Effects of Environmental Demands on Locomotion After Traumatic Brain Injury

Article Outline

• Abstract
• Methods
  • Participants
  • Instrumentation
  • Protocol
  • Data Analyses
• Results
  • Reading Time
  • Walking Speed
  • Stride Length
  • Clearance Margin
• Discussion
  • General Performance on the Reading Task
  • General Locomotor Capacity Without Division of Attention
  • Division of Attention During Locomotion
  • Study Limitations
• Conclusions
• References
• Copyright

Abstract 

Vallée M, McFadyen BJ, Swaine B, Doyon J, Cantin J-F, Dumas D. Effects of environmental demands on locomotion after traumatic brain injury.

Objective

To determine the effects of increasingly demanding environments related to simultaneous visual tasks and physical obstructions on the locomotor ability of people with traumatic brain injury (TBI).

Design

Group comparison study.

Setting

Gait analysis laboratory within a postacute rehabilitation facility.

Participants

Volunteer sample of 9 people (8 men, 1 woman; age, 39.3±13.0y) with moderate to severe TBI and a comparison group of 9 subjects without neurologic problems matched for age and sex (8 men, 1 woman; age, 39.7±12.3y).

Interventions

Not applicable.

Main Outcome Measures

Reading times for the Stroop bar and Stroop word tasks, walking speeds, stride lengths, and obstacle clearance margin.

Results

The TBI group was slower than the control group in performing the Stroop bar task during sitting (P=.002), and while avoiding the narrow obstacle (P=.05), and in performing the Stroop word task while avoiding the wide obstacle (P=.019). Despite their relatively normal gait speeds on level ground, subjects with TBI walked more slowly than control subjects for the narrow (P=.024) and the wide (P=.019) obstacle conditions and for the most complex dual task (P=.042). Greater lead-limb clearance margins were observed for the TBI group than for control subjects for all conditions whereas no differences were found for the trail limb except at the far end of the wide obstacle.

Conclusions

Despite their good recovery of locomotor function, with respect to normal level walking speeds and ability to avoid obstacles, subjects with moderate and severe TBI showed residual deficits in relation to greater difficulties in dealing with environments that challenge their locomotor and attentional abilities. The use of such naturally based dual tasks may help identify some of the environmental obstructions to social participation after TBI.

Key Words:  Attention , Gait , Rehabilitation

PEOPLE WHO HAVE SUFFERED a traumatic brain injury (TBI) have both physical and cognitive deficits that interfere with their locomotor ability. With time and rehabilitation, recovery is possible but varies in relation to different personal and social factors.¹ Katz et al² showed that the recovery of independent walking after TBI takes on average 1.5 months, with 94% of people recovering after 3 months. Despite such quick recovery of basic ambulation, however, the capacity to perform challenging locomotor tasks may still be compromised. In a group of subjects with severe TBI, Swaine and Sullivan³ reported that 60% to 65% were unable to complete locomotor activities such as running, hopping, and jumping. Williams et al⁴ noted that high-level mobility is poorly understood after TBI, and proper clinical measures to assess higher levels of mobility are required. However, there is still little known about the basic residual locomotor capacity after TBI, particularly in relation to combined physical and cognitive constraints imposed by daily environments.

Motor ability after TBI has not been studied as well as for other populations with neurologic impairments (eg, stroke). It is known, however, that people with TBI show a diminution in the control of static⁵, ⁶, ⁷ and dynamic⁸ balance that depends on visual information. Ochi et al⁹ reported that level walking speeds for subjects with TBI represent approximately half of those for control subjects. McFadyen et al¹⁰ showed that subjects with mild to severe TBI walked more slowly and had shorter stride lengths than control subjects. Basford et al¹¹ found that despite a seemingly good recovery according to clinical evaluation, people with mild to severe TBI reduced their sagittal plane movement as a strategy to maintain their balance while walking.

Stepping over obstacles generally requires specific locomotor adaptations related to lower-limb coordination, foot clearance, and speed.¹², ¹³, ¹⁴, ¹⁵ Chou et al¹⁶ found slower walking speeds and shorter stride lengths in subjects with TBI compared with control subjects for both unobstructed and obstructed walking. McFadyen¹⁰ showed similar results, although these variables were not significantly different from those for control subjects for obstacle crossing. There was also a consistent tendency for subjects with TBI to adopt greater foot clearance margins. The researchers concluded that despite their functional walking ability, subjects with TBI used increased caution in this demanding environment.

