Neuropsychologic and Functional Outcome After Complicated Mild Traumatic Brain Injury

Article Outline

• Abstract
• Methods
  • Participants
  • Procedures
  • Neuropsychologic Measures
    • Rey Auditory Verbal Learning Test
    • Logical memory test
    • Block design test
    • Trail-Making Test
    • Digit span backward test
    • Controlled Oral Word Association Test
    • Symbol Digit Modalities Test, oral version
    • Wisconsin Card Sorting Test
  • Functional Status and Outcome Measures
    • FIM instrument
    • Disability Rating Scale
    • Community Integration Questionnaire
  • Statistical Analyses
• Results
• Discussion
  • Study Limitations
• Conclusions
• References
• Copyright

Abstract 

Kashluba S, Hanks RA, Casey JE, Millis SR. Neuropsychologic and functional outcome after complicated mild traumatic brain injury.

Objective

To investigate the extent to which neuropsychologic and functional outcome after complicated mild traumatic brain injury (TBI) parallels that of moderate TBI recovery.

Design

A longitudinal study comparing neuropsychologic and functional status of persons with complicated mild TBI and moderate TBI at discharge from inpatient rehabilitation and at 1 year postinjury.

Setting

Rehabilitation hospital with a Traumatic Brain Injury Model System.

Participants

Persons with complicated mild TBI (n=102), each with an intracranial brain lesion documented through neuroimaging and a highest Glasgow Coma Scale (GCS) score in the emergency department between 13 and 15, and 127 persons with moderate TBI.

Interventions

Not applicable.

Main Outcome Measures

FIM instrument, Disability Rating Scale, Community Integration Questionnaire, Wechsler Memory Scale logical memory I and II, Rey Auditory Verbal Learning Test, Trail-Making Test, Controlled Oral Word Association Test, Symbol Digit Modalities Test, Wisconsin Card Sorting Test, and block design.

Results

Few differences in neuropsychologic performance existed between the TBI groups. Less severely impaired information processing speed and verbal learning were seen in the complicated mild TBI group at rehabilitation discharge and 1 year postinjury. Despite overall improvement across cognitive domains within the complicated mild TBI group, some degree of impairment remained at 1 year postinjury on those measures that had identified participants as impaired soon after injury. No differences on functional ability measures were found between the TBI groups at either time period postinjury, with both groups exhibiting incomplete recovery of functional status at the 1-year follow-up.

Conclusions

When classifying severity of TBI based on GCS scores, consideration of a moderate injury designation should be given to persons with an intracranial bleed and a GCS score between 13 and 15.

Key Words: Brain injuries, Neuropsychological tests, Rehabilitation

TRAUMATIC BRAIN INJURY (TBI) represents a substantial public health problem and is a leading cause of death and disability in North America. TBI places a substantial burden on society because of its primary (eg, medical costs) and secondary (eg, lost productivity) effects.¹ Similarly, the physical, cognitive, and psychosocial functioning impairments in persons after TBI are far reaching and can result in a significant change in the afflicted person's life-course, an extreme loss of income or earning potential, and a profound disruption to the family unit.², ³, ⁴ The dynamics of recovery from TBI are multifold and largely depend on the severity of the ensuing TBI, as illustrated by the well-established dose-response relationship between TBI severity and cognitive outcome.⁵ The ability to reliably predict impairment and outcome after TBI is of critical importance for capitalizing on early interventions and ensuring the most appropriate long-term treatment planning.

The most commonly used classification of brain injury severity is based on the Glasgow Coma Scale (GCS),⁶ a metric originally designed for the assessment of depth of coma. Mild TBI is classified based on a GCS score in the range of 13 to 15, as well as a loss of consciousness (LOC) of less than 30 minutes and posttraumatic amnesia (PTA) less than 24 hours. In contrast, moderate TBI is classified by GCS scores ranging from 9 to 12 with PTA of approximately 1 week.

Outcome after mild TBI is typically associated with neuropsychologic performance indistinguishable from healthy controls and symptom resolution by 3 months postinjury.⁵, ⁷, ⁸, ⁹ However, a modest body of research indicates that mild TBI is not a homogeneous classification, such that mild TBI complicated by an intracranial lesion may result in poorer outcome. Specifically, findings indicate that neuropsychologic outcomes after complicated mild TBI, as defined by GCS score and the presence of an intracranial lesion, may be more similar to outcomes in patients who have sustained a TBI in the moderate range of severity.¹⁰, ¹¹, ¹², ¹³, ¹⁴ The terms high-risk mild TBI¹⁰ and complicated mild TBI¹¹ are used to refer to this subgroup of patients. Based on these empirical findings, reconsideration of the conventional classification of brain injury severity has been proposed to include a more precise classification of injuries within the GCS score range of 13 to 15. Thus, a new TBI classification would include mild, complicated mild, moderate, and severe injury groups.

