressure on key area or a blockade of delivery of critical
cellular substrate. Diffuse neuronal dysfunction precipitates coma by abnormal-
ities only at the cellular level and may be further divided into two general cate-
gories: toxic and metabolic. In a toxin-induced coma, an exogenous substance
is responsible for the clinical findings; in a metabolic coma, a perturbation of an
endogenous process, such as temperature or sodium regulation, has gone awry.
This classification, although useful, does have limitations. A metabolic process,
such as hypoglycemia or hypoxia, may initially produce coma through diffuse neuronal
dysfunction; however, if the process is uncorrected and cell death occurs, the cause
of coma becomes structural. Similarly, a diffuse neuronal process, such as cerebral
edema, may become a structural problem if the edema occludes vessels in the pos-
terior circulation and produces brainstem ischemia.

CAUSES

A causal overview of coma is presented in Table 1, categorized based on this logic,

and includes coma mimics, which are several disorders that may be easily mistaken

Table 1
Causal overview of coma and coma mimics

Coma
Diffuse Neuronal Dysfunction
Structural Toxic Metabolic Coma Mimics
Neoplasia Sedative-hypnotics Respiratory Locked-in syndrome
Hydrocephalus agents insufficiency Neuromuscular paralysis
Intracranial Opioids Dysthermia Akinetic mutism
hemorrhage Dissociative agents Dysglycemia Psychogenic unresponsiveness
Vascular Carbon monoxide Electrolyte disorders
occlusion Toxic alcohols Infection
Antidepressants Hypothyroidism
Antiepileptics Thiamine deficiency
Agents of Nonconvulsive status
histotoxic hypoxia epilepticus
Simple asphyxiants
Serotonin syndrome
Neuroleptic malignant
syndrome
Clonidine
 Initial Diagnosis and Management of Coma 3

for coma but do not involve interruption of the ARAS-thalamic-cortical pathway. For
the purposes of this article, the focus is on relatively common entities that may pre-
sent with coma, rather than those that are uncommon or in which coma is a late
finding.

STRUCTURAL CAUSES OF COMA

Tumors
Tumors may cause coma by exerting pressure on either a key area (eg, the brainstem)
or by causing a diffuse increase in intracranial pressure. More commonly, however,
patients with tumors have a slow progression of neurologic findings. Abrupt onset
of coma in such patients often results from hemorrhage into an expanding mass.
Even small tumors, however, may cause obstructive hydrocephalus or focal infarc-
tions, each of which may in turn lead to the relatively abrupt onset of coma.

Acute Hydrocephalus
There is approximately 100 to 150 mL of cerebrospinal fluid (CSF) in the adult brain.
CSF is produced predominantly in the choroid plexus, circulates through the ventric-
ular system, and empties into the subarachnoid space where it is absorbed predom-
inantly into the venous system through the arachnoid villi.4 Occlusion of this flow via
tumor, clotting of intraventricular blood, or dysfunction of the arachnoid villi may
lead to an increase in intraventricular CSF, with a concurrent increase in intracranial
pressure and resultant coma.

Intracranial Hemorrhage
Central nervous system (CNS) hemorrhage resulting in coma may have 1 of 4 causes.

Spontaneous subarachnoid hemorrhage

Spontaneous subarachnoid hemorrhage (SAH) usually results from the rupture of an
aneurysm in the Circle of Willis (often referred to as a berry aneurysm). Thunderclap
headache on presentation is present in more than 95% of patients.5 Coma in the
setting of SAH may be due to acute hydrocephalus or anoxic-ischemic injury.

Subdural hemorrhage
Subdural hemorrhage (SDH) is an accumulation of blood between the dura and the
arachnoid membrane. SDH is often associated with a trauma but may also be asso-
ciated with low intracranial pressure, as occurs after lumbar puncture.6 SDH may
occur because of either shearing of bridging veins7 or arterial interruption.8 The use
of both antiplatelet agents and anticoagulants increase the risk of SDH.9 SDH may
produce a rapid shift of brain parenchyma, resulting in compression of the thalamus
and pressure on the brainstem. Seizures, including nonconvulsive status epilepticus
(NCSE), may mimic structural injury and are more often seen after hematoma
evacuation.

Epidural hemorrhage
Epidural hemorrhage (EDH) is most often due to blunt force trauma that disrupts an
epidural artery, with blood collecting in the potential space between the dura and
the skull. Patients may present with initial confusion or loss of consciousness from
which they recover, only to subsequently “talk and deteriorate.” This lucid interval oc-
curs in approximately half of all EDH patients.10 Coagulopathy is associated with a
poorer outcome in patients with EDH.11 Similar to SDH, brain parenchymal shift, brain-
stem pressure, and seizures may result.
4 Traub & Wijdicks

Intraparenchymal hemorrhage
Intraparenchymal hemorrhage (IPH) is usually due to longstanding hypertension and
associated vascular changes, although amyloid angiopathy and coagulopathy are
other possible causes.12 Coma from an IPH may be caused by the disruption of key
tracts or a general increase in intracranial pressure, depending on the location of
the lesion.

Vascular Occlusion
Arterial vascular occlusion may be either thrombotic or embolic; both may produce
coma if critical structures are affected. Of note, arterial vascular occlusion causing
coma is usually a posterior circulation event, with occlusion in the vertebrobasilar sys-
tem leading to hypoperfusion of crucial structures within the ARAS. Arterial occlusion
in the anterior circulation is an uncommon cause of coma because bilateral cortical
disruption is required to produce the requisite depression of consciousness. This
may occur, however, in patients who have suffered a stroke on one side of the brain
and subsequently suffer an acute arterial vascular occlusion on the other.

DIFFUSE NEURONAL DYSFUNCTION CAUSES OF COMA: METABOLIC

Respiratory Insufficiency
Respiratory insufficiency may produce coma in two ways. First, the brain is particularly
sensitive to the effects of hypoxia, with coma possible within minutes of acute oxygen
deprivation. Second, hypercarbia may cause coma; the exact mechanism is unclear,
but may involve an alteration in neurotransmitter levels or changes in intracranial pres-
sure as increases in carbon dioxide levels are associated with increases in cerebral
blood flow.

