Coma Scales

TE Gofton and GB Young, Western University, London, ON, Canada

r 2014 Elsevier Inc. All rights reserved.
This article is a revision of the previous edition article by Brian G Young, volume 1, pp 745–746, r 2003, Elsevier Inc.

Introduction consistently in different hospitals, and the later versions of the

GCS have not been adequately tested for reliability. The GCS
Scoring systems were developed for the quantification and was designed for the initial assessment of patients with head
standardization of the severity of acute illnesses and for the injury. Problems with the use of the GCS arise when patients
prediction of outcome. Motivations for their development are intubated and cannot respond verbally or if the eyes are
included the need for reliable and standardized data for re- swollen shut, preventing ocular assessment. A theoretical dis-
source allocation, quality assurance, and improvement initia- advantage is the three-dimensional assessment: The total score
tives. Such clinimetrics rest on pragmatic and clinical is obtained by adding the values for three motor activities –
observational data that can be treated in a scientific fashion to eye opening, best motor response, and best verbal response.
develop scoring systems, which can be validated and tested for These are assumed to be independent variables but they are
clinical relevance. Coma scales serve the following purposes: not. Because they covary, their addition may not be valid.
(1) standardization of level of consciousness for clinical re- Furthermore, to achieve a score of 6–12, there are more than
search (e.g., inclusion in a series and allowing comparison of 10 simple combinations of variables, each with very different
studies), (2) monitoring the course of illness, (3) prognostic clinical profiles. It seems unlikely that all patients with specific
estimations, and (4) management decisions. scores ranging from 6 to 10 will be equivalent in disease se-
The ideal coma scale should have good validity, ordinal verity. Additionally, there is little difference in outcome over
arrangement of degree of severity, linearity (i.e., there should several different score values (e.g., between 10 and 15). The
be an equal weighting to individual units of score), correlation GCS is often insufficiently sensitive for the detection of
with outcome, ease of use (brevity, simplicity, unambiguity, changes in the level of consciousness in patients following
and practicality), and a little redundancy. head injury or with masses and risk of herniation when they
are in lighter stages of impaired consciousness. Furthermore,
in the application of the GCS to patients who have been in the
ICU for an extended period of time, eye opening does not
Specific Scales equal conscious awareness because patients with persistent
vegetative state (VS) may show this and patients with seizures
Glasgow Coma Scale show spontaneous eye opening.
However, the GCS has been the standard scoring assess-
The Glasgow Coma Scale (GCS) (Table 1) is almost uni-
ment throughout the world for 20 years. It seems unlikely that
versally used in emergency rooms and intensive care units
it will be easily replaced, even by potentially superior scoring
(ICUs), and is by far the most common coma scale cited in
systems. The Innsbruck and the Edinburgh-2 Coma Scales
the neurosurgical literature. However, it has not been used
have some of the same problems as the GCS but the Reaction
Level Scale (RLS 85) has a number of advantages over the
others.
Table 1 Glasgow Coma Scale

Item Factor Score

Reaction Level Scale
Best motor response Obeys 6
The RLS 85 (Table 2), developed in Sweden in 1985, is the
Localizes 5
eight-grade single-line ordinal scale for assessment of patients
Withdraws (flexion) 4
Abnormal flexion 3 in the ICU. It can be applied to patients who are intubated or
Extensor response 2 whose eyes have swollen shut. There is no addition of cov-
Nil 1 arying values. Although the numerical values are not neces-
sarily separated by steps of equal value, the order appears to be
Verbal response Oriented 5
valid. The test compares favorably with the GCS and outcome
Confused conversation 4
is inversely related to the achieved score. It is reliable and
Inappropriate words 3
Incomprehensible sounds 2 reproducible, and has also shown good interrater reliability
Nil 1 regardless of the etiology for coma. Furthermore, any change
in the RLS 85 is related to a significant change in patient status
Eye opening Spontaneous 4 and it is superior to the GCS in this respect.
To speech 3
The scale cannot be applied to cases in which the patient is
To pain 2
clinically or pharmacologically paralyzed but alert (e.g.,
Nil 1
polyneuropathy, spinal cord lesion, locked-in syndrome, and

Encyclopedia of the Neurological Sciences, Volume 1 doi:10.1016/B978-0-12-385157-4.00317-1 833

834 Coma Scales

Table 2 Reaction Level Scale 85

Score Clinical descriptor Qualifying factors

1 – Normal Alert Alert; not drowsy; oriented

No delay in response Intubated patient: No delay in reaction
2 – Mildly drowsy or confused Drowsy or confused Drowsy: Drowsy or shows delay in reaction
Responsiveness to light stimulation Confused: Wrong answer to: What is your
name? Where are you? What year and
month is it?
3 – Very drowsy or confused; response to Strong stimulation: Loud verbal or painful Mental responsiveness: Arousable
strong stimulation stimulation
Performs at least one of the following: Obeys
commands (including nonverbal response,
e.g., ‘‘Lift up your arms;’’ orienting eye
movements; wards off painful stimulus)
4 – Unconscious: localizes but does not ward Unconscious: No mental response Localizes
off pain Cannot perform any activity defined under Retromandibular pain: Patient moves one
mental responsiveness hand above chin level; with pain to nail
beds, the other arm crosses the midline

5 – Unconscious: Withdraws Withdrawing movement: With With nail bed pressure, patient makes
retromandibular pressure, patient turns withdrawal movement with abduction at
head away shoulder
6 – Unconscious: Stereotypic flexion Stereotypic movement: With retromandibular The movement is ‘mechanical’ and clearly
movement or nail bed pressure, patient slowly different from the withdrawal response
assumes decorticate posture described previously
7 – Unconscious: Stereotyped extension Stereotyped extension movements: With If there is a mixture of extension and flexion,
movements retromandibular or nail bed pressure, the flexion (best response) is recorded
patient extends the limbs