The natural course of recovery after complicated mild TBI remains unclear. To date, research on complicated mild TBI has focused on cognitive outcome only in the short-term postinjury (ie, 1–3mo postinjury), whereas assessment of functional outcome after complicated mild TBI has solely relied on global outcome measures (eg, Glasgow Outcome Scale [GOS]¹⁵) that fail to consider the multifaceted context of outcome after brain injury. As such, little is known about cognitive recovery after complicated mild TBI in the longer term, and the issue of functional outcome after complicated mild TBI remains largely uninvestigated. Rather than rehabilitation of functional impairments, treatment for uncomplicated mild TBI is typically geared toward issues of pain, dizziness, fatigue, irritability, and sensory disorders, with psychoeducation and support emphasizing the expectation of symptom resolution by a few weeks to 3 months postinjury.¹⁶, ¹⁷ Symptom persistence beyond 3 months after mild TBI is considered primarily psychogenic in nature.¹⁸, ¹⁹, ²⁰ Although effective for patients with uncomplicated mild TBI, this treatment approach may be inappropriate for use with patients after complicated mild TBI given that the severity and duration of their cognitive impairments have been shown to more closely approximate those of patients with moderate TBI within 3 months postinjury. Extending these findings, it is reasonable that more severe functional impairments may exist after complicated mild TBI, such that rehabilitation of functional deficits may in fact be warranted with this subgroup of patients. For instance, investigations using the GOS at 6 months¹⁰, ¹¹ and 1 year²¹ postinjury have reported increased disability among complicated mild TBI patients compared with their uncomplicated mild TBI counterparts. Assessment of functional status is a critical measure of outcome given the movement toward providing a continuum of health care from injury through to community integration, and the functional assessment of TBI patients within the scope of the rehabilitation setting is a primary means of justifying treatment effectiveness within the managed health care environment.²²

The purpose of the present study was to investigate outcome after complicated mild TBI by comparing neuropsychologic and functional status soon after injury (ie, at discharge from inpatient rehabilitation) and at 1 year postinjury in patients with complicated mild TBI and patients of moderate TBI status. Based on empirical evidence to date, it was expected that the complicated mild TBI and moderate TBI groups would exhibit similar levels of functional disability and cognitive performance at the time of discharge from inpatient rehabilitation. Investigation of functional and neuropsychologic outcome after complicated mild TBI at 1 year postinjury was primarily exploratory in nature given the paucity of research in this TBI population. Taking into consideration the established dose-response relationship between severity of injury and cognitive outcome based on the conventional classification of TBI severity (ie, mild, moderate, severe), it was hypothesized that the complicated mild TBI group would exhibit better overall outcomes with respect to cognitive and functional status compared with the moderate TBI group at 1 year postinjury.

Back to Article Outline

Methods 

Participants 

Data were provided from 229 participants who were enrolled in the Southeastern Michigan Traumatic Brain Injury System (SEMTBIS), a participating center in the Traumatic Brain Injury Model Systems (TBIMS) project. Inclusion criteria for the TBIMS project require that participants (1) be at least 16 years of age at the time of injury, (2) arrived at a TBIMS level I trauma center within 24 hours of injury, (3) received both acute care and inpatient rehabilitation in hospitals designated as TBIMS centers, and (4) provided informed consent to participate in the TBIMS project. In addition, participants must have sustained a medically documented TBI defined as injury to brain tissue caused by an external mechanical force as evidenced by LOC due to brain trauma, posttraumatic confusion or skull fracture or by objective neurologic findings that are reasonably attributable to TBI on a physical or mental status examination. Excluded were participants with primary anoxic encephalopathy or lacerations and/or bruises of the scalp or forehead without other criteria listed above. Neuropsychologic and functional status data used in the current study were collected from 1989 through 2002, with the dates of injury for participants ranging from 1989 to 2001. Approval from each of the affiliated research ethics boards was obtained for the current study.

The complicated mild TBI group was comprised of 102 participants, each with an intracranial brain lesion documented through neuroimaging (ie, computed tomography scan) and a highest GCS score in the emergency department (ED) between 13 and 15 (mean ± standard deviation [SD], 14.4±.71). One hundred twenty-seven participants with GCS scores ranging from 9 to 12 (mean, 10.6±.98) were in the moderate TBI group. To increase the generalizability of the study sample, participants with a prior history of brain injury, substance use, or premorbid psychiatric or neurologic disorder were not excluded. There were no between-group differences with respect to the number of participants with a premorbid history of brain injury (10%), substance use (28%), or psychiatric or neurologic disorder (7%) (P>.05). The most common causes of TBI in both groups included blunt assault (37%), motor vehicle collisions (21%), falls (17%), being struck as a pedestrian (14%), and gunshot wounds (8%). Most participants within each group were men, African Americans, and single persons (table 1).

Table 1. Demographic and Injury Severity Statistics for the Complicated Mild TBI and Moderate TBI Groups

Variable	Complicated Mild TBI	Moderate TBI
Age (y)⁎	48.58±16.42	36.24±13.80
Sex (% male)	81	78
Education (%)
< high school or GED	40.6	52.8
≥ high school	59.4	47.2
Ethnicity (%)
Black	74	70
White	23	28
Other	3	2
Marital status (%)
Single	49	58
Married	15	13
Divorced	23	13
Separated	6	7
Widowed	7	8
Highest GCS score⁎	14.44±0.71	10.55±0.98
Duration of LOC⁎ (d)	3.68±5.50	6.79±7.10
Duration of PTA⁎ (d)	25.48±22.66	39.42±23.87

NOTE. Values are mean ± standard deviation (SD) or as otherwise indicated.

Abbreviation: GED, General Educational Development diploma.