Dysthermia
Extremes of body temperature may accompany other primary causes of coma or be a
primary cause. Although the exact temperature at which coma occurs will vary by in-
dividual, loss of consciousness in hypothermic patients generally occurs around
28 C13 and hyperthermia-induced coma generally does not occur below a tempera-
ture of 40 C.14

Hypertension
Rarely, severe hypertension may result in a loss of vascular epithelial integrity in small
vessels of the brain, resulting in a patchwork pattern of vascular narrowing and vaso-
dilation, resulting in cerebral edema.15 This condition, called posterior reversible en-
cephalopathy syndrome, may present with significant alterations in consciousness.

Dysglycemia
Hypoglycemia may produce virtually any neurologic sign, symptom, or syndrome,
including coma. Hypoglycemia is most common in diabetic patients who are taking
hypoglycemic agents, such as insulin or sulfonylureas, and the rate of coma in such
patients is about 1% to 2% per year.16,17 Hyperglycemia may also cause coma,
most commonly in the setting of a hyperosmolar hyperglycemic state (HHS) in which
glucose levels are greater than 600 mg/dL and osmolality greater than
320 mOsm/kg.18 Coma is more common in HHS than diabetic ketoacidosis (DKA).
Serum osmolality is the driver of mental status changes in hyperglycemic states and
HHS is associated with higher serum osmolality levels than DKA.19
 Initial Diagnosis and Management of Coma 5

Electrolyte Disorders
Disorders of sodium hemostasis, particularly when they are acute, may produce
coma. Hyponatremia produces an imbalance of intracellular versus extracellular
osmolality, the flow of free water into the brain parenchyma, and the development
of cerebral edema.20 Hypernatremia may also cause coma, and overly-rapid correc-
tion (particularly when the hypernatremia develops acutely) may lead to either demy-
elination or intracranial hemorrhage due to abrupt changes in intraparenchymal
volume.21
Hypercalcemia is common in patients with advanced malignancy, occurring in 10%
to 20% of such patients.22 Although the most common neurologic presentations of hy-
percalcemia are confusion, delirium, or lethargy, coma is reported.23
Infection
Coma in the setting of infection may be due to one of several CNS infections. Profound
coma is a rare presentation of meningitis24 but is more commonly seen in fulminant
cases. In one series, approximately 10% of subjects with encephalitis presented
with coma (defined by the investigators as Glasgow Coma Scale [GCS] 8) and these
subjects had poorer outcomes.25
Systemic non-CNS infections, such as sepsis, may also produce coma. The myriad
biochemical and microcirculatory changes involved in sepsis-induced coma are
incompletely understood but seem to activate neuroinflammatory and ischemic path-
ways culminating in dysfunction of the brain parenchyma.26
Thyroid Disorders
Myxedema coma is a severe form of hypothyroidism in which alterations in cerebral
blood flow and glucose metabolism may lead to significant changes in mental status27
and coma.28 The alterations in mental status that accompany hyperthyroidism classi-
cally include nervousness and anxiety,29 but decreases in mental status (which may
include coma) may occur and are more common in the elderly.30,31
Renal Failure
Renal failure produces neurologic findings that include uremic encephalopathy, which
in severe cases may manifest as coma.32 The molecular basis of uremic encephalop-
athy is not fully elucidated but it is likely a multifactorial process that includes the accu-
mulation of false neurotransmitters.
Hepatic Failure
Hepatic failure may lead to an encephalopathic state caused by either an accumula-
tion of endogenous toxins (including ammonia) or cerebral edema.33
Hyperammonemia
Although hyperammonemia is a common finding in hepatic failure, nonhepatic hyper-
ammonemia may cause coma as well. Valproic acid therapy in the setting of carnitine
deficiency, infection with urease-producing bacteria, recent surgery (particularly lung
transplantation, bariatric surgery or ureterosigmoidostomy), hyperalimentation, and
errors of metabolism are also potential causes.34
Thiamine Deficiency
Thiamine deficiency is a common problem in malnourished patients. In the emergency
department, thiamine deficiency is of particular concern in patients with alcohol-
related presentations, not only in alcoholics but in binge drinkers as well.35 Severe
6 Traub & Wijdicks

thiamine deficiency, usually seen in the context of alcoholism, may lead to Wernicke
encephalopathy (characterized by encephalopathy, oculomotor dysfunction, and
gait ataxia) or Korsakoff psychosis (a chronic amnestic condition). Coma as a present-
ing symptom of thiamine deficiency, however, is very uncommon.36,37

Nonconvulsive Status Epilepticus

NCSE is an epileptogenic condition in which the classic manifestations of seizure (eg,
focal or general motor activity) are absent. NCSE may be mistaken for coma or unre-
sponsiveness, and is an under-recognized cause of altered mental status in the emer-
gency department.38

DIFFUSE NEURONAL DYSFUNCTION CAUSES OF COMA: TOXINS

Sedative-Hypnotic Agents
Sedative-hypnotic agents are a broad class of drugs that include ethanol,
benzodiazepines, barbiturates, baclofen, gamma-hydroxybutyrate, and others.
Most sedative-hypnotic agents act by facilitating the effect of the neurotransmitter
gamma-aminobutyric acid (GABA), hyperpolarizing neurons either through an in-
crease in chloride conductance (GABAA)39 or through an increase in potassium
conductance (GABAB).40 Ethanol, in addition to interacting with the GABA system,41
also produces some effects via interference with the excitatory neurotransmitter
N-methyl-D-asparate (NMDA).42

Opioids
Opioids (ie, heroin, morphine, oxycodone, hydrocodone, and others) may produce
profound decreases in mental status, including coma, in addition to other clinical find-
ings such as respiratory depression. Opioid receptors are coupled to G proteins,
which may exert their effects via adenylate cyclase, calcium channels, or potassium
channels.43 Opioids have multiple receptor subtypes, with the mu receptor respon-
sible for coma.

