320
Agitation and Restlessness After Closed Head Injury: A Prospective Study of 100Consecutive Admissions Marvin M. Brooke, MD, Kent A. Questad, PhD, David R. Patterson, PhD, Karen J. Bashak, MHSc ABSTRACT. Brooke MM, Questad KA, Patterson DR, Bashak KJ. Agitation and restlessness after closed head injury: a prospective study of 100 consecutive admissions. Arch Phys Med Rehabil 1992;73:320-3. l Agitation and restlessness are two of the most striking and problematic behaviors for patients with traumatic brain injury (TBI), their caregivers, and their families. These behaviors are often treated with physical and chemical restraints which have potentially harmful side effects. There are, however, few prospective studies which clearly define agitation and restlessness in a representative sample of TBI patients. Subjects for this study were 100 consecutive patients with traumatic, closed head injury (CHI) admitted to a regional Level I Trauma Center with a Glasgow Coma Scale score of less than 8, who had more than one hour of coma, and who required more than one week of hospitalization. Agitation was defined as episodic motor or verbal behavior which interfered with patient care or clearly required physical or chemical restraints to prevent damage to persons or property. This variable was rated on the Overt Aggression Scale, a lbitem scale, in four categories: verbal aggression; physical aggression against objects; physical aggression against self; or physical aggression directed at others. Systematic direct observations, caregiver interviews, and chart reviews were used to determine the frequency and duration of agitation. Patients were also monitored for restlessness, which was defined as behavior that interfered with staff or required some action by staff, such as change of activity, but either did not meet the severity criteria for agitation, or was continuous. Only 11 of the 100 subjects exhibited episodic agitation which met the criteria. Eight subjects were agitated for one week, one for two weeks, one for three weeks, and one for four weeks. Only one subject went directly from being unresponsive to being agitated. Restlessness, however, occurred in 35 patients. The results suggest that agitation is not frequently or predictably a phase of recovery from head injury. The precise definition and expected duration of these behaviors should be considered carefully before potentially harmful treatment is prescribed. 0 1992 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation KEY M.ORDS: Bruin injury; Ps~rhmotor
ugitution; Stati.stic.c
Agitation and restlessness are well-known sequelae of traumatic brain injury’,2 and are two of the most problematic behaviors to evaluate and treat. These behaviors are distressing to patients, families, and caregivers. They can also interfere with therapy and patient safety. Agitation and restlessness are, at times, managed with physical or chemical restraints, although clinicians should be cautious and aware of the potential harmful side effects of such interventions. There is little researchlm4which carefully defines these behaviors, accurately describes their frequency and duration, and is based on an adequate number of cases without sampling biases. This lack of information compounds the difficulties in assessment and management. Some research has suggested that traumatic brain injury From the Department of Rehabilitation Medicine, Tufts University School of Medicine (Dr. Brooke) and Department of Rehabilitation Medicine, University of Washington School of Medicine (Drs. Questad. Patterson. Ms. Bashak). This research was supported in part by the Harborview Injury Prevention and Research Center. Center for Disease Control grant CCR49-002570. and the National Institute on Disability and Rehabilitation Research grant GOO830076, Department of Education. Washington, DC. Submitted for publication May 22. 1990. Accepted in revised form February I I, 1991. No commercial party having a direct or indirect interest in the subject matter ofthis article has conferred or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Marvin M. Brooke, MD, Department of Rehabilitation Medicine. Tufts University School of Medicine. I36 Harrison Avenue, Boston, MA 02 1 I I. GJ 1992 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation OcO3-9993/92/7304-0003$3.00/0
Arch Phys Med Rehabil Vol73, April 1992
patients go through a stage of agitation and restlessness as they recover from their head injuries.3-5 The literature is limited both by a lack of clarity about the definitions of agitation and restlessness and by small or biased sampling of subjects. This lack of data about the definition, frequency, and duration of agitation makes it difficult to study. Patients, families, and caregivers would like to know if it is likely to occur, if it will be severe, and if it will last long enough to justify treatment interventions. The frequency and duration must be known before research studies can be designed for treatment effectiveness, and before appropriate plans can be made for facility use, staff time, and resource allocation. The purpose of this study was to determine the frequency and duration of agitation and restlessness by systematically studying all the consecutive patients with closed head injury (CHI) who met predefined criteria for severity. The study was designed to define carefully and to separate restlessness and agitation. METHOD Subjects One hundred consecutively admitted subjects with traumatic CHI between 18 and 55 years of age were studied prospectively. All of the subjects had been admitted to a
AGITATION
AND RESTLESSNESS
Regional Level I Trauma Center, the only such facility in the four-state area, and which, therefore, receives almost all patients with moderate and severe traumatic brain injuries. Most subjects were injured in motor vehicle accidents which resulted in acceleration/deceleration injuries, and a few were injured in falls or altercations. The study included only patients with traumatic CHIs causing more than one hour of unconsciousness and an emergency room admission Glasgow Coma Scale score of less than 8. We excluded subjects who died, who had loss of consciousness for reasons other than traumatic CHI, who did not stay in the hospital at least one week, or who had a history of drug, alcohol, or psychiatric problems severe enough to require hospitalization at any point before this admission for CHI.
