Journal of Mental Health and Human Behaviour

: 2015  |  Volume : 20  |  Issue : 2  |  Page : 59--64

A comparative study of cognitive function following traumatic brain injury: Significance of initial Glasgow coma scale score to predict cognitive outcome

Pradipta Majumder1, SK Khandelwal2, M Sood3, A Nehra3, BS Sharma4,  
1 Department of Child and Adolescent Psychiatry, Children's Hospital of Philadelphia, 3440 Civic Center Blvd, Philadelphia, PA 19104, USA
2 Department of Psychiatry, All India Institute of Medical Sciences, New Delhi, India
3 Department of Clinical Neuro Psychology, Neuro Science Center, All India Institute of Medical Sciences, New Delhi, India
4 Department of Neuro Surgery, All India Institute of Medical Sciences, New Delhi, India

Correspondence Address:
Pradipta Majumder
Department of Child and Adolescent Psychiatry, Children«SQ»s Hospital of Philadelphia, 3440 Civic Center Boulevard, Philadelphia, PA 19104


Introduction: Traumatic brain injury (TBI) is a leading cause of death and disability all over the world. It is associated with diversities of outcomes including cognitive deficits. The worse cognitive outcome is often associated with more severe degree of TBI as measured by initial Glasgow Coma Scale (GCS) score. Materials and Methods: Present study compared the cognitive function of TBI patients having initial GCS score 9-10 with those having the initial GCS score 11-12. The assessment on Postgraduate Institute battery of brain dysfunction was conducted when the patients came for their follow-up visit at a tertiary health care facility between 6 months and 12 months of sustaining TBI. Results: There was moderate degree of cognitive dysfunction in the group with initial GCS score of 9-10 and no dysfunction in the group with initial GCS score of 11-12. Conclusion: The initial GCS score of 10 may be critical to predict cognitive deficits among TBI patients during 6-12 months of recovery period.

How to cite this article:
Majumder P, Khandelwal S K, Sood M, Nehra A, Sharma B S. A comparative study of cognitive function following traumatic brain injury: Significance of initial Glasgow coma scale score to predict cognitive outcome.J Mental Health Hum Behav 2015;20:59-64

How to cite this URL:
Majumder P, Khandelwal S K, Sood M, Nehra A, Sharma B S. A comparative study of cognitive function following traumatic brain injury: Significance of initial Glasgow coma scale score to predict cognitive outcome. J Mental Health Hum Behav [serial online] 2015 [cited 2020 Aug 5 ];20:59-64
Available from:

Full Text


Traumatic brain injury (TBI) is defined as extracranial mechanical force to the brain that leads to any period of loss of consciousness (LOC) or any anterograde or retrograde amnesia or any alteration in mental state at the time of the event. [1] A TBI occurs every 15 s in the United States and is rated as the most frequent cause of death and disability for persons <45 years of age. The incidence of hospitalization for TBI in older adults is approximately 140 per 100,000 population, [2] and as a result, TBI is an important public health issue. In India, 30,000 persons die and 125,000 persons are disabled each year due to TBI. [3] However, there may be a large number of cases who do not receive due medical care, or are treated locally in rural areas with no formal registration of case either in the health facility or police records. [4] It has been estimated that around 25% of people sustaining mild or moderate TBI do not seek for medical attention. [5] TBI is often been labeled as the "silent" or "hidden" epidemic because many individuals are not identified by the health care system and their neurological, neuropsychological and neurobehavioral symptoms, and functional difficulties are attributed to etiologies other than brain injury. [4] The population at highest risk of TBI are 15-24-year-old with male to female ratio approximately 3:2. [6] Until recently, most TBIs were the result of motor vehicle accidents, but the Center of Disease Control's most recent report reveals that falls are currently the leading cause of TBI, with rates highest for children under 4 and adults over 75 years of age. [7]

There are a number of risk factors for TBI, which may include substance abuse, psychiatric condition with impulsive behavior like bipolar disorder, cluster B personality disorder, and attention deficit hyperkinetic disorder. Elderly group, arteriosclerosis, and alcoholism are additional risk factors for neuropsychiatric disturbances following TBI. These all delay the reparative process within the central nervous system. [8]

