- Email: [email protected]
Patterns of Recovery of Posttraumatic Confusional State in Neurorehabilitation Admissions After Traumatic Brain Injury
1749
ORIGINAL ARTICLE
Patterns of Recovery of Posttraumatic Confusional State in Neurorehabilitation Admissions After Traumatic Brain Injury Mark Sherer, PhD, Stuart A. Yablon, MD, Risa Nakase-Richardson, PhD ABSTRACT. Sherer M, Yablon SA, Nakase-Richardson R. Patterns of recovery of posttraumatic confusional state in neurorehabilitation admissions after traumatic brain injury. Arch Phys Med Rehabil 2009;90:1749-54. Objective: To provide preliminary descriptions of patterns of resolution of symptoms of acute confusion after traumatic brain injury (TBI). Design: Prospective, descriptive, cohort study. Setting: Inpatient neurorehabilitation unit. Participants: Patients (N⫽107) meeting criteria for posttraumatic confusional state at admission to inpatient rehabilitation. Interventions: Not applicable. Main Outcome Measure: Patterns of resolution of posttraumatic confusional state symptoms over the first 3 confusion assessment protocol evaluations for patients with mild, moderate, and severe confusion. Results: Posttraumatic confusional state symptoms resolving earliest were psychotic-type symptoms, decreased daytime arousal, and nighttime sleep disturbance. Fluctuation and cognitive impairment were the 2 most persistent symptoms. Seventy-three percent of patients showed improvement of 1 or more symptoms from the first to third evaluation. Confusion severity groups did not significantly differ on indices of injury severity (Glasgow Coma Scale score, time to follow commands) but did differ on functional status at discharge from inpatient rehabilitation. Conclusions: While posttraumatic confusional state is a heterogeneous disorder, there is a predictable pattern of symptom resolution. Differences in patients’ confusion severity and patterns of symptoms may relate to differing underlying neural injury. Key Words: Brain injuries; Confusion; Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine ARLY RECOVERY FOR persons with significant TBI is E characterized by a period of disorientation and impaired cognitive abilities with other neurobehavioral disturbances.
1,2
This phase of recovery is most commonly called PTA, but more recently has been called the posttraumatic confusional state.1 We prefer the term posttraumatic confusional state to
From the TIRR Memorial Hermann, Houston, TX (Sherer); Baylor College of Medicine, Houston, TX (Sherer); Methodist Rehabilitation Center, Jackson, MS (Yablon); and James A. Haley Veterans Affairs Medical Center, Tampa, FL (NakaseRichardson). Supported by the U.S. Department of Education National Institute on Disability and Rehabilitation Research (grant nos. H133A020514 and H133A070043). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Mark Sherer, PhD, ABPP-Cn, TIRR Memorial Hermann, Dept of Research, 1333 Moursund, Houston, TX 77030, e-mail: Mark.Sherer@ memorialhermann.org. 0003-9993/09/9010-00113$36.00/0 doi:10.1016/j.apmr.2009.05.011
PTA because it more accurately indicates the clinical presentation of early recovery from TBI as a confusional state as opposed to simply an amnestic state. Confused patients pose significant problems for clinical management. Such patients cause increased risk for injury to themselves and others, have poor compliance with treatment, and are emotionally upsetting to family/significant others. Duration of posttraumatic confusional state has been used as an index of injury severity and is predictive of outcome.3-5 We recently showed that severity of confusion at a set time postinjury is also predictive of outcome and that, in preliminary analyses, the presence of particular symptoms of confusion has specific implications for late outcome.6 These findings indicate that there is importance in understanding the process of symptomatic recovery from confusion and the associations of various symptoms of confusion with each other. Ideally, improved understanding of this clinical state would have implications for clinical management. Previous investigations provide a limited description of specific aspects of early recovery for confused patients after TBI,1,7-9 primarily focusing on the resolution of disorientation and cognitive symptoms.7,8 Orientation recovers in a typical pattern with orientation to person recovering first, then place, and finally, time.7 Ability to perform simple attentional tasks improves before vigilance, orientation, and performance of complex attentional tasks with a working memory component. Ability to form and retain new episodic memory recovers later.1 Posttraumatic agitation generally begins resolving after general cognitive functions have started to improve, but before full orientation returns.8 The present investigation provides a novel preliminary description of early recovery from TBI by examining a broad range of symptoms and by using an instrument with demonstrated validity for assessment of confusion after TBI. Our previous work identified 7 key symptoms of confusion in early recovery from TBI.2 These are disorientation, cognitive impairment, fluctuation in presentation, agitation, nighttime sleep disturbance, decreased daytime arousal, and psychotic-type symptoms. We have demonstrated the prognostic importance of presence and severity of confusion and of presence of specific symptoms of confusion for early functional status and late productivity outcome for persons with TBI.2,6 The primary goal of this investigation was to provide an initial description of the patterns of resolution of these symptoms in persons in early recovery from TBI.
List of Abbreviations CAP DRS GCS LOS PTA TBI TBIMS
Confusion Assessment Protocol Disability Rating Scale Glasgow Coma Scale length of stay posttraumatic amnesia traumatic brain injury Traumatic Brain Injury Model Systems
Arch Phys Med Rehabil Vol 90, October 2009
1750
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
METHODS Study Population The study population for this investigation has been previously described.6 Participants were all qualified persons with TBI admitted to a rehabilitation hospital from April 2003 through February 2006. Data collection was approved by the relevant institutional review board. Participants were recruited as part of the National Institute for Rehabilitation and Research TBIMS research program. Inclusion criteria were medically documented TBI, admission to the affiliated level 1 trauma center within 24 hours of injury, receipt of inpatient rehabilitation at the TBIMS-affiliated rehabilitation unit, age 16 years or older at time of injury, recovery to a responsive state prior to discharge from inpatient rehabilitation, and provision of informed consent by the person with TBI or appropriate proxy. Data Collection Demographic (age, years of education, sex) and injury characteristics (GCS score at admission to the emergency department, time in days from injury to recovery of ability to follow commands, acute care LOS [in days], rehabilitation LOS [in days]) were abstracted from medical records or obtained in interview with family or significant others of the person with TBI. Symptoms of confusion were assessed using the CAP. The CAP was generally administered on the day after admission to rehabilitation and approximately every 2 days thereafter. Functional status at rehabilitation hospital admission and discharge was rated using the DRS. Measures The CAP2,6 was developed to provide diagnostic criteria for posttraumatic confusional state. This measure provides a structured and repeatable method for measuring and tracking 7 symptoms of posttraumatic confusional state. Serial administration of the CAP provides information on the presence, severity, duration, and patterns of symptoms of posttraumatic confusional state in persons in early recovery from TBI. Four symptoms (fluctuation, nighttime sleep disturbance, daytime decreased arousal, psychotic-type symptoms) are measured with examiner-rated items. Fluctuation refers to change in clinical presentation over a period of minutes or