Measures Agitation was operationally defined as episodic motor or verbal behavior which interfered with patient care, or clearly required physical or chemical restraints to prevent damage to persons or property. The behavior had to be severe enough to be rated on the Overt Aggression Scale (OAS),(j require restraints, require medication or disrupt patient care. The OAS was completed by the trained research nurse based on patient observations (two or three times a week), chart review, and staff interviews. The OAS is a 16-item scale which assesses four categories of behavior: (I) verbal aggression: (2) physical aggression against objects: (3) physical aggression against selfi or (4) physical aggression directed at others. It requires evaluating whether each of 16 specific behaviors was observed. The measure was used in this way for each episode of agitation and the time and duration were noted. Data were also collected on the staff interaction with the patient, the use of medication, and the use of restraints. In this way, it was possible to define the agitation in objective, quantifiable terms. This also allowed us to document the episodic nature of agitation: interference with patient care, therapy, or safety: and in the future, effects of treatment. We defined restlessness separately as behavior which interfered or required some action on the part of staff, such as change of activity, but did not meet the criteria for agitation. Restlessness was either not severe enough or did not interfere with care, therapy, or safety at sufficient frequency to show up on the above measures. In addition. if the behavior was continuous rather than episodic, such as continuous pulling at a splint, it was also defined as restlessness. The Galveston Orientation and Amnesia Test (GOAT)’ was administered each week to assess the cognitive status of patients. The same research nurse observer completed this scale by interviewing the patient. The GOAT scores were then compared with the OAS agitation scores to see if agitation was predictably related to the degree of recovery of mental status. We did not collect more detailed measures of functional outcome or status at discharge.
AFTER CHI, Brooke
321
Subjects Review Committee. It was possible to include all of the 100 consecutive admissions that met the criteria by reviewing the list of admissions and by surveying each floor and interviewing the staff. Each patient was observed, patient and staff were interviewed, and medical records were reviewed two or three times each week. This process was continued through the patient’s acute neurologic surgery care and inpatient rehabilitation until discharge from the hospital.
RESULTS All 100 CHI subjects were admitted immediately after their head injuries and were successfully studied through their hospitalization, including inpatient rehabilitation, until stability of their medical condition warranted discharge either to a lesser level of care (eg. home or nursing home) or another rehabilitation facility. The numbers of patients remaining in the hospital by week postinjury were as follows: week I, 100: week 2,90; week 3.83; week 4, 34: week 5,25; week 6, 16; week 7, 15; week 8, 14: week 9.8; week 10,6; 11 weeks or more, 4. Data were gathered for a total of 184 days. The average age of subjects was 31. There were 87 men and 13 women. Subjects, generally, had severe head injuries. Only 50% were initially responsive to commands and, therefore, out of coma by the first week of hospitalization. Thirty-four subjects were still in the hospital at the end of four weeks. Only 65% of these 34 subjects were responsive to commands at the fourth week. Because less than 10% of the sample remained after 10 weeks, the data from subjects who stayed longer than 10 weeks were not analyzed.
Agitation Only 11 of the 100 patients exhibited episodic agitation. meeting the criteria described in the methods section. Eight of these patients were rated as agitated for one week. Five of those eight, however, were in the hospital for only one week. One patient showed agitation ratings for two weeks, one for three weeks, and one for four weeks. The subjects with agitation lasting two and four weeks showed this behavior at least once a week for their entire hospital stay. The onset of agitation postinjury occurred within one week in
Table 1: Subjects Remaining in Hospital by Week Postinjury Week
Number ______.--___
of Subjects
Procedure Subjects who met the admission criteria were studied according to the guidelines set forth by the University Human Arch Phys Med Rehabil Vol7’3, April 1992
AGITATION AND RESTLESSNESS AFTER CHI, Brooke
322
Table 2: Week of Onset Postinjury for Agitation (n = 100) Week Postinjury
Subjects
(n or 90) 6 3 1 1 TOTAL I I
I ? 5 4
six patients, within two weeks in three, within three weeks in one, and within four weeks in the 11th patient.