Within daily community activities, one must not only adapt his/her walking patterns but also attend to different simultaneous stimuli (eg, street signs). Problems of attention are one of the most frequent sequela after a TBI¹⁷, ¹⁸ related to increased distractibility¹⁹, ²⁰ and problems in sustained²¹ and divided attention.²², ²³ In terms of combined physical and cognitive tasks, Geurts et al⁸ showed that subjects with TBI tended toward greater errors in a simple dual arithmetic task with effects tending to be stronger with increased balance-task complexity. Another study on the effects of a dual task on standing balance showed greater interference in subjects with severe TBI, particularly with simple nonspatial tasks.²⁴

Only 1 study to date has looked at the effects of a concurrent cognitive task on gait in people with TBI. Parker et al²⁵ found that both subjects with mild TBI and controls had slower walking speeds and shorter stride lengths during a dual-task condition as compared with just walking. Subjects with TBI, however, had even shorter stride lengths than control subjects when attention was divided. Studying the center of mass (COM) motion, the researchers observed a conservative strategy in the mild TBI group, with a deficit in dual-task management.

Given their importance to activities of daily living, studies of simultaneous locomotor and cognitive tasks after TBI are required not only to improve our understanding of general mobility constraints after TBI but also to provide information related to the facilitators and obstacles to social participation. The purpose of this study was to determine the effects of increasingly demanding environments on locomotor ability of independently ambulating people with moderate or severe TBI as compared with control subjects. Specific environmental factors considered in this study were with respect to a nonlocomotor-related reading task and obstacle avoidance. It was expected that, regardless of their autonomy and locomotor recovery, subjects with TBI would show both decreased locomotor and cognitive performances in the more demanding environments compared with control subjects.

Back to Article Outline

Methods 

Participants 

We recruited people who had suffered TBI from the TBI unit of the Québec Rehabilitation Institute (QRI). All subjects suffered only 1 TBI, with severity ratings of moderate or severe based on a combination of the hospital admission Glasgow Coma Scale (GCS) score, duration of posttraumatic amnesia (PTA), length of the loss of consciousness, and interpretation of the neuroradiologic examination.²⁶ Subjects also had to be able to walk at a speed greater than 0.7m/s without assistance or walking device (eg, cane). We excluded subjects if they had a skull fracture or cerebral lesion caused by perforation, cognitive or behavioral problems that could adversely affect their participation, or any other neurologic or musculoskeletal problems affecting their locomotion. For comparison purposes, we also recruited control subjects with no self-reported physical or neurologic problems who were matched on average for age and sex to the TBI group. The ethics committee of the QRI approved this study, and all subjects signed a consent form before participating.

Instrumentation 

Subjects wore their own comfortable walking shoes, shorts, and a T-shirt for the study. We fixed triads of noncolinear infrared markers on the 3 lower-limb segments bilaterally and on the pelvis, trunk, and head. We collected kinematic data using 3 Optotrak sensor bars^a at a frequency of 75Hz. We statically digitized the heel and toe in relation to foot markers for reconstruction after data collection. Subjects’ verbal responses to the cognitive task (described later) were recorded with a microphone placed in front of the mouth and fixed on headphones that played white noise to minimize distraction from ambient sounds in the laboratory (eg, keyboard tapping). We used a channel mixer^b to amplify the voice, which was then captured on both a computer (1000Hz) and as part of a video recording of the trials. We placed a set of five 43.2-cm (17-in) flat-screen computer monitors,^c 2 pairs forming the sides of the walking path (width, ≈1.6m) and 1 at the end. The monitors were connected to the computer providing the visual stimuli using a video splitter.^d

Protocol 

All subjects passed separate clinical tests of executive function and attention as well as of balance and gait ability, but we only concentrate on the laboratory results here. Visual acuity of all subjects was assessed with the Snellen test to ensure that they had a normal vision (20/20). The laboratory session was composed of 3 different physical conditions combined with 3 different conditions related to concurrent visual tasks. The simultaneous visual stimuli were adapted versions of the Stroop bar and word tests used in neuropsychology²⁷ and shown to be valid for testing attention deficits in a TBI population.²⁸ Specifically, we modified the paper tests using 2 columns of either colored bars or words that we projected simultaneously on all of the computer monitors placed along the walkway in clear view for the subjects. This arrangement allowed subjects to get visual information without moving the head too much from their desired focal point. When bars were presented, subjects had simply to sequentially name the color (red, blue, green) of each of the 8 bars. To increase the level of complexity of the task, we presented words indicating the same 3 colors but in a different color than their lexical meaning. In this case, subjects had to name the color of the ink and ignore the meaning of the word. We chose the Stroop tests because they provide a simple reading task not requiring memory and they could be incremented in complexity. We also chose them to enable future comparisons between the standard neuropsychologic paper versions and those used during the locomotor testing among subjects.