hours. As an example, a patient who cooperates with therapy at the beginning of a session but stops cooperating and declines to respond at the end of the session is rated as showing fluctuation. Fluctuation is assessed through direct examination of the patient and interviews with therapy and nursing personnel. Sleep disturbance is indicated by frequent nighttime awakenings and/or extended periods of no sleep at night. Sleep disturbance is assessed through review of nursing notes, examination of sleep charts, and interview with family members who stay in patient rooms at night. Decreased daytime arousal refers to decreased ability to maintain a wakeful state during the daytime, particularly during therapy activities. Decreased daytime arousal is assessed through direct examination of the patient and interviews with therapy and nursing personnel. Psychotictype symptoms refer to frank delusions or hallucinations or misperceptions of reality that are similar to delusions or hallucinations such as unusual thoughts or visual illusions. Psychotic-type symptoms are assessed through direct examination of the patients and interviews with family members, therapists, and nurses. Orientation is measured with the Galveston Orientation and Amnesia Test.10 The standard cut-off of 75 or less is used to determine presence of disorientation. Cognitive impairment is Arch Phys Med Rehabil Vol 90, October 2009
a composite score based on 7 attentional tasks including simple attention (eg, counting to 20), vigilance (eg, indicating when a target letter is spoken in a series of letters), working memory (eg, reciting the months of the year in reverse order), visual recognition memory (eg, identifying target previously viewed pictures from foils), and auditory comprehension (eg, indicating yes or no for the accuracy of sentences). The items are scored using criteria developed for the CAP. Scores can range from 0 to 28, and scores of 18 or lower indicate substantial cognitive impairment. Agitation is measured with the Agitated Behavior Scale.11 Based on analyses completed in CAP development studies, scores greater than 17 indicate significant restlessness. Criteria for each of the 7 symptoms are applied to determine presence or absence of each symptom for each examination. Accordingly, CAP scores range from 0 (no symptoms of confusion present) to 7 (all 7 symptoms of confusion present). Persons with 4 of 7 symptoms present, or 3 symptoms if 1 is disorientation, are classified as in posttraumatic confusional state. Note that many patients who do not meet criteria for posttraumatic confusional state are not normal because they have a variety of cognitive impairments seen in early recovery from TBI such as memory impairment, distractibility, or slowed cognitive processing speed. The version of the CAP used in the present study6 had a minor modification from the original version.2 In the original version, patients could be scored as having psychotic-type symptoms if they showed delusions, hallucinations, or symptoms suggestive of possible delusions or hallucinations or if they showed evidence of thought process abnormalities. This latter criterion was eliminated in the present investigation because of concern that thought process abnormalities could not be reliably and meaningfully distinguished from the highly distractible, disjointed conversation expected in acutely confused patients. The DRS12 is a 30-point scale with 8 areas of functioning rated: eye opening; verbalization; motor response; level of cognitive ability for daily activities of feeding, toileting, and grooming; overall level of dependence; and employability. Each area of functioning is rated on a scale of 0 to 3, 0 to 4, or 0 to 5, with a higher score representing a lower level of functioning. Scores on each item are summed to yield a total score between 0 and 29, with a higher score indicating greater disability. Analysis Descriptive data (age, years of education, sex, GCS, time to follow commands, acute care LOS, rehabilitation LOS, interval from injury to first CAP evaluation, DRS at admit and DRS at discharge, and DRS discharge) are presented as quartiles for continuous variables and percentages for categorical variables. Participants were classified as in posttraumatic confusional state or not in posttraumatic confusional state based on the initial CAP after admission to rehabilitation. Patients in posttraumatic confusional state were compared with patients not in posttraumatic confusional state on descriptive measures using Mann-Whitney U tests, and all subsequent analyses described only those in posttraumatic confusional state. Confused participants with 3 or 4 symptoms on the initial CAP were classified as mildly confused, those with 5 symptoms as moderately confused, and those with 6 or 7 symptoms as severely confused. Mildly, moderately, and severely confused patients were compared on the descriptive variables using KruskalWallis tests. In order to describe the process of symptom resolution in patients over time, the first 3 CAP evaluations for each confused patient were analyzed. Patients for whom there were fewer than 3 completed CAP evaluations were excluded from
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
analyses requiring these data. Extension of this analysis to the fourth CAP would have required exclusion of a large number of additional patients. Data recorded and presented in tables and figures included number of symptoms improved between evaluations, number of symptoms worsened between evaluations, number of patients showing no change between evaluations, symptoms most likely to improve, and symptoms most likely to worsen for each of the 3 confusion severity groups. RESULTS Comparison of Nonconfused and Confused Patients During the study period, 195 potential participants were admitted to the study site. Of these, 16 did not recover to a responsive state (ie, remained in vegetative or minimally conscious state) and 3 had clinical presentations that were substantially influenced by non-TBI disorders (spinal cord injury, dementia, deafness). An additional 5 potential participants declined to give consent, and 3 participants who were enrolled were inadvertently missed for collection of key study measures. The resulting study sample was comprised of 168 participants. Table 1 shows the comparison of these 168 patients classified as not in posttraumatic confusional state (n⫽61) and in posttraumatic confusional state (n⫽107) based on the initial CAP after rehabilitation admission. Those in posttraumatic confusional state had lower preinjury years of education, had lower initial GCS, took longer to follow commands, had longer acute and rehabilitation LOS, had a longer interval from injury to initial CAP, and had poorer functional status, as measured with the DRS, at rehabilitation admission and discharge. All subsequent analyses pertain only to patients in posttraumatic confusional state.
Table 1: Comparison of Nonconfused and Confused Patients as Determined by the First CAP Evaluation After Admission to Inpatient Rehabilitation (nⴝ168)
Continuous descriptors Age (y) Education* (y) GCS score* TFC‡ (d) Acute LOS† (d) Rehabilitation LOS‡ (d) CAP days† Admit DRS‡ Discharge DRS‡ Categorical descriptors Sex Men Women GCS Severe (3–8) Moderate (9–12) Mild (13–15)
Nonconfused (n⫽61) Median (25th, 75th Percentiles)
Confused (N⫽107) Median (25th, 75th Percentiles)
24.4 (19.7, 42.9) 12.0 (10.0, 14.0) 9.0 (5.5, 14.0) 2.0 (0.5, 5.0) 14.0 (10.0, 25.5) 14.0 (10.0, 18.0) 17.0 (12.0, 26.5) 8.0 (6.0, 10.0) 5.0 (4.0, 7.0)
28.6 (20.2, 49.7) 11.0 (10.0, 12.0) 7.0 (5.0, 10.0) 6.0 (2.0, 16.0) 20.0 (15.0, 28.0) 16.0 (14.0, 24.0) 21.0 (16.0, 30.0) 14.0 (11.0, 17.0) 7.5 (6.0, 11.0)
n (%)
n (%)
38 (62) 23 (38)
75 (70) 30 (30)
27 (44) 12 (20) 22 (36)
69 (64) 23 (21) 15 (14)
NOTE. Two confused participants had missing values for to follow commands with TFC, and 1 confused patient was missing values for both admit and discharge DRS. Abbreviations: CAP days, interval from injury to initial CAP evaluation; TFC, time to follow commands. *P⬍.05; †P⬍.01; ‡P⬍.001.
