THE EARLY STAGES OF recovery from traumatic brain

521 BRIEF REPORT Sequence of Recovery During the Course of Emergence From the Minimally Conscious State Christine M. Taylor, BAppSc, Vanessa H. Aird, BAppSc, Robyn L. Tate, MPsychol, PhD, Michele H. Lammi, BAppSc ABSTRACT. Taylor CM, Aird VH, Tate RL, Lammi MH. Sequence of recovery during the course of emergence from the minimally conscious state. Arch Phys Med Rehabil 27;88: 521-5. Objective: To document and examine recovery profiles of persons during the course of emergence from a minimally conscious state (MCS) after traumatic brain injury (TBI). Design: Case series. Setting: Participants in an inpatient brain injury rehabilitation program. Participants: Nine people with TBI who emerged from MCS. Interventions: Not applicable. Main Outcome Measure: The Western Neuro Sensory Stimulation Profile. Results: In all cases, stereotyped (nonreflexive) movement was the first behavior to resolve, yet by itself it was never sufficient to indicate emergence from the MCS. Two of the 9 patients showed consistent object manipulation before all 3 elements of functional communication. Conversely, in 2 patients, all 3 components of functional communication emerged before object manipulation. In the remaining 5 patients, object manipulation occurred along with components of functional communication. Conclusions: During the course of data collection, issues were raised about differentiation between the MCS and posttraumatic amnesia. The data are also discussed within the context of the work of Giacino et al and the development of their scale to measure duration of the MCS. These data support the theory that functional communication and object manipulation are the 2 key indicators of emergence from MCS. Key Words: Brain injuries; Minimally conscious state; Rehabilitation. 27 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation THE EARLY STAGES OF recovery from traumatic brain injury (TBI) are characterized by various states of altered consciousness, including coma, and, depending on injury severity, the vegetative state (VS) and minimally conscious state (MCS). Patients with TBI also subsequently transit through a period of posttraumatic amnesia (PTA). Although operational definitions have been applied successfully to coma and VS, the MCS and its surrounding terminology have had a more recent history. The MCS is a condition that follows coma and is defined by the Aspen From the Brain Injury Rehabilitation Service, Royal Rehabilitation Centre Sydney, Ryde, Australia (Taylor, Aird, Lammi); and Rehabilitation Studies Unit, University of Sydney, Sydney, Australia (Tate). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Christine M. Taylor, BAppSc, Brain Injury Rehabilitation Service, Royal Rehabilitation Centre Sydney, PO Box 6, Ryde NSW 16, Australia, e-mail: christine.taylor@royalrehab.com.au. 3-9993/7/84-11149$32./ doi:1.116/j.apmr.27.1.13 workgroup as severely altered consciousness in which the person demonstrates minimal but definite behavioral evidence of self or environmental awareness. 1(p84),2 This expert group further outlined 5 criteria to characterize entry into the MCS. One or more of the following behaviors must be observed, even if inconsistently, for a diagnosis of MCS: (1) comprehension of simple commands, (2) manipulation of objects, (3) gestural or verbal yes-no response, (4) intelligible verbalizations, and (5) stereotyped (but nonreflexive) movements. Although clear diagnostic guidelines for entry into the MCS have been established, criteria for emergence from the MCS are less clear. In 22, Giacino et al 2 proposed 2 essential features for emergence, comprising reliable and accurate (1) functional interactive communication and (2) functional use of 2 different objects. The criteria are consensus based, and the authors point to the limited availability of empirical evidence at this stage. In their group of 15 patients, they reported that in 47% functional object use emerged first, in 2% functional communication emerged first, and in the remaining 33% they coincided. The interval between emergence for object use and functional communication ranged from 5 to 14 days. Our group studied the natural history of recovery from the MCS in a small group of 21 patients and reported on the postacute neurobehavioral profiles 3 and longer-term functional and psychosocial recovery. 