Procedures 

Participants were first administered the neuropsychologic test battery and functional disability measures during their inpatient rehabilitation once they had emerged from PTA. Emergence from PTA was defined by obtaining a score of 76 on the Galveston Orientation and Amnesia Test¹⁵ over the course of 72 hours. Each participants returned for a follow-up neuropsychologic assessment within 2 months of the 1-year anniversary of discharge from inpatient rehabilitation (mean ± SD, 333.9±40.2d). Measures of functional disability and outcome were also completed at the 1-year follow-up.

The neuropsychologic tests and functional assessment measures were administered and scored according to standardized test instructions by experienced neuropsychologists and supervised technicians. Multidisciplinary hospital TBI team members ensured completion of functional measures at the time of inpatient discharge, and SEMTBIS research staff assisted with the completion of functional measures at the 1-year follow-up through an interview with each participant and/or significant caregiver. Follow-up interviews were conducted in person or by telephone.

Neuropsychologic Measures 

The neuropsychologic assessment battery included standard measures used to evaluate a broad range of cognitive abilities that are typically assessed after TBI including memory for verbal information, visuospatial ability, attention, information processing speed, working memory, novel problem solving, and cognitive flexibility.

The Rey Auditory Verbal Learning Test (RAVLT)²³ is a verbal learning and recall task assessing a participant's ability to learn a list of 15 unrelated words over 5 learning trials and recall these words after administration of a second interference word list. The total number of words recalled over the 5 learning trials was the score used in the current study.

The Wechsler Memory Scale logical memory test²⁴ is a story recall task requiring participants to recall 2 orally presented stories immediately after their presentation (logical memory I) and after a 30-minute delay (logical memory II). The number of story elements correctly recalled immediately and after the 30-minute delay after the presentation of the stories was used for analysis in the current study.

The block design test²⁵ is a measure of visuoconstruction ability whereby participants construct block designs matching those shown on a stimulus card within a specified time period. Raw scores based on the number of successfully constructed designs were used in the current study's analyses.

The Trail-Making Test (TMT)²⁶ is a visuomotor tracking and information processing speed task. Part A of this test requires participants to draw lines connecting circled numbers in consecutive order. Part B requires participants to draw lines connecting letters and numbers in an alternating ascending sequence. Each part of the test was scored as the total time to complete the task (in seconds).

The digit span backward test²⁵ is a working memory task requiring participants to repeat orally presented numbers in reverse order. The total number of correct trials was used in the current analyses.

The Controlled Oral Word Association Test (COWAT)²⁷ is a word initiation task whereby participants are asked to generate aloud as many words as possible beginning with a specified letter, without using proper nouns or similar root words with different endings, over three 1-minute trials. The total number of words produced across the 3 trials was used in the current analyses.

The oral version of the Symbol Digit Modalities Test (SDMT)²⁸ is an information processing speed task consisting of a visual key of symbols that matches the numbers 1 through 9. A series of symbols and empty boxes are located below the key. Participants are required to verbalize the number that matches each symbol in the key at the top of the page. The test was scored as the total number of correctly stated numbers within a 90-second time period.

The Wisconsin Card Sorting Test (WCST)²⁹ is a measure assessing problem solving and flexibility of thinking by requiring participants to match cards to 1 of 4 stimulus cards. Participants received feedback regarding the accuracy of each choice as an aid to determining the next appropriate match. Matches are made on the basis of color, form, or number. The concept for matching the cards changes without explicit feedback after a participant correctly matches 10 consecutive cards. A perseverative error represents an incorrect match that is repeated despite feedback that it does not match the correct matching principle. The number of perseverative errors was used in the current study's analyses.

Functional Status and Outcome Measures 

The FIM instrument³⁰ is an 18-item, observer-rated instrument used to measure functional ability. Ratings were based on consensus from a multidisciplinary TBI team consisting of physiatrists, psychologists, nurses, and physical and occupation therapists. Individual item scores on the FIM range from 1 to 7, with an item score of 7 categorized as “complete independence” and an item score of 1 as “total assist.” Previous research on the FIM has defined 2 statistically and clinically unique domains: motor functioning and cognitive functioning.³¹ Separate cognitive and motor domain FIM scores can be calculated, with a total cognitive score ranging from 5 to 35 and a total motor score ranging from 13 to 91. Higher scores reflect less disability.

The Disability Rating Scale (DRS)³² is a TBI-specific outcome measure that was developed as a means of assessing recovery from brain injury by monitoring change in disability from time of coma to reintegration into the community. The DRS uses a 30-point continuous scale ranging from death (score of 30) to no disability (score of 0) and is based on 8 items in 4 categories: (1) arousal and awareness, (2) cognitive ability and ability to handle self-care, (3) physical dependence on others, and (4) psychosocial adaptation to home, school, or work activities. Compared with the GOS, the DRS has been shown to be more sensitive to clinical changes after TBI.³³ The DRS was administered in interview form to each participant and/or a significant other in the current study.

The Community Integration Questionnaire (CIQ)³⁴ was developed for use with people after brain injury and is the most commonly used comprehensive measure of community integration after TBI.³⁴, ³⁵ The CIQ is administered in interview form, and scoring results in a CIQ total score (range, 0–29) and 3 subscale scores (home integration, social integration, productivity). The basis for scoring the CIQ lies primarily in the frequency of performing roles or activities, with secondary weighting given to whether activities are performed jointly with others. Higher scores are consistent with greater overall community reintegration, with lower scores reflective of poorer functional outcome. The patient-report version of the CIQ was used in the current study.