Restlessness We also looked at restlessness, which was defined as behavior which interfered with care, therapy, or safety, but either did not meet the severity criteria for agitation, or was continuous. Thirty-five patients showed this behavior at some point during their hospital stay. Twenty-three patients demonstrated it for only one week, three for two weeks, six for three weeks, two for four weeks, and one for six weeks. Restlessness developed in 29 patients within one week postinjury; in two patients within two weeks: and in four other patients within four, six, eight, and 10 weeks postinjury.
Subjects hospitalized more than 4 weeks We further examined the incidence of these behaviors in the subjects who remained in the hospital for four or more weeks. There were 34 subjects who required four or more weeks of hospitalization. Only three subjects (8.8%) demonstrated agitation, one for one week, one for three weeks, and one for four weeks. Restlessness occurred more frequently. Ten subjects (29.4%) were restless for one week, three subjects for two weeks, five for three weeks, three for four weeks, and one for six weeks. Eleven subjects did not show either behavior during their stay. The course of agitation over time was examined to see if agitation occurred at a predictable time or was related to a particular stage of responsiveness or mental status during recovery. The OAS agitation scores were compared with the GOAT mental status scores using a Spearman rank order comparison and there was no correlation between agitation and the degree of recovery of mental status. Subjects were then divided by the GOAT criteria into two groups: normal and abnormal orientation. There was no substantial or significant difference in the frequency of agitation in the normal and abnormal orientation groups. Only one subject was unresponsive at first, then became agitated, and finally was oriented and not agitated.
these studies, however, are limited by methodologic problems including definition of terms and the selection and size of the sample. Denny-Brown’ stated that “restlessness” is a natural part of the recovery process after closed head injury. Levin and Grossman’ described a constellation of behaviors that consisted of disinhibited movement, restlessness, thrashing, and aggressiveness, and these were exhibited by one third of head injured patients on a neurosurgery service. These behaviors required the use of restraints to prevent patient injuries. Corrigan and Mysiw3 studied 18 patients admitted to their rehabilitation facility for a period varying from two weeks to nine weeks of observation. They used the Agitated Behavior Scale to characterize patients as either “high agitation” or “low agitation,” and included many forms of restlessness in this definition. Items such as “short attention span, ” “uncooperative,” and “sudden changes of mood” were scored as being present in a slight, moderate, or extreme degree, or absent. They found that “10 (55%) of the 18 patients were highly agitated” when they began monitoring them. Four of these ten patients, or 22% of the total sample, were agitated “at the time the monitoring was discontinued.” They compared agitation to scores on the Orientation Group Monitoring System and Mini-Mental State, and found a significant correlation. They argued that the results provided some equivocal support for a period of confusion and agitation as a discreet stage of recovery from traumatic head injury. Reyes and associates4 studied 87 patients with traumatic head injury, admitted to their rehabilitation unit, for five years. They defined agitation as constant, uninhibited movement, and they defined restlessness as constant activity with the ability to briefly inhibit movement. This was a diverse group of patients, eight of whom were in prolonged coma and seven of whom were dying. They reported that I2 patients were agitated on admission and 32 were restless. They stated that these patients, who show agitation or restlessness, still improved. They also stated that restlessness and agitation were related to improved physical functioning at discharge, but inversely related to psychologic adjustment. Malkmus and associates’ described the phenomena of restlessness and agitation within the context of the Ranch0 Los Amigos Levels of Cognitive Functioning. Level IV is
Table 3: Week of Onset Postinjuty for Restlessness (II = 100) Subjects (n or o/u)
2 0 I
Arch Phys Med Rehabil Vol73, April 1992
TOTAL
0 I Y I 0 I 35
~~____
1 3 5
29
DISCUSSION Agitation has been described as a frequent complication seen in the recovering head injury patient. It is often viewed as a part of the natural recovery process. Some authors suggest that it is a stage that most patients go through. Some of
