Cognitive, functional, and psychosocial outcome after severe traumatic brain injury: A cross-sectional study at a tertiary care trauma center
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.121920
Source of Support: None, Conflict of Interest: None
Background: Traumatic brain injury (TBI) constitutes a significant public health problem. Objectives: To assess cognitive, functional, and psychosocial outcome in patients with severe TBI (STBI). Materials and Methods: A total of 77 survivors of STBI treated at our center were prospectively assessed in the outpatient department 1 year after the injury. These patients were assessed for cognitive, functional, and psychosocial outcome using cognitive outcome tests, dysfunctional analysis questionnaire, and personality trait inventory, respectively. Cognitive and functional outcome was graded as average and above average recovery (good recovery) or below average recovery (poor recovery). Psychosocial outcome was assessed as average recovery (good recovery) or mild impairment and severe impairment (poor to very poor recovery). Statistical analysis was done using Chi-square and Fisher's exact tests. Results: The ability to learn new things was most affected aspect of cognitive function and only 1% of patients had good recovery in this domain. However, 44% had good outcome for simple memory. A total of 61% showed good recovery in several aspects of functional status. At the end of 1 year, 62.5% patients still remained unemployed. 45.4% patients had good recovery, while 37% had severe impairment for various personality traits and 40% of patients had impaired emotional stability, while 57% patients showed impaired recovery in depressive tendency. Conclusion: In this study 61% patients with STBI had good recovery in functional outcome and 45.4% in psychosocial outcome at 1 year follow-up. However, improvement in cognitive outcome was not so optimistic with the ability to learn new things being most affected.
Keywords: Cognitive outcome, functional outcome, glasgow outcome score, psychosocial outcome, traumatic brain injury
Traumatic brain injury (TBI) is one of leading causes of mortality and disability worldwide and is estimated to surpass many diseases by 2020. , The mortality in severe TBI (STBI) ranges from 35 to 45%. ,,, Survivors of STBI suffer from physical, functional, cognitive, and psychosocial disabilities. The prognosis of STBI can be assessed over time by various functional outcome scales. ,,,,,,,
Persons with STBI are likely to present with deficits in social interaction, problem solving, memory, oral expression, and comprehension. There is a general trend towards better physical and functional recovery as compared to cognitive skills. However, functional recovery largely depends upon environmental stimulation and social interaction, which induces multiple biological effect on brain.  The study aims to assess cognitive outcome, functional status, and psychosocial outcome at 1 year after STBI in Indian population and to assess the effect of various clinical parameters on cognitive, functional, and psychosocial outcome.
This cross-sectional study involved patients with STBI (Glasgow Coma Scale (GCS) score 3-8) treated at our center. Seventy-seven consecutive patients who presented at 1 year postinjury and met inclusion and exclusion criterion were enrolled. The ethical clearance was obtained from Institute Ethics Committee, All India Institute of Medical Sciences, New Delhi; prior to start of the study. Written informed consent was taken from patients or family members, to participate in the study.
The details of these patients were retrieved from computerized database. Demographic profile and various clinical details like type, mode, and severity of injury; GCS at admission and at discharge; treatment given; details of surgical treatment; length of hospital stay; and complications during hospitalization were reviewed. The patients who presented initially with history of mild to moderate TBI, associated spinal injury and or polytrauma, patients aged <20 or >46 years, history of mental retardation and unknown patients were excluded.
Functional and psychosocial outcome analysis was done by either direct interviewing patient or his/her attendant. The cognitive outcome analysis was done by interviewing patients.
Cognitive assessment was done using tests from Postgraduate Institute Memory Scale (PGIMS), which is a structured verbal test directed towards temporal sequencing (mental balance), attention and concentration (digital span forward and backward), simple memory (verbal memory for similar pairs), and ability to learn new things (verbal memory for dissimilar pairs).  Raw score was computed as per the test rules which were further matched with their respective age and educational level norms. Then the scores of each test were measured on scale from 0 to 100 using percentile ranks. The results of cognitive outcome were analyzed by categorizing cognitive outcome into: Below average recovery (20-40), average (40-60), and above average recovery (60-100). Average and above average recovery were considered as good outcome.