Statistical Analyses 

As is typical in prospective studies of clinical factors, missing data and partially observed predictors are commonplace. In the current study, participants having data available for both time periods varied across measures (ie, ranging from 100%–21%, with an average of 59% of participants having data for all measures of interest). To avoid the inefficiency and inherent bias of using complete case analysis, which necessitates dropping participants with any missing observations, the current study used the statistical modeling technique of multiple imputation³⁶ using R statistical software.^a The premise behind multiple imputation modeling is the ability to estimate missing data by generating multiple complete datasets based on plausible values for missing observations that reflect uncertainty about the model. These values are then used to fill in (ie, impute) the missing data. The results of each newly generated dataset are combined so that statistical analyses conducted on the imputed dataset take into account the uncertainty of the imputation. Thus, the multiple imputed data sets can be analyzed using standard procedures for complete data. Multiple imputation is considered a powerful method for handling missing data and is advantageous for use when applied to longitudinal data.³⁷

Binary logistic regression analyses using the imputed data model were conducted to determine how patients with complicated mild TBI differed from those of moderate TBI status on specific measures of functional disability and neuropsychologic performance. Functional disability and neuropsychologic performance were considered using separate analyses corresponding to the 2 time periods of interest after injury. For descriptive purposes, it was of further interest to examine and compare the clinical significance of each TBI group's performance over time. As such, the degree of functional and neuropsychologic impairment within each group at discharge and 1 year postinjury was explored. Impairment ratings for the neuropsychologic measures were taken from normative data and were based on a prototypical patient within each TBI group.

Back to Article Outline

Results 

Independent t tests and chi-square analyses of demographic factors showed no statistically significant differences (P>.05) between the complicated mild TBI and moderate TBI groups on level of education (Fisher exact test, P=.081), sex (Fisher exact test, P=.622), ethnicity (χ32_,N=229=2.56), or marital status (χ52_N=229=5.19). Conversely, the complicated mild TBI group was significantly older than the moderate TBI group (t₂₂₇=6.18, P<.001).

Comparison of injury severity characteristics, including highest GCS score and duration of LOC and PTA, showed statistically significant differences between the groups all in the expected direction of greater injury severity in the moderate TBI group (t₂₂₅=34.72, t₂₁₈=−3.66, t₁₉₈=−4.20, respectively; P<.01). Descriptive statistics for the complicated mild TBI and moderate TBI groups' performances on neuropsychologic and functional disability measures at rehabilitation discharge and at 1 year postinjury are provided in table 2.

Table 2. Descriptive Statistics for Functional and Neuropsychologic Performance at Rehabilitation Discharge and 1 Year Postinjury

NOTE. Values are mean ± SD and 95% confidence interval.

A binary logistic regression using as covariates age, length of PTA, education, and scores on the DRS, FIM motor domain, and FIM cognitive domain at the time of discharge from inpatient rehabilitation found only age and length of PTA to be significant predictors (P<.01), indicating that increased age and shorter PTA were associated with the complicated mild TBI group. Similarly, results of a logistic regression conducted on 1-year–postinjury data using the same covariates, with the addition of the CIQ home integration and productivity subscales, again showed age and length of PTA as the only 2 significant predictors. Thus, no differences on functional disability measures were found between the groups at discharge from rehabilitation or at the 1-year follow-up.

In terms of cognitive performance, results of binary logistic regression analyses used to determine group differences on neuropsychologic measures showed that in addition to age and duration of PTA (P<.01), performance on 2 neuropsychologic measures differentiated the complicated mild TBI and moderate TBI groups at discharge from rehabilitation (block design, WCST), and scores on 3 neuropsychologic measures differentiated the groups at 1 year postinjury (COWAT, SDMT, WCST). Specifically, poorer performance on the block design test was predictive of the moderate TBI group at the time of rehabilitation discharge (P<.05), whereas better performance on the SDMT (oral) was predictive of the complicated mild TBI group at 1 year postinjury (P<.05). Unexpectedly, worse performance on the WCST was predictive of the complicated mild TBI group at both time periods (P<.05) as determined by more perseverated responses on this task, although the associated odds of the complicated mild TBI group committing more perseverative responses on this test were less than 2% at both time periods. Also unexpected was the finding that fewer words generated on the COWAT was more predictive of the complicated mild TBI group 1 year after injury, although the odds were again relatively small at .96. Table 3 lists the odds ratios for the statistically significant neuropsychologic predictor variables.

Table 3. Significant Neuropsychologic Predictor Variables at Inpatient Rehabilitation Discharge and 1 Year Postinjury

Abbreviation: OR, odds ratio.