Dissociative Agents
Phencyclidine and ketamine depress (and therefore interrupt) thalamic-cortical
tracts,44 producing a temporary state in which cardiorespiratory functions are pre-
served but in which the patient is dissociated from his or her higher functions. Disso-
ciative agents likely exert most of their effects via NMDA antagonism but also have
effects on opiate receptors and sympathetic neurotransmission.45

Carbon Monoxide
Carbon monoxide (CO) poisoning is alarmingly prevalent, accounting for approxi-
mately 50,000 visits per year to US emergency departments46 and, in severe cases,
presenting with coma.47 CO is a complex toxin that affects oxyhemoglobin dissocia-
tion, increases oxidative stress, interrupts cellular respiration, and leads to the gener-
ation of reactive oxygen species. All of these may contribute to the development of
neurologic impairment.48

Serotonin Syndrome and Neuroleptic Malignant Syndrome

Serotonin syndrome (SS) and neuroleptic malignant syndrome (NMS) are distinct en-
tities with overlapping presentations that, when severe, include profound alterations in
mental status, muscular rigidity, and hyperthermia. SS results from an excess of cen-
tral and peripheral serotonin activity, often when two or more serotonergic agents are
 Initial Diagnosis and Management of Coma 7

used together49 NMS (although less well understood) likely results from central dopa-
minergic blockade.50

Miscellaneous Toxins
Several other toxins may produce coma. Toxic alcohols, such as methanol and
ethylene glycol, are CNS depressants that produce coma in a manner similar to
ethanol. Psychiatric medications, such as tricyclic antidepressants and serotonin se-
lective reuptake inhibitors, may produce coma as an exaggeration of their normal
pharmacologic effects. Simple asphyxiants, such as nitrogen, act by displacing oxy-
gen and producing hypoxia. Agents of histotoxic hypoxia, such as cyanide, interfere
with aerobic metabolism and the generation of adenosine triphosphate. Clonidine al-
ters central sympathomimetic neurotransmission.

Coma Mimics
Four conditions deserve mention as coma mimics. The locked-in syndrome describes
paralysis of all voluntary muscles of the body save the eyes, usually as a result of
ischemia or infarction to key CNS tracts often involving the pons, with preservation
of consciousness and higher cortical functions.51 Neuromuscular paralysis may be iat-
rogenic, after the administration of succinycholine or curare-like drugs, or may arise
from varied environmental sources, such as the toxin of Clostridium botulinum, the
venom of elapid snakes,52 or tetrodotoxin-producing organisms such as the blue-
ringed octopus.53 Akinetic mutism usually results from injury to the frontal or prefrontal
motor cortex, in which patients cannot initiate voluntary motor movements.54 In all 3 of
these diseases, consciousness is preserved. The fourth coma mimic is psychogenic
unresponsiveness, a complex disorder in which there is no neurologic insult; the con-
dition resolves spontaneously.55

INITIAL STABILIZATION

The initial stabilization of comatose patients is the same as that for that of all emer-
gency department patients and consists of securing the patients airway (with attention
to the cervical spine), breathing, and circulation.
Decisions regarding airway management are often very difficult, driven by
gestalt rather than algorithmic decision making, and are based on several factors.
Mechanism of coma is important; although a GCS of 8 or less in a trauma patient
is often viewed as an indication for intubation,56,57 poisoned patients with such
GCS levels can be managed without intubation and with low levels of complica-
tions.58,59 Monitoring concerns may also enter into the decision-making process.
Patients who require significant time out of the department for diagnostic imaging,
as may occur during a computerized tomography (CT) scan, may require intuba-
tion; whereas patients who remain in an acute care area might be managed
expectantly, even at the same level of consciousness. Expected clinical course,
particularly in poisoned patients, is also a factor. The patient with an isolated al-
prazolam ingestion will likely do well without intubation, whereas a patient with
carbamazepine ingestion is more likely to have a complicated course and require
airway intervention.
Concurrently with airway management, the cervical spine must be stabilized when-
ever there is a possibility that the patient’s alteration in mental status has a traumatic
cause. Cervical spine injuries are commonly associated with alterations mental status
of traumatic cause, occurring in 5% or more of such patients.60,61
8 Traub & Wijdicks

HISTORY

Comatose patients by definition cannot give details of their illness, so it is crucial that
the provider actively seek alternative sources of information. Emergency medical ser-
vice responders often provide the most valuable information. They can relay informa-
tion obtained from family members or bystanders, describe the patient’s initial level of
consciousness and how that has changed en route, provide a description of how the
patient was found, and contribute important situational information such as the pres-
ence or absence of prescription bottles or drug paraphernalia. Ideally, the belongings
of comatose patients should also be examined for clues, such as pill bottles in coat
pockets or a purse, a pharmacy phone number, or a medication list in a wallet. Finally,
medical alert bracelets or information from the institution’s medical record may shed
light on an otherwise confusing presentation.
To the extent possible, the clinician should establish the rate of progression of
symptoms. Coma because of subarachnoid hemorrhage, cerebellar infarction,
and IPH is usually of abrupt onset, whereas coma from infection may evolve
over hours to days. Coma from poisoning may occur over minutes if caused
by a large dose of drug (eg, an opioid) or a rapidly acting toxin (eg, volatilized
cyanide after malicious mixing with a soft drink). Alternatively, it may occur
over the course of hours if the patient over-titrates a self-administered substance
(eg, ethanol) or ingests a substance that may have delayed toxicity (eg,
baclofen).
Importantly, when obtaining historical information, providers must avoid the errors
of premature closure and diagnostic anchoring.62 For example, alcohol is ubiquitous
in many western countries and its presence may be coincident with, rather than caus-
ative of, coma. SDH and hypoglycemia are life-threatening causes of coma that may
coexist with ethanol intoxication.63,64 As with any condition, physicians evaluating pa-
tients with coma must be willing to reconsider initial impressions when additional data,
such as a fever or classic physical examination findings, are inconsistent with the initial
working diagnosis.