Physical conditions included walking straight ahead for 11m while unobstructed, while stepping over a narrow obstacle, and while stepping over a wide obstacle. To make comparisons across the greater variability in locomotor abilities originally expected in the TBI group, we fixed the obstacle dimensions as a ratio of each subject’s maximum step height and length. With this method, the difficulty of the physical task was adapted to each subject’s personal capacity. We first asked subjects to walk in a fluid manner while stepping as high as possible and then recorded foot coordinate data. We repeated the same procedure by asking subjects to walk in a fluid manner while making the longest step length possible. We calculated the averages of maximum step heights and lengths over 2 or 3 consecutive steps and set the height (and depth for the wide obstacle) to 30% of these values. The narrow obstacle was 2.8cm wide for all participants.

All subjects first performed 3 trials of each type of Stroop reading task while sitting. Subjects were then asked to walk on the walkway 2 or 3 times to familiarize themselves with all the equipment and to find their natural walking speed. For the subsequent walking trials, we first exposed subjects to blocks of 5 trials of each different physical condition (unobstructed, narrow and wide obstacles). Finally, we again collected separate blocks of 10 trials of each physical condition but with the visual stimuli conditions also randomly presented. We informed subjects of the context of the visual stimulus before each trial and instructed them to read the projected page as fast as possible, starting with the left-hand column and going from the top to the bottom. If they finished reading the projected page before reaching the end of the walkway, we instructed subjects to start reading again. We asked subjects to always walk at their natural speeds, and they were free to choose with which limb to initiate clearance when obstructed.

Data Analyses 

We filtered marker coordinate data using a second-order Butterworth, zero-lag filter with a cutoff frequency of 6Hz. We defined the lead limb as that which cleared the obstacle first with the trail limb following. An average of each dependent variable was calculated across trials for each subject to be used for statistical analyses. For each condition, we grouped the data according to the level of difficulty of the task; we considered the wide obstacle the most difficult physical task and the Stroop word test the most difficult simultaneous visual task. In this respect, we considered the combination of the wide obstacle with the Stroop word task to be the most difficult dual task. Dependent variables that we kept for analyses were average reading time per word (calculated as total reading time divided by the number of items read), walking speed (average speed of the body COM between consecutive lead-limb foot contacts), stride length of each limb (normalized to lower-limb length and beginning at contact just before the obstacle by the lead and trail limbs in the avoidance conditions), and foot clearance margin (distance of the foot above the obstacle normalized to obstacle height). For the wide obstacle conditions, we also calculated clearance margin at the far end of the obstacle using the heel of the lead limb and the toe of the trail limb according to the normal foot trajectory of each limb.

We used nonparametric statistical tests to compare conditions within (Friedman, Wilcoxon) and between (Mann-Whitney U) groups because they are more conservative than their parametric counterparts and because of the questionable homogeneity of the data among the TBI group. We set significance levels to .05. Although we made a number of comparisons, any corrections used to minimize type I errors would have also increased the risk of type II errors. However, we declared all α values for all comparisons.

Back to Article Outline

Results 

We compared 9 subjects (8 men, 1 woman) with a moderate or severe TBI (table 1) with 9 control subjects matched for age (39.7±12.3y) and sex (8 men, 1 woman). The clinical gait speeds of the TBI group were quite good.

Table 1. Characteristics of Subjects With TBI

Abbreviations: F, female; M, male; NA, not available; SD, standard deviation.

Reading Time 

We found a general effect (P<.001) for the average reading time per word showing that subjects with TBI took longer in general to read during both Stroop tasks. Both groups took longer to perform the Stroop word task than the Stroop bar task across all conditions (P=.004) (fig 1). The TBI group was significantly slower than the control group in performing the Stroop bar task (see fig 1A) during sitting (P=.002) and while avoiding the narrow obstacle (P=.05). When performing the Stroop word task, subjects with TBI were only significantly slower than the control group during avoidance of the wide obstacle (P=.019) (see fig 1B). Across conditions, control subjects needed the same time to perform the Stroop word task while sitting, walking, and crossing over obstacles. Subjects with TBI, however, needed more time to complete this secondary task for wide obstacle avoidance as compared with the sitting position (P=.012) and unobstructed walking (P=.027).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Average time per item to read (A) the Stroop bar and (B) the Stroop word tasks while sitting, walking unobstructed (NO), avoiding the narrow obstacle (NrO), and avoiding the wide obstacle (WdO) for the subjects with TBI (dark boxes) and control subjects (pale boxes). The boxplots indicate medians (thick horizontal bars) and the 75th (top of box) and 25th (bottom of box) percentile ranges; the whiskers indicate full ranges. Outliers (∘) represent values that were at least 1.5 times greater or less than the interquartile range (IQR). Significant differences (P≤.05) are shown within groups (^___) and between groups (*).