A descriptive analysis of cognitive impairment levels was conducted to determine whether any clinically meaningful group differences existed in neuropsychologic performance. By comparing each group's mean score on the neuropsychologic measures versus each measure's demographically appropriate normative data, a level of impairment for each neuropsychologic variable was determined. To do so, a prototypical patient was created based on the mean demographic characteristics of each group (ie, a 49-year-old man with 12 years of education in the complicated mild TBI group; a 36-year-old man with 12 years of education in the moderate TBI group). Table 4 lists levels of impairment for each group across neuropsychologic measures. Qualitative examination of impairment levels on neuropsychologic measures at both rehabilitation discharge and 1 year postinjury showed more severely impaired information processing speed (eg, SDMT) and verbal learning (eg, logical memory I and II, RAVLT) in the moderate TBI group. However, examination of impairment levels failed to find differences in performance on those neuropsychologic measures identified as statistically different on logistic regression analysis. For instance, block design performance no longer differentiated the groups at rehabilitation discharge, because both groups' performances were in the mildly impaired range. Similarly, despite increased odds of poorer performance by the complicated mild TBI group at 1 year postinjury on the COWAT, both groups performed within normative limits on this task. The complicated mild TBI group's performance on the WCST at rehabilitation discharge was indeed marginally more impaired compared with that of the moderate TBI group (ie, moderately impaired vs mildly to moderately impaired), although no differences in level of impairment on this task existed by 1 year postinjury. Finally, despite overall improvement across cognitive domains within the complicated mild TBI group, some degree of impairment remained at 1 year postinjury on those tasks identified as impaired soon after injury.

Table 4. Level of Impairment on Neuropsychologic Measures Based on Group Means Applied to a Prototypical Group Patient

To gain a better appreciation of functional outcome and level of independence after complicated mild TBI, a descriptive analysis of the functional measures was conducted. Because the FIM considers a score of 6 or 7 for any of the motor and cognitive domain items as independence in performing that function, summing across domain items results in a total score of 78 or more on the FIM motor domain (ie, 13 motor function items multiplied by 6) and a total score of 30 or more on the FIM cognitive domain (ie, 5 cognitive function items multiplied by 6) as equivalent to full domain independence. Using these cutoff scores, at the time of rehabilitation discharge 51% of patients in the complicated mild TBI group were considered independent on the FIM motor domain, whereas only 15% of this group was independent on the FIM cognitive domain. By 1 year postinjury, independence on the FIM motor and cognitive domains improved to 95% and 71%, respectively, indicating near resolution of physical disability but continued difficulty in terms of everyday comprehension, expression, social interaction, problem solving, and memory in nearly one third of the group. Functional outcome after complicated mild TBI using the broader measures of employability status and community integration were also explored. Based on the dichotomy of employable versus unemployable from the DRS, only 7% of patients in the complicated mild TBI group were considered employable at discharge from rehabilitation versus 69% at 1 year postinjury. Unexpectedly, degree of community integration within the complicated mild TBI group at 1 year postinjury, as reflected by the CIQ total score, was comparable to CIQ scores obtained in patients with moderate and severe TBI.³⁸, ³⁹ No differences were found between the complicated mild TBI and moderate TBI groups in terms of FIM motor and cognitive domain independence or employability status at either time period (Fisher exact test, P>.05), and an independent t test showed no between-group differences on the CIQ total score (t₁₂₁=−1.57, P>.05).

These results were based on the classification of TBI groups according to each patient's highest GCS scores. To control for the possibility that a patient's highest GCS scores did not accurately represent the severity of TBI ensued (eg, a patient's GCS score deteriorated to a level indicative of a more severe injury), the statistical analyses described were also conducted on groups classified using GCS scores at admission to the ED (complicated mild TBI, n=80; moderate TBI, n=76). Logistic regression analyses again showed no between-group differences on functional outcome measures at rehabilitation discharge or at 1 year postinjury. Likewise, regression analyses using neuropsychologic performances as covariates yielded similar results to those obtained based on highest GCS scores, except that more perseverative responses on the WCST at rehabilitation discharge and fewer words generated on the COWAT at 1-year follow-up were no longer associated with the complicated mild TBI group.

Back to Article Outline

Discussion 

Results of the current study support the hypothesis of similar levels of cognitive and functional outcome between complicated mild TBI and moderate TBI patients soon after injury. Unexpectedly, the finding of equivalent functional dependence and similar levels of cognitive impairment at 1 year postinjury did not support the hypothesis of better outcome after complicated mild TBI in the longer term. Overall, this study showed few differences in outcome between patients with mild TBI complicated by an intracranial lesion and patients of moderate TBI status in both the short- and long-term postinjury.

Previous studies have shown better neuropsychologic performance after uncomplicated mild TBI compared with mild TBI complicated by a brain lesion despite similar GCS scores and durations of LOC and PTA between these 2 TBI populations. For instance, Borgaro et al¹⁴ found that patients with complicated mild TBI performed significantly worse than uninjured control participants on all cognitive measures of the Barrow Neurological Institute Screen for Higher Cerebral Functions at 1 month postinjury, whereas GCS score–matched, uncomplicated mild TBI patients' performances were comparable with those of uninjured controls. Likewise, Williams et al¹¹ found similar impairment levels within 3 months postinjury between complicated mild TBI and moderate TBI patients on measures of information processing, memory, and word generation compared with unimpaired performances in patients with uncomplicated mild TBI whose duration of LOC and PTA were equivalent to those of the complicated mild TBI group. The current study extends previous findings by showing that neuropsychologic performances on only 2 measures at rehabilitation discharge and 3 measures at 1 year postinjury statistically differentiated the complicated mild TBI group from the moderate TBI group. Moreover, qualitative analysis of levels of impairment based on normative data effectively nullified any group differences on these identified neuropsychologic measures.