4 The purpose of this study was to contribute additional empirical data on patterns of emergence from the MCS. Specifically, like Giacino, we examine the sequence of emergence of specific domains. We also examine latencies among emergence in specific domains. METHODS The sample has been described previously and comprised 21 patients admitted to a regionally based brain injury rehabilitation unit from an acute facility, 18 of whom had participated in the follow-up study reported by Lammi et al. 4 In brief, all 21 patients had sustained an extremely severe TBI and were participating in their initial period of postacute rehabilitation. They were admitted to the rehabilitation unit a median of 39 days posttrauma (range, 17 216d). The group was predominantly male (n 17 [81%]), aged 39 years when injured (range, 15 69y), with the injuries occurring as a result of road traffic crashes (n 13 [62%]) or falls (n 8 [38%]). On admission to the rehabilitation unit, each patient presented with a severely altered state of consciousness (one was in coma; the others were deemed to be in an MCS). All patients were routinely administered the Western Neuro Sensory Stimulation Profile (WNSSP), 5 with administration frequency determined by clinical need. This scale was specifically designed to monitor and predict change in the slow-to-recover population as well as in those patients who are functioning between levels II and V on the Rancho Los Amigos Scale. 6 The 33 items of the WNSSP cover 9 domains: arousal/attention, auditory response, auditory comprehension, expressive communication, visual tracking, visual comprehension, tactile response, object manipulation, and olfactory response. This tool was developed before the more recent developments in refining nomenclature in this subject group; however, it does specifically measure the defining MCS behaviors. Arch Phys Med Rehabil Vol 88, April 27

522 EMERGENCE FROM THE MINIMALLY CONSCIOUS STATE, Taylor Table 1: Descriptive Data for Day Posttrauma of Emergence of Specific Behaviors (N 9) Characteristic Median Range Interquartile Range Estimated duration of MCS 72 46 158 36 Day admitted to rehabilitation 34 17 44 12 Stereotyped (nonreflexive) movement 27 138 19 Object manipulation 54 46 242 32 Intelligible verbalization 54 39 158 36 rehension 64 46 138 32 Yes-no response 64 46 138 32 Functional communication 64 46 158 35 Sequence (irrespective of domain) First 27 138 19 Second 47 39 138 22 Third 64 46 138 29 Fourth 64 46 158 36 Fifth 72 46 242 35 The 5 agreed behaviors that signal entry to MCS (ie, comprehension of simple commands, manipulation of objects, gestural or verbal yes-no response, intelligible verbalizations, stereotyped nonreflexive movements) were monitored with the WNSSP throughout the course of each patient s recovery. We used these same 5 behaviors to identify the point of emergence from MCS, requiring that each of the behaviors was performed consistently on more than 1 test occasion. Specifically, as per the following WNSSP criteria: functional interactive communication was assessed by (1) the comprehension of simple commands using the auditory and visual comprehension subtests (eg, shake my hand ), in which the correct response was required; (2) yes-no response used the expressive communication subtest yes-no response (eg, Do you live in ), but the correct response was not required, as per the WNSSP manual; and (3) intelligible verbalization was assessed by the expressive communication subtest vocalization, which required intelligible, relevant speech. Stereotyped nonreflexive movements were examined via the arousal/attention, visual tracking, tactile response, and olfactory subtests. Functional object use was assessed with the object manipulation subtest (eg, This is a. Show me how you would use it ) in which the correct response was required. Although these criteria for emergence appear to differ from those of Giacino et al, 2 the differences reflect surface labeling issues (ie, both our functional interactive communication and that of Giacino require accurate comprehension to be shown). For the precise day of emergence in specific domains, we used the daily documentation that was made in the patient s medical record. RESULTS Duration of the MCS was able to be determined in 17 patients and was a median of 54 days (range, 27 615d). The remaining 4 patients had either died before emerging (n 1) or not emerged from the MCS when they were last contacted between 4 to 5 years posttrauma (see Lammi et al 4 ). Only nonreflexive movement was ever shown during the course of observations in any of these 4 patients, occurring at 58, 136, and 77 days posttrauma in 3 patients whom we documented as still in the MCS at 141, 