1751
Comparison of Patients With Mild, Moderate, and Severe Confusion Table 2 shows the comparison of patients with mild, moderate, and severe posttraumatic confusional state as determined by the initial CAP after rehabilitation admission. Other than number of symptoms of confusion (CAP total), confusion severity groups differed only on DRS at rehabilitation discharge, with those with more severe confusion showing poorer functional status. Resolution of Symptoms of Confusion Table 3 presents data on the resolution of symptoms of posttraumatic confusional state for the 3 injury severity groups. These patterns of symptoms resolution are shown graphically in figure 1. Three participants (2 mildly, 1 moderately confused) had only 1 CAP evaluation, and 7 additional participants (4 mildly, 2 moderately, 1 severely confused) had only 2 CAP evaluations. This affected the numbers included for changes from CAP 1 to CAP 2 (104/107 [97%] patients were reported on) and from CAP 2 to CAP 3 (97/107 [93%] patients were reported on). The interval (mean ⫾ SD) from the first to the second CAP evaluation was 1.8⫾1.3 days, and the interval (mean ⫾ SD) from the second to the third CAP evaluation was 2.5⫾1.6 days. Across confusion severity groups, from the first to second CAP, 49% to 63% of participants showed improvement as indicated by a decreased number of symptoms of confusion. Patients with initial moderate and severe confusion continued to improve at about the same rates, 50% to 56%, respectively, between the second and third CAPS. Only 25% of those with initial mild confusion showed continued improvement, however, indicating a possible floor effect, because these patients had only a few remaining symptoms of confusion. Worsening confusion, as indicated by an increased number of symptoms, was rare, ranging from as few as 5% of those with initial severe confusion showing worsening between the first and second CAPS to 14% of those with initial mild confusion showing worsening between the first and second evaluations. Proportions of patients who stayed the same (continued with the same number of symptoms between evaluations) from CAP 1 to 2 and from CAP 2 to 3 ranged from 26% to 69% across confusion severity groups. Sixty-nine percent of patients with mild confusion showed no change in the number of symptoms between the second and third CAPs. Almost half (46%) of those with severe confusion showed no change between the first and second CAPs. For all patients in posttraumatic confusional state, from the first to the third CAP, 73% improved, while only 27% remained the same or worsened. Early recovery for severely confused patients was marked by resolution of disturbances of arousal (nighttime sleep disturbance and decreased daytime arousal) and by resolution of psychotic-type symptoms. Although initially present in a smaller proportion of patients, these symptoms were also likely to resolve early for those with moderate confusion. Psychotictype symptoms and nighttime sleep disturbance fluctuated in patients with moderate and severe confusion, because these symptoms were among the most likely to improve but were also the most likely to worsen. Disorientation and agitation tended to resolve next, with some patients also showing resolution of cognitive impairment and fluctuation. Cognitive impairment and fluctuation were the most persistent symptoms for all confused patients, although a significant number also remained with disorientation by the third CAP assessment. Arch Phys Med Rehabil Vol 90, October 2009
1752
POSTTRAUMATIC CONFUSIONAL STATE, Sherer Table 2: Comparison of Confusion Severity Groups as Determined by the First CAP Evaluation After Admission to Inpatient Rehabilitation (Nⴝ107)
Continuous descriptors Age (y) Education (y) GCS score TFC (d) Acute LOS (d) Rehabilitation LOS (d) CAP days CAP total* Admit DRS Discharge DRS† Categorical descriptors Sex, n (%) Male Female GCS, n (%) Severe (3–8) Moderate (9–12) Mild (13–15)
Mildly Confused (n⫽38) Median (25th, 75th Percentiles)
Moderately Confused (n⫽28) Median (25th, 75th Percentiles)
Severely Confused (41) Median (25th, 75th Percentiles)
27.0 (21.4, 53.2) 11.5 (10.0, 12.0) 7.0 (5.8, 10.2) 4.5 (1.0, 12.0) 19.0 (14.8, 25.8) 16.0 (14.0, 23.2) 20.5 (16.8, 29.2) 4.0 (3.0, 4.0) 12.5 (9.8, 15.2) 7.0 (5.0, 8.0)
28.1 (18.5, 48.2) 11.5 (10.0, 12.0) 7.0 (4.2, 10.8) 6.5 (3.0, 18.5) 20.0 (15.2, 30.2) 21.0 (5.0, 29.8) 21.0 (16.2, 31.0) 5.0 (5.0, 5.0) 14.0 (12.0, 17.0) 7.5 (6.0, 10.0)
30.4 (20.6, 50.8) 11.0 (10.0, 12.0) 7.0 (4.0, 10.0) 8.0 (2.0, 16.5) 23.0 (15.0, 29.5) 16.0 (13.5, 23.0) 24.0 (16.0, 30.5) 7.0 (6.0, 7.0) 14.5 (12.0, 19.0) 8.5 (6.2, 13.0)
31 (82) 7 (18)
19 (68) 9 (32)
25 (61) 16 (39)
23 (60) 10 (26) 5 (13)
19 (68) 5 (18) 4 (14)
27 (66) 8 (19) 6 (15)
Abbreviations: CAP days, the interval from injury to the initial CAP evaluation; CAP total, the number of symptoms of confusion found to be present on the CAP evaluation; TFC, time to follow commands. *P⬍.001; †P⬍.01.
DISCUSSION The present preliminary investigation produced novel findings regarding the symptomatic course of recovery for confused patients early after TBI. As expected, confusion has a recovering course, with almost three quarters of patients showing resolution of some symptoms over the approximately 4-day period between the first and third evaluations. Disturbances of arousal and psychotic-type symptoms resolved first. Because most participants had severe TBI and, thus, some period of coma, resolution of disturbances of arousal may represent a continuation of recovery indicated by emergence from the minimally conscious or vegetative states. This recovery may be supported by improving function of neural pathways projecting from the brain stem to the frontal lobes. The temporal association of resolution of psychotic-type symptoms may suggest overlap in the neural substrate for disturbed arousal and misperceptions of the external world. The temporal association between resolution of disorientation and agitation, a finding observed previously,8 suggests that both are associated with profound disturbances of attentional processes producing disorganized behavior and inability to attend to even basic elements of the immediate environment. This interpretation is consistent with the finding that cognitive impairments generally resolved after disorientation and agitation. The cognitive measures used in the CAP include tasks requiring vigilance and working memory that require more intact attentional abilities than simple orientation. Current findings are consistent with elements of the conceptualization by Stuss et al1 of the confused state as a global disturbance of attentional processes. Finally, we found that fluctuation in presentation is the most persistent symptom of confusion in early recovery from TBI. In this investigation, fluctuation was manifested as change in the adequacy of behavioral performances from one time of day to another. From clinical observation, these variations seemed to be a result of changes in level of arousal or alterations in the Arch Phys Med Rehabil Vol 90, October 2009
degree of distractibility and restlessness. Return of the ability of the brain to produce consistent behavior in the face of fatigue or other situational stresses indicates a later stage of recovery from TBI. As expected, confused patients differed from nonconfused patients on a number of factors. Confused patients had more severe injuries as indicated by initial level of responsiveness (as measured by the GCS), time to follow commands, and hospital LOS. It was surprising that these factors did not differ among the 3 confusion severity groups. These groups did not differ on initial GCS, time to follow commands, or hospital LOS, although more severely confused patients did have poorer functional status at rehabilitation discharge. While visual inspection of median times to follow commands and acute hospital LOSs for the 3 groups suggests that longer intervals were associated with greater confusion, this difference was not statistically significant because of substantial within-group variability causing substantial overlap between groups for this interval. This finding that greater confusion did not have a direct relationship to accepted indices of greater injury severity is provocative. It may suggest that there are differences in the underlying neuropathology among the confusion groups. In this speculative scenario, patients in each group had sufficient TBI severity to result in a general degree of neuropathology sufficient to cause confusion. However, the specific brain areas and pathways affected were sufficiently variable between patient groups to result in different patterns of symptoms and differential associated courses and durations13 of recovery. Study Limitations The present investigation has a number of limitations. First, participants were studied only after admission to inpatient rehabilitation. Persons with TBI not coming to rehabilitation were not studied, and the implications of these findings for these patients are unknown. Also, very early recovery of con-
POSTTRAUMATIC CONFUSIONAL STATE, Sherer Table 3: Evolution of Symptoms of Confusion Over the First 3 CAP Assessments After Admission to Inpatient Rehabilitation (nⴝ97) Patients With Severe Confusion on Initial CAP (n⫽41)
No. improved (%) By 1 symptom By 2 symptoms By 3 or more symptoms No. worse (%) By 1 symptom No. same (%) Exactly Symptoms most likely to improve (n) Symptom most likely to worsen (n)
1st CAP to 2nd CAP (n⫽41)
2nd CAP to 3rd CAP (n⫽40)
20 (49) 11 (27) 6 (15) 3 (7) 2 (5) 2 (5) 19 (46) 18 (44) Arouse (11), Psych (11), Sleep (9) Psych (4)
20 (50) 9 (22) 8 (20) 3 (8) 5 (12) 5 (12) 15 (38) 15 (38) Sleep (8), Arouse (8), Psych (6) Sleep (3)
1753
fusion occurring at the trauma center was not studied, and no statements can be made about the course of recovery prior to rehabilitation admission. A longer epoch of evaluation with a greater number of assessments might have been interesting, but would have introduced an even greater injury severity bias, excluding patients with more moderate TBI. We were not able to report on change over 3 evaluations in all patients because only 93% of patients had the required evaluations. Aspects of confusion studied and the metrics of measurement were those reflected in the CAP. There may be other important symptoms or finer/composite measurement instruments incorporating symptom severity that would have provided additional findings of interest.