Walking Speed 

Even though subjects with TBI walked more slowly than control subjects overall, they showed good locomotor recovery with near-normal walking speeds both in the clinic (see table 1) and in the laboratory, where we found no significant differences between groups during unobstructed walking (fig 2A). Despite their relatively normal unobstructed gait speeds, subjects with TBI were significantly slower than control subjects for both the narrow (P=.024) (see fig 2B) and wide (P=.019) (see fig 2C) obstacles and for the most complex dual-task environment (wide obstacle with Stroop word task) (P=.042). Compared with unobstructed walking, only subjects in the TBI group decreased their crossing speeds while stepping over the wide obstacle in combination with division of attention (P=.02 for the Stroop bar task, P=.027 for the Stroop word task) (see fig 2C). Finally, we observed that speeds were affected more by the Stroop word than the Stroop bar task during unobstructed walking for control subjects (P=.004) (see fig 2A) and for the narrow and wide obstacle conditions for subjects with TBI (P=.023 and P=.004, respectively) and for control subjects (P=.027 and P=.02, respectively) (see figs 2B, 2C).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  Walking speeds during (A) unobstructed walking (NO) and (B) while avoiding the narrow obstacle (NrO) and (C) the wide obstacle (WdO). Data are indicated for subjects with TBI (dark boxes) and control subjects (pale boxes) without any division of attention and with simultaneous Stroop bar (B) or Stroop word (W) tasks. The boxplots indicate medians (thick horizontal bars) and the 75th (top of box) and 25th (bottom of box) percentile ranges; the whiskers indicate full ranges. Outliers (∘) represent values that were at least 1.5 times greater or less than the IQR. Significant differences (P≤.05) are shown within groups (^___) and between groups (*).

Stride Length 

We observed that subjects with TBI adopted shorter stride lengths than control subjects in general for both limbs (fig 3). For the lead limb, significant differences were observed specifically for the 2 obstacle conditions (see figs 3B, 3C), either alone (P<.001) or with the Stroop bar task (P=.002) and the Stroop word task (P=.05 for the narrow, P=.003 for the wide obstacles). Across conditions, we found that the narrow obstacle caused a decrease of the lead-limb stride length compared with unobstructed walking (P=.012) only in the TBI group. In both groups, the Stroop word task affected the lead-limb stride lengths more than the Stroop bar task during unobstructed walking (P=.039 and P=.012 for TBI and control groups, respectively).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Stride lengths for (A, B, C) the lead and (D, E, F) the trail limbs while walking unobstructed (NO), while avoiding the narrow obstacle (NrO), and while avoiding the wide obstacle (WdO) without any division of attention and with simultaneous Stroop bar (B) or Stroop word (W) tasks for the subjects with TBI (dark boxes) and control subjects (pale boxes). The boxplots indicate medians (thick horizontal bars) and the 75th (top of box) and 25th (bottom of box) percentile ranges; the whiskers indicate full ranges. Outliers (∘) represent values that were at least 1.5 times greater or less than the IQR. Significant differences (P≤.05) are shown within groups (^___) and between groups (*).

For the trail limb (see figs 3D–F), subjects with TBI had shorter stride lengths than control subjects when stepping over obstacles only (P=.014 for the narrow obstacle [see fig 3E], P=.008 for the wide obstacle [see fig 3F]) and when stepping over the wide obstacle with both levels of visual stimuli (P=.011). During unobstructed walking, although the addition of the Stroop word task caused a decrease of the trail-limb stride lengths in both groups, the Stroop bar task affected trail-limb stride length only in the TBI group (P=.039). The Stroop word task decreased trail-limb stride lengths more than the Stroop bar task during unobstructed walking in the control group (P=.004). For the wide obstacle conditions, however, only the TBI group significantly decreased their trail-limb stride lengths for the Stroop word task compared with both the Stroop bar and nondistracted avoidance tasks (P=.039).

