Descriptive Analysis of Children and Adolescents in a Low Response State Following Severe Brain Injury Following Severe Brain Injury

By: Peter D. Patrick, Ph.D., Julia Buch Wamstad, Psy.D., Kenneth Norwood, M.D., Jodi Darring, B.S., Lisa Cantore, NP, RN, BSN, MSN, CCRN, James Blackman, M.D. M.P.H.

At the University of Virginia Kluge Children’s Rehabilitation Center, a primary focus is on addressing the needs of children and adolescents who cannot, or do not, regain functional arousal, awareness and communication following a severe brain injury. These children remain in a static, prolonged low response state characteristic of Rancho Los Amigos Levels I, II and III. To enter the program they must have been in this state for a minimum of thirty days. Because of their low response state they do not “qualify for rehabilitation” under the typical criteria from most insurance companies. This often results in difficult choices for the acute medical team and families.

We have not used terms like vegetative or minimally conscious at KCRC since 1998 due to the controversy at the time. We have used RLA scores as a more descriptive level of identification. However, Table 1 provides the percentages of children who would meet criteria for three traditional groupings, namely, vegetative, minimally conscious and akinetic mutism (Cartlidge, 2001). 100% of the comparison group had admitting RLA above 3. 

Table 1

Diagnosis

Percentage of patients

Vegetative State

41%

Minimally Conscious State

44%

Akinetic Mutism

14%

Of the LRG patients who were admitted with a RLA of 2, 75% progressed to a RLA of 3 or more during their inpatient hospitalization, 13% did not progress beyond a RLA of 2, and discharge RLA data for 13% was not available.

The KCRC low response protocol admits the child into an “acute medical bed” and initially addresses comorbid medical conditions or iatrogenic factors that may be contributing to their “complicated emergence” from the low response state. Once comorbid or iatrogenic factors are addressed, some children will spontaneously recover functional mental skills and move through the RLA levels (Figure 1). However, the large majority of the children we see remain in a low response state even after these factors have been minimized or eliminated. These children are considered to be in a “neuropathic state” that is the primary influence in their failure to regain basic cognitive skills (Patrick, 2006). Children in a persistently low response state are considered for pharmacological trials to improve arousal and awareness. When appropriate, and with family consent, the children are started on a six to eight week trial of a dopamine enhancing medication - either amantadine or mirapex if they were determined to be in a vegetative state or minimally conscious state.  Children who were diagnosed with akinetic mutism were treated with Ritalin/Concerta. There are patients in our data base who progressed from amantadine or mirapex and once demonstrating awareness and communication were moved to Ritalin/Concerta due to an adynamic profile. Throughout the medication trial, the children are monitored for side effects. Changes are assessed using the Western NeuroSensory Stimulation Profile (Ansell, 1989) (administered by a speech-language pathologist) and serial neurobehavioral assessments (administered by a neuropsychologist). 

In addition to the pharmacological trial, the children receive a full array of medical, nursing and traditional therapies (occupational, physical, and speech therapy) as part of their activities. While addressing medical need, basic therapies also address range of motion, positioning and daily stimulation.

 Figure 1

 

 

The goal of this study is to compare subjects who were in a low response state to subjects who sustained TBIs or ABIs, but who were not in a persistently low response state (gained RLA IV or more within 30 days post injury). In addition, we were interested in comparing the subjects who were admitted in a low response state and eventually progressed to rehabilitation to subjects who were admitted in a low response state but were not able to improve enough to progress to rehabilitation.

Comparison of patients in a low response state to a comparison group

We currently have a database of 59 children (39 males, 20 females) recently enrolled in our low response protocol. Fifty subjects in the low response group (LRG) sustained traumatic brain injuries (TBI) and 9 subjects sustained other types of acquired brain injury (ABI), including asphyxias. The average earliest recorded GCS was 3.97. According to Grados (2001), children with deep brain lesions after TBI are likely highly vulnerable to neurological and functional disability. We used the Grados classification system (Table 2) to determine the depth of lesion for each patient. Children in the LR group had a mean Grados of 4.33, while children in the comparison group had a mean Grados of 3.09. Average age at admission to the protocol was 15.4 (range 1.4-22.5). On average, subjects were 1120.5 days post injury (range 44.0-3650.0 days). Thirty eight (64%) subjects recovered sufficiently to qualify for rehabilitation services, while 21 (36%) were not able to regain functional mental skills that met criteria for rehabilitation. 

