The Glasgow Coma Scale in the assessment of consciousness

This has come up in Question 22 from the first paper of 2012. The college specifically wanted a discussion of the limitations of the GCS.

The Glasgow Coma Scale undergoes a thorough dissection in LITFL.

Additionally, a good editorial on this topic is available, and summarises the problems pretty well.

Advantages of the Glasgow Coma Scale

• The most widely recognised of all conscious level scoring systems in the world.
• Reproduceable by well-trained staff
• Easy to perform with minimal training
• LITFL mention that it has "face validity" (i.e. it looks like it should work)
• It has prognostic value: the motor score particularly has a significant impact on the prognosis.
• The motor score findings (2,3,4) have specific pathophysiological correlations
• It is used to categorise traumatic brain injury into mild, moderate and severe.
• It is used to determine the need for an EVD in a patient with traumatic brain injury in the absence of any CT abnormalities (at a GCS of 8, you'd want an EVD)
• It can be used to indicated a depth of coma at which one's airway reflexes are likely to become unreliable
• It has been incorporated into the APACHE-II scoring system

Disadvantages of the Glasgow Coma Scale

Apart from being confused by the presence of drugs, the GCS has a few important problems.

• It is only a scale for assessing consciousness, not a stand-alone assessment tool for brain trauma - Teasdale and Jennett even said so themselves in their answer to the criticism that their scale was somehow "incomplete". 
• It is unreliable in patients in the middle range of 9-12
• People dont know how to use it. Only 15% of military physicians were able to calculate it correctly.
• Even when calculated correctly, It has high inter-observer variability: even trained emergency staff get a different score on the same patient in 38% of cases. 6-17% of scores were 2 or more points apart.
• Its inter-observer variability means we should always report the exact findings rather than the number which the patient has scored.
• It is inadequate to assess higher cortical functions, and there may be a lot of variability at the upper range of the score. The delirious person scoring 14 could have a massively impaired cognition, or a mild confusion.
• It is inadequate to assess brainstem reflexes.
  • Therefore, it cannot be used as a trigger for intubation (GCS of 8)
• The eye score is unreliable if the eyes are damaged. Alternatively, it is possible to score an E4 even if one is braindead, provided one's eyes are open. Intelligence in interpretation is called for, and perhaps because of this the GCS is not ideal as a screening tool among partially trained staff.
• The total score is meaningless:
  • The components are more important individually
  • Depending on the individual component score, the prognosis may be very different for patients with the same total score.
• It is affected by drugs and alcohol.
  • However, it is still used in assessing drug overdose patients.
• It is affected by language barriers
• Intubation makes a mockery of its verbal conponent
• It needs to be modified for use in young children.