16, and 1825 days after trauma, respectively. The data from 8 of the 17 patients who emerged from the MCS were not suitable for detailed study of patterns of emergence because of impairments confounding the examinations (one with severe aphasia; one with anarthria and spastic tetraplegia) or the patients emerged from the MCS fairly rapidly after admission to the rehabilitation unit (n 6). In these latter cases, the WNSSP had only been administered on 1 or 2 test occasions, and thus there were insufficient data to meaningfully examine patterns of emergence. The final 9 patients emerging from the MCS during inpatient rehabilitation were assessed on multiple occasions with the WNSSP in relation to the 5 behavioral domains, and they were monitored until the point of emergence. Descriptive data are presented in table 1. Specific comparison was made regarding the time taken to emerge of the 2 behavioral domains suggested by Giacino, 2 namely, functional object use (at a median of day 54 after trauma) and functional communication (median of day 64). The result of a Wilcoxon signed-rank test was not statistically significant (z.51, P.5). Table 2 shows the sequence of emergence, irrespective of the behavioral domain. Repeated measures analyses performed by using Wilcoxon tests compared day posttrauma of emergence of adjacent categories (1st vs 2nd behavioral domain, 2nd vs 3rd, and so on). Results showed a significant difference between day of injury (as well as admission to the rehabilitation unit) and emergence of the first of the defining behaviors (day posttrauma: z 2.67, P.9; day postadmission z 2.68, P.8), but the remaining analyses did not meet the Bonferroni-adjusted level of P less than.1 (although 3/4 comparisons were significant at the P.5 level). These data show that emergence generally occurred over approximately a 6- to 12-week period posttrauma and a period of 4 to 6 weeks after admission to rehabilitation. Case ID (Lammi et al 4 ) Table 2: Day Posttrauma at Which the MCS Defining Behaviors Occurred in Individual Cases Estimated Duration of MCS lications OM Yes-No Funct Comm A 1 44 81 Trache 53 6 81 81 74 81 B 3 35 158 Trache, ROM 138 242 158 138 138 158 C 4 34 47 Trache 42 47 49 47 47 49 D 5 17 72 None 27 46 65 72 72 72 E 9 31 5 Trache 36 47 47 5 47 5 F 11 27 54 Trache, dysarth 39 54 54 47 47 54 G 13 35 79 None 79 54 64 64 64 H 14 42 87 None 52 77 87 77 84 87 I 16 27 46 None 32 46 39 46 46 46 Abbreviations:, admission to rehabilitation;, comprehension; dysarth, dysarthria; Funct Comm, functional communication;, nonreflexive movement; OM, object manipulation; ROM, range of motion;, verbalization; Trache, tracheostomy intubation. Arch Phys Med Rehabil Vol 88, April 27

EMERGENCE FROM THE MINIMALLY CONSCIOUS STATE, Taylor 523 Individual data are presented in table 2 and are also depicted graphically in figure 1. In all patients, stereotyped (nonreflexive) movement was the first behavior to be identified, yet by itself it was never sufficient to indicate emergence from the MCS. Nonetheless, Spearman correlation coefficients between duration of the MCS and the 5 behaviors were all highly statistically significant (r range,.92.99), including nonreflexive movements (r.92, P.1). This latter result indicates that there is a close association between occurrence of nonreflexive movements and emergence from the MCS. Moreover, the descriptive data show that in this small sample, for the 8 patients who were not physically restricted and emerged from the MCS, this occurred within approximately 6 weeks of first showing nonreflexive movements. In other words, from the point in time that these particular patients were classified as nonvegetative and noncomatose, it is weeks rather than months later that they emerged from the MCS. We do, however, recognize that other patients may remain in the MCS for many months or years despite having nonreflexive movements, and we described 3 such patients from this series in our outcome study. 