Clearance Margin 

When we combined all conditions, subjects with TBI had higher clearance margins at the leading edge of the obstructions than control subjects for the lead (P<.001) and trail (P=.011) limbs (figs 4A, 4B, 4D, 4E). Closer analyses showed, however, significant differences between groups for the lead limb only (P<.02) (see figs 4A, 4B). Similar trends were evident across conditions for each group, but only control subjects significantly increased lead-limb clearance margin with the increasing Stroop task complexity during narrow obstacle avoidance (P=.004 when we added the Stroop bar or the Stroop word tasks, P=.039 between Stroop tasks). Lead-limb clearance margins were affected more by the Stroop word than the Stroop bar task only during narrow obstacle avoidance in control subjects (P=.039).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Clearance margins for (A, B, C) the lead and (D, E, F) the trail limbs when stepping over the front end of the narrow (NrO) and wide (WdO) obstacles and over the far end of the wide obstacle (FWdO) without any division of attention and with simultaneous Stroop bar (B) or Stroop word (W) tasks for the subjects with TBI (dark boxes) and control subjects (pale boxes). The boxplots indicate medians (thick horizontal bars) and the 75th (top of box) and 25th (bottom of box) percentile ranges; the whiskers indicate full ranges. Outliers (∘) represent values that were at least 1.5 times greater or less than the IQR. Significant differences (P≤.05) are shown within groups (^___) and between groups (*).

For the clearance margin at the far end of the wide obstacle, we observed a general effect of the Stroop tasks and of obstacle presence for both limbs. Specifically, subjects with TBI kept a greater (P=.001) heel-clearance margin of the lead limb (see fig 4C) compared with control subjects during crossing both without the concurrent Stroop task and with the Stroop bar task. For the toe of the trail limb (see fig 4F), we observed significant differences only for combined Stroop and obstacle conditions (P=.011 for both Stroop tasks). Control subjects slightly but significantly decreased the trail-limb clearance margin during dual tasks as compared with obstacle avoidance alone (P=.039 for the Stroop bar, P=.027 for the Stroop word), but this was not the case for subjects with TBI who already had higher clearance margins than control subjects.

Back to Article Outline

Discussion 

The results of the present study provide insight into the residual deficits of people with a moderate or severe TBI who can ambulate independently. Although in unobstructed environments these subjects had levels of locomotor function similar to those of control subjects, we found group differences for dependent variables related to increases in complexity in both the cognitive and locomotor tasks. There did not appear to be a simple change in behavior in direct relation to additional physical and cognitive aspects of the environment. Subjects with TBI, however, were affected more by the presence of obstacles and when performing the Stroop tasks and were clearly affected more by the most complex environment combining the wide obstacle with the simultaneous Stroop word task.

General Performance on the Reading Task 

The control of posture and locomotion is not completely automatic. There is an attentional cost that decreases performance with increasing difficulty of both the motor and simultaneous cognitive tasks.²⁹, ³⁰ People with TBI often have limited attentional ability³¹ and the present results support this. The wide range of scores for the Stroop word reading performance for both groups will contribute to the lack of significant differences between the groups (although the small sample size must be considered too). However, when we combined this task with the wide obstacle, people with TBI clearly showed difficulties under this condition. Avoidance of a wide obstacle is more difficult and likely requires more attention to be done safely. It is possible that tasks like the Stroop word test combined with different physical tasks of increasing complexity could eventually be used to discriminate attentional ability among people with TBI and to train those who have attained a good level of locomotor recovery.

General Locomotor Capacity Without Division of Attention 

Although other studies have shown slower unobstructed walking speeds for subjects with TBI,⁹, ¹⁰, ¹⁶ there were no significant differences between groups for this condition in the present study. The same observation holds true for lead- and trail-limb stride lengths when unobstructed. Overall, however, the clear trend toward slower walking speeds and shorter stride lengths during the unobstructed walking condition was still obvious in the present study. Lack of significance in the present study may be related to a number of factors including better locomotor recovery, or perhaps a better matching of age and sex between groups. It seems that a large variability in the control group’s performance was also partly responsible for the present lack of significant differences (see fig 2). The TBI group of the present study also walked at greater average speeds compared with those reported by Chou et al,¹⁶ and this may be due in part to the fact that the subjects of the latter study walked barefoot.

Postural control is more challenging during obstacle avoidance than during level gait given the required adjustments to the trajectories of the feet. Both the present work and a previous study¹⁶ have shown that subjects with TBI cross obstacles more slowly compared with control subjects. The shorter stride lengths observed both here and previously¹⁰, ¹⁶ for subjects with TBI explains such slower crossing speeds. Further analyses of the present data also show that subjects with TBI significantly decreased their speeds compared with unobstructed walking for both the narrow and wide obstacles (P=.004) but that control subjects slowed only for the wide obstacle. These observations show that subjects with TBI were affected by the presence of an obstacle regardless of its geometry.