Though the TBI may lead to diverse consequences in the form of physical impairments including headaches, sleep problems, fatigue, blurred vision, dizziness, loss of hearing, and seizure disorders, [9],[10] the major sequelae of TBI lies in cognitive impairment and behavioral problems. The diffuse axonal injury has profound effect on the cognitive functioning as it involves a wide processing than a focal one. There may be defects in executive cognitive function with minimal or no neurological or focal cortical deficits. There is decreased speed in information processing, decreased attention, increased distractibility, deficits in rapid decision-making and problem-solving, and deficits in the ability of sustaining efforts. There are also deficits of memory and difficulty in learning information. Language disabilities may also accompany.

TBI severity is defined by the duration of LOC, altered mental status (e.g., confusion) or posttraumatic amnesia (PTA). [11] The severity of TBI is often graded according to the initial Glasgow Coma Scale (GCS) score as mild with 13-15, moderate with 9-12, and severe with scores below 9.

It has been shown in a recent study that a higher incidence of age ≥35 years, low initial GCS score, at least unilateral pupil dilatation, and neurological deficit predict a less favorable outcome following TBI. The other predictive factors have been identified as age, intracerebral lesion, cause of injury, injury severity score, hypoxia, hypotension, the Traumatic Coma Data Bank computerized tomography classification, and traumatic subarachnoid hemorrhage. [12] Studies conducted so far give heterogeneous results on the predictive validity of GCS score on the outcome. A study conducted by Lieh-Lai et al., showed that a low GCS score may not always accurately predict the outcome of severe TBI. In the absence of hypoxic-ischemic injury, children with TBI, and GCS scores of 3-5 can recover independent function. [13] In another study by Davis et al., it was found that GCS score is highly predictive of outcome of TBI. Further study analysis also documented that a change of GCS score from field (at the time of injury) and GCS score at the arrival at the hospital is also highly important in predicting the outcome. [14]

Cognitive outcome depends on a number of factors, such as degree of diffuse axonal injury, duration of LOC and PTA, clinical evidence of brain stem dysfunction at the time of injury, and presence and size of focal hemispheric injury. [8]

The review of literature indicated the possible influence of initial GCS score at the cognitive outcome of the patients sustaining TBI. [13],[14] However, the present study was planned to find whether there is any initial GCS score that may be critical in predicting the cognitive outcome following TBI.

 Materials and Methods

Rationale of study

The review of literature states that TBI adversely affect the cognitive function that is adversely affected by the poorer GCS score at admission. However, studies had paid little attention in predicting any GCS score that may be critical in predicting the cognitive adversities in between 6 months and 12 months following the traumatic event. The present study was carried out with patients sustaining moderate degree of TBI as defined by 9-12 as the initial GCS score at admission. However, we attempted to find out if there is any score or range of scores that may predict worsened cognitive outcome following TBI.

Aim and objective

The present study aimed at comparing the cognitive function of patients with initial GCS score 9-10 and those with initial GCS score 11-12 after 6-12 months of sustaining the TBI being treated surgically.

Universe of the study

Thirty subjects, treated surgically for moderate TBI reporting to the Neurosurgery Clinic at All India Institute of Medical Sciences (AIIMS), New Delhi for the follow-up at 6-12 months postsurgery. AIIMS is a public funded Tertiary Care Teaching Hospital.

Thirty surgically treated patients of TBI of moderate severity meeting the inclusion and exclusion criteria listed below were taken for the study. The present study was a cross-sectional comparison of the cognitive function of subjects coming for follow-up after 6 months to 1-year following surgical treatment of their moderate degree of TBI. The ethical clearance was taken from the AIIMS Ethical Committee.

Inclusion and exclusion criteria

Patients (aged 18-50 years) who had sustained moderate TBI (defined as GCS score of 9-12 at the time of admission) and treated surgically were informed about the study during their follow-up visit at 6-12 months, and those providing valid informed consent were included provided his/her physical condition allowed him/her to be cooperative enough to undergo assessment.