Week Postinjury __
4 5 6 7 8 Y IO ~__
____.
____-.
~_.~.._~~_
AGITATION
AND RESTLESSNESS
AFTER CHI, Brooke
323
dehned as confused-agitated, heightened activity. decreased having the behaviors this investigation distinguished as either restlessness or agitation. Corrigan and Mysiw3 reported ability to process new information and respond to events, aggressiveness, attempts to remove restraints or tubes, in- 55% with “high” agitation on referral and a decrease to 22% at the end of monitoring. Reyes and associates4 reported a ability to cooperate with treatment, and confabulation. form of agitation separate from restlessness in 14% of subThey stated that the confused-agitated patient has a markedly reduced abilitv to respond to simple commands. It is jects on referral. The operational definition of agitation and restlessness is important if the frequencies reported by this implied that agitation is a natural stage of recovery and investigation1 1% agitation and 35% restlessness-are to there is no differentiation of restlessness, confusion. and be compared to other studies of similar patient populations. agitation. Our study was different from other studies in several CONCLUSION ways. We examined 100 consecutive closed head injury admissions to a regional Trauma Center and included prospecThese results may be more representative of the experitively defined criteria for severity. Studies which examined ence of traumatic brain injury patients because our study only those patients referred to a secondary or tertiary re- was done with a consecutive sample from a predefined ferral center might be expected to be confounded by a numcatchment area. These results suggest that when agitation is ber of potential selection biases. Because the sample in our more carefully defined, its frequency and predictability study was limited to almost all of the moderate and severe may be less than generally believed. Most subjects did not head injuries for the region, the possibility of selection bias experience either agitation or restlessness according to our was minimized. This makes it more likely that these results criteria. Only 1 I % experienced agitation. Although the freare representative of the total population of CHI patients quency of restlessness was greater, neither agitation nor with injuries of similar severity. These results may not be restlessness lasted for more than a week’s hospital stay in applicable to other centers if the referral patterns are differmost cases. This is not to say that these behaviors are not ent. Further, there may have been characteristics of treatdistressing or important in patient management. This inment in our hospital which changed the frequency of agita- formation may be helpful and reassuring to patients, famition relative to that of the nation as a whole: for example, lies, and caregivers. Since physical and chemical restraints managing the environment by increasing or decreasing the have potentially negative side effects, their use should be amount of stimulation (eg, by limiting visitors, radio, or TV carefully justified since agitation may typically be a tranuse, in the room) to reduce any signs of discomfort is a sient behavior with low-frequency occurrence. Often it may standard part of treatment. Our results also cannot be combe more prudent to simply “wait out” episodes of agitation pared to those studies of agitation and restlessness in pa- if they are, indeed, transient in nature. In any case, it aptients with mild CHIs. pears that when agitation is more carefully defined and studAnother problem with the results of this study is the re- ied, it is not as universal or predictable a phenomenon in sult of the high attrition rate of subjects; approximately half patients recovering from traumatic CHI as it was thought the subjects were monitored for only the one or two weeks to be. that they were hospitalized. Patients who are discharged References from the hospital. usually because they had shown suffi1. Denny-Brown D. Disability arising from closed head injury. cient improvement in their medical condition, were no JAMA 1945; 127:429-36. longer included as subjects to be studied. The results might 7 have been different if all the subjects had been studied for a -. Levin HS. Grossman RG. Behavioral sequelae of closed head injury: a quantitative study. Arch Neural 1978:35:7X)-7. month or more. It is notable, however, that the frequency of 3. Corrigan JD. Mysiw WJ. Agitation following traumatic head agitation (8.8% vs 1 1%) and restlessness (29.4% vs 35%) in injury: equivocal evidence for a discreet stage of cognitive rethe 34 subjects that were in the hospital for a month or covery. Arch Phys Med Rehabil 1988;69:487-92. more was similar to that found in the total 100 patients. 4. Reyes RL. Bhattacharyya AK, Heller D. Traumatic head inThe duration of agitation, however, was substantially jury: restlessness and agitation as prognosticators of physical and psychologic improvement in patients. Arch Phys Med Relonger. but the small number (three) of agitated subjects in habil 198 1:62:20-3. the subgroup makes it difficult to draw any conclusions about agitated patients in general. Attrition may have seri- 5. Malkmus D, Booth BJ, Kodimer C. Rehabilitation of the head injured adult: comprehensive cognitive management. Dowously affected our estimate of the duration of agitation, if it ney. CA: Professional Staff Association of Ranch0 Los Amigos occurs. Further research designed to follow patients who Hospital, Inc., 1980. become agitated is needed to provide a more valid estimate 6. Yudofsky SC, Silver JM, Jackson W, lndicott J. Williams D. of agitation duration. The overt aggression scale for the objective rating of verbal and Another difference in this study was the careful distincphysical aggression. Am J Psychiatry 1986; 143:35-9. tion that was made between restlessness and agitation in 7. Levin HS, O’Donnell VM. Grossman KG. The Galveston orientation and amnesia test: a practical scale to assess cognition operational terms that were objective and quantifiable. after head injury. J Nerv Ment Dis 1979: 167:675-X4. Levin and Grossman2 reported one third of their subjects as
Arch Phys Med Rehabil Vol73, April 1992