With respect to clearance margins, the higher values adopted by subjects with TBI for the lead limb over both obstacles (and both ends of the wide obstacle) indicate increased caution within this group, as previously suggested.¹⁰ This increased safety strategy used by the TBI group is also supported by the slower gait speeds and shorter stride lengths. It has been suggested³² that safety concerns probably dominate over energy costs when stepping over obstacles. As discussed previously,¹⁰ the question remains as to the level of trainability of these patterns through physical therapy (eg, mobility training in different environments) and neuropsychologic interventions (eg, better planning, decreased anxiety). Certainly, there is a need to integrate and consider more complex activities and environments in gait training and assessment after TBI.⁴ Although very little work exists to guide us with respect to the exact environmental characteristics and training protocols to be targeted, a recent case report³³ suggests progressively challenging both motor and attention capacities of people with brain injury with respect to real-world activities.

The reason that both groups showed similar trail-limb clearance margins and that subjects with TBI did not produce the same exaggerated safety margin as for the lead limb is likely because the trail-limb trajectory is not visually controlled in the same way as for the lead limb during crossing.³⁴ When subjects with TBI relied less on visual information during trail-limb clearance, they performed similarly to the control group. This may suggest that subjects with TBI are better able to implement somatosensory information when limb trajectory control relies less on vision. However, there was a tendency for the TBI group to have higher clearances over the narrow obstacle and at the far end of the wide obstacle compared with control subjects, suggesting that increased caution cannot be completely ruled out. In fact, group differences became significant for the far end of the wide obstacle with the Stroop tasks because clearance heights were maintained for the TBI group but decreased for the control group. This result lends greater support to the theory of increased caution by the TBI group and suggests that these subjects have the capacity to use feed-forward information, previously suggested as important for obstacle clearance.³⁴

Division of Attention During Locomotion 

Increasing the difficulty of the cognitive task (from Stroop bar to Stroop word) always caused an increase in reading time but affected the locomotor performance more variably. Specifically, we observed that speed decreased from adding the Stroop bar to adding the Stroop word test for both groups for all conditions (except during unobstructed walking for the TBI group), whereas stride length and particularly clearance margin were more robust, with little change between the Stroop conditions. With respect to group differences for each dual-task condition, only the most complex environment with the combined wide obstacle and Stroop word task showed a significant slowing in the TBI group for speed and reading time and a decrease in stride length. These observations suggest that gait speed is more sensitive to dual-task complexity, and subjects with TBI who have good locomotor function show definite decrements in cognitive and locomotor performance in the more complex environmental contexts. This is important when considering outcome measures of both physical capacity and social reintegration.

The effects we observed here on both reading time and locomotor behavior (for both groups) are contrary to the results reported by Lajoie et al³⁵ for young adults. These researchers showed that only the cognitive task was affected when a reaction time task was performed during gait. This could be because we used a secondary task that necessitated greater attentional processing, as in daily visual activities, than a simple reaction time task. Also, although we placed several screens at strategic places to minimize visual scanning by the subjects, there was probably still some visual interference between looking at the screens and the obstacle. However, this is an activity commonly found in natural environments.

To date only 1 other study²⁵ has investigated a dual-task paradigm during walking (unobstructed only) in people with a brain injury—in this case, subjects with mild TBI (concussions). Similar results for both that and the present study with respect to shorter stride lengths suggest that regardless of the differences in head injury severity, there are similarities in behavior between people suffering from a concussion and those with good locomotor function after a moderate to severe TBI. One must be cautious about making too many inferences about similarities across severity levels at this point, however. The abstract cognitive tasks used by Parker et al²⁵ (spelling 5-letter words in reverse, subtraction by 7 or reciting months of the year in reverse order) probably used different cognitive abilities than our visual Stroop tasks. Future work on severity levels using the same dual-task paradigms will be important.

Study Limitations 

Some of the variability in the present results could be due to the smaller number of subjects recruited in the study. As can be seen in table 1, 1 subject had a GCS score of 13. Although this is generally considered mild, we included the subject based on cerebral lesions in 2 cortical sites clearly seen by radiologic tests that indicated a trauma of moderate severity according to the classification system in the Province of Quebec in Canada.²⁶ It has also been recently shown¹⁰ that the GCS score alone does not have a simple relation to locomotor ability.