The functional outcome was assessed by dysfunctional analysis questionnaire (DAQ), which studies five domains namely social, vocational, family, personal, and cognitive.  Each domain was rated by patient or his significant family member on a 5 point scale, which ranged from better than before (scored as 2), same as before (scored as 4), deteriorated mildly (scores as 6), deteriorated moderately (scored as 8), and deteriorated severely (scored as 10). The scores of each subscale were then added. The maximum and minimum scores with DAQ are 500 and 100 respectively; with lower scores indicating better results. Each domain was tested and scored accordingly. DAQ scores were grouped into three categories, that is, below average recovery (60-100), average (40-60), and above average recovery (20-40). Good outcome included patients with above average and average recovery.
The psychosocial analysis was assessed by personality trait inventory (PTI) scale,  which is a short Indian version for Minnesota Multiphasic Personality Inventory.  It has 90 questions which should be answered in yes or no response. Based on these answers, personality traits like depression, cyclothymia, activity, paranoid tendency, superego, dominance, emotional instability, introversion, and social desirability were tested. For each trait, 10 questions were asked and only positively marked responses were noted.
Score of each domain were calculated as per the formula, for mean and standard deviation. These values were grouped into three groups: Average recovery (± 1 standard deviation (SD)), mildly impaired (± 2 SD), severely impaired (± 3SD).
Assuming recovery rate of STBI to be 40% (which varies between 30 and 80%), 77 patients needed to be enrolled. All analyses were performed using Statistical Package for Social Sciences (SPSS) version 17 and data was analyzed with Chi-square and Fisher's exact tests. Probability values <0.05 in these analyses were considered statistically significant.
Seventy-seven consecutive patients meeting inclusion and exclusion criterion were enrolled after screening 168 patients who presented 1-year postinjury. Mean age of patients was 30.4 years, 67 (87%) were males and 10 (13%) were females. Most common cause of STBI was road traffic accident (84.5%), followed by falls and assaults (13 and 2.5%, respectively).
At admission, GCS score below 5 was present in 19 (25%) patients, while 58 (75%) had GCS score above 5. Most common computed tomography (CT) scan finding was acute subdural hemorrhage (SDH), seen in 39 (51%) patients, followed by multiple or single contusion in 15 (20%) patients, and diffuse axonal injury in seven (9%) patients.
Intracranial pressure (ICP) monitoring was done in all patients. A total of 60 (78%) patients underwent some surgical intervention. Depending upon type of lesion, patients either underwent decompressive craniectomy (DC) alone, DC with contusectomy or epidural hematoma (EDH) evacuation. Remaining 17 (22%) patients were managed conservatively with medical therapy and ICP monitoring. Sixty-six (86%) patients were discharged within 1 month and 11 (14%) patients remained in hospital for more than 1 month.
Eight patients had penetrating STBI, and 69 (90%) had closed STBI. Thirteen (16%) patients had one or multiple episode of convulsive seizure during hospitalization.
Approximately 56-99% patients had poor outcome in various domains of cognition. Good recovery in learning ability for new things was seen in only one (1%) patient. Maximum number of patients (44%) had good recovery in simple memory [Table 1].
Most consistent factors influencing cognitive outcome were: GCS at discharge, length of hospital stay, preinjury educational level, and hydrocephalus (P = 0.01). All the four domains were significantly influenced, except the new learning ability, where association was statistically nonsignificant. GCS (less than 5) at admission was not a statistically significant factor associated with cognitive outcome.
The laterality of lesion had a significant correlation with simple memory only (P = 0.04). The patients with right-sided lesion had better recovery with 15/24 (62.5%) patients showing good recovery. In contrast, only 11/36 patients with left-sided lesions (30.5%) had good recovery in their cognitive outcome.
Forty-seven (61%) patients had good recovery in all domains of DAQ, while 30 (39%) patients showed poor recovery. At the end of 1 year, 48 (62.5%) patients still remained unemployed [Table 2].
Factors significantly associated with functional recovery were gender (P = 0.02), GCS at admission (P = 0.04), duration of hospital stay (P = 0.001), presence of hydrocephalus (P = 0.01), GCS at discharge (P = 0.001), level of education (P = 0.001), pupilary reaction at admission (P = 0.001), and side of lesion (P = 0.05).