Patients With Moderate Confusion on Initial CAP (n⫽28)
No. improved (%) By 1 symptom By 2 symptoms By 3 or more symptoms No. worse (%) By 1 symptom By 2 symptoms No. same (%) Exactly Symptoms most likely to improve (n) Symptom most likely to worsen (n)
1st CAP to 2nd CAP (n⫽27)
2nd CAP to 3rd CAP (n⫽25)
17 (63) 9 (33) 5 (19) 3 (11)
14 (56) 8 (32) 4 (16) 2 (8)
3 (11) 2 (7) 1 (4) 7 (26) 5 (19) Agit (7), Sleep (6), Cog (5), Psych (5) Psych (3)
2 (8) 2 (8) — 9 (36) 7 (28) Disorient (6), Sleep (6), Fluct (4), Arouse (4) Sleep (3)
Patients With Mild Confusion on Initial CAP (n⫽38) 1st CAP to 2nd CAP (n⫽36)
2nd CAP to 3rd CAP (n⫽32)
No. improved (%) 18 (50) 8 (25) By 1 symptom 11 (31) 7 (22) By 2 symptoms 7 (19) — By 3 or more symptoms — 1 (3) No. worse (%) 5 (14) 2 (6) By 1 symptom 5 (14) 2 (6) No. same (%) 13 (26) 22 (69) Exactly 10 (28) 22 (69) Symptoms most likely to Disorient (9), Cog (2), Agit (2), Fluct (2), Arouse (2), improve (n) Agit (5), Sleep (2) Sleep (4) Symptoms most likely to Arouse (3) Disorient (1), Agit (1) worsen (n) NOTE. The terms exactly, improved, same, and worse are defined as follows: exactly, the patient had the exact same pattern of symptoms on the subsequent CAP evaluation; improved, the patient had fewer symptoms on the subsequent CAP evaluation; same, the patient had the same number of symptoms on the subsequent CAP evaluation; worse, the patient had more symptoms on the subsequent CAP evaluation. Abbreviations: Agit, agitation; Arouse, decreased daytime arousal; Cog, cognitive impairment; Disorient, disorientation; Fluct, fluctuation; Psych, psychotic-type symptoms; Sleep, nighttime sleep disturbance.
Fig 1. Percentages of patients with TBI with (A) severe, (B) moderate, and (C) mild confusion showing each of 7 symptoms over the first 3 CAP evaluations after admission to rehabilitation (nⴝ41, nⴝ28, nⴝ38, respectively). Abbreviations: Agit, agitation; Arouse, decreased daytime arousal; Cog, cognitive impairment; Disorient, disorientation; Fluct, fluctuation; Psych, psychotic-type symptoms; Sleep, nighttime sleep disturbance.
Arch Phys Med Rehabil Vol 90, October 2009
1754
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
CONCLUSIONS While heterogeneous, posttraumatic confusion is not manifested in a random mishmash of cognitive and behavioral manifestations with relevance limited to duration of symptoms and associated injury severity. Early recovery from post-TBI confusion has a somewhat predictable course that likely reflects the underlying neural substrate of the injury. Improved understanding of the patterns of symptoms of confusion and their causes may lead to improved early treatment for patients with TBI. As an example, the apparent association between arousal disturbances and psychotic-type symptoms may motivate clinicians to treat nighttime sleep disturbance and decreased daytime arousal aggressively in hopes of also addressing psychotictype symptoms. However, we are unaware of any data at this time that would confirm that such aggressive treatment of arousal disturbances would decrease the incidence or shorten the duration of psychotic-type symptoms. References 1. Stuss DT, Binns MA, Carruth FG, et al. The acute period of recovery from traumatic brain injury: posttraumatic amnesia or posttraumatic confusional state? J Neurosurg 1999;90:635-43. 2. Sherer M, Nakase-Thompson R, Yablon SA, Gontkovsky ST. Multidimensional assessment of acute confusion after traumatic brain injury. Arch Phys Med Rehabil 2005;86:896-904. 3. Ellenberg JH, Levin HS, Saydjari C. Posttraumatic amnesia as a predictor of outcome after severe closed head injury. Arch Neurol 1996;53:792-1. 4. Sherer M, Sander A, Nick TG, High WM Jr, Malec JF, Rosenthal M. Early cognitive status and productivity outcome after traumatic brain injury: findings from the TBI model systems. Arch Phys Med Rehabil 2002;83:183-92.
Arch Phys Med Rehabil Vol 90, October 2009
5. Nakase-Richardson R, Yablon SA, Sherer M. Prospective comparison of acute confusion severity with duration of post-traumatic amnesia in predicting employment outcome after traumatic brain injury. J Neurol Neurosurg Psychiatr 2007;78:872-6. 6. Sherer M, Yablon SA, Nakase-Richardson R, Nick T. Effect of severity of posttraumatic confusion and its constituent symptoms on outcome after traumatic brain injury. Arch Phys Med Rehabil 2008;89:42-7. 7. High WM Jr, Levin HS, Gary HE. Recovery of orientation following closed-head injury. J Clin Exp Neuropsychol 1990; 12:703-14. 8. Corrigan JD, Mysiw WJ. Agitation following traumatic head injury: equivocal evidence for a discrete stage of cognitive recovery. Arch Phys Med Rehabil 1988;69:487-92. 9. Weir N, Doig EJ, Fleming FM, Wiemers A, Zemljic C. Objective and behavioral assessment of the emergence from post-traumatic amnesia (PTA). Brain Inj 2006;20:927-35. 10. Levin HS, O’Donnell VM, Grossman RG. The Galveston Orientation and Amnesia Test: a practical scale to assess cognition after head injury. J Nerv Ment Dis 1979;167:675-84. 11. Corrigan JD. Development of a scale for assessment of agitation following traumatic brain injury. J Clin Exp Neuropsychol 1989; 11:261-77. 12. Rappaport M, Hall KM, Hopkins K, Belleza T, Cope DN. Disability rating scale for severe head trauma: coma to community. Arch Phys Med Rehabil 1982;63:118-23. 13. Nakase-Richardson R, Sherer M, Yablon SA, Evans CC. Initial neuroimaging is predictive of confusion at one-month post traumatic brain injury. J Int Neuropsychol Soc 2006;12(Suppl 1):215. Available at: http://journals.cambridge.org/action/displayJournal?jid⫽INS#. Accessed September 15, 2009.