PHYSICAL EXAMINATION

A complete physical examination will provide clues to the diagnosis of coma and help
streamline the patient’s diagnostic evaluation. Crucial physical examination findings,
and the important causes of coma associated with them, are listed below.

VITAL SIGNS
Pulse
Bradycardia may occur in the context of sympatholytic drugs, such as clonidine; in the
setting of sedative hypnotic toxicity, particularly with barbiturates and gamma-
hydroxybutyrate; and with increases in intracranial pressure, characteristically accom-
panied by systemic hypertension. Tachycardia is common with psychotropic drug
poisoning, ketamine intoxication, adrenergic hyperactivity from intracranial hemor-
rhage, and 3,4-methylenedioxymethamphetamine (MDMA) intoxication, which pro-
duces coma via symptomatic hyponatremia.65

Blood Pressure
Hypotension may occur in sepsis and many poisonings, particularly tricyclic antide-
pressants, sedative-hypnotic agents, and clonidine. Hypertension is a prerequisite
for the diagnosis of hypertensive encephalopathy and is common in ketamine or
 Initial Diagnosis and Management of Coma 9

phencyclidine intoxication, MDMA intoxication, and in the setting of increased

intracranial pressure. In the cases of elevated intracranial pressure, the hyperten-
sion is usually accompanied by bradycardia, a combination known as Cushing
response.

Respiratory Rate
Tachypnea is common with metabolic acidosis of any cause. Bradypnea may be seen
in both opioid and sedative-hypnotic toxicity but is much more pronounced in the
former. Bradypnea in the setting of structural CNS disease suggests medullary
involvement and is often a preterminal event.

Temperature
Hyperthermia may be due to environmental exposure, infection, NMS, SS, salicylate
poisoning, and several primary CNS disorders, including subarachnoid hemorrhage
and hypothalamic injury. Hypothermia is commonly due to environmental exposure
but may also be seen with hypoglycemia of any cause, sedative-hypnotic toxicity, hy-
pothyroidism, and overwhelming infection.

Physical Examination
Head, eyes, ears, nose, and throat
Head Examination of the head may show obvious signs of deformity, such as crepitus
or bony step-offs in the setting of a skull fracture. Other signs suggestive of basilar
skull fracture that should be specifically evaluated are bilateral orbital ecchymosis
(Raccoon Eyes) and postauricular bruising (Battle’s sign).

Eyes Miosis is commonly seen in opioid and clonidine toxicity, and may also be seen
in the setting of pontine hemorrhage. Mydriasis is common in poisoning with com-
pounds with anticholinergic properties (eg, tricyclic antidepressants) and MDMA. Hor-
izontal nystagmus is common in ethanol intoxication and with toxicity from
antiepileptic drugs. Vertical nystagmus is uncommon and suggests dissociative
agents (eg, phencyclidine and ketamine) or brainstem dysfunction. Gaze deviation
may be in response to an ipsilateral hemispheric lesion, a contralateral pontine lesion,
or a seizure focus. Roving eye movements do not specifically localize a lesion but sug-
gest that the brainstem is intact. When ice water irrigation of the ear is performed in
normal patients, there is tonic deviation of the eyes toward the side of the irrigation
followed by rapid nystagmus away from the side of the irrigation. Loss of the former
suggests a midbrain or pontine lesion, loss of the latter suggests a cortical lesion,
and loss of neither suggests psychogenic coma.
Ears Hemotympanum may be seen in approximately 50% of basilar skull fractures.66

Throat Dry mucous membranes are suggestive of profound dehydration or anticholin-

ergic toxicity, whereas increased salivation may be seen with ketamine toxicity.

Skin
Excessively dry skin suggests poisoning with a drug with anticholinergic properties,
such as tricyclic antidepressants. Diaphoresis may be seen in any hyperadrenergic
state. Coma bullae are classically associated with barbiturate toxicity but may also
be seen in other settings, such as infection.67 Small linear areas of pinprick-size
trauma over veins (track marks) suggest the ongoing abuse of intravenous drugs,
particularly opioids, but cannot be used in and of themselves to diagnose the cause
of a comatose state.
10 Traub & Wijdicks

Bowel Sounds
Markedly decreased bowel sounds are associated with both anticholinergic toxicity
and opioid toxicity.

The Toxidrome-Oriented Physical Examination

Syndromes are a constellation of signs or symptoms that together suggest a certain
disease entity. In humans, certain toxins or classes of toxins produce stereotypical
physical examination findings. These toxic syndromes, or toxidromes, can provide
invaluable information. In comatose patients, the use of vital signs and examination
of the eye, throat (mucous membranes), skin, and bowel sounds may yield a pattern
that strongly suggests a given class of toxins (Table 2).

IMAGING AND LABORATORY TESTING

Although a thorough history and physical examination will often generate a refined dif-
ferential diagnosis, imaging studies and laboratory testing play an important role in the
diagnosis of coma. Such interventions, however, should serve to refine clinical impres-
sions and should not be ordered indiscriminately as a substitute for thoughtful patient
evaluation.
When coma is obviously caused by diffuse neuronal dysfunction, such as hypogly-
cemia or a known ingestion, CT is rarely if ever necessary. The diagnostic yield of CT in
this setting is exceedingly low.68 When a structural cause is suspected, particularly in
the setting of trauma, CT of the head is essential and will often identify the lesion. In the
emergency department, most patients with a structural cause of coma will have a
hemorrhagic syndrome.
MRI scans allow greater definition of cortical and subcortical structures and may
show cortical injury, laminar necrosis, or white matter disease that is not apparent
on a CT scan. Magnetic resonance angiography allows for excellent visualization of
the arterial system and is the preferred method for visualization of suspected acute
basilar artery occlusion; however, this may not be possible at all institutions. In such
instances, CT angiogram or conventional angiography can allow for rapid visualization
of the arterial system. The major drawbacks of MRI scans include accessibility, cost,
and time necessary to complete the scan.
The authors believe that basic laboratory testing should include the following at a
minimum: a complete blood count, to assess for leukocytosis as corroborating evi-
dence of infection and to assess the numerical adequacy of platelets in the event of
CNS hemorrhage; an electrocardiogram, to exclude toxin-induced conduction system
abnormalities; serum glucose, preferably via a rapid bedside test; serum electrolytes,