Agitation and Restlessness After Closed Head Injury: A Prospective Study of 100Consecutive Admissions Marvin M. Brooke, MD, Kent A. Questad, PhD, David R. Patterson, PhD, Karen J. Bashak, MHSc ABSTRACT. Brooke MM, Questad KA, Patterson DR, Bashak KJ. Agitation and restlessness after closed head injury: a prospective study of 100 consecutive admissions. Arch Phys Med Rehabil 1992;73:320-3. l Agitation and restlessness are two of the most striking and problematic behaviors for patients with traumatic brain injury (TBI), their caregivers, and their families. These behaviors are often treated with physical and chemical restraints which have potentially harmful side effects. There are, however, few prospective studies which clearly define agitation and restlessness in a representative sample of TBI patients. Subjects for this study were 100 consecutive patients with traumatic, closed head injury (CHI) admitted to a regional Level I Trauma Center with a Glasgow Coma Scale score of less than 8, who had more than one hour of coma, and who required more than one week of hospitalization. Agitation was defined as episodic motor or verbal behavior which interfered with patient care or clearly required physical or chemical restraints to prevent damage to persons or property. This variable was rated on the Overt Aggression Scale, a lbitem scale, in four categories: verbal aggression; physical aggression against objects; physical aggression against self; or physical aggression directed at others. Systematic direct observations, caregiver interviews, and chart reviews were used to determine the frequency and duration of agitation. Patients were also monitored for restlessness, which was defined as behavior that interfered with staff or required some action by staff, such as change of activity, but either did not meet the severity criteria for agitation, or was continuous. Only 11 of the 100 subjects exhibited episodic agitation which met the criteria. Eight subjects were agitated for one week, one for two weeks, one for three weeks, and one for four weeks. Only one subject went directly from being unresponsive to being agitated. Restlessness, however, occurred in 35 patients. The results suggest that agitation is not frequently or predictably a phase of recovery from head injury. The precise definition and expected duration of these behaviors should be considered carefully before potentially harmful treatment is prescribed. 0 1992 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation KEY M.ORDS: Bruin injury; Ps~rhmotor
ugitution; Stati.stic.c
Agitation and restlessness are well-known sequelae of traumatic brain injury’,2 and are two of the most problematic behaviors to evaluate and treat. These behaviors are distressing to patients, families, and caregivers. They can also interfere with therapy and patient safety. Agitation and restlessness are, at times, managed with physical or chemical restraints, although clinicians should be cautious and aware of the potential harmful side effects of such interventions. There is little researchlm4which carefully defines these behaviors, accurately describes their frequency and duration, and is based on an adequate number of cases without sampling biases. This lack of information compounds the difficulties in assessment and management. Some research has suggested that traumatic brain injury From the Department of Rehabilitation Medicine, Tufts University School of Medicine (Dr. Brooke) and Department of Rehabilitation Medicine, University of Washington School of Medicine (Drs. Questad. Patterson. Ms. Bashak). This research was supported in part by the Harborview Injury Prevention and Research Center. Center for Disease Control grant CCR49-002570. and the National Institute on Disability and Rehabilitation Research grant GOO830076, Department of Education. Washington, DC. Submitted for publication May 22. 1990. Accepted in revised form February I I, 1991. No commercial party having a direct or indirect interest in the subject matter ofthis article has conferred or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to Marvin M. Brooke, MD, Department of Rehabilitation Medicine. Tufts University School of Medicine. I36 Harrison Avenue, Boston, MA 02 1 I I. GJ 1992 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation OcO3-9993/92/7304-0003$3.00/0
Arch Phys Med Rehabil Vol73, April 1992
patients go through a stage of agitation and restlessness as they recover from their head injuries.3-5 The literature is limited both by a lack of clarity about the definitions of agitation and restlessness and by small or biased sampling of subjects. This lack of data about the definition, frequency, and duration of agitation makes it difficult to study. Patients, families, and caregivers would like to know if it is likely to occur, if it will be severe, and if it will last long enough to justify treatment interventions. The frequency and duration must be known before research studies can be designed for treatment effectiveness, and before appropriate plans can be made for facility use, staff time, and resource allocation. The purpose of this study was to determine the frequency and duration of agitation and restlessness by systematically studying all the consecutive patients with closed head injury (CHI) who met predefined criteria for severity. The study was designed to define carefully and to separate restlessness and agitation. METHOD Subjects One hundred consecutively admitted subjects with traumatic CHI between 18 and 55 years of age were studied prospectively. All of the subjects had been admitted to a
AGITATION
AND RESTLESSNESS
Regional Level I Trauma Center, the only such facility in the four-state area, and which, therefore, receives almost all patients with moderate and severe traumatic brain injuries. Most subjects were injured in motor vehicle accidents which resulted in acceleration/deceleration injuries, and a few were injured in falls or altercations. The study included only patients with traumatic CHIs causing more than one hour of unconsciousness and an emergency room admission Glasgow Coma Scale score of less than 8. We excluded subjects who died, who had loss of consciousness for reasons other than traumatic CHI, who did not stay in the hospital at least one week, or who had a history of drug, alcohol, or psychiatric problems severe enough to require hospitalization at any point before this admission for CHI.