Back to Article Outline

Conclusions 

Despite their good locomotor recovery, subjects with moderate to severe TBI showed residual deficits for walking within environments with physical obstructions and dual tasks. Given the similarity of such dual tasks to daily locomotor activities, they have the potential to be used in the assessment and rehabilitation of people with TBI.

Suppliers

Back to Article Outline

References 

1. Lezak MD , Howieson DB , Loring DW . In: Neuropsychological assessment . 4th ed.. New York: Oxford Univ Pr; 2004;
   • View In Article
2. Katz DI , White DK , Alexander MP , Klein RB . Recovery of ambulation after traumatic brain injury . Arch Phys Med Rehabil . 2004;85:865–869
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (188 KB)
   • CrossRef
3. Swaine BR , Sullivan SJ . Longitudinal profile of early motor recovery following severe traumatic brain injury . Brain Inj . 1996;10:347–366
   • View In Article
   • MEDLINE
   • CrossRef
4. Williams G , Robertson V , Greenwood K . Measuring high-level mobility after traumatic brain injury . Am J Phys Med Rehabil . 2004;83:910–920
   • View In Article
   • MEDLINE
   • CrossRef
5. Wade LD , Canning CG , Fowler V , Felmingham KL , Baguley IJ . Changes in postural sway and performance of functional tasks during rehabilitation after traumatic brain injury . Arch Phys Med Rehabil . 1997;78:1107–1111
   • View In Article
   • Abstract
   • Full-Text PDF (475 KB)
   • CrossRef
6. Lehmann JF , Boswell S , Price R , et al.   Quantitative evaluation of sway as an indicator of functional balance in post-traumatic brain injury . Arch Phys Med Rehabil . 1990;71:955–962
   • View In Article
   • MEDLINE
7. Wöber C , Oder W , Kollegger H , et al.   Posturographic measurement of body sway in survivors of severe closed head injury . Arch Phys Med Rehabil . 1993;74:1151–1156
   • View In Article
   • MEDLINE
8. Geurts AC , Ribbers GM , Knoop JA , van Limbeek J . Identification of static and dynamic postural instability following traumatic brain injury . Arch Phys Med Rehabil . 1996;77:639–644
   • View In Article
   • Abstract
   • Full-Text PDF (743 KB)
   • CrossRef
9. Ochi F , Esquenazi A , Hirai B , Talaty M . Temporal-spatial feature of gait after traumatic brain injury . J Head Trauma Rehabil . 1999;14(2):105–115
   • View In Article
   • MEDLINE
   • CrossRef
10. McFadyen BJ , Swaine B , Dumas D , Durand A . Residual effects of a traumatic brain injury on locomotor capacity . J Head Trauma Rehabil . 2003;18:512–525
    • View In Article
    • MEDLINE
    • CrossRef
11. Basford JR , Chou LS , Kaufman KR , et al.   An assessment of gait and balance deficits after traumatic brain injury . Arch Phys Med Rehabil . 2003;84:343–349
    • View In Article
    • Abstract
    • Full-Text PDF (132 KB)
    • CrossRef
12. McFadyen BJ , Winter DA . Anticipatory locomotor adjustments during obstructed human walking . Neurosci Res Commun . 1991;9:37–44
    • View In Article
13. Patla AE , Prentice SD , Robinson C , Neufeld J . Visual control of locomotion (strategies for changing direction and for going over obstacles) . J Exp Psychol Hum Percept Perform . 1991;17:603–634
    • View In Article
    • MEDLINE
    • CrossRef
14. Patla AE , Rietdyk S , Martin C , Prentice S . Locomotor patterns of the leading and the trailing limbs as solid and fragile obstacles are stepped over (some insights into the role of vision during locomotion) . J Mot Behav . 1996;28:35–47
    • View In Article
    • CrossRef
15. McFadyen BJ , Carnahan H . Anticipatory locomotor adjustments for accommodation versus avoiding level changes in humans . Exp Brain Res . 1997;114:500–506
    • View In Article
    • MEDLINE
    • CrossRef
16. Chou LS , Kaufman KR , Walker-Rabatin AE , Brey RH , Basford JR . Dynamic instability during obstacle crossing following TBI . Gait Posture . 2004;20:245–254
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (218 KB)
    • CrossRef
17. Stierwalt JA , Murray LL . Attention impairment following traumatic brain injury . Semin Speech Lang . 2002;23:129–138
    • View In Article
    • MEDLINE
    • CrossRef
18. Ponsford JL , Kinsella G . Attentional deficits following closed head injury . J Clin Exp Neuropsychol . 1992;14:822–838