4 At an individual level, 2 of the 9 patients (cases A, D in fig 1, table 2) showed consistent object manipulation before any of the 3 elements of functional communication, with a time lag of 21 and 26 days, respectively. Conversely, in 2 patients (cases B, G), each of the 3 components of functional communication emerged before object manipulation, by 84 and 15 days, respectively, but in 1 of these (case B) object manipulation was compromised by severely restricted range of movement. In the remaining 5 cases (cases C, E, F, H, I), object manipulation occurred along with components of functional communication. Within the domain of functional communication, the pattern of emergence of the 3 constituent behaviors did not show a consistent trend. Three of the 9 patients (cases D, G, I) showed intelligible verbalizations as the first domain to emerge, which occurred 7 to 1 days before other functional communication behaviors. For 1 patient (case A), yes-no response emerged 1 week before other behaviors and for another (case H) comprehension emerged 1 week before the next communication behavior. In 2 patients, cases B and F, comprehension and yes-no response emerged simultaneously, 1 week or more before intelligible verbalization. In the remaining 2 patients (cases C, E), all 3 components emerged within a few days of each other. In determining the timeframe from early communication attempts (ie, the first of the 3 functional communication behaviors to occur) to the establishment of a consistent functional communication repertoire (ie, showed consistency in all 3 behaviors), the latency between the first and the third of these was calculated. The results indicated variable latencies between 2 and 2 days (median, 7d). DISCUSSION Data from this small, independent sample contribute to those published by Giacino et al. 2 In agreement with their findings, we observed variability in emergence in the 2 key domains: functional object use emerged before the repertoire of functional communication in 2 of 9 (22%) patients, vice versa in another 2 of 9 (22%), and over the same period of time in 5 of 9 (56%). In all patients, nonreflexive movements were the first of the behaviors to emerge. These included turning or looking toward the source of a voice or to a tactile stimulus or visual tracking of a person, object, or mirror. Although on its own nonreflexive behavior was never an indicator of emergence, those patients who did go on to emerge showed the other behaviors within weeks of the onset of nonreflexive movement. Moreover, the correlation coefficient between return of nonreflexive movement and duration of MCS was very high (r.92). Therefore, it may be important to routinely monitor the latency from the onset of nonreflexive movement to the commencement of functional object use and functional communication because it may provide important prognostic information regarding emergence from MCS. In a slightly different vein by using functional magnetic resonance imaging with a patient in the VS, Owen et al 7 showed that specific brain areas showed differential activation when the patient was asked to imagine she was playing tennis (supplementary motor area) or imagine that she was walking through her home (parahippocampal gyrus, posterior parietal cortex, lateral premotor cortex). These kinds of data suggest she had at least some capacity for comprehension and response to spoken language. If this occurs in patients in VS, then it reinforces the absolute importance of monitoring patients in the MCS at a very early stage of this period of recovery. In the course of monitoring recovery during the MCS, a number of issues arose regarding the process of emergence from the MCS that deserve further examination. Although the impact of aphasia and dyspraxia on a person s performance has been addressed, 8 the effect of PTA on MCS duration warrants discussion. Patients in PTA are defined as being confused, amnesic, and likely to evidence behavioral disturbance. 9(p675) Unreliability and inaccuracy of response is a characteristic feature of the patient in PTA. There continues to be debate about many aspects of PTA, 1 including defining features and nomenclature, 11 as well as operational definitions of exit. 12,13 Although Levin et al 9 suggested that PTA was measured from the end of coma, the convention has been to measure PTA from the date of the injury. This clearly encompasses the period of MCS. At a clinical level, however, we believe it is important to make a clear differential diagnosis between PTA and resolved MCS, which, like coexisting impairments in speech and dyspraxia, may be difficult for those patients whose executive and regulative abilities are compromised because of the presence of PTA. From many perspectives including counseling families, rehabilitation program planning, and accuracy of nomenclature, it is important to know whether the patient is still in MCS or has transited to the next stage of recovery, PTA. Clinicians need to know whether they are testing for MCS or PTA; families welcome this obvious sign of progress toward recovery. More specifically, in our rehabilitation setting, it is the emergence from MCS that signifies a readiness to increase participation in rehabilitation, which will often involve PTA testing. If a patient is able to participate in a question-answer dyad, follow simple instructions, and shows some attempt to consider the response, regardless of accuracy, they would be classified as having emerged from the MCS. There is no longer... minimal but definite behavioral evidence... 