Measures Agitation was operationally defined as episodic motor or verbal behavior which interfered with patient care, or clearly required physical or chemical restraints to prevent damage to persons or property. The behavior had to be severe enough to be rated on the Overt Aggression Scale (OAS),(j require restraints, require medication or disrupt patient care. The OAS was completed by the trained research nurse based on patient observations (two or three times a week), chart review, and staff interviews. The OAS is a 16-item scale which assesses four categories of behavior: (I) verbal aggression: (2) physical aggression against objects: (3) physical aggression against selfi or (4) physical aggression directed at others. It requires evaluating whether each of 16 specific behaviors was observed. The measure was used in this way for each episode of agitation and the time and duration were noted. Data were also collected on the staff interaction with the patient, the use of medication, and the use of restraints. In this way, it was possible to define the agitation in objective, quantifiable terms. This also allowed us to document the episodic nature of agitation: interference with patient care, therapy, or safety: and in the future, effects of treatment. We defined restlessness separately as behavior which interfered or required some action on the part of staff, such as change of activity, but did not meet the criteria for agitation. Restlessness was either not severe enough or did not interfere with care, therapy, or safety at sufficient frequency to show up on the above measures. In addition. if the behavior was continuous rather than episodic, such as continuous pulling at a splint, it was also defined as restlessness. The Galveston Orientation and Amnesia Test (GOAT)’ was administered each week to assess the cognitive status of patients. The same research nurse observer completed this scale by interviewing the patient. The GOAT scores were then compared with the OAS agitation scores to see if agitation was predictably related to the degree of recovery of mental status. We did not collect more detailed measures of functional outcome or status at discharge.
AFTER CHI, Brooke
321
Subjects Review Committee. It was possible to include all of the 100 consecutive admissions that met the criteria by reviewing the list of admissions and by surveying each floor and interviewing the staff. Each patient was observed, patient and staff were interviewed, and medical records were reviewed two or three times each week. This process was continued through the patient’s acute neurologic surgery care and inpatient rehabilitation until discharge from the hospital.
RESULTS All 100 CHI subjects were admitted immediately after their head injuries and were successfully studied through their hospitalization, including inpatient rehabilitation, until stability of their medical condition warranted discharge either to a lesser level of care (eg. home or nursing home) or another rehabilitation facility. The numbers of patients remaining in the hospital by week postinjury were as follows: week I, 100: week 2,90; week 3.83; week 4, 34: week 5,25; week 6, 16; week 7, 15; week 8, 14: week 9.8; week 10,6; 11 weeks or more, 4. Data were gathered for a total of 184 days. The average age of subjects was 31. There were 87 men and 13 women. Subjects, generally, had severe head injuries. Only 50% were initially responsive to commands and, therefore, out of coma by the first week of hospitalization. Thirty-four subjects were still in the hospital at the end of four weeks. Only 65% of these 34 subjects were responsive to commands at the fourth week. Because less than 10% of the sample remained after 10 weeks, the data from subjects who stayed longer than 10 weeks were not analyzed.
Agitation Only 11 of the 100 patients exhibited episodic agitation. meeting the criteria described in the methods section. Eight of these patients were rated as agitated for one week. Five of those eight, however, were in the hospital for only one week. One patient showed agitation ratings for two weeks, one for three weeks, and one for four weeks. The subjects with agitation lasting two and four weeks showed this behavior at least once a week for their entire hospital stay. The onset of agitation postinjury occurred within one week in
Table 1: Subjects Remaining in Hospital by Week Postinjury Week
Number ______.--___
of Subjects
Procedure Subjects who met the admission criteria were studied according to the guidelines set forth by the University Human Arch Phys Med Rehabil Vol7’3, April 1992
AGITATION AND RESTLESSNESS AFTER CHI, Brooke
322
Table 2: Week of Onset Postinjury for Agitation (n = 100) Week Postinjury
Subjects
(n or 90) 6 3 1 1 TOTAL I I
I ? 5 4
six patients, within two weeks in three, within three weeks in one, and within four weeks in the 11th patient.