Of the 67 male patients, 43 (62.8%); and four (40%) of the 10 female patients had good outcome in functional recovery. The patients with right-sided lesions had better functional recovery than patients with left-sided lesions. Of the 25 patients with right-sided lesion, 19 (77%) patients showed good recovery in the functional status. Whereas of the 39 patients with with left-sided lesion, only 19 (51%) patients showed good recovery. Higher GCS at admission was associated with good recovery in functional status (P < 0.05). Nineteen patients had GCS below 5 at admission, of these seven (35.2%) patients made good recovery. Fifty-eight patients had GCS above 5 at admission, of these 40 (71.8%) patients had good functional recovery. Poor recovery was seen in 90% (9/10) patients hospitalized for more than 30 days (P = 0.001).
GCS at discharge had a significant effect (P = 0.001) on functional outcome. Out of 49 patients with GCS of 13-15 at discharge, 35 (71%) patients had good functional recovery, while none with GCS below 8 made good recovery.
Of the total patients, 45% patients had good outcome, while 37% patients had poor recovery of various personality traits assessing psychosocial outcome and 40% patients had impaired emotional stability, while 57% patients showed impaired recovery in depressive tendency [Table 3].
Improvement in personality traits was significantly related to GCS at admission (P = 0.03), duration of hospital stay (P = 0.001), GCS at discharge (P = 0.001), fever (P = 0.002), meningitis (P = 0.02), presence of hydrocephalus (P = 0.01), pre-injury education status (P = 0.01), pupillary reaction (P = 0.02), and alcohol intoxication (P = 0.001).
The incidence of hydrocephalus was 6.6%. All patients with hydrocephalus had below average recovery in their cognitive and functional outcome and severely impaired personality traits.
Fever, alcohol intoxication, and meningitis had adverse effect on psychosocial outcome measures only but had no significant impact on functional and cognitive outcome measures. All personality traits were either mildly or severely impaired in patients who were under influence of alcohol at time of injury (P < 0.05).
Pre-injury education status had a significant effect on cognitive (P = 0.002), functional outcome (P = 0.001), and different traits of personality (P = 0.001). Patient with high education level tend to recover better than those who were less educated.
Thirty-four patients had bilaterally equal and reacting pupils at time of admission, 23 (67.5%) of them showed good recovery in functional outcome. However, only one (9%) had good recovery out of 11 patients with bilaterally dilated and nonreacting pupil at time of admission (P = 0.001). Similar results were observed while studying significance of pupillary reactivity on psychosocial outcome.
Type of injury (either penetrating or closed) or lesion (acute SDH, diffuse axonal injury, or contusion) did not show any correlation with either functional, cognitive, or psychosocial outcome.
Outcome measures of STBI are not well-defined. It is difficult to assess outcome of STBI because survivors of STBI suffer from multiple problems in terms of physical deformity, functional disability, memory disturbances, cognitive dysfunction, and difficulty in performing executive function. , It is virtually impossible to select a single parameter which measures all aspects of dysfunction in STBI. Bagiella et al.,  recommends nine outcome measures to study two different areas of recovery: Functional and cognitive. These measures are Extended Glasgow Outcome Scale (GOS-E); Controlled Oral Word Association Test (COWT);  Trail Making Test, Parts A and B; California Verbal Learning Test-II;  Wechsler Adult Intelligence Scale-III Digit Span subtest; Wechsler Adult Intelligence Scale-III Processing Speed Index;  and Stroop Color-Word Matching Test, Parts 1 and 2.  Chamoun et al.,  used mortality and GCS as outcome measure for analysis of outcome for head injury with GCS of 3. However, neuropsychological tests of executive function are cornerstones in evaluation of TBI. Mammi et al.,  used Functional Independence Measure, Disability Rating Scale, and community integration questionnaire as outcome measures after STBI. All these outcome measures precisely assess different domains of disability. However, all these outcome measures were created with basic understanding that patient is a citizen of developed country. These parameters were not standardized for Indian population. In present study, we have attempted to measure functional and cognitive outcome using tests which are standardized for Indian population.