Further consideration of levels of cognitive impairment across groups showed more impaired information processing speed and learning and memory functioning in the moderate TBI group at both time periods of interest. This finding is consistent with prior research², ⁴⁰ showing that existing residual cognitive deficits after TBI are most commonly exhibited as slowed processing speed and compromised learning and memory. That the moderate group exhibited more impaired performances compared with the complicated mild TBI group in these cognitive domains even shortly after injury suggests appreciable differences in the vulnerability of the associated neural systems between these 2 TBI populations. Furthermore, that complete resolution of cognitive functioning to unimpaired levels did not occur within the complicated mild TBI group by 1 year postinjury provides additional evidence that despite similar durations of LOC and PTA, neuropsychologic outcome for complicated mild TBI patients does indeed differ from that of patients with uncomplicated mild TBI, whose neuropsychologic status is expected to return to baseline levels by 3 months postinjury.⁷, ⁸, ⁹

In terms of functional status, the current study found no between-group differences at either time period of interest. Previous studies using the GOS found that outcome after complicated mild TBI was poorer compared with uncomplicated mild TBI, such that outcome scores after complicated mild TBI more closely approximated those of patients with moderate TBI.¹⁰, ¹¹, ²¹ By using more sensitive measures of TBI outcome and assessing a broader range of functional abilities, the current results extend previous findings by showing equivalent outcomes in terms of levels of physical (ie, FIM motor domain) and cognitive (ie, FIM cognitive domain) independence, employability (eg, DRS), and degree of community integration (ie, CIQ) between complicated mild TBI and moderate TBI patients. As with neuropsychologic outcome, recovery of functional status was far from complete by 1 year postinjury, thus providing further evidence of disparate outcomes after complicated mild TBI vis-à-vis uncomplicated mild TBI. For instance, within the complicated mild TBI group total scores on the DRS corresponded to the partially impaired range, 29% of patients remained in the dependent range on the FIM cognitive domain, nearly one third of the group remained unemployable, and level of community integration at 1 year postinjury was similar to that of patients with moderate to severe TBI.³⁷, ³⁸

The established differences in level of cognitive impairment with respect to information processing speed and learning and memory functioning across time between the complicated mild TBI and moderate TBI groups did not translate into discrepancies in the groups' ability to participate within a broad range of everyday physical and mental activities, routines, and social roles as assessed by measures of functional ability. Equally deficient functional recovery at 1 year postinjury in both TBI groups despite more impaired cognitive performance on some tasks by moderate TBI patients is likely partially explained by the complexity of cognitive processes required to adequately engage in a functional activity, such that impairments within a specific cognitive domain do not correspond to specific functional disability. Studies to date have shown an important role for the domain of executive functioning in the recovery of functional ability post-TBI. For example, using the CIQ at 1 year post-TBI, Ross et al⁴¹ found that age, information processing, memory, and complex attention were related to levels of social integration. Little et al⁴² reported that processing speed, cognitive flexibility, and novel problem solving were significantly related to postacute TBI rehabilitation discharge scores on the DRS. Similarly, in a mixed sample of rehabilitation inpatients, Hanks et al⁴³ found that measures of executive control predicted rehabilitation outcome on community re-entry beyond information regarding sensory and motor dysfunction. In addition to cognitive performance, the prediction of functional outcome post-TBI is complicated by a wide array of other moderating variables. To further our understanding of functional outcome and enhance our ability to better predict functional outcome after TBI, additional research exploring the interaction between intrinsic (eg, premorbid personality, mood, pain) and extrinsic (eg, vocational history, family, and social support) factors and cognitive status post-TBI is required.

Results of the current study show sufficient parallels in outcome after complicated mild TBI and moderate TBI to indicate that the incorporation of a new subclassification of TBI is not warranted and that patients with GCS scores of 13 to 15 and evidence of an intracranial bleed should be classified as having moderate TBI. Consequently, the focus of treatment and rehabilitation after complicated mild TBI should differ from that of uncomplicated mild TBI. Expectations regarding the cognitive and functional limitations after complicated mild TBI should be conveyed to patients, family members, and employers so as to promote optimal accommodations and assistance with transitioning back to the larger community setting as recovery progresses. Similarly, behavioral interventions pertaining to supervision, decision making, work environment, and issues related to mood and adjustment should be addressed accordingly with this TBI population.

Moreover, these findings highlight the limitations of relying on GCS scores in the classification of TBI severity given that some persons falling within the conventional mild range of injury severity sustain intracranial lesions that place them at risk for poorer outcome. Studies¹², ⁴⁴ have shown that patient presentations characterized by GCS scores of 15 and brief LOC and PTA do not preclude the existence of intracranial abnormalities. Greater consideration should be given to factors such as duration of LOC, length of PTA, and time to follow commands in the classification of TBI severity. Likewise, the incorporation of results from intracranial radiologic imaging should be considered, even with seemingly milder injuries.

Study Limitations 

Given the exploratory nature of outcome after complicated mild TBI at 1 year postinjury, replication of the findings in other complicated mild TBI samples is necessary. Of further consideration is the extent to which the complicated mild TBI group in the current study represents a subset of the most severely injured patients within this TBI population, given that their injuries were of sufficient severity to warrant inpatient rehabilitation. Future research should include patients with complicated mild TBI who sustain less severe bodily injuries such that hospitalization is minimal postinjury. Notably, the patients with complicated mild TBI in the current study all endured intracranial bleeds; thus, patients who sustain epidural bleeds or surface scalp abrasions should not be considered equivalent with this group. Finally, although the overrepresentation of men, African Americans, and single persons in the current study is consistent with previous studies of TBI, caution should be exercised in generalizing the findings to TBI populations who differ demographically.