Table 2
Physical examination findings for ingestions producing coma

Finding Opiates Sedative-Hypnotic Agents Dissociative Agents

Pulse Normal vs decreased Normal vs decreased Increased
Blood pressure Normal vs decreased Normal vs decreased Increased
Respiratory rate Markedly decreased Normal vs decreased Normal
Mucous membranes Normal Normal Salivation (ketamine)
Eyes Miosis Normal Nystagmus
Skin Normal Normal Normal
Bowel sounds Markedly decreased Normal Normal
 Initial Diagnosis and Management of Coma 11

including calcium; and tests of renal and hepatic function. Additional high-yield testing
in the correct clinical setting may include a serum ammonia level as a marker for he-
patic encephalopathy, or as a primary hyperammonemic disorder that may occur in
the setting of valproic acid treatment.69 An arterial or venous blood gas may also be
helpful, particularly when poisoning with salicylates, toxic alcohols, or CO is
suspected.
The authors discourage the routine use of urine drug screen (UDS) testing, for
several reasons. First, such tests do not measure intoxication or toxicity but, instead,
the presence of a drug or drug metabolite, meaning that findings on a positive UDS
may be coincident with, rather than causative of, coma. Second, UDS tests may
show false negative results. For example, not all benzodiazepines react with
commonly used commercial benzodiazepine assays.70 In such cases, over-reliance
on a UDS may cause a clinician to reject a diagnosis that is clinically apparent. Third,
UDS tests may generate false-positive results; this is particularly true with the phency-
clidine assay, which is often positive in the setting of commonly used medications
such as the cough suppressant dextromethorphan.71 Finally, and perhaps most
importantly, studies suggest that UDS testing does not significantly affect clinical de-
cision making.72,73
Electroencephalographic (EEG) testing may be useful in patients who have a history
of a seizure disorder, or who have had several seizures before the development of
coma, to exclude NCSE. The authors strongly recommend against emergency depart-
ment EEG testing when an alternative diagnosis is clear, such as a large subdural he-
matoma, or strongly suspected in the case of patients with clear alcohol intoxication.
When EEG is used in the ED, the authors suggest that an abbreviated protocol be used
if at all possible.74
Lumbar puncture should be performed in any case in which meningitis or en-
cephalitis is suspected. Historical context, such as a close contact with meningi-
tis; symptoms, such as photophobia or neck pain; or physical signs, such as a
fever or meningismus, may all suggest a CNS infection. Unfortunately, historical
and physical examination findings are neither sensitive nor specific for the diag-
nosis of CNS infection.75 Therefore, the decision to perform a lumbar puncture
must be made on a case-by-case basis by the treating physician. Generally, the
authors do not advocate for routine lumbar puncture in all comatose patients,
particularly in those for whom an alternative diagnosis is established or strongly
suspected.

GRADING SYSTEMS

Grading systems allow providers to quickly convey a general sense of the patient’s
condition. This is particularly important when one provider cannot examine the patient,
as may occurs in telephone discussions between providers or during emergency med-
ical service communications.
Two major grading systems are used to assess the depths of coma (Table 3). The
GCS, first described in 1974,76 is a 15-point composite score of eye, motor, and ver-
bal responses developed to assess patients with head trauma. The Full Outline of
Unresponsiveness (FOUR) score, first described in 2005, is a 16-point scale that as-
sesses eye, motor, brainstem reflexes, and respirations.77 Both rely on clinical find-
ings, have high inter-rater reliability,78 and can be performed quickly. The FOUR
score has a benefit over GCS in that one of three of the GCS criteria is unreliable
in intubated patients. The FOUR score is also a better predictor of mortality in criti-
cally ill patients.79
12 Traub & Wijdicks

Table 3
Glasgow coma scale and full outline of unresponsiveness (FOUR) score

Glasgow Coma Scale Full Outline of Unresponsiveness Score

Eye Opening Response Eye response
Spontaneously, 4 Open and tracking or blinking to command, 4
To speech, 3 Open but not tracking, 3
To pain, 1 Closed but open to loud voice, 2
Best Verbal Response Closed but open to pain, 1
Oriented, 5 Closed with pain, 0
Disoriented, 5 Motor response
Inappropriate words, 4 Thumbs up, fist or peace sign, 4
Incomprehensible sounds, 3 Localizes pain, 3
No response, 1 Flexion response to pain, 2
Best Motor Response Extension response to pain, 1
Obeys commands, 6 No response to pain or myoclonic status, 0
Localizes pain, 5 Brainstem reflexes
Withdrawal from pain, 4 Pupil and corneal reflexes present, 4
Decorticate flexion, 3 1 pupil wide and fixed, 3
Decerebrate extension, 2 Pupil or corneal reflexes absent, 2
No movement, 1 Pupil and corneal reflexes absent, 1
Absent pupil, corneal and cough reflex, 0
Respiration
Not intubated, regular pattern, 4
Not intubated, Cheyne-Stokes pattern, 3
Not intubated, irregular breathing, 2
Breathes above ventilator rate, 1
Breathes at ventilator rate or apnea, 0

Importantly, both systems exist as scales, meaning that coma is rarely a yes or no
delineation. Although many investigators define coma as a GCS of less than 8, this
cutoff represents an arbitrary choice more than a significant physiologic distinction.