The average recovery rate for cognitive outcome varied between 1 and 44% for different domains of cognition and there were only three (4%) patients with above average cognitive outcome recovery for mental balance domain. The poor recovery in cognitive outcome has been well-documented. , Brooks and Aughton  showed patients with STBI had difficulty in learning and constructional tasks. In our study, 99% showed poor recovery in verbal memory for dissimilar parts, which measures the ability to learn new things. These cognitive skills have been shown to depend on intact cholinergic pathways. However, mechanism of cholinergic deficiency is not known yet, and so cholinergic drugs are not routinely used in STBI.
The data regarding correlation of patient factors like age, GCS, etc., and cognitive outcome is controversial. Jeon et al.,  found that demographic factors like age, pupillary reactivity, and CT scan finding do not affect neurocognitive outcome in STBI. Another study showed no significant correlation between GCS and cognitive outcome in STBI.  In our study, GCS at discharge, educational status of patient, and length of hospital stay were significantly correlated with cognitive outcome.
Cognitive outcome after severe head injury has also been correlated with length of coma, which reflects severity of brain damage;  which is also concordant with the findings in our study. Various other studies have shown a persistent decline in cognitive function of severe head injured patients.  A number of explanations for cognitive decline are possible like brain vulnerability in a particular area (e.g., prefrontal cortex) or type of tissue damage (e.g., white matter). Various studies have shown that progression of atrophy in white matter after TBI is caused by amyloid deposition or neuroinﬂammatory processes affecting white matter.  A second possible explanation is that severity of TBI (regardless of location or tissue type) would predict decline. In a model of recovery and rehabilitation after brain injury, large lesions showed reduced recovery potential than medium and small-sized lesions.  Cortical areas distant from injury have been shown to undergo major neuroanatomic reorganization over time. Loss of projections to distant areas functionally related to injured areas, combined with increased disuse of functionally severed areas over time, could provide an underlying mechanism for a delayed and gradual decline after TBI recovery. 
Lastly, subject and/or environmental variables may predict decline. Low education, older age, and cognitive inactivity have been associated with a faster rate of cognitive decline in TBI. 
Similar to cognitive outcome, there have been many studies which have shown improvement in functional outcome in relation to STBI. , Sousa et al.,  showed that many patients with good functional outcome had a significant improvement in their motor function, but continued to have neurobehavioral and cognitive deficits. Ponsford et al.,  showed most patients were physically independent and competent in personal and domestic activities of daily living. In an another study, 64% of patients had good recovery after STBI as measured using GOS-E at the end of 1 year.  We have found similar results in our study with 60% of our patients having good functional recovery. Utomo et al.,  showed that age and low GCS had a significant effect on mortality and functional outcome of STBI. Similar results were also confirmed in our study. Apart from these influencing factors, we found correlation of other factors, such as hospital stay, hydrocephalus, pupillary reactivity, and side of lesion with functional outcome which is concordant with other similar studies in literature. ,,, It has been shown that functional recovery of brain largely depends upon environmental stimulation and social interaction which induces multiple biological effects on brain.  The patients with STBI have inadequate social integration, highly depressive tendencies, and frequent mood changes and symptoms like anxiety, loss of temper, difficulty in taking interest, and childish behavior. , In our study, 40% patients had impaired emotional stability, while 57% patients showed impaired recovery in depressive tendency.
We did not find any significant correlation between type of injury or lesion and functional or cognitive outcome after STBI. However, Gennarelli et al.,  showed marked heterogeneity within severe head-injury group and pointed out that patients with same GCS have markedly different outcomes, depending on causative lesion. Acute subdural hematoma with GCS of 3-5 was uniformly the worst problem (74% mortality and 8% good recovery); whereas, diffuse injury coma of 6-24 h with GCS of 6-8 had 9% mortality and 68% incidence of good recovery.
The limitations of the study were that sample size was small and the data reflected a form of hospital-based registry. The heterogeneity existing among the patients with STBI necessitates further stratification in an outcome-based analysis, based on the type of injury and magnetic resonance imaging (MRI) findings such as diffuse axonal injury and long tract disruption. Besides this, a longer duration of follow-up would be useful to study improvements in cognitive and functional outcomes. In future, larger multicentric studies with longer follow-up would have to be designed to further study impact of TBI on cognitive, functional, and psychosocial outcome.
We acknowledge Prof. R M Pandey, Department of Biostatistics, AIIMS, New Delhi, India; for statistical analysis.
[Table 1], [Table 2], [Table 3]