1(p84) but rather an ability to interact consistently and to more actively engage in the rehabilitation process. At this stage, it may be more relevant clinically to describe the impact of PTA and confusion on a patient s performance, as opposed to MCS. We consider that the continuum of the MCS needs to be further described to include a description at the higher end of functioning in this state (ie, at the ceiling of Rancho level IV [confusedagitated]). A patient at this level may be engaging in functional activities but showing more pronounced evidence of confusion or amnesia, contributing to their difficulty in accurately producing some responses. An important next step will be to obtain a more detailed examination of the natural history of the transition period between emergence from the MCS and entry to PTA. CONCLUSIONS The contribution of the present data set from an independent sample lends strength to the exit criteria from the MCS proposed by Giacino. 2 That is, even with the use of different assessment tools, it is clear that functional communication and Arch Phys Med Rehabil Vol 88, April 27

524 EMERGENCE FROM THE MINIMALLY CONSCIOUS STATE, Taylor 1 6 2 2 2 16 12 Case A (Tracheostomy) Case B (Tracheostomy; range of mvt) 1 1 6 2 6 2 Case C (Tracheostomy) Case D 1 1 6 2 6 2 Case E (Tracheostomy) Case F (Tracheostomy) Fig 1. Pattern of emergence from MCS in 9 participants. The heavy line denotes emergence from MCS. Abbreviations:, admission to rehabilitation;, comprehension;, nonreflexive movement;, object manipulation;, verbalization. Arch Phys Med Rehabil Vol 88, April 27

EMERGENCE FROM THE MINIMALLY CONSCIOUS STATE, Taylor 525 1 1 6 2 6 2 Case G Case H 1 6 2 Case I Fig 1. (Continued) functional object use are useful indicators of emergence from the MCS. However, we suggest that a clearer distinction between PTA and MCS be made so that length of MCS does not become artificially inflated because of the well known and distinct syndrome of PTA. References 1. Giacino JT, Zasler ND, Katz DI, Kelly JP, Rosenberg JH, Filley CM. Development of practice guidelines for assessment and management of the vegetative and minimally conscious states. J Head Trauma Rehabil 1997;12:79-89. 2. Giacino JT, Ashwal S, Childs N, et al. The minimally conscious state: definition and diagnostic criteria. Neurology 22;58:349-53. 3. Smith VH, Taylor CM, Lammi MH, Tate RL. Recovery profiles of cognitive-sensory modalities in patients in the minimally conscious state following traumatic brain injury. Brain Impair 21; 2:29-38. 4. Lammi MH, Smith VH, Tate RL, Taylor CM. The minimally conscious state and recovery potential: a follow-up study 2 to 5 years after traumatic brain injury. Arch Phys Med Rehabil 25; 86:746-54. 5. Ansell BJ, Keenan JE, de la Rocha O. Western Neuro Sensory Stimulation Profile: a tool for assessing slow-to-recover headinjured patients. Tustin: Western Neuro Care Center; 1989. 6. Malkmus D, Booth B, Kodimer C. Rehabilitation of the head-injured adult: comprehensive cognitive management. Downey: Professional Staff Association, Rancho Los Amigos Hospital; 19. 7. Owen AM, Coleman MR, Boly M, Davis MH, Laureys S, Pickard JD. Detecting awareness in the vegetative state. Science 26;313:12. 8. Giacino JT, Whyte J. The vegetative and minimally conscious states: current knowledge and remaining questions. J Head Trauma Rehabil 25;2:3-5. 9. 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. 1. Tate RL, Pfaff A. Problems and pitfalls in the assessment of posttraumatic amnesia. Brain Impair 2;1:116-29. 11. Stuss DT, Binns MA, Carruth F, et al. The acute period of recovery from traumatic brain injury: posttraumatic amnesia or posttraumatic confusional state? J Neurosurg 1999;9:635-43. 12. Tate RL, Pfaff A, Jurjevic L. Resolution of disorientation and amnesia during posttraumatic amnesia. J Neurol Neurosurg Psychiatry 2;68:178-85. 13. Tate RL, Pfaff A, Baguley IJ, et al. A multicentre, randomised trial examining the effect of test procedures on emergence from post-traumatic amnesia. J Neurol Neurosurg Psychiatry 26;77: 841-9. Arch Phys Med Rehabil Vol 88, April 27