ORIGINAL ARTICLE
Patterns of Recovery of Posttraumatic Confusional State in Neurorehabilitation Admissions After Traumatic Brain Injury Mark Sherer, PhD, Stuart A. Yablon, MD, Risa Nakase-Richardson, PhD ABSTRACT. Sherer M, Yablon SA, Nakase-Richardson R. Patterns of recovery of posttraumatic confusional state in neurorehabilitation admissions after traumatic brain injury. Arch Phys Med Rehabil 2009;90:1749-54. Objective: To provide preliminary descriptions of patterns of resolution of symptoms of acute confusion after traumatic brain injury (TBI). Design: Prospective, descriptive, cohort study. Setting: Inpatient neurorehabilitation unit. Participants: Patients (N⫽107) meeting criteria for posttraumatic confusional state at admission to inpatient rehabilitation. Interventions: Not applicable. Main Outcome Measure: Patterns of resolution of posttraumatic confusional state symptoms over the first 3 confusion assessment protocol evaluations for patients with mild, moderate, and severe confusion. Results: Posttraumatic confusional state symptoms resolving earliest were psychotic-type symptoms, decreased daytime arousal, and nighttime sleep disturbance. Fluctuation and cognitive impairment were the 2 most persistent symptoms. Seventy-three percent of patients showed improvement of 1 or more symptoms from the first to third evaluation. Confusion severity groups did not significantly differ on indices of injury severity (Glasgow Coma Scale score, time to follow commands) but did differ on functional status at discharge from inpatient rehabilitation. Conclusions: While posttraumatic confusional state is a heterogeneous disorder, there is a predictable pattern of symptom resolution. Differences in patients’ confusion severity and patterns of symptoms may relate to differing underlying neural injury. Key Words: Brain injuries; Confusion; Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine ARLY RECOVERY FOR persons with significant TBI is E characterized by a period of disorientation and impaired cognitive abilities with other neurobehavioral disturbances.
1,2
This phase of recovery is most commonly called PTA, but more recently has been called the posttraumatic confusional state.1 We prefer the term posttraumatic confusional state to
From the TIRR Memorial Hermann, Houston, TX (Sherer); Baylor College of Medicine, Houston, TX (Sherer); Methodist Rehabilitation Center, Jackson, MS (Yablon); and James A. Haley Veterans Affairs Medical Center, Tampa, FL (NakaseRichardson). Supported by the U.S. Department of Education National Institute on Disability and Rehabilitation Research (grant nos. H133A020514 and H133A070043). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Mark Sherer, PhD, ABPP-Cn, TIRR Memorial Hermann, Dept of Research, 1333 Moursund, Houston, TX 77030, e-mail: Mark.Sherer@ memorialhermann.org. 0003-9993/09/9010-00113$36.00/0 doi:10.1016/j.apmr.2009.05.011
PTA because it more accurately indicates the clinical presentation of early recovery from TBI as a confusional state as opposed to simply an amnestic state. Confused patients pose significant problems for clinical management. Such patients cause increased risk for injury to themselves and others, have poor compliance with treatment, and are emotionally upsetting to family/significant others. Duration of posttraumatic confusional state has been used as an index of injury severity and is predictive of outcome.3-5 We recently showed that severity of confusion at a set time postinjury is also predictive of outcome and that, in preliminary analyses, the presence of particular symptoms of confusion has specific implications for late outcome.6 These findings indicate that there is importance in understanding the process of symptomatic recovery from confusion and the associations of various symptoms of confusion with each other. Ideally, improved understanding of this clinical state would have implications for clinical management. Previous investigations provide a limited description of specific aspects of early recovery for confused patients after TBI,1,7-9 primarily focusing on the resolution of disorientation and cognitive symptoms.7,8 Orientation recovers in a typical pattern with orientation to person recovering first, then place, and finally, time.7 Ability to perform simple attentional tasks improves before vigilance, orientation, and performance of complex attentional tasks with a working memory component. Ability to form and retain new episodic memory recovers later.1 Posttraumatic agitation generally begins resolving after general cognitive functions have started to improve, but before full orientation returns.8 The present investigation provides a novel preliminary description of early recovery from TBI by examining a broad range of symptoms and by using an instrument with demonstrated validity for assessment of confusion after TBI. Our previous work identified 7 key symptoms of confusion in early recovery from TBI.2 These are disorientation, cognitive impairment, fluctuation in presentation, agitation, nighttime sleep disturbance, decreased daytime arousal, and psychotic-type symptoms. We have demonstrated the prognostic importance of presence and severity of confusion and of presence of specific symptoms of confusion for early functional status and late productivity outcome for persons with TBI.2,6 The primary goal of this investigation was to provide an initial description of the patterns of resolution of these symptoms in persons in early recovery from TBI.
List of Abbreviations CAP DRS GCS LOS PTA TBI TBIMS
Confusion Assessment Protocol Disability Rating Scale Glasgow Coma Scale length of stay posttraumatic amnesia traumatic brain injury Traumatic Brain Injury Model Systems
Arch Phys Med Rehabil Vol 90, October 2009
1750
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
METHODS Study Population The study population for this investigation has been previously described.6 Participants were all qualified persons with TBI admitted to a rehabilitation hospital from April 2003 through February 2006. Data collection was approved by the relevant institutional review board. Participants were recruited as part of the National Institute for Rehabilitation and Research TBIMS research program. Inclusion criteria were medically documented TBI, admission to the affiliated level 1 trauma center within 24 hours of injury, receipt of inpatient rehabilitation at the TBIMS-affiliated rehabilitation unit, age 16 years or older at time of injury, recovery to a responsive state prior to discharge from inpatient rehabilitation, and provision of informed consent by the person with TBI or appropriate proxy. Data Collection Demographic (age, years of education, sex) and injury characteristics (GCS score at admission to the emergency department, time in days from injury to recovery of ability to follow commands, acute care LOS [in days], rehabilitation LOS [in days]) were abstracted from medical records or obtained in interview with family or significant others of the person with TBI. Symptoms of confusion were assessed using the CAP. The CAP was generally administered on the day after admission to rehabilitation and approximately every 2 days thereafter. Functional status at rehabilitation hospital admission and discharge was rated using the DRS. Measures The CAP2,6 was developed to provide diagnostic criteria for posttraumatic confusional state. This measure provides a structured and repeatable method for measuring and tracking 7 symptoms of posttraumatic confusional state. Serial administration of the CAP provides information on the presence, severity, duration, and patterns of symptoms of posttraumatic confusional state in persons in early recovery from TBI. Four symptoms (fluctuation, nighttime sleep disturbance, daytime decreased arousal, psychotic-type symptoms) are measured with examiner-rated items. Fluctuation refers to change in clinical presentation over a period of minutes or hours. As an example, a patient who cooperates with therapy at the beginning