Back to Article Outline

Conclusions 

The current study extends findings of short-term cognitive and functional outcome after complicated mild TBI and provides a preliminary view of longer-term recovery. Sufficient parallels in outcome after complicated mild TBI and moderate TBI were found to indicate that when classifying severity of TBI based on GCS scores, consideration of a moderate injury designation should be given to persons with an intracranial bleed and a GCS score between 13 and 15.

Supplier

Back to Article Outline

References 

1. National Institutes of Health Consensus Developmental Panel on Rehabilitation of Persons with Traumatic Brain Injury. Rehabilitation of persons with traumatic brain injury. JAMA. 1999;282:974–981
   • View In Article
   • MEDLINE
   • CrossRef
2. Dikmen SS, Ross BL, Machamer JE, Temkin NR. One year psychosocial outcome in head injury. J Int Neuropsychol Soc. 1995;1:67–77
   • View In Article
   • MEDLINE
   • CrossRef
3. Cicerone KD, Mott T, Azulay JA, Friel J. Community integration and satisfaction with functioning after intensive cognitive rehabilitation for traumatic brain injury. Arch Phys Med Rehabil. 2004;85:943–950
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (87 KB)
   • CrossRef
4. Sander AM, Kreutzer JS, Rosenthal M, Delmonico R, Young ME. A multicenter longitudinal investigation of return to work and community integration following traumatic brain injury. J Head Trauma Rehabil. 1996;11(5):70–84
   • View In Article
   • CrossRef
5. Dikmen SS, McLean A, Temkin N. Neuropsychologic and psychosocial consequences of minor head injury. J Neurol Neurosurg Psychiatry. 1986;49:1227–1232
   • View In Article
   • MEDLINE
   • CrossRef
6. Teasdale G, Jennett B. Assessment of coma and impaired consciousness (A practical scale). Lancet. 1974;2:81–84
   • View In Article
   • CrossRef
7. Levin HS, Mattis S, Ruff RM, et al. Neurobehavioral outcome of minor head injury: a three center study. J Neurosurg. 1987;66:234–243
   • View In Article
   • MEDLINE
   • CrossRef
8. Kashluba S, Paniak C, Blake T, Reynolds S, Toller-Lobe G, Nagy J. A longitudinal, controlled study of patient complaints following treated mild traumatic brain injury. Arch Clin Neuropsychol. 2004;19:805–816
   • View In Article
   • MEDLINE
   • CrossRef
9. Ponsford J, Willmott C, Rothwell A, et al. Factors influencing outcome following mild traumatic brain injury in adults. J Int Neuropsychol Soc. 2000;6:568–579
   • View In Article
   • MEDLINE
   • CrossRef
10. Hsiang JN, Yeung T, Ashley LM, Poon WS. High-risk mild head injury. J Neurosurg. 1997;87:234–238
    • View In Article
    • MEDLINE
    • CrossRef
11. Williams DH, Levin HS, Eisenberg HM. Mild head injury classification. Neurosurgery. 1990;27:422–428
    • View In Article
    • MEDLINE
    • CrossRef
12. Vollmer DG, Dacey RG. The management of mild and moderate head injuries. Neurosurg Clin N Am. 1991;2:437–455
    • View In Article
    • MEDLINE
13. Goldstein FC, Levin HS, Goldman WP, Clark AN, Kenehan-Altonen TK. Cognitive and neurobehavioral functioning after mild versus moderate traumatic brain injury in older adults. J Int Neuropsychol Soc. 2001;7:373–383
    • View In Article
    • MEDLINE
    • CrossRef
14. Borgaro SR, Prigatano GP, Kwasnica C, Rexer JL. Cognitive and affective sequelae in complicated and uncomplicated mild traumatic brain injury. Brain Inj. 2003;17:189–198
    • View In Article
    • MEDLINE
    • CrossRef
15. Jennett B, Bond M. Assessment outcome after severe brain damage. Lancet. 1975;1:480–484
    • View In Article
    • MEDLINE
16. Paniak C, Toller-Lobe G, Durand A, Nagy J. A randomized trial of two treatments for mild traumatic brain injury. Brain Inj. 1998;12:1011–1023
    • View In Article
    • MEDLINE
    • CrossRef
17. Gronwall D. Rehabilitation programs for patients with mild head injury: components, problems, and evaluation. J Head Trauma Rehabil. 1986;1(2):53–62
    • View In Article
    • CrossRef
18. Binder LM, Rohling ML, Larrabee GJ. A review of mild head trauma: part I (Meta-analytic review of neuropsychological studies). J Clin Exp Neuropsychol. 1997;19:421–431
    • View In Article
    • MEDLINE
    • CrossRef
19. Mittenberg W, Strauman S. Diagnosis of mild head injury and the postconcussion syndrome. J Head Trauma Rehabil. 2000;15:783–791
    • View In Article
    • MEDLINE
    • CrossRef
20. Kashluba S, Paniak C, Casey JE. Persistent symptoms associated with factors identified by the WHO task force on mild traumatic brain injury. Clin Neuropsychol. 2007;27:1–14
    • View In Article