 

Table 2. Grados Depth of Lesion Ratings

Grados Rating

Depth of Lesion Group

1

Frontotemporal

2

Frontotemporal/Brain Stem-cerebellum

3

Corpus callosum

4

Basal ganglia

5

Thalamus

6

Brain stem-cerebellum

All 59 children had medical comorbidities that required treatment. In addition to medications to improve cognition, subjects were on a variety of other medications, including agents for spasticity (63%), seizures (46%), and storming (42%). Examples of medications used to treat spasticity include baclofen and diazepam. “Storming” medications were used for a cluster of symptoms including hypertension, tachycardia, and hyperthermia (see reference for Paroxysmal Autonomic Instability with Dystonia in Blackman, 2004 and Diagnosing Dysautonomia after TBI in Baguley, 2009). 

Of the LRG, fifty five (93%) were admitted into the pharmacological trial phase. The additional 4 subjects either spontaneously improved prior to medication introduction or had other medical complications that did not allow for the pharmacological trial phase.  

The children in our database were compared to a comparison group of patients who also sustained TBIs or ABIs, but who were not in a persistently low response state (gained RLA IV or more within 30 days post injury) (Tables 3 and 4). The groups were similar in age/age distribution and gender distribution. When comparing the LRG to the comparison group, there were some interesting differences. The LRG sustained more severe injuries based on initial GCS. Also, the LRG spent significantly longer as “acute” patients (M=59.7 days) than the comparison group (M=5.4 days). The LRG had more subjects on medications for seizures, spasticity, and storming than the comparison group.

Table 3. Demographics

 

Low Response

Comparison

N

59

41

Gender

39 Males, 20 Females

26 Males, 15 Females

Age at Injury

15.04 years

14.65 years

Injury Type

50 TBI, 9 ABI

34 TBI, 7 ABI

Average GCS

3.97

6.88

Grados

4.33

3.09

Enrolled in rehabilitation

64%

100%

Days Post Injury

1120.46 days

979.13 days

Acute LOS

59.73 days

5.37 days

Rehab LOS

33.36 days

28.78 days

Total LOS

94.24 days

34.15 days

 

Table 4. Medical Comorbidities

 

Low Response

Comparison

Medical Comorbidities

100%

93%

LR Medication

93%

0%

Seizure Medication

46%

32%

Spasticity Medication

63%

7%

Storming Medication

42%

10%

 

Comparison of patients in a LR state who progressed to rehabilitation and those who did not progress to rehabilitation

Thirty eight of the LRG (64%) entered rehabilitation services. In order to qualify for rehabilitation services, children must be able to sufficiently participate with therapists as well as tolerate the required amount of therapy. All but two children who qualified for rehabilitation had a TBI.

LRG subjects who went to rehabilitation (LRG+rehab) were compared to those who did not go to rehabilitation or who only had very short rehabilitation stays and were then disqualified due to medical complications (LRG-rehab) (Table 5). The groups were comparable with initial GCS 4.00 and 3.90 as well as Grados Levels 4.31 and 4.37. All subjects in the LRG had other medical comorbidities. However, LRG-rehab subjects were more likely to be taking medication for seizures, spasticity, and/or storming than LRG+rehab, with the biggest difference noted with medication for spasticity.

Table 5. Rehabilitation status comparisons

 

LRG+rehab

LRG-rehab

Comparison

N

38

21

41

GCS

4.00

3.90

6.88

Grados

4.31

4.37

3.09

Total LOS

95.38 days

92.30 days

34.15 days

Medical Comorbidities

100%

100%

93%

Low Response Medication

97%

86%

0%

Seizure Medication

42%

52%

32%

Spasticity Medication

58%

71%

7%

Storming Medication

42%

43%

10%

In order to compare functional outcomes of children in the LRG compared to the comparison group, we used the WeeFIM instrument. The goal of the WeeFIM instrument is to “measure changes in function over time to weight the burden of care in terms of physical, technologic, and financial resources” (Braun, 1991).