    • View In Article
    • MEDLINE
    • CrossRef
19. Whyte J , Fleming M , Polansky M , Cavallucci C , Coslett HB . The effect of visual distraction following traumatic brain injury . J Int Neuropsychol Soc . 1998;4:127–136
    • View In Article
    • MEDLINE
    • CrossRef
20. Whyte J , Schuster K , Polansky M , Adams J , Coslett HB . Frequency and duration of inattentive behavior after traumatic brain injury (effects of distraction, task and practice) . J Int Neuropsychol Soc . 2000;6:1–11
    • View In Article
    • MEDLINE
21. Whyte J , Polansky M , Fleming M , Coslett HB , Cavallucci C . Sustained arousal and attention after traumatic brain injury . Neuropsychologia . 1995;33:797–813
    • View In Article
    • MEDLINE
    • CrossRef
22. Stuss DT , Stethem LL , Hugenholtz H , Picton T , Pivik J , Richard MT . Reaction time after head injury (fatigue, divided and focused attention, and consistency of performance) . J Neurol Neurosurg Psychiatry . 1989;52:742–748
    • View In Article
    • MEDLINE
    • CrossRef
23. Park NW , Moscovitch M , Robertson IH . Divided attention impairments after traumatic brain injury . Neuropsychologia . 1999;37:1119–1133
    • View In Article
    • MEDLINE
    • CrossRef
24. Brauer SG , Broom A , Stone C , Clewett S , Herzig P . Simplest tasks have greatest dual task interference with balance in brain injured adults . Hum Mov Sci . 2004;23:489–502
    • View In Article
    • MEDLINE
    • CrossRef
25. Parker TM , Osternig LR , Lee HJ , Donkelaar P , Chou LS . The effect of divided attention on gait stability following concussion . Clin Biomech (Bristol, Avon) . 2005;20:389–395
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (135 KB)
    • CrossRef
26. Gadoury M . In: Cadre de référence clinique pour l’élaboration de programmes de réadaptation pour la clientèle ayant subi un traumatisme cranio-cérébral (adulte) . 2nd ed.. Quebec City: Société de l’Assurance Automobile du Québec; 2001;
    • View In Article
27. Golden CJ . Identification of brain disorders by the Stroop Color and Word Test . J Clin Psychol . 1976;32:654–658
    • View In Article
    • MEDLINE
    • CrossRef
28. Stuss DT , Ely P , Hugenholtz H , et al.   Subtle neuropsychological deficits in patients with good recovery after closed head injury . Neurosurgery . 1985;17:41–47
    • View In Article
    • MEDLINE
    • CrossRef
29. Lajoie Y , Teasdale N , Bard C , Fleury M . Attentional demands for static and dynamic equilibrium . Exp Brain Res . 1993;97:139–144
    • View In Article
    • MEDLINE
30. Dault MC , Frank JS , Allard F . Influence of a visuo-spatial, verbal and central executive working memory task on postural control . Gait Posture . 2001;14:110–116
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (178 KB)
    • CrossRef
31. Schmitter-Edgecombe M . Effects of traumatic brain injury on cognitive performance (an attentional resource hypothesis in search of data) . J Head Trauma Rehabil . 1996;11(2):17–30
    • View In Article
    • CrossRef
32. Patla AE . Adaptability of human gait (implications for the control of locomotion) . Amsterdam: Elsevier Science; 1991;
    • View In Article
33. Tappen RS . Rehabilitation for balance and ambulation in a patient with attention impairment due to intracranial hemorrhage . Phys Ther . 2002;82:473–484
    • View In Article
    • MEDLINE
34. Mohagheghi AA , Moraes R , Patla AE . The effects of distant and on-line visual information on the control of approach phase and step over an obstacle during locomotion . Exp Brain Res . 2004;155:459–468
    • View In Article
    • MEDLINE
    • CrossRef
35. Lajoie Y , Teasdale N , Bard C , Fleury M . Upright standing and gait (are there changes in attentional requirements related to normal aging?) . Exp Aging Res . 1996;22:185–198
    • View In Article
    • MEDLINE
    • CrossRef

• a Model 3020; Northern Digital Inc, 103 Randall Dr, Waterloo, ON N2V 1C5, Canada.
• b MDR 6; Samson Technologies Corp, 575 Underhill Blvd, Syosset, NY 11791.
• c Compaq 1720; Hewlett-Packard USA, 20555 SH 249, Houston, TX 77070.
• d ST128PRO 8; StarTech.com, 45 Artisans Cres, London, ON N5V 5E9, Canada.