of a session but stops cooperating and declines to respond at the end of the session is rated as showing fluctuation. Fluctuation is assessed through direct examination of the patient and interviews with therapy and nursing personnel. Sleep disturbance is indicated by frequent nighttime awakenings and/or extended periods of no sleep at night. Sleep disturbance is assessed through review of nursing notes, examination of sleep charts, and interview with family members who stay in patient rooms at night. Decreased daytime arousal refers to decreased ability to maintain a wakeful state during the daytime, particularly during therapy activities. Decreased daytime arousal is assessed through direct examination of the patient and interviews with therapy and nursing personnel. Psychotictype symptoms refer to frank delusions or hallucinations or misperceptions of reality that are similar to delusions or hallucinations such as unusual thoughts or visual illusions. Psychotic-type symptoms are assessed through direct examination of the patients and interviews with family members, therapists, and nurses. Orientation is measured with the Galveston Orientation and Amnesia Test.10 The standard cut-off of 75 or less is used to determine presence of disorientation. Cognitive impairment is Arch Phys Med Rehabil Vol 90, October 2009
a composite score based on 7 attentional tasks including simple attention (eg, counting to 20), vigilance (eg, indicating when a target letter is spoken in a series of letters), working memory (eg, reciting the months of the year in reverse order), visual recognition memory (eg, identifying target previously viewed pictures from foils), and auditory comprehension (eg, indicating yes or no for the accuracy of sentences). The items are scored using criteria developed for the CAP. Scores can range from 0 to 28, and scores of 18 or lower indicate substantial cognitive impairment. Agitation is measured with the Agitated Behavior Scale.11 Based on analyses completed in CAP development studies, scores greater than 17 indicate significant restlessness. Criteria for each of the 7 symptoms are applied to determine presence or absence of each symptom for each examination. Accordingly, CAP scores range from 0 (no symptoms of confusion present) to 7 (all 7 symptoms of confusion present). Persons with 4 of 7 symptoms present, or 3 symptoms if 1 is disorientation, are classified as in posttraumatic confusional state. Note that many patients who do not meet criteria for posttraumatic confusional state are not normal because they have a variety of cognitive impairments seen in early recovery from TBI such as memory impairment, distractibility, or slowed cognitive processing speed. The version of the CAP used in the present study6 had a minor modification from the original version.2 In the original version, patients could be scored as having psychotic-type symptoms if they showed delusions, hallucinations, or symptoms suggestive of possible delusions or hallucinations or if they showed evidence of thought process abnormalities. This latter criterion was eliminated in the present investigation because of concern that thought process abnormalities could not be reliably and meaningfully distinguished from the highly distractible, disjointed conversation expected in acutely confused patients. The DRS12 is a 30-point scale with 8 areas of functioning rated: eye opening; verbalization; motor response; level of cognitive ability for daily activities of feeding, toileting, and grooming; overall level of dependence; and employability. Each area of functioning is rated on a scale of 0 to 3, 0 to 4, or 0 to 5, with a higher score representing a lower level of functioning. Scores on each item are summed to yield a total score between 0 and 29, with a higher score indicating greater disability. Analysis Descriptive data (age, years of education, sex, GCS, time to follow commands, acute care LOS, rehabilitation LOS, interval from injury to first CAP evaluation, DRS at admit and DRS at discharge, and DRS discharge) are presented as quartiles for continuous variables and percentages for categorical variables. Participants were classified as in posttraumatic confusional state or not in posttraumatic confusional state based on the initial CAP after admission to rehabilitation. Patients in posttraumatic confusional state were compared with patients not in posttraumatic confusional state on descriptive measures using Mann-Whitney U tests, and all subsequent analyses described only those in posttraumatic confusional state. Confused participants with 3 or 4 symptoms on the initial CAP were classified as mildly confused, those with 5 symptoms as moderately confused, and those with 6 or 7 symptoms as severely confused. Mildly, moderately, and severely confused patients were compared on the descriptive variables using KruskalWallis tests. In order to describe the process of symptom resolution in patients over time, the first 3 CAP evaluations for each confused patient were analyzed. Patients for whom there were fewer than 3 completed CAP evaluations were excluded from
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
analyses requiring these data. Extension of this analysis to the fourth CAP would have required exclusion of a large number of additional patients. Data recorded and presented in tables and figures included number of symptoms improved between evaluations, number of symptoms worsened between evaluations, number of patients showing no change between evaluations, symptoms most likely to improve, and symptoms most likely to worsen for each of the 3 confusion severity groups. RESULTS Comparison of Nonconfused and Confused Patients During the study period, 195 potential participants were admitted to the study site. Of these, 16 did not recover to a responsive state (ie, remained in vegetative or minimally conscious state) and 3 had clinical presentations that were substantially influenced by non-TBI disorders (spinal cord injury, dementia, deafness). An additional 5 potential participants declined to give consent, and 3 participants who were enrolled were inadvertently missed for collection of key study measures. The resulting study sample was comprised of 168 participants. Table 1 shows the comparison of these 168 patients classified as not in posttraumatic confusional state (n⫽61) and in posttraumatic confusional state (n⫽107) based on the initial CAP after rehabilitation admission. Those in posttraumatic confusional state had lower preinjury years of education, had lower initial GCS, took longer to follow commands, had longer acute and rehabilitation LOS, had a longer interval from injury to initial CAP, and had poorer functional status, as measured with the DRS, at rehabilitation admission and discharge. All subsequent analyses pertain only to patients in posttraumatic confusional state.
Table 1: Comparison of Nonconfused and Confused Patients as Determined by the First CAP Evaluation After Admission to Inpatient Rehabilitation (nⴝ168)
Continuous descriptors Age (y) Education* (y) GCS score* TFC‡ (d) Acute LOS† (d) Rehabilitation LOS‡ (d) CAP days† Admit DRS‡ Discharge DRS‡ Categorical descriptors Sex Men Women GCS Severe (3–8) Moderate (9–12) Mild (13–15)
Nonconfused (n⫽61) Median (25th, 75th Percentiles)
Confused (N⫽107) Median (25th, 75th Percentiles)
24.4 (19.7, 42.9) 12.0 (10.0, 14.0) 9.0 (5.5, 14.0) 2.0 (0.5, 5.0) 14.0 (10.0, 25.5) 14.0 (10.0, 18.0) 17.0 (12.0, 26.5) 8.0 (6.0, 10.0) 5.0 (4.0, 7.0)
28.6 (20.2, 49.7) 11.0 (10.0, 12.0) 7.0 (5.0, 10.0) 6.0 (2.0, 16.0) 20.0 (15.0, 28.0) 16.0 (14.0, 24.0) 21.0 (16.0, 30.0) 14.0 (11.0, 17.0) 7.5 (6.0, 11.0)
n (%)
n (%)
38 (62) 23 (38)
75 (70) 30 (30)
27 (44) 12 (20) 22 (36)
69 (64) 23 (21) 15 (14)
NOTE. Two confused participants had missing values for to follow commands with TFC, and 1 confused patient was missing values for both admit and discharge DRS. Abbreviations: CAP days, interval from injury to initial CAP evaluation; TFC, time to follow commands. *P⬍.05; †P⬍.01; ‡P⬍.001.