TREATMENT

The ultimate treatment of coma depends on the cause. In general, there are three
overarching themes regarding the treatment of coma.
First, coma from structural causes may be catastrophic and untreatable. However,
when the cause is treatable, it can be treated surgically or with geographically targeted
pharmacologic or mechanical intervention. The authors advocate for the early involve-
ment of neurosurgical specialists for patients with coma from intracranial hemorrhage
or hydrocephalus because early intervention can have an extraordinary effect on both
mortality and long-term outcomes. Patients with ischemic cerebrovascular disease
should be promptly assessed, ideally by a team that includes a neurologist, to assess
their candidacy for either intravenous or intra-arterial thrombolysis.
Second, in patients whose coma is a result of a metabolically induced diffuse
neuronal dysfunction, treatment involves progress toward homeostasis. In some
cases, such as hypoglycemia and respiratory insufficiency, the goal is expeditious
normalization of values such as serum glucose or the partial pressure of oxygen or car-
bon dioxide in the blood. In other cases, such as hypertensive encephalopathy and
hyponatremia, the correct initial treatment involves only a partial correction, and an
abrupt return to normal could be clinically devastating.80
Third, in patients whose coma is a result of a toxin-induced diffuse neuronal
dysfunction, the most important intervention, established more than 50 years ago, is
 Initial Diagnosis and Management of Coma 13

the provision of appropriate supportive care.81 Initial supportive care includes

securing the airway, assuring adequate oxygenation and ventilation, and assuring
appropriate circulation with intravenous fluids and, if necessary, vasopressors.
Advanced supportive care may include altering systemic or compartmental pH to
reduce drug toxicity or increase drug excretion, such as administering sodium bicar-
bonate for tricyclic antidepressant or salicylate toxicity, or the administration of intra-
venous lipid emulsion to alter drug distribution.82 Although there are specific toxins
that induce coma for which specific antidotal therapy may be critical or lifesaving
(eg, fomepizole for toxic alcohols, hydroxocobalamin for cyanide, or naloxone for opi-
oids), antidotal therapy plays little or no role in treating most toxins that may produce
coma.

SUMMARY

Coma represents a true medical emergency. Although drug intoxications are a leading
cause of coma in patients who present to the emergency department, other metabolic
disturbances and traumatic brain injury are common causes as well. The general
emergency department approach to the patient with coma begins with stabilization
of airway, breathing, and circulation, followed by a thorough physical examination
to generate a limited differential diagnosis that is then refined by focused testing.
Definitive treatment is ultimately disease-specific.

REFERENCES

1. Edlow BL, Takahashi E, Wu O, et al. Neuroanatomic connectivity of the human

ascending arousal system critical to consciousness and its disorders.
J Neuropathol Exp Neurol 2012;71(6):531–46.
2. Sarasso S, Rosanova M, Casali AG, et al. Quantifying cortical EEG responses to
TMS in (un)consciousness. Clin EEG Neurosci 2014;45(1):40–9.
3. Widjicks EFM. The comatose patient. Oxford University Press; 2014.
4. Han CY, Backous DD. Basic principles of cerebrospinal fluid metabolism and
intracranial pressure homeostasis [review]. Otolaryngol Clin North Am 2005;
38(4):569–76.
5. Gorelick PB, Hier DB, Caplan LR, et al. Headache in acute cerebrovascular dis-
ease. Neurology 1986;36(11):1445–50.
6. Gaucher DJ Jr, Perez JA Jr. Subdural hematoma following lumbar puncture. Arch
Intern Med 2002;162(16):1904–5.
7. Miller JD, Nader R. Acute subdural hematoma from bridging vein rupture: a po-
tential mechanism for growth [review]. J Neurosurg 2014;120(6):1378–84.
8. Maxeiner H, Wolff M. Pure subdural hematomas: a postmortem analysis of their
form and bleeding points. Neurosurgery 2007;61(Suppl 1):267–72 [discussion:
272–3].
9. De Bonis P, Trevisi G, de Waure C, et al. Antiplatelet/anticoagulant agents and
chronic subdural hematoma in the elderly. PLoS One 2013;8(7):e68732.
10. Bullock MR, Chesnut R, Ghajar J, et al, Surgical Management of Traumatic Brain
Injury Author Group. Surgical management of acute epidural hematomas [re-
view]. Neurosurgery 2006;58(Suppl 3):S7–15 [discussion: Si–Siv].
11. Mayr R, Troyer S, Kastenberger T, et al. The impact of coagulopathy on the
outcome of traumatic epidural hematoma. Arch Orthop Trauma Surg 2012;
132(10):1445–50.
12. Aguilar MI, Freeman WD. Spontaneous intracerebral hemorrhage [review]. Semin
Neurol 2010;30(5):555–64.
14 Traub & Wijdicks