21. van der Naalt J, van Zomeren AH, Sluiter WJ, Minderhound JM. One year outcome in mild to moderate head injury: the predictive value of acute injury characteristics related to complaints and return to work. J Neurol Neurosurg Psychiatry. 1999;66:207–213
    • View In Article
    • MEDLINE
    • CrossRef
22. Hall KM, Bushnik T, Lakisic-Kazazic B, Wright J, Cantagallo A. Assessing traumatic brain injury outcome measures for long-term follow-up of community based individuals. Arch Phys Med Rehabil. 2001;82:367–374
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (96 KB)
    • CrossRef
23. Rey A. L'examen clinique en psychologie. Paris: Presses Universitaires de France; 1958;
    • View In Article
24. Wechsler D. Memory scale–revised manual. San Antonio: Psychological Corp; 1987;
    • View In Article
25. Wechsler D. WAIS-R manual. New York: Psychological Corp; 1981;
    • View In Article
26. Army individual test battery: manual of directions and scoring. Washington (DC): War Department, Adjutant General's Office; 1944;
    • View In Article
27. Benton AL, Hamsher K. Multilingual Aphasia Examination. Iowa City: AJA Assoc; 1989;
    • View In Article
28. Smith A. Symbol Digits Modalities Test Manual. (revised) Los Angeles: Western Psychological Services; 1982;
    • View In Article
29. Grant DA, Berg EA. A behavioral analysis of the degree of reinforcement and ease of shifting to new responses in a Weigl-type card sorting problem. J Exp Psychol. 1948;38:404–411
    • View In Article
    • CrossRef
30. Uniform Data System for Medical Rehabilitation. Uniform data system functional independence manual. Version 4.0 Buffalo: Uniform Data System for Medical Rehabilitation; 1993;
    • View In Article
31. Linacre JM, Heinemann AW, Wright BD, Granger CV, Hamilton BB. The structure and stability of the Functional Independence Measure. Arch Phys Med Rehabil. 1994;75:127–132
    • View In Article
    • MEDLINE
32. Rappaport M, Hall K, Hopkins K, Belleza T, Cope DN. Disability Rating Scale for severe head trauma: coma to community. Arch Phys Med Rehabil. 1982;63:118–123
    • View In Article
    • MEDLINE
33. Hall KM, Cope DN, Rappaport M. Glasgow Outcome Scale and Disability Rating Scale: comparative usefulness in following recovery in traumatic brain injury. Arch Phys Med Rehabil. 1985;66:35–37
    • View In Article
    • MEDLINE
34. Willer B, Rosenthal M, Kreutzer JS, Gordon WA, Rempel R. Assessment of community integration following rehabilitation for traumatic brain injury. J Head Trauma Rehabil. 1993;8(2):75–87
    • View In Article
    • CrossRef
35. Dijkers M. Measuring the long-term outcomes of traumatic brain injury: a review of Community Integration Questionnaire studies. J Head Trauma Rehabil. 1997;12(6):74–91
    • View In Article
    • CrossRef
36. Rubin DB. Multiple imputation for nonresponse in surveys. New York: John Wiley & Sons; 1987;
    • View In Article
37. Tabachnick BG, Fidell LS. Using multivariate statistics. 4th ed.. Boston: Allyn & Bacon; 2001;
    • View In Article
38. Sander AM, Fuchs KL, High WM, Hall KM, Kreutzer JS, Rosenthal M. The Community Integration Questionnaire revisited: an assessment of factor structure and validity. Arch Phys Med Rehabil. 1999;80:1303–1308
    • View In Article
    • Abstract
    • Full-Text PDF (896 KB)
    • CrossRef
39. Novack TA, Alderson AL, Bush BA, Meythaler JM, Canupp K. Cognitive and functional recovery at 6 and 12 months post-TBI. Brain Inj. 2000;14:987–996
    • View In Article
    • MEDLINE
    • CrossRef
40. Levin HS, Gary HE, Eisenberg HM, et al. Neurobehavioral outcome 1 year after severe head injury. J Neurosurg. 1990;73:699–709
    • View In Article
    • MEDLINE
    • CrossRef
41. Ross SR, Millis SR, Rosenthal M. Neuropsychological prediction of psychosocial outcome after traumatic brain injury. Appl Neuropsychol. 1997;4:165–170
    • View In Article
    • MEDLINE
    • CrossRef
42. Little AJ, Templer DI, Persel CS, Ashley MJ. Feasibility of the neuropsychological spectrum in prediction of outcome following head injury. J Clin Psychol. 1996;52:455–460
    • View In Article
    • MEDLINE
    • CrossRef
43. Hanks RA, Rapport LJ, Millis SR, Deshpande SA. Measures of executive functioning as predictors of functional ability and social integration in a rehabilitation sample. Arch Phys Med Rehabil. 1999;80:1030–1037
    • View In Article
    • Abstract
    • Full-Text PDF (1096 KB)
    • CrossRef
44. Jeret J, Mandell M, Anziska B, et al. Clinical predictors of abnormality disclosed by computed tomography after mild head injury. Neurosurgery. 1993;32:9–16
    • View In Article
    • MEDLINE
    • CrossRef

• a R Project for Statistical Computing. Available at: http://www.r-project.org. Accessed January 31, 2008.