The study subjects were assessed on the basis of their demographic variables and clinical parameters. The data on demographic variables were collected in a semi-structured performa. Subsequently the cognitive functions were assessed by means of Hindi Mental Status Examination (HMSE) (adapted from Mini-Mental Status Examination) [15],[16] and Postgraduate Institute battery of brain dysfunction (PGIBBD). [17] The HMSE was a product of Indo-US Cross-National Dementia Epidemiological Study and mainly devised for the illiterate Hindi speaking population. [16] The instrument has been validated to screen cognitive function in diverse range of setting and disease conditions. PGIBBD was constructed by Parshad and Verma in the year 1990 at PGIMER, Chandigarh. The PGIBBD is administered in Hindi and has been developed and validated for use in the Hindi speaking population. It helps in finding degree of deficit in memory, intelligence, and/or percepto-motor equity separately. [17] This instrument is also widely used in India for detailed evaluation of cognitive function.


This study was carried out in the Neurosurgery Clinic, Neuroscience Center, AIIMS, New Delhi. A purposive sample of 30 subjects attending the Neurosurgery Clinic was chosen who met the inclusion and exclusion criteria.

Four hundred and two patients were screened, and 366 were excluded for reasons like not fulfilling the inclusion and exclusion criteria, patient's refusal, and inability to finish the entire assessment. The remaining 30 subjects formed the sample of this study. The discharge summary of the patients provided information regarding the GCS score at admission, the surgical intervention carried out, postsurgery complications and other relevant data. Every patient was assessed for a 1-time assessment on a convenient day of his or her choice. Important demographic and clinical information was gathered from the case note as well as from the patient and the accompanying person(s) and was recorded on a semi-structured proforma designed for this study. The surgical procedure, any complication(s), history of seizure in pre- or post-surgical period, and any significant co-morbidity were noted. These 30 patients were divided into two groups, Group A with initial GCS score 9-10 and Group B with GCS score 11-12.


The two groups were named as Group A with an initial GCS score of 9-10 and Group B with an initial GCS score of 11-12. The number of people in Group A was 11 and that of the Group B was 19. SPSS version 12 ( was used for the statistical analysis. T-test and Chi-square test were used for the quantitative and qualitative data, respectively.

Sociodemographic variables

The mean age of Group A was 38.45 (Standard deviation [SD] 9.761) and that of Group B was 31.74 (SD 9.204) showing no statistically significant difference (P = 0.52). In Group A, all of the subjects were male whereas in Group B, the number of males was 16 and the number of females was 3 having no statistically significant difference (P = 0.176).

Clinical variables

Five patients (3 in Group A and 2 in Group B) were found to be under the influence of alcohol at the time of their trauma. No statistically significant difference was noted among the two groups (P = 0.326).

Ten patients (4 in Group A and 6 in Group B) had developed seizures following TBI.

There was no statistically significant difference in this variable also.

Comparison of the cognitive function

The mean HMSE score at the time of assessment of Group A (GCS score 9-10) and Group B (GCS score 11-12) were 27.28 and 29.00 during the time of assessment. Though the mean values of these two groups independently did not show any significance in terms of cognitive decline, there was statistically significant difference in between the two groups (P = 0.035 and t = -2.213).

The two groups were then compared on the basis of the total PGIBBD dysfunction, and the findings are shown in [Table 1] and [Table 2].{Table 1}{Table 2}

The data derived from PGIBBD showed that there was statistically significant difference (P < 0.001) in the dysfunction rating of the two groups, patient in Group A (mean = 22) being more dysfunctional than Group B (mean = 11.37). More specifically, on comparing the data on different sub tests of PGIBBD, Group A had significantly more dysfunction in remote memory (P = 0.007), recent memory (0.032), delayed recall (P = 0.006), retention of similar pairs (P = 0.015), retention of dissimilar pairs (P = 0.047), visual retention (P = 0.007), P/K × 100 (P = 0.005) (i.e., the ratio of performance on pass-along/P and Koh's Block/K), performance quotient (PQ) (P = 0.004), difference in between PQ and verbal quotient (VQ) (0.001), Nahor-Benson test (P = 0.033), and Bender-Gestalt test (P = 0.044).

However, the mean dysfunction of subtest 2 (recent memory) in Group A was 0.947 and that in Group B was 0.181. Both of these scores individually did not show any moderate dysfunction; the mean scores of the two groups had statistically significant difference so far.