1751
Comparison of Patients With Mild, Moderate, and Severe Confusion Table 2 shows the comparison of patients with mild, moderate, and severe posttraumatic confusional state as determined by the initial CAP after rehabilitation admission. Other than number of symptoms of confusion (CAP total), confusion severity groups differed only on DRS at rehabilitation discharge, with those with more severe confusion showing poorer functional status. Resolution of Symptoms of Confusion Table 3 presents data on the resolution of symptoms of posttraumatic confusional state for the 3 injury severity groups. These patterns of symptoms resolution are shown graphically in figure 1. Three participants (2 mildly, 1 moderately confused) had only 1 CAP evaluation, and 7 additional participants (4 mildly, 2 moderately, 1 severely confused) had only 2 CAP evaluations. This affected the numbers included for changes from CAP 1 to CAP 2 (104/107 [97%] patients were reported on) and from CAP 2 to CAP 3 (97/107 [93%] patients were reported on). The interval (mean ⫾ SD) from the first to the second CAP evaluation was 1.8⫾1.3 days, and the interval (mean ⫾ SD) from the second to the third CAP evaluation was 2.5⫾1.6 days. Across confusion severity groups, from the first to second CAP, 49% to 63% of participants showed improvement as indicated by a decreased number of symptoms of confusion. Patients with initial moderate and severe confusion continued to improve at about the same rates, 50% to 56%, respectively, between the second and third CAPS. Only 25% of those with initial mild confusion showed continued improvement, however, indicating a possible floor effect, because these patients had only a few remaining symptoms of confusion. Worsening confusion, as indicated by an increased number of symptoms, was rare, ranging from as few as 5% of those with initial severe confusion showing worsening between the first and second CAPS to 14% of those with initial mild confusion showing worsening between the first and second evaluations. Proportions of patients who stayed the same (continued with the same number of symptoms between evaluations) from CAP 1 to 2 and from CAP 2 to 3 ranged from 26% to 69% across confusion severity groups. Sixty-nine percent of patients with mild confusion showed no change in the number of symptoms between the second and third CAPs. Almost half (46%) of those with severe confusion showed no change between the first and second CAPs. For all patients in posttraumatic confusional state, from the first to the third CAP, 73% improved, while only 27% remained the same or worsened. Early recovery for severely confused patients was marked by resolution of disturbances of arousal (nighttime sleep disturbance and decreased daytime arousal) and by resolution of psychotic-type symptoms. Although initially present in a smaller proportion of patients, these symptoms were also likely to resolve early for those with moderate confusion. Psychotictype symptoms and nighttime sleep disturbance fluctuated in patients with moderate and severe confusion, because these symptoms were among the most likely to improve but were also the most likely to worsen. Disorientation and agitation tended to resolve next, with some patients also showing resolution of cognitive impairment and fluctuation. Cognitive impairment and fluctuation were the most persistent symptoms for all confused patients, although a significant number also remained with disorientation by the third CAP assessment. Arch Phys Med Rehabil Vol 90, October 2009
1752
POSTTRAUMATIC CONFUSIONAL STATE, Sherer Table 2: Comparison of Confusion Severity Groups as Determined by the First CAP Evaluation After Admission to Inpatient Rehabilitation (Nⴝ107)
Continuous descriptors Age (y) Education (y) GCS score TFC (d) Acute LOS (d) Rehabilitation LOS (d) CAP days CAP total* Admit DRS Discharge DRS† Categorical descriptors Sex, n (%) Male Female GCS, n (%) Severe (3–8) Moderate (9–12) Mild (13–15)
Mildly Confused (n⫽38) Median (25th, 75th Percentiles)
Moderately Confused (n⫽28) Median (25th, 75th Percentiles)
Severely Confused (41) Median (25th, 75th Percentiles)
27.0 (21.4, 53.2) 11.5 (10.0, 12.0) 7.0 (5.8, 10.2) 4.5 (1.0, 12.0) 19.0 (14.8, 25.8) 16.0 (14.0, 23.2) 20.5 (16.8, 29.2) 4.0 (3.0, 4.0) 12.5 (9.8, 15.2) 7.0 (5.0, 8.0)
28.1 (18.5, 48.2) 11.5 (10.0, 12.0) 7.0 (4.2, 10.8) 6.5 (3.0, 18.5) 20.0 (15.2, 30.2) 21.0 (5.0, 29.8) 21.0 (16.2, 31.0) 5.0 (5.0, 5.0) 14.0 (12.0, 17.0) 7.5 (6.0, 10.0)
30.4 (20.6, 50.8) 11.0 (10.0, 12.0) 7.0 (4.0, 10.0) 8.0 (2.0, 16.5) 23.0 (15.0, 29.5) 16.0 (13.5, 23.0) 24.0 (16.0, 30.5) 7.0 (6.0, 7.0) 14.5 (12.0, 19.0) 8.5 (6.2, 13.0)
31 (82) 7 (18)
19 (68) 9 (32)
25 (61) 16 (39)
23 (60) 10 (26) 5 (13)
19 (68) 5 (18) 4 (14)
27 (66) 8 (19) 6 (15)
Abbreviations: CAP days, the interval from injury to the initial CAP evaluation; CAP total, the number of symptoms of confusion found to be present on the CAP evaluation; TFC, time to follow commands. *P⬍.001; †P⬍.01.
DISCUSSION The present preliminary investigation produced novel findings regarding the symptomatic course of recovery for confused patients early after TBI. As expected, confusion has a recovering course, with almost three quarters of patients showing resolution of some symptoms over the approximately 4-day period between the first and third evaluations. Disturbances of arousal and psychotic-type symptoms resolved first. Because most participants had severe TBI and, thus, some period of coma, resolution of disturbances of arousal may represent a continuation of recovery indicated by emergence from the minimally conscious or vegetative states. This recovery may be supported by improving function of neural pathways projecting from the brain stem to the frontal lobes. The temporal association of resolution of psychotic-type symptoms may suggest overlap in the neural substrate for disturbed arousal and misperceptions of the external world. The temporal association between resolution of disorientation and agitation, a finding observed previously,8 suggests that both are associated with profound disturbances of attentional processes producing disorganized behavior and inability to attend to even basic elements of the immediate environment. This interpretation is consistent with the finding that cognitive impairments generally resolved after disorientation and agitation. The cognitive measures used in the CAP include tasks requiring vigilance and working memory that require more intact attentional abilities than simple orientation. Current findings are consistent with elements of the conceptualization by Stuss et al1 of the confused state as a global disturbance of attentional processes. Finally, we found that fluctuation in presentation is the most persistent symptom of confusion in early recovery from TBI. In this investigation, fluctuation was manifested as change in the adequacy of behavioral performances from one time of day to another. From clinical observation, these variations seemed to be a result of changes in level of arousal or alterations in the Arch Phys Med Rehabil Vol 90, October 2009
degree of distractibility and restlessness. Return of the ability of the brain to produce consistent behavior in the face of fatigue or other situational stresses indicates a later stage of recovery from TBI. As expected, confused patients differed from nonconfused patients on a number of factors. Confused patients had more severe injuries as indicated by initial level of responsiveness (as measured by the GCS), time to follow commands, and hospital LOS. It was surprising that these factors did not differ among the 3 confusion severity groups. These groups did not differ on initial GCS, time to follow commands, or hospital LOS, although more severely confused patients did have poorer functional status at rehabilitation discharge. While visual inspection of median times to follow commands and acute hospital LOSs for the 3 groups suggests that longer intervals were associated with greater confusion, this difference was not statistically significant because of substantial within-group variability causing substantial overlap between groups for this interval. This finding that greater confusion did not have a direct relationship to accepted indices of greater injury severity is provocative. It may suggest that there are differences in the underlying neuropathology among the confusion groups. In this speculative scenario, patients in each group had sufficient TBI severity to result in a general degree of neuropathology sufficient to cause confusion. However, the specific brain areas and pathways affected were sufficiently variable between patient groups to result in different patterns of symptoms and differential associated courses and durations13 of recovery. Study Limitations The present investigation has a number of limitations. First, participants were studied only after admission to inpatient rehabilitation. Persons with TBI not coming to rehabilitation were not studied, and the implications of these findings for these patients are unknown. Also, very early recovery of con-
POSTTRAUMATIC CONFUSIONAL STATE, Sherer Table 3: Evolution of Symptoms of Confusion Over the First 3 CAP Assessments After Admission to Inpatient Rehabilitation (nⴝ97) Patients With Severe Confusion on Initial CAP (n⫽41)
No. improved (%) By 1 symptom By 2 symptoms By 3 or more symptoms No. worse (%) By 1 symptom No. same (%) Exactly Symptoms most likely to improve (n) Symptom most likely to worsen (n)
1st CAP to 2nd CAP (n⫽41)
2nd CAP to 3rd CAP (n⫽40)
20 (49) 11 (27) 6 (15) 3 (7) 2 (5) 2 (5) 19 (46) 18 (44) Arouse (11), Psych (11), Sleep (9) Psych (4)
20 (50) 9 (22) 8 (20) 3 (8) 5 (12) 5 (12) 15 (38) 15 (38) Sleep (8), Arouse (8), Psych (6) Sleep (3)
1753
fusion occurring at the trauma center was not studied, and no statements can be made about the course of recovery prior to rehabilitation admission. A longer epoch of evaluation with a greater number of assessments might have been interesting, but would have introduced an even greater injury severity bias, excluding patients with more moderate TBI. We were not able to report on change over 3 evaluations in all patients because only 93% of patients had the required evaluations. Aspects of confusion studied and the metrics of measurement were those reflected in the CAP. There may be other important symptoms or finer/composite measurement instruments incorporating symptom severity that would have provided additional findings of interest.