Restlessness We also looked at restlessness, which was defined as behavior which interfered with care, therapy, or safety, but either did not meet the severity criteria for agitation, or was continuous. Thirty-five patients showed this behavior at some point during their hospital stay. Twenty-three patients demonstrated it for only one week, three for two weeks, six for three weeks, two for four weeks, and one for six weeks. Restlessness developed in 29 patients within one week postinjury; in two patients within two weeks: and in four other patients within four, six, eight, and 10 weeks postinjury.
Subjects hospitalized more than 4 weeks We further examined the incidence of these behaviors in the subjects who remained in the hospital for four or more weeks. There were 34 subjects who required four or more weeks of hospitalization. Only three subjects (8.8%) demonstrated agitation, one for one week, one for three weeks, and one for four weeks. Restlessness occurred more frequently. Ten subjects (29.4%) were restless for one week, three subjects for two weeks, five for three weeks, three for four weeks, and one for six weeks. Eleven subjects did not show either behavior during their stay. The course of agitation over time was examined to see if agitation occurred at a predictable time or was related to a particular stage of responsiveness or mental status during recovery. The OAS agitation scores were compared with the GOAT mental status scores using a Spearman rank order comparison and there was no correlation between agitation and the degree of recovery of mental status. Subjects were then divided by the GOAT criteria into two groups: normal and abnormal orientation. There was no substantial or significant difference in the frequency of agitation in the normal and abnormal orientation groups. Only one subject was unresponsive at first, then became agitated, and finally was oriented and not agitated.
these studies, however, are limited by methodologic problems including definition of terms and the selection and size of the sample. Denny-Brown’ stated that “restlessness” is a natural part of the recovery process after closed head injury. Levin and Grossman’ described a constellation of behaviors that consisted of disinhibited movement, restlessness, thrashing, and aggressiveness, and these were exhibited by one third of head injured patients on a neurosurgery service. These behaviors required the use of restraints to prevent patient injuries. Corrigan and Mysiw3 studied 18 patients admitted to their rehabilitation facility for a period varying from two weeks to nine weeks of observation. They used the Agitated Behavior Scale to characterize patients as either “high agitation” or “low agitation,” and included many forms of restlessness in this definition. Items such as “short attention span, ” “uncooperative,” and “sudden changes of mood” were scored as being present in a slight, moderate, or extreme degree, or absent. They found that “10 (55%) of the 18 patients were highly agitated” when they began monitoring them. Four of these ten patients, or 22% of the total sample, were agitated “at the time the monitoring was discontinued.” They compared agitation to scores on the Orientation Group Monitoring System and Mini-Mental State, and found a significant correlation. They argued that the results provided some equivocal support for a period of confusion and agitation as a discreet stage of recovery from traumatic head injury. Reyes and associates4 studied 87 patients with traumatic head injury, admitted to their rehabilitation unit, for five years. They defined agitation as constant, uninhibited movement, and they defined restlessness as constant activity with the ability to briefly inhibit movement. This was a diverse group of patients, eight of whom were in prolonged coma and seven of whom were dying. They reported that I2 patients were agitated on admission and 32 were restless. They stated that these patients, who show agitation or restlessness, still improved. They also stated that restlessness and agitation were related to improved physical functioning at discharge, but inversely related to psychologic adjustment. Malkmus and associates’ described the phenomena of restlessness and agitation within the context of the Ranch0 Los Amigos Levels of Cognitive Functioning. Level IV is
Table 3: Week of Onset Postinjuty for Restlessness (II = 100) Subjects (n or o/u)
2 0 I
Arch Phys Med Rehabil Vol73, April 1992
TOTAL
0 I Y I 0 I 35
~~____
1 3 5
29
DISCUSSION Agitation has been described as a frequent complication seen in the recovering head injury patient. It is often viewed as a part of the natural recovery process. Some authors suggest that it is a stage that most patients go through. Some of