The same was true for retention of similar pairs, the ratio of performance on pass-along/P and Kohs Block/K, PQ, Nahor-Benson tests.


The present study was a cross-sectional comparison of patients coming for their follow-up at the Neurosurgery Outpatient Departments of one of the tertiary teaching hospitals in North India following surgical intervention in case of moderate degree of TBI, and presenting in between 6 months and 12 months following the traumatic event. The instruments used in this study included HMSE and PGIBBD, which are well-validated and widely used instruments used in Indian patients to assess cognitive function. [15],[16],[17],[18],[19],[20]

The study group was divided on the basis of initial GCS score at admission, as Group A (with GCS score 9-10) and Group B (GCS score of 11-12). No significant statistical difference was found in their demographic variables. The cognitive deficit of the present study population was done in two steps. First of all, the cognitive function was assessed by means of HMSE score. Initial screening by HMSE did not find any significant dysfunction in the mean values in each GCS group. This might be due to the fact that, primarily HMSE was designed for the illiterate Hindi speaking people of India whereas all of our study populations were educated. [16] However, the mean score of each of these groups had statistically significant difference with each other.

Subsequently the cognitive functions as found in the PGIBBD of the groups were compared. On comparing the data on different subtests, the two groups showed significant difference in dysfunction rating of remote memory, recent memory, delayed recall, retention of similar pairs, retention of dissimilar pairs, visual retention, the ratio of performance on pass-along/P and Kohs Block/K, PQ, difference in between PQ and VQ, Nahor-Benson test, and Bender-Gestalt test. The mean dysfunction rating in recent memory, retention of similar pairs, the ratio of performance on pass-along/P and Kohs Block/K, PQ, Nahor-Benson test in each of the groups did not have any significant deficits, but the difference in the mean dysfunction of the two groups was statistically significant. In other cases, the mean dysfunction of Group A that is group with more dysfunction had moderate dysfunction according to the PGIBBD.

Additionally, in certain other parameters like remote memory, recent memory, delayed recall, retention of similar pairs, retention of dissimilar pairs, visual retention, the ratio of performance on pass-along/P and Kohs Block/K, PQ, difference in between PQ and VQ, Nahor-Benson test, and Bender-Gestalt test, the two GCS groups showed significant difference. But, the mean dysfunction rating of both of the groups in recent memory, retention of similar pairs, the ratio of performance on pass-along/P and Kohs Block/K, PQ, Nahor-Benson test independently were less than moderate (i.e., the dysfunction rate of these subtests were <2), but the difference of the mean scores of the two GCS groups were statistically significant.

Moderate dysfunction was found out in at admission GCS score 9-10 groups in the areas of remote memory, delayed memory, retention of dissimilar pairs, visual retention, and difference in PQ and VQ, and in Bender-Gestalt test. These findings indicate that the patients with poor "at admission GCS" may have moderate deficits in specific domains of cognitive functions even after 6-12 months following the traumatic event.

The mean overall dysfunction rating of the Group A was found to be 22 which denote a moderate dysfunction. The mean overall dysfunction of Group B was 11.37 which failed to achieve any significance. This implies that those individuals having their initial GCS score <10 had deficits in their cognitive function and those above 10 did not.

The finding of statistically significant difference in between Group A (GCS score 9-10) and Group B (GCS score 11-12) also indicates that the degree of dysfunction also depends upon the severity of head injury (defined by at admission GCS score). The mean cognitive dysfunction rating of each group also indicated that those having initial GCS score <10 had at least moderate degree of dysfunction rating as measured by the PGIBD following 6-12 months of sustaining TBI. Those with GCS score more than 10 did not have any cognitive dysfunction in our study.