13. Brown DJ, Brugger H, Boyd J, et al. Accidental hypothermia [review]. N Engl J
Med 2012;367(20):1930–8 [Erratum appears in N Engl J Med 2013;368(4):394].
14. Bouchama A, Knochel JP. Heat stroke [review]. N Engl J Med 2002;346(25):
1978–88.
15. Strandgaard S, Paulson OB. Cerebral blood flow and its pathophysiology in hy-
pertension. Am J Hypertens 1989;2(6 Pt 1):486–92.
16. Schloot NC, Haupt A, Schütt M, et al, German/Austrian DPV Initiative. Risk of se-
vere hypoglycemia in sulfonylurea-treated patients from diabetes centers in Ger-
many/Austria: how big is the problem? which patients are at risk? Diabetes Metab
Res Rev 2016;32(3):316–24.
17. Kurien M, Mollazadegan K, Sanders DS, et al. A nationwide population-based
study on the risk of coma, ketoacidosis and hypoglycemia in patients with celiac
disease and type 1 diabetes. Acta Diabetol 2015;52(6):1167–74.
18. Pasquel FJ, Umpierrez GE. Hyperosmolar hyperglycemic state: a historic review
of the clinical presentation, diagnosis, and treatment [review]. Diabetes Care
2014;37(11):3124–31.
19. Lorber D. Nonketotic hypertonicity in diabetes mellitus [review]. Med Clin North
Am 1995;79(1):39–52.
20. Verbalis JG, Goldsmith SR, Greenberg A, et al. Diagnosis, evaluation, and treat-
ment of hyponatremia: expert panel recommendations. Am J Med 2013;126(10
Suppl 1):S1–42.
21. Sterns RH. Disorders of plasma sodium–causes, consequences, and correction
[review]. N Engl J Med 2015;372(1):55–65.
22. Legrand SB. Modern management of malignant hypercalcemia [review]. Am J
Hosp Palliat Care 2011;28(7):515–7.
23. Ben-Asuly S, Horne T, Goldschmidt Z, et al. Coma due to hypercalcemia in a pa-
tient with Paget’s disease and multiple parathyroid adenomata. Am J Med Sci
1975;269(2):267–75.
24. Lucas MJ, Brouwer MC, van der Ende A, et al. Outcome in patients with bacterial
meningitis presenting with a minimal Glasgow Coma Scale score. Neurol Neuro-
immunol Neuroinflamm 2014;1(1):e9.
25. Singh TD, Fugate JE, Rabinstein AA. The spectrum of acute encephalitis: causes,
management, and predictors of outcome. Neurology 2015;84(4):359–66.
26. Adam N, Kandelman S, Mantz J, et al. Sepsis-induced brain dysfunction [review].
Expert Rev Anti Infect Ther 2013;11(2):211–21.
27. Constant EL, de Volder AG, Ivanoiu A, et al. Cerebral blood flow and glucose
metabolism in hypothyroidism: a positron emission tomography study. J Clin En-
docrinol Metab 2001;86(8):3864–70.
28. Kwaku MP, Burman KD. Myxedema coma [review]. J Intensive Care Med 2007;
22(4):224–31.
29. Devereaux D, Tewelde SZ. Hyperthyroidism and thyrotoxicosis [review]. Emerg
Med Clin North Am 2014;32(2):277–92.
30. Boelaert K, Torlinska B, Holder RL, et al. Older subjects with hyperthyroidism pre-
sent with a paucity of symptoms and signs: a large cross-sectional study. J Clin
Endocrinol Metab 2010;95(6):2715–26.
31. Ghobrial MW, Ruby EB. Coma and thyroid storm in apathetic thyrotoxicosis [re-
view]. South Med J 2002;95(5):552–4.
32. Seifter JL, Samuels MA. Uremic encephalopathy and other brain disorders asso-
ciated with renal failure [review]. Semin Neurol 2011;31(2):139–43.
33. Bernal W, Wendon J. Acute liver failure. N Engl J Med 2013;369(26):2525–34.
 Initial Diagnosis and Management of Coma 15

34. LaBuzetta JN, Yao JZ, Bourque DL, et al. Adult nonhepatic hyperammonemia: a
case report and differential diagnosis [review]. Am J Med 2010;123(10):885–91.
35. Li SF, Jacob J, Feng J, et al. Vitamin deficiencies in acutely intoxicated patients in
the ED. Am J Emerg Med 2008;26(7):792–5.
36. Wallis WE, Willoughby E, Baker P. Coma in the Wernicke-Korsakoff syndrome.
Lancet 1978;2(8086):400–1.
37. Gibb WR, Gorsuch AN, Lees AJ, et al. Reversible coma in Wernicke’s encepha-
lopathy. Postgrad Med J 1985;61(717):607–10.
38. Zehtabchi S, Abdel Baki SG, Malhotra S, et al. Nonconvulsive seizures in patients
presenting with altered mental status: an evidence-based review [review]. Epi-
lepsy Behav 2011;22(2):139–43.
39. Olsen RW, Yang J, King RG, et al. Barbiturate and benzodiazepine modulation of
GABA receptor binding and function [review]. Life Sci 1986;39(21):1969–76.
40. Emson PC. GABA(B) receptors: structure and function [review]. Prog Brain Res
2007;160:43–57.
41. Valenzuela CF, Jotty K. Mini-review: effects of ethanol on GABAA receptor-
mediated neurotransmission in the cerebellar cortex–recent advances. Cere-
bellum 2015;14(4):438–46.
42. He Q, Titley H, Grasselli G, et al. Ethanol affects NMDA receptor signaling at
climbing fiber-Purkinje cell synapses in mice and impairs cerebellar LTD.
J Neurophysiol 2013;109(5):1333–42.
43. Waldhoer M, Bartlett SE, Whistler JL. Opioid receptors [review]. Annu Rev Bio-
chem 2004;73:953–90.
44. Miyasaka M, Domino EF. Neural mechanisms of ketamine-induced anesthesia. Int
J Neuropharmacol 1968;7(6):557–73.
45. Stahl SM. Mechanism of action of ketamine. CNS Spectr 2013;18(4):171–4.
46. Hampson NB, Weaver LK. Carbon monoxide poisoning: a new incidence for an
old disease. Undersea Hyperb Med 2007;34(3):163–8.
47. Ku CH, Hung HM, Leong WC, et al. Outcome of patients with carbon monoxide
poisoning at a far-east poison center. PLoS One 2015;10(3):e0118995.
48. Weaver LK. Clinical practice. Carbon monoxide poisoning [review]. N Engl J Med
2009;360(12):1217–25.
49. Boyer EW, Shannon M. The serotonin syndrome [review]. N Engl J Med 2005;
352(11):1112–20 [Erratum appears in N Engl J Med 2009;361(17):1714; N Engl
J Med 2007;356(23):2437].
50. Caroff SN, Mann SC. Neuroleptic malignant syndrome [review]. Med Clin North
Am 1993;77(1):185–202.
51. Cardwell MS. Locked-in syndrome. Tex Med 2013;109(2):e1.
52. Bawaskar HS, Bawaskar PH. Envenoming by the common krait (Bungarus caer-
uleus) and Asian cobra (Naja naja): clinical manifestations and their management
in a rural setting. Wilderness Environ Med 2004;15(4):257–66.
53. Cavazzoni E, Lister B, Sargent P, et al. Blue-ringed octopus (Hapalochlaena sp.)
envenomation of a 4-year-old boy: a case report. Clin Toxicol (Phila) 2008;46(8):
760–1.
54. Ackermann H, Ziegler W. Akinetic mutism–a review of the literature [review].
Fortschr Neurol Psychiatr 1995;63(2):59–67 [in German].
55. Baxter CL, White WD. Psychogenic coma: case report. Int J Psychiatry Med
2003;33(3):317–22.
56. Dunham CM, Barraco RD, Clark DE, et al, EAST Practice Management Guide-
lines Work Group. Guidelines for emergency tracheal intubation immediately after
traumatic injury [review]. J Trauma 2003;55(1):162–79.
16 Traub & Wijdicks