The WeeFIM scores of LRG+rehab were compared to the LRG-rehab group (Table 6). The LRG-rehab group had lower scores on all WeeFIM measures. We also compared rates of change over days of hospitalization using the WeeFIM scores (Table 7). While all patients generally show rates of change, LRG-rehab subjects improved at a slower rate than LRG+rehab.

Table 6. Mean WeeFIM Scores

 

Rehab

No Rehab

Comparison

Admit WeeFIM Total

29.59

20.78

46.05

Admit WeeFIM Self Care

11.67

8.00

18.07

Admit WeeFIM Mobility

8.06

6.22

11.54

Admit WeeFIM Communication

10.74

6.44

16.44

DC WeeFIM Total

56.20

35.22

88.80

DC WeeFIM Self Care

24.50

13.22

39.39

DC WeeFIM Mobility

16.67

10.67

24.68

DC WeeFIM Communication

17.17

11.33

24.73

Table 7. WeeFIM Rates of Change

 

Rehab

No Rehab

Comparison

WeeFIM Total

.31

.14

2.31

WeeFIM Self Care

.16

.05

1.16

WeeFIM Mobility

.09

.04

.73

WeeFIM Communication

.07

.05

.42

The WNSSP can be used for treatment planning, monitoring changes in performance, diagnosing deficits, and studying recovery patterns (Ansell, 1989). We have found the WNSSP to be an instrument that is sensitive and specific to readiness and response in therapy (Patrick, 2009).          

The WNSSP scores of LRG+rehab were compared to the LRG-rehab group (Table 8). While the LRG-rehab group had lower DC WNSSP scores (16.94 point difference) than the LRG+rehab group, it is noteworthy that the two groups change at about the same rate over the course of their hospitalization, according to therapist ratings (.37 and .33) (Table 9).

Table 8. Mean WNSSP Scores

 

Rehab

No Rehab

Comparison

Admit WNSSP

15.43

15.60

45.57

DC WNSSP

58.36

41.42

109.20

Table 9. WNSSP Rate of Change

 

Rehab

No Rehab

Comparison

WNSSP Rate of Change

.37

.33

.61

Post acute services received after discharge from inpatient hospitalization

After sustaining a TBI, patients often have many appointments to attend after they are discharged from inpatient hospitalization. We looked at the types of post acute services that the LRG and the comparison group received both less than 30 days post KCRC discharge (short term phase) as well as more than 30 days post KCRC discharge (long term phase). We looked at a total of 37 different possible subspecialties.

Short Term Phase Visits (<30 days post KCRC discharge)
Within the first 30 days post KCRC discharge, the most common outpatient visits did not differ much between the LRG and the comparison group. Within this period, the most common outpatient visits for the LRG are OT, PT, and speech therapy, as well as neurosurgery and pediatric surgery. Similarly, within 30 days post KCRC discharge, the most common outpatient visits for the comparison group were OT, PT, and speech therapy, as well as neurosurgery.

In the short term phase, patients in both the LRG and the comparison group saw anywhere from 0-8 different subspecialties. The average number of subspecialties seen by the LRG was .93 and the average number of subspecialties seen by the comparison group was 1.17.

While the most common outpatient visits did not differ much between the LRG and the comparison group, we have captured the subspecialties that differed more than 5% in the percentage of visits from each group (Table 10).

Table 10. Noteworthy differences in rate of subspecialty visits <30 days post KCRC discharge

 

Low Response

Comparison

OT Treatment

5.1%

12.20%

Speech Treatment

3.4%

9.76%

Neurosurgery

8.5%

24.39%

Long Term Phase Visits (>30 days post KCRC discharge)
After 30 days post KCRC discharge, the most common outpatient visits for the LRG included OT, PT, and the Acquired Brain Injury (ABI) clinic. The most common outpatient visits for the comparison group were also OT, PT, and the ABI clinic, as well as neurosurgery clinic visits.

The number of outpatient appointments that patients in the LRG and the comparison group attended did not differ significantly. In the long term phase, patients in the LRG saw between 0-16 different subspecialties (M=5.31), and patients in the comparison group saw between 0-15 different subspecialties (M=5.95).

Other noteworthy differences between the LRG and the comparison group are described in Table 11.