Among pediatric patients, a critical point of the GCS set at 5 was most strongly correlated with outcome of pediatric TBI. [21] Jiang [22] et al., in 2002, also documented that the outcome varies with the initial GCS score. They have shown that, in the good recovery group, the percentage of patients with GCS score in range of 8-3 were 39.88% with GCS of 8 points, 37.61% with GCS of 7 points, 32.16% with GCS of 6 points, 21.69% of 5 points, 16.18% of 4 points, and 6.98% of 3 points and they concluded that lower GCS scores correlated with lower chances of good recovery. [22] In another study conducted by Corral (2007) [23] had shown that, the improvement in Glasgow Outcome Scale scores between 6 months and 1-year was greater in the high GCS score at admission group than in the low score group. MRC CRASH trial collaboration also documented that GCS has a clear linear relation with mortality. [24]

Again, in the present study, relatively higher education rate of the study population might have influenced favorably the cognitive function of our population. Studies also support the positive role of education in predicting better cognitive outcome following TBI as already has been discussed. [25],[26]

Our finding supports the impression that cognitive dysfunction do occur following TBI. The association of poorer GCS scores at the time of admission and the worse cognitive function also supports the previous researches.

Strength of the study

The present study dealt with a very relevant topic involving TBI patients who are known to suffer from cognitive decline that may persist for variable time following TBI.

The present study had many strengths. The study clearly defined moderate degree of TBI and collected retrospective data from objective source like discharge summary given to the patient after initial hospitalization, thus minimizing recall bias. Further, sampling bias was also minimized because at the time of admission, assessment of the patients was done by the treating surgeons, not by the investigators.

The inclusion and exclusion criteria were also stringent and it included patients treated surgically following TBI and continued to come for follow up though there are a number of studies involving patients sustaining TBI but, most of them either have included mild degree of TBI or have included moderate and severe TBI. The present study assessed exclusively the moderate degree of TBI patients.

The present study used instruments that are valid and reliable and had been widely used among the Indian patients. Moreover, the study instruments have a valid Hindi version thereby minimizing the interviewer's bias during trans-coding the verbatim from native language to English.

The 30 individuals were selected after screening 402 patients which suggest the effort that was made to include as homogeneous sample as possible.

Above all, the in-depth assessment that was attempted in the present study to obtain a comprehensive picture of the cognitive function is another strength of the present study.

Limitations of the study

The present study has some limitations. A sample of convenience was chosen and, therefore, proper randomization was not achieved. Those with more severe illness or those with more co-morbid illnesses might have already died before the assessment was carried out leading to a late-look bias. However, the present study aimed at assessing only moderate degree of TBI and assessing more severe form of TBI was out of the scope of the present study. People coming to a government run hospital might also restrict sample to one segment of the society.

The study was a cross-sectional assessment. Like most of the cross-sectional study design, the present study also has some other limitations as for example, rapport establishment may be difficult in this short time span. Study population as well as their relatives may try to attribute any form of cognitive deficit to the TBI.

Though the present study attempted to exclude patients with any significant medical illnesses that may confound the results, no investigations were carried out to rule them out. So, the chance of recall bias and presence of these illnesses (e.g., hypothyroidism) in subclinical form could not be completely excluded.