8 – Unconscious: No response With retromandibular or nail bed pressure,

there is no response of limbs or face

use of neuromuscular blocking agents) or to those with psy- Table 3 Full Outline of UnResponsiveness Score
chogenic unresponsiveness. These are usually not difficult to
identify and exclude. Category Score

Eye response E0 – eyelids remain closed with pain

E1 – eyelids closed, opens to pain, not tracking
E2 – eyelids closed, opens to loud voice, not tracking
Full Outline of UnResponsiveness Scale E3 – eyelids open but not tracking
E4 – eyelids open or opened, tracking or blinking to
The Full Outline of UnResponsiveness (FOUR) scale (Table 3) command
was developed to provide a more comprehensive and clinic-
ally relevant scoring system for use in comatose patients. The Motor M0 – no response to pain or generalized myoclonus
response status epilepticus
FOUR score assesses eye reflexes, motor reflexes, brainstem
M1 – extensor posturing
reflexes, and respiratory pattern. There is no assessment of
M2 – flexion response to pain
verbal responses, which is appropriate, because comatose pa- M3 – localizing to pain
tients are generally intubated. The FOUR score remains per- M4 – thumbs up, fist, or peace sign to command
tinent in intubated patients and has the ability to identify the
locked-in and VSs. It has been validated for use (in multiple Brainstem B0 – absent pupil, corneal, and cough reflex
reflexes B1 – pupil and corneal reflexes absent
languages) in medical ICU and emergency departments by
B2 – pupil or corneal reflexes absent
physicians and nurses, for adult and pediatric populations,
B3 – one pupil wide and fixed
patients with traumatic brain injury, stroke, and comatose B4 – pupil and corneal reflexes present
survivors of cardiac arrest. Interrater reliability is equivalent to
the GCS among the neurologists, other physicians, residents, Respiration R0 – breathes at ventilator rate or apnea
and neuroscience nurses. Evidence shows that in serial as- R1 – breathes above ventilator rate
R2 – not intubated, irregular breathing pattern
sessments, the FOUR score is useful in predicting outcome in a
R3 – not intubated, Cheyne–Stokes breathing pattern
similar manner to the GCS. The FOUR score accurately pre-
R4 – not intubated, regular breathing pattern
dicts outcome following cardiac arrest, with patients scoring
 Coma Scales 835

Table 4 Coma Recovery Scale-Revised up, fist, and peace sign) is required. The brainstem reflex cat-
egory requires testing of the cough, pupillary, and corneal
Function scale Score
reflexes, thereby eliciting more information with regard to
Auditory function scale 0 – none brainstem involvement. Assessment of respiration involves
1 – auditory startle accounting for intubation and for breathing pattern.
2 – localization to sound Although, the FOUR score was designed with the intention
3 – reproducible movement to of assessing comatose patients in a more detailed manner,
command criticisms of the scale suggest that it would benefit from greater
4 – consistent movement to command standardization of administration and scoring to improve its
Visual function scale 0 – none administration and use in daily practice.
1 – visual startle
2 – fixation
Coma Recovery Scale-Revised
3 – visual pursuit
4 – object localization: reaching The Coma Recovery Scale-Revised (CRS-R) is a 23-item scale,
5 – object recognition which has been validated in multiple languages. It was de-
Motor function scale 0 – none/flaccid veloped and implemented in a rehabilitation medicine setting.
1 – abnormal posturing The validation studies were performed in groups of patients
2 – flexion withdrawal under comprehensive acute rehabilitation care following se-
3 – localization to noxious stimulation vere brain injury or various etiologies. The original CRS was
4 – object manipulation revised to improve differentiation and monitoring of patients
5 – automatic motor response in a VS, a minimally conscious state (MCS), and emerging
6 – functional object use from MCS. The scale demonstrates good interrater and
Oromotor/verbal function 0 – none test–retest reliability, and has been shown to be a reliable
scale 1 – oral reflexive movement measure of neurobehavioral response. The CRS-R has not
2 – vocalization/oral movement been tested in patients with severe brain injury during the
3 – intelligible verbalization acute hospitalization and ICU period; thus, its role in an acute
Communication scale 0 – none care hospital setting is unknown (Table 4).
1 – nonfunctional: intentional
2 – functional: accurate
Arousal scale 0 – unarousable See also: Coma. Coma, Hyperosmolar. Coma, Postoperative.
1 – eye opening with stimulation Consciousness. Endozepines and Coma. Hyperglycemia and Coma.
2 – eye opening without stimulation Traumatic Brain Injury; Overview
3 – attention

48 on days 3–5 after cardiac arrest being more likely to sur-

Further Reading
vive to hospital discharge.
Kalmar K and Giacino JT (2005) The JFK Coma Recovery Scale-Revised.
In the FOUR score, there are four categories for assessment Neuropsychological Rehabilitation 15: 454–460.
(eye response, motor response, brainstem reflexes, and res- Report of the American Congress of Rehabilitation Medicine, Brain Injury-
piration), each with a maximum of four points (0 ¼worst Interdisciplinary Special Interest Group, Disorders of Consciousness Task Force
score, 4¼best score). Assessment of eye response involves (2010) Assessment scales for disorders of consciousness: Evidence-based
recommendations for clinical practice and research. Archives of Physical
assessment of tracking or blinking to command in addition to Medicine and Rehabilitation 91: 1795–1813.
eye opening. For a maximal score in the motor responses
category, demonstration of specific hand positions (thumbs