Question: Does an initial Glasgow Coma Scale (GCS) score of 15 mean there was no brain injury? Also, please explain why there seems to be so much confusion about the GCS scoring for persons with TBI.
Answer: Brain Injury Source, Ask the Doctor
By Nathan D. Zasler, MD, FAAPM&R, FAADEP, DAAPM, CIME
Does an initial Glasgow Coma Scale (GCS) score of 15 mean there was no brain injury? Also, please explain why there seems to be so much confusion about the GCS scoring for persons with TBI.
Thank you for your question. First, I would note that the GCS score is quite entrenched in neuroscience literature and remains the standard initial assessment paradigm for persons with TBI relative to producing an objective reproducible and relatively simple means of measuring an individual's presenting level of consciousness. This scale also avoids the use of words that often are not understood clearly across multiple professions (e.g., nursing, physicians, EMS crews) such as semi-comatose, comatose, vegetative, lethargic and stuporous, among others.
Jennett and Teasdale first introduced the scale in the journal Lancet in 1974. The GCS scale has three evaluation categories: eye opening, verbal response and motor response. Eye opening scores range from one to four. Specifically, a score of one equals none, a score of two equals pain, a score of three equals voice and a score of four is spontaneous. Verbal responses are scored from one to five. A score of one equals none, a score of two is incomprehensible sound, a score of three is inappropriate words, a score of four is confusion and a score of five is orientation. Last, motor responses are scored from one to six. A score of one equals none, a score of two equals abnormal extension, a score of three equals abnormal flexion, a score of four equals withdrawal to pain, a score of five equals localization to pain and a score of six equals following commands.
I often am asked what GCS is the most predictive of outcome. The answer to this is a post-resuscitation GCS. This is because artefactually lower scores on the GCS may be generated, independent of the severity of the traumatic brain injury, because of a variety of different phenomena of alcohol and/or drug intoxication, shocks, hypothermia and/or hypoxia. Post-resuscitation GCS typically refers to the best GCS score obtained within the first six to eight hours after injury following non-surgical resuscitation.
Although arbitrary, the GCS scores can be grouped to classify mild, moderate and severe TBIs. In most literature, mild brain injury is defined as GCS scores ranging from 13-15, moderate injury, 9-12, and severe injury, 3-8. It is important to note that the GCS score may be challenging to obtain in persons who present with certain early clinical findings including oral or tracheal intubation, the latter via the oropharynx or through a tracheostomy and/or significant orbital swelling secondary to periorbital/facial trauma. This latter clinical finding prohibits the accurate assessment of eye opening responses.
Certainly, as previously noted, medications-particularly sedatives and neuromuscular blockers-will create a significant perturbation in the GCS score and, therefore, one cannot determine accurate GCS once these drugs are administered. The other thing that is important to remember is that the motor components of the scale are a particularly strong predictor of outcome at the lower end of this scale, holding less significance to the upper end of the scale.
Certainly, people can have documented brain injuries and still present in the field and/or emergency room with a GCS of 15. I certainly have seen many an individual-both during my time in the emergency room in training, as well as in the context of follow-up care-who presented with rather significant brain injury but had a "normal" GCS (e.g., 15) at the time of his/her initial assessment. Certainly, such a score does not rule in or out a history of TBI, either recent or in the distant past. I would note that in my own experience, one can have significant focal cortical contusions and still present oriented and "grossly intact," by GCS scoring to the emergency room. When one has more significant diffuse axonal injury, however, one typically does not expect to see a GCS of 15. Persons with brief periods of loss of consciousness without significant periods of associated amnesia, however, can still present after such events with GCS scores of 15 to the emergency room. One of the other problems with the GCS is that in the middle range scores, one tends to see the greatest number of scoring errors. There still are concerns on the part of some practitioners that the GCS score may be of less value than previously thought (Koziol et al., 1990).
Ultimately, one also must remember that although the GCS is one of the better single measures that we have to predict both acute and long-term outcome, it is not infallible as far as outcome prediction. That is, high scores do not predict good outcome and low scores do not predict bad outcome when one is looking at larger populations of individuals or, for that matter, even a single person. It is important to remember that GCS scores do not follow a normal distribution and studies that have analyzed outcome via GCS scores have fallen disappointingly short in this area. I also would note that there are ongoing studies examining the utility of expanded scoring systems for the GCS for persons with mild TBI (e.g., GCS scores between 13-15). There also have been modifications to the GCS score to attempt to address some of the perceived shortcomings of the GCS. In particular, one such attempt is the Glasgow Liege Scale Score, in which there is inclusion of brain stem reflexes (Borne, 1985).
In conclusion, even though the GCS has it's problems and there remains some confusion about its utility, I would say that it still remains the "gold standard" with regard to the acute neurosurgical/neurological evaluation of the person with TBI, regardless of the severity.
Refernces:
Borne MR, Albert A, Hans P et al.: Relative prognostic value of best motor response in brain stem reflexes in patients with severe head injury. Neurosurgery. 16:595-601, 1985
Jennett B, Teasdale G, Braakman R et al.: Predicting Outcome in Individual Patients After Severe Head Injury. 1:1031-1034, 1976
Koziol JA & Hackey W.: Multi-various data reduction by principal component with applications in neurological scoring instruments. J Neurol. 237:461-464, 1990
Teasdale G & Jennett B: Assessment of coma and impaired consciousness: A practical scale. Lancet 2:81-84, 1974