57. Davis DP. Prehospital intubation of brain-injured patients [review]. Curr Opin Crit
Care 2008;14(2):142–8.
58. Dietze P, Horyniak D, Agius P, et al. Effect of intubation for gamma-hydroxybutyric
acid overdose on emergency department length of stay and hospital admission.
Acad Emerg Med 2014;21(11):1226–31.
59. Duncan R, Thakore S. Decreased Glasgow Coma Scale score does not mandate
endotracheal intubation in the emergency department. J Emerg Med 2009;37(4):
451–5.
60. Schenarts PJ, Diaz J, Kaiser C, et al. Prospective comparison of admission
computed tomographic scan and plain films of the upper cervical spine in trauma
patients with altered mental status. J Trauma 2001;51(4):663–8 [discussion:
668–9].
61. Diaz JJ Jr, Gillman C, Morris JA Jr, et al. Are five-view plain films of the cervical
spine unreliable? A prospective evaluation in blunt trauma patients with altered
mental status. J Trauma 2003;55(4):658–63 [discussion: 663–4].
62. Eva KW, Link CL, Lutfey KE, et al. Swapping horses midstream: factors related to
physicians’ changing their minds about a diagnosis. Acad Med 2010;85(7):
1112–7.
63. Heninger M. Subdural hematoma occurrence: comparison between ethanol and
cocaine use at death. Am J Forensic Med Pathol 2013;34(3):237–41.
64. Williams HE. Alcoholic hypoglycemia and ketoacidosis [review]. Med Clin North
Am 1984;68(1):33–8.
65. Traub SJ, Hoffman RS, Nelson LS. The “ecstasy” hangover: hyponatremia due to
3,4-methylenedioxy-methamphetamine. J Urban Health 2002;79(4):549–55.
66. Liu-Shindo M, Hawkins DB. Basilar skull fractures in children. Int J Pediatr Otorhi-
nolaryngol 1989;17(2):109–17.
67. Bosco L, Schena D, Colato C, et al. Coma blisters in children: case report and
review of the literature [review]. J Child Neurol 2013;28(12):1677–80.
68. Forsberg S, Höjer J, Ludwigs U, et al. Metabolic vs structural coma in the ED–an
observational study. Am J Emerg Med 2012;30(9):1986–90.
69. Cuturic M, Abramson RK. Acute hyperammonemic coma with chronic valproic
acid therapy. Ann Pharmacother 2005;39(12):2119–23.
70. Bertol E, Vaiano F, Borsotti M, et al. Comparison of immunoassay screening tests
and LC-MS-MS for urine detection of benzodiazepines and their metabolites: re-
sults of a national proficiency test. J Anal Toxicol 2013;37(9):659–64.
71. Rengarajan A, Mullins ME. How often do false-positive phencyclidine urine
screens occur with use of common medications? Clin Toxicol (Phila) 2013;
51(6):493–6.
72. Brett AS. Implications of discordance between clinical impression and toxicology
analysis in drug overdose. Arch Intern Med 1988;148(2):437–41.
73. Bast RP, Helmer SD, Henson SR, et al. Limited utility of routine drug screening in
trauma patients. South Med J 2000;93(4):397–9.
74. Bautista RE, Godwin S, Caro D. Incorporating abbreviated EEGs in the initial
workup of patients who present to the emergency room with mental status
changes of unknown etiology. J Clin Neurophysiol 2007;24(1):16–21.
75. Brouwer MC, Thwaites GE, Tunkel AR, et al. Dilemmas in the diagnosis of acute
community-acquired bacterial meningitis [review]. Lancet 2012;380(9854):
1684–92.
76. Teasdale G, Jennett B. Assessment of coma and impaired consciousness. A
practical scale. Lancet 1974;2(7872):81–4.
 Initial Diagnosis and Management of Coma 17

77. Wijdicks EF, Bamlet WR, Maramattom BV, et al. Validation of a new coma scale:
the FOUR score. Ann Neurol 2005;58(4):585–93.
78. Sadaka F, Patel D, Lakshmanan R. The FOUR score predicts outcome in patients
after traumatic brain injury. Neurocrit Care 2012;16(1):95–101.
79. Wijdicks EF, Kramer AA, Rohs T Jr, et al. Comparison of the full outline of unre-
sponsiveness score and the Glasgow Coma Scale in predicting mortality in crit-
ically ill patients. Crit Care Med 2015;43(2):439–44.
80. Rafat C, Flamant M, Gaudry S, et al. Hyponatremia in the intensive care unit: how
to avoid a Zugzwang situation? Ann Intensive Care 2015;5(1):39.
81. Cemmmesen C, Nilson E. Therapeutic trends in the treatment of barbiturate
poisoning. The Scandinavian method. Clin Pharmacol Ther 1961;2:220–9.
82. Ozcan MS, Weinberg G. Intravenous lipid emulsion for the treatment of drug
toxicity [review]. J Intensive Care Med 2014;29(2):59–70.