The cognitive deficits found in surgically treated patients with moderate degree of TBI are subtle; however, the deficit is moderate with low GCS score and it persists for 6-12 months following TBI. The GCS score of 10 at admission may be considered as crucial to predict cognitive outcome at 6-12 months following the TBI.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Kim E, Lauterbach EC, Reeve A, Arciniegas DB, Coburn KL, Mendez MF, et al. Neuropsychiatric complications of traumatic brain injury: A critical review of the literature (a report by the ANPA Committee on Research). J Neuropsychiatry Clin Neurosci 2007;19:106-27.
2Frankel JE, Bean JF, Frontera WR. A follow-up study of older adults with traumatic brain injury: Taking into account decreasing length of stay. Arch Phys Med Rehabil 2006;87:57-62.
3Sudarsanan S, Chaudhary S, Pawar AA, Srivastava K. Psychiatric effects of traumatic brain injury. Med J Armed Forces India 2006;62:259-63.
4Ashman TA, Gordon WA, Cantor JB, Hibbard MR. Neurobehavioral consequences of traumatic brain injury. Mt Sinai J Med 2006;73:999-1005.
5Sosin DM, Sniezek JE, Thurman DJ. Incidence of mild and moderate brain injury in the United States, 1991. Brain Inj 1996;10:47-54.
6Kim E. Neuropsychiatric aspects of traumatic brain injury. Special report. Neuropsychiatry 2006;23:1-3. Available from: [Last accessed on 2008 06 30].
7Langlois JA, Rutland-Brown W, Thomas KE. The incidence of traumatic brain injury among children in the United States: Differences by race. J Head Trauma Rehabil 2005;20:229-38.
8Rao V, Lyketsos C. Neuropsychiatric sequelae of traumatic brain injury. Psychosomatics 2000;41:95-103.
9Dikmen S, McLean A, Temkin N. Neuropsychological and psychosocial consequences of minor head injury. J Neurol Neurosurg Psychiatry 1986;49:1227-32.
10Kraus JF, McArthur DL. Epidemiologic aspects of brain injury. Neurol Clin 1996;14:435-50.
11Jennett B. Scale and scope of the problem. In: Rosenthal M, Bond M, Griffith E, Miller JF, editors. Rehabilitation of the Adult and Child with Traumatic Brain Injury. Philadelphia, PA: F. Davis; 1990. p. 3-7.
12Park JE, Kim SH, Yoon SH, Cho KG, Kim SH. Risk factors predicting unfavorable neurological outcome during the early period after traumatic brain injury. J Korean Neurosurg Soc 2009;45:90-5.
13Lieh-Lai MW, Theodorou AA, Sarnaik AP, Meert KL, Moylan PM, Canady AI. Limitations of the Glasgow Coma Scale in predicting outcome in children with traumatic brain injury. J Pediatr 1992;120 (2 Pt 1):195-9.
14Davis DP, Serrano JA, Vilke GM, Sise MJ, Kennedy F, Eastman AB, et al. The predictive value of field versus arrival Glasgow Coma Scale score and TRISS calculations in moderate-to-severe traumatic brain injury. J Trauma 2006;60:985-90.
15Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-98.
16Ganguli M, Ratcliff G, Chandra V, Sharma S, Gilby J, Pandav R, et al. A Hindi version of the MMSE: The development of a cognitive screening instrument for a largely illiterate rural elderly population in India. Int J Geriatr Psychiatry 1995;10:367-77.
17Prasad D, Verma SK. PGI Battery of Brain Dysfunctions. Agra: National Psychological Corporation; 1990.
18Sheehan DV, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): The development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59 Suppl 20:22-33.
19de Azevedo Marques JM, Zuardi AW. Validity and applicability of the Mini International Neuropsychiatric Interview administered by family medicine residents in primary health care in Brazil. Gen Hosp Psychiatry 2008;30:303-10.
20Chiu WT, Huang SJ, Hwang HF, Tsauo JY, Chen CF, Tsai SH, et al. Use of the WHOQOL-BREF for evaluating persons with traumatic brain injury. J Neurotrauma 2006;23:1609-20.
21Chung CY, Chen CL, Cheng PT, See LC, Tang SF, Wong AM. Critical score of Glasgow Coma Scale for pediatric traumatic brain injury. Pediatr Neurol 2006;34:379-87.
22Jiang JY, Gao GY, Li WP, Yu MK, Zhu C. Early indicators of prognosis in 846 cases of severe traumatic brain injury. J Neurotrauma 2002;19:869-74.
23Corral L, Ventura JL, Herrero JI, Monfort JL, Juncadella M, Gabarrós A, et al. Improvement in GOS and GOSE scores 6 and 12 months after severe traumatic brain injury. Brain Inj 2007;21:1225-31.
24MRC CRASH Trial Collaborators, Perel P, Arango M, Clayton T, Edwards P, Komolafe E, et al. Predicting outcome after traumatic brain injury: Practical prognostic models based on large cohort of international patients. BMJ 2008;336:425-9.
25Kesler SR, Adams HF, Blasey CM, Bigler ED. Premorbid intellectual functioning, education, and brain size in traumatic brain injury: An investigation of the cognitive reserve hypothesis. Appl Neuropsychol 2003;10:153-62.
26Jeon IC, Kim OL, Kim MS, Kim SH, Chang CH, Bai DS. The effect of premorbid demographic factors on the recovery of neurocognitive function in traumatic brain injury patients. J Korean Neurosurg Soc 2008;44:295-302.