Week Postinjury __
4 5 6 7 8 Y IO ~__
____.
____-.
~_.~.._~~_
AGITATION
AND RESTLESSNESS
AFTER CHI, Brooke
323
dehned as confused-agitated, heightened activity. decreased having the behaviors this investigation distinguished as either restlessness or agitation. Corrigan and Mysiw3 reported ability to process new information and respond to events, aggressiveness, attempts to remove restraints or tubes, in- 55% with “high” agitation on referral and a decrease to 22% at the end of monitoring. Reyes and associates4 reported a ability to cooperate with treatment, and confabulation. form of agitation separate from restlessness in 14% of subThey stated that the confused-agitated patient has a markedly reduced abilitv to respond to simple commands. It is jects on referral. The operational definition of agitation and restlessness is important if the frequencies reported by this implied that agitation is a natural stage of recovery and investigation1 1% agitation and 35% restlessness-are to there is no differentiation of restlessness, confusion. and be compared to other studies of similar patient populations. agitation. Our study was different from other studies in several CONCLUSION ways. We examined 100 consecutive closed head injury admissions to a regional Trauma Center and included prospecThese results may be more representative of the experitively defined criteria for severity. Studies which examined ence of traumatic brain injury patients because our study only those patients referred to a secondary or tertiary re- was done with a consecutive sample from a predefined ferral center might be expected to be confounded by a numcatchment area. These results suggest that when agitation is ber of potential selection biases. Because the sample in our more carefully defined, its frequency and predictability study was limited to almost all of the moderate and severe may be less than generally believed. Most subjects did not head injuries for the region, the possibility of selection bias experience either agitation or restlessness according to our was minimized. This makes it more likely that these results criteria. Only 1 I % experienced agitation. Although the freare representative of the total population of CHI patients quency of restlessness was greater, neither agitation nor with injuries of similar severity. These results may not be restlessness lasted for more than a week’s hospital stay in applicable to other centers if the referral patterns are differmost cases. This is not to say that these behaviors are not ent. Further, there may have been characteristics of treatdistressing or important in patient management. This inment in our hospital which changed the frequency of agita- formation may be helpful and reassuring to patients, famition relative to that of the nation as a whole: for example, lies, and caregivers. Since physical and chemical restraints managing the environment by increasing or decreasing the have potentially negative side effects, their use should be amount of stimulation (eg, by limiting visitors, radio, or TV carefully justified since agitation may typically be a tranuse, in the room) to reduce any signs of discomfort is a sient behavior with low-frequency occurrence. Often it may standard part of treatment. Our results also cannot be combe more prudent to simply “wait out” episodes of agitation pared to those studies of agitation and restlessness in pa- if they are, indeed, transient in nature. In any case, it aptients with mild CHIs. pears that when agitation is more carefully defined and studAnother problem with the results of this study is the re- ied, it is not as universal or predictable a phenomenon in sult of the high attrition rate of subjects; approximately half patients recovering from traumatic CHI as it was thought the subjects were monitored for only the one or two weeks to be. that they were hospitalized. Patients who are discharged References from the hospital. usually because they had shown suffi1. Denny-Brown D. Disability arising from closed head injury. cient improvement in their medical condition, were no JAMA 1945; 127:429-36. longer included as subjects to be studied. The results might 7 have been different if all the subjects had been studied for a -. Levin HS. Grossman RG. Behavioral sequelae of closed head injury: a quantitative study. Arch Neural 1978:35:7X)-7. month or more. It is notable, however, that the frequency of 3. Corrigan JD. Mysiw WJ. Agitation following traumatic head agitation (8.8% vs 1 1%) and restlessness (29.4% vs 35%) in injury: equivocal evidence for a discreet stage of cognitive rethe 34 subjects that were in the hospital for a month or covery. Arch Phys Med Rehabil 1988;69:487-92. more was similar to that found in the total 100 patients. 4. Reyes RL. Bhattacharyya AK, Heller D. Traumatic head inThe duration of agitation, however, was substantially jury: restlessness and agitation as prognosticators of physical and psychologic improvement in patients. Arch Phys Med Relonger. but the small number (three) of agitated subjects in habil 198 1:62:20-3. the subgroup makes it difficult to draw any conclusions about agitated patients in general. Attrition may have seri- 5. Malkmus D, Booth BJ, Kodimer C. Rehabilitation of the head injured adult: comprehensive cognitive management. Dowously affected our estimate of the duration of agitation, if it ney. CA: Professional Staff Association of Ranch0 Los Amigos occurs. Further research designed to follow patients who Hospital, Inc., 1980. become agitated is needed to provide a more valid estimate 6. Yudofsky SC, Silver JM, Jackson W, lndicott J. Williams D. of agitation duration. The overt aggression scale for the objective rating of verbal and Another difference in this study was the careful distincphysical aggression. Am J Psychiatry 1986; 143:35-9. tion that was made between restlessness and agitation in 7. Levin HS, O’Donnell VM. Grossman KG. The Galveston orientation and amnesia test: a practical scale to assess cognition operational terms that were objective and quantifiable. after head injury. J Nerv Ment Dis 1979: 167:675-X4. Levin and Grossman2 reported one third of their subjects as
Arch Phys Med Rehabil Vol73, April 1992