Patients With Moderate Confusion on Initial CAP (n⫽28)
No. improved (%) By 1 symptom By 2 symptoms By 3 or more symptoms No. worse (%) By 1 symptom By 2 symptoms No. same (%) Exactly Symptoms most likely to improve (n) Symptom most likely to worsen (n)
1st CAP to 2nd CAP (n⫽27)
2nd CAP to 3rd CAP (n⫽25)
17 (63) 9 (33) 5 (19) 3 (11)
14 (56) 8 (32) 4 (16) 2 (8)
3 (11) 2 (7) 1 (4) 7 (26) 5 (19) Agit (7), Sleep (6), Cog (5), Psych (5) Psych (3)
2 (8) 2 (8) — 9 (36) 7 (28) Disorient (6), Sleep (6), Fluct (4), Arouse (4) Sleep (3)
Patients With Mild Confusion on Initial CAP (n⫽38) 1st CAP to 2nd CAP (n⫽36)
2nd CAP to 3rd CAP (n⫽32)
No. improved (%) 18 (50) 8 (25) By 1 symptom 11 (31) 7 (22) By 2 symptoms 7 (19) — By 3 or more symptoms — 1 (3) No. worse (%) 5 (14) 2 (6) By 1 symptom 5 (14) 2 (6) No. same (%) 13 (26) 22 (69) Exactly 10 (28) 22 (69) Symptoms most likely to Disorient (9), Cog (2), Agit (2), Fluct (2), Arouse (2), improve (n) Agit (5), Sleep (2) Sleep (4) Symptoms most likely to Arouse (3) Disorient (1), Agit (1) worsen (n) NOTE. The terms exactly, improved, same, and worse are defined as follows: exactly, the patient had the exact same pattern of symptoms on the subsequent CAP evaluation; improved, the patient had fewer symptoms on the subsequent CAP evaluation; same, the patient had the same number of symptoms on the subsequent CAP evaluation; worse, the patient had more symptoms on the subsequent CAP evaluation. Abbreviations: Agit, agitation; Arouse, decreased daytime arousal; Cog, cognitive impairment; Disorient, disorientation; Fluct, fluctuation; Psych, psychotic-type symptoms; Sleep, nighttime sleep disturbance.
Fig 1. Percentages of patients with TBI with (A) severe, (B) moderate, and (C) mild confusion showing each of 7 symptoms over the first 3 CAP evaluations after admission to rehabilitation (nⴝ41, nⴝ28, nⴝ38, respectively). Abbreviations: Agit, agitation; Arouse, decreased daytime arousal; Cog, cognitive impairment; Disorient, disorientation; Fluct, fluctuation; Psych, psychotic-type symptoms; Sleep, nighttime sleep disturbance.
Arch Phys Med Rehabil Vol 90, October 2009
1754
POSTTRAUMATIC CONFUSIONAL STATE, Sherer
CONCLUSIONS While heterogeneous, posttraumatic confusion is not manifested in a random mishmash of cognitive and behavioral manifestations with relevance limited to duration of symptoms and associated injury severity. Early recovery from post-TBI confusion has a somewhat predictable course that likely reflects the underlying neural substrate of the injury. Improved understanding of the patterns of symptoms of confusion and their causes may lead to improved early treatment for patients with TBI. As an example, the apparent association between arousal disturbances and psychotic-type symptoms may motivate clinicians to treat nighttime sleep disturbance and decreased daytime arousal aggressively in hopes of also addressing psychotictype symptoms. However, we are unaware of any data at this time that would confirm that such aggressive treatment of arousal disturbances would decrease the incidence or shorten the duration of psychotic-type symptoms. References 1. Stuss DT, Binns MA, Carruth FG, et al. The acute period of recovery from traumatic brain injury: posttraumatic amnesia or posttraumatic confusional state? J Neurosurg 1999;90:635-43. 2. Sherer M, Nakase-Thompson R, Yablon SA, Gontkovsky ST. Multidimensional assessment of acute confusion after traumatic brain injury. Arch Phys Med Rehabil 2005;86:896-904. 3. Ellenberg JH, Levin HS, Saydjari C. Posttraumatic amnesia as a predictor of outcome after severe closed head injury. Arch Neurol 1996;53:792-1. 4. Sherer M, Sander A, Nick TG, High WM Jr, Malec JF, Rosenthal M. Early cognitive status and productivity outcome after traumatic brain injury: findings from the TBI model systems. Arch Phys Med Rehabil 2002;83:183-92.
Arch Phys Med Rehabil Vol 90, October 2009
5. Nakase-Richardson R, Yablon SA, Sherer M. Prospective comparison of acute confusion severity with duration of post-traumatic amnesia in predicting employment outcome after traumatic brain injury. J Neurol Neurosurg Psychiatr 2007;78:872-6. 6. Sherer M, Yablon SA, Nakase-Richardson R, Nick T. Effect of severity of posttraumatic confusion and its constituent symptoms on outcome after traumatic brain injury. Arch Phys Med Rehabil 2008;89:42-7. 7. High WM Jr, Levin HS, Gary HE. Recovery of orientation following closed-head injury. J Clin Exp Neuropsychol 1990; 12:703-14. 8. Corrigan JD, Mysiw WJ. Agitation following traumatic head injury: equivocal evidence for a discrete stage of cognitive recovery. Arch Phys Med Rehabil 1988;69:487-92. 9. Weir N, Doig EJ, Fleming FM, Wiemers A, Zemljic C. Objective and behavioral assessment of the emergence from post-traumatic amnesia (PTA). Brain Inj 2006;20:927-35. 10. Levin HS, O’Donnell VM, Grossman RG. The Galveston Orientation and Amnesia Test: a practical scale to assess cognition after head injury. J Nerv Ment Dis 1979;167:675-84. 11. Corrigan JD. Development of a scale for assessment of agitation following traumatic brain injury. J Clin Exp Neuropsychol 1989; 11:261-77. 12. Rappaport M, Hall KM, Hopkins K, Belleza T, Cope DN. Disability rating scale for severe head trauma: coma to community. Arch Phys Med Rehabil 1982;63:118-23. 13. Nakase-Richardson R, Sherer M, Yablon SA, Evans CC. Initial neuroimaging is predictive of confusion at one-month post traumatic brain injury. J Int Neuropsychol Soc 2006;12(Suppl 1):215. Available at: http://journals.cambridge.org/action/displayJournal?jid⫽INS#. Accessed September 15, 2009.