 

 

Table 11. Noteworthy differences in rate of subspecialty visits >30 days post KCRC discharge

 

Low Response

Comparison

No post acute care at UVA

22.0%

9.76%

OT Evaluation

50.8%

56.10%

OT Treatment

33.9%

48.78%

Speech Evaluation

44.1%

53.66%

Neurosurgery

11.9%

29.27%

ABI Clinic

55.9%

82.93%

Neuropsychology clinic visit

39.0%

56.10%

PM&R

10.2%

0.00%

Surgery

10.2%

4.88%

Ophthalmology

20.3%

36.59%

Nutrition

11.9%

17.07%

Radiology

1.7%

7.32%

Dentistry

16.9%

4.88%

Lessons Learned

We have had great interest in understanding the course of events for the child/adolescent in a persistent low response state. In addition to our modest research efforts, we now have a database system to collect ongoing differences and similarities for the low response child compared to a child or adolescent who sustains a traumatic brain injury but is not in a prolonged low response state. The effort to measure clinical changes over time and to better understand the distribution of services will hopefully better prepare us to respond to the clinical decision making, as well as, family and patient needs.

  • 64% of patients who come to KCRC in a low response state are eventually able to progress to rehabilitation
  • When compared to a comparison group, patients in a low response state spend longer as “acute status” patients. However, once they progress to rehabilitation, the time spent in rehabilitation does not differ from the comparison group.
  • When compared to a comparison group, patients in the LRG are on more medication to treat storming. Within the LRG, whether or not a child is on medication for storming does not appear to predict whether or not they are able to participate in rehabilitation.
  • Patients in the LRG are much more likely to be on medication for spasticity. Within the LRG, patients who did not progress to rehabilitation were more likely to be on medication for spasticity than those who did progress to rehabilitation. This observation suggests that more aggressive and systematic management of tone abnormalities may improve outcome.
  • When patients who are initially in a low response state enter rehabilitation, they do make progress. However, their rate of change is slower when compared to a comparison group. This implies that this patient population will need intense rehabilitation for a longer period of time in order to make similar gains to the comparison group. This finding may also indicate that there is a group of children who do not progress to rehabilitation during the initial hospitalization but who would benefit from rehabilitation at a later date. Unfortunately, a later admission to rehabilitation is often not allowed by insurance plans, a gap in coverage that should be addressed.
  • Upon discharge from rehabilitation, patients who were in a LR state see about the same number of subspecialties, and for the most part, there are no major differences in the types of subspecialties most often seen.

 

References

Ansell, B.J., Keenan, J.E., & De La Rocha, O. (1989). Western Neuro Sensory Stimulation Profile (Tustin, California: Western Neuro Care Center).

Baguley, I.J., Nott, M.T., Slewa-Younan, S., Heriseanu, R.E., & Perkes, I.E. (2009). Diagnosing dysautonomia after acute traumatic brain injury: Evidence for overresponsiveness to afferent stimuli. Archives of Physical Medicine and Rehabilitation, 90(4):580-586.

Blackman, J.A., Patrick P.D., Buck, M.L., & Rust, R.S. (2004). Paroxysmal Autonomic Instability with Dystonia after brain injury. Archives of Neurology, 61:321-328.

Braun, S., & Granger, C.V. (1991). A practical approach to functional assessment in pediatrics. Occup Ther Pract, 2:46-51.

Cartlidge, N. (2001). States related to or confused with coma. J Neurol Neurosurg Psychiatry, 71(supp I):i18-i19.

Grados, M.A., Slomine, B.S., Gerring, J.P., Vasa, R., Bryan, N., & Denckla, M.B. (2001) Depth of lesion model in children and adolescents with moderate to severe traumatic brain injury: Use of SPGR MRI to predict severity and outcome. Journal of Neurology, Neurosurgery, and Psychiatry 70(3):350-358.

Patrick, P.D., Blackman, J.A., Mabry, J.L., Buck, M.L., Gurka, M.J., & Conaway, M.R. (2006)Dopamine agonist therapy in low-response children following traumatic brain injury. Journal of Child Neurology 21(10):879-885.

Patrick, P.D.,  Wamstad, J.B., Mabry J.L., Smith-Janik, S, Gurka, M.J., Buck, M.L., Blackman, J.A.. (2009) Assessing the relationship between the WNSSP and therapeutic participation in adolescents in low response states following server traumatic brain injury. Brain Injury 23(6):528-534.