Use of Donepezil for Memory and Executive Disorders Following Severe Traumatic Brain Injury in Children and Adolescents

By: Peter D Patrick, Ph.D., University of Virginia School of Medicine. Jennifer Mabry, Ed.D., Radford University. Matt Gurka, Ph.D., West Virginia University. Marcia Buck, Pharm.D., University of Virginia School of Medicine. Howard Goodkin, Ph.D.,M.D., University of Virginia School of Medicine. Rob Rust, M.D., University of Virginia School of Medicine

Severe traumatic brain injury (TBI) in children and adolescents has been associated with prolonged adverse changes in cognitive and neurobehavioral recovery.  Review of literature would suggest that while sensory motor challenges are difficult, family and survivors of TBI endorse the significant frustration and limitations imposed upon them by alteration in the cognitive and behavioral competency when compared to their pre-injury state (Kade HD, 2009).

There have been multiple efforts to use remedial services, traditional mental health services, and more recently cognitive remediation resources to repair and restore cognitive and behavioral competency (Sohlberg MM, 2000, M Ylyvisaker, 2007, Laatsch, 2007).  Medications have also been used in their traditional manner to address emotional/motivational changes associated with recovery from TBI.  More recently there have been investigations exploring the cognitive enhancing use of various agents (Newsom, 2009).

Donepezil (Aricept®) is a peiperidine-based agent that is a reversible acetylcholinesterase inhibitor and has been used for its cognitive enhancing effects in adults.  Donepezil has repeatedly been found advantageous in graded restoration of memory function, especially when addressing changes in memory due to Alzheimer’s disease (M Shigeta, 2001).  Additional use of donepezil has been explored. Whelan, et al (FJ Whelan, 2000) investigated the use of donepezil with adult TBI as did Taverni (JP Taverni, 1998).  Both found improved mental function, full scale IQ, and clinician ratings.

Donepezil has recently been used “off-label” therapeutically in the pediatric population, particularly those diagnosed with ADHD (J Beiderman, 2000).  The manufacturers of Aricept® (donepezil), Eisai Inc., are seeking an indication for treatment of ADHD in children and adolescents.  In 2003, a study explored the use of donepezil in 43 children diagnosed with autism with favorable results (Michael G Chez, 2003).

C. W. Popper (2000) suggested that the use of donepezil in children may have more general effects than addressing just the core elements of attention/concentration, and may help where traditional stimulants do not by affecting “organizational capacity” and possibly other executive functions.  It may be the more general effects of choline systems on learning that are reflected in Popper’s comments.  Cholinergic systems of the brain have been found to be associated with memory and learning (DD Cardenas, 1987).  These systems both in animal models as well as in human clinical presentations have been associated with memory disorders.  In the adult population, widely recognizable disorders such as Alzheimer and delirium have been associated with hypo-cholinergic states (CM Smith, 1979).

Our study investigated the use of a choline agonist (Aricept®/donepezil) in children/adolescents following severe TBI with memory impairment and disorders of executive function.

Subjects

Following IRB reviewed consenting procedures, the Inclusion/Exclusion Criteria were

  1. Suffered a medically documented severe traumatic brain injury,
  2. initial Glasgow Coma Scale Score of 8 or less,
  3. 8 to 21 years of age,
  4. meeting criteria for frontal lobe syndrome (ICD9, 310.1),
  5. free of medications that would interfere with testing (sedation),
  6. no active medical co-morbid condition (e.g. seizures) and
  7. able to participate in neuropsychological examination.

Sixteen (16) subjects were enrolled, 12 subjects completed the full profile (one subject was withdrawn prior to starting any agents due to unstable seizures, three were withdrawn due to complaints of side effects, wherein one was found to have unrelated side effects due to other medications. When the research codes were broken, two of the three that were withdrawn due to side effects were, in fact, on placebo at the time of complaints).  Twelve completed the protocol with a total of seven males and five females.  Although we used a randomized assignment table which was administered from Pharmacy and PI/evaluators blinded, five males were assigned to the Arm 1 (medication followed by placebo) and Arm 2 (placebo followed by medication) contained five females and two males.

Table 1

  1. N=16 were enrolled
  1. One withdrew at baseline due to unstable seizures
  1. Three withdrew due to side effect complaints (One had unrelated SE, two were on placebo at the time they withdrew for what they thought were SEs)
  1. One withdrew following 20 weeks due to greater than 10% noncompliance with taking medication/placebo
  1. 12 completed full twenty week protocol
  1. 15 years average age ( 8 to 18 years of age)
  1. 6 males, 5 females
  1. 5 Arm 1 ( medication followed by placebo)
  1. 7 Arm 2 (placebo followed by medication)

Study Design

This was a twenty week randomized crossover prospective study following severe (as defined by initial Glasgow Coma Scale Score) TBI.  The twenty week trial has two phases: (Phase 1) Ten week active medication, (Phase 2) preceded or followed by a ten week placebo phase.  The subjects are randomly assigned by Pharmacy to one of two arms.  Arm 1 is ten weeks of active medication (five weeks at 5 mg per day, then based on side effects, five weeks at 10 mg per day). Arm 1 finishes on a ten week placebo phase.  Five subjects who were randomly assigned to Arm 1 completed the full protocol.  All were males.

Arm 2 is ten weeks of placebo followed by a ten week medication phase (5 mg per day, then 10 mg per day).  Seven subjects who were randomly assigned to Arm 2 completed the protocol, one male and six females.

Table 2: The Use of Donepezil in Recovery of Executive Function Skills with Children/Adolescents Following Traumatic Brain Injury

k

Data Collection

Consenting, baseline information gathering, and medical examination were conducted to determine admission criteria. 

Baseline medical examination consisted of review of chart, electrocardiogram, and neurological screening.  Baseline side effect measures were all conducted prior to medication or placebo arms.

Neuropsychological examination was conducted on weeks nine and eighteen.

  1. Memory testing using the Wide Range Assessment of Memory and Learning.
  2. Executive function testing using the selected subtests of Delis Kaplan Executive Function System.

Serial parent rating was also conducted on weeks 1, 5, 9, 17, and 19.

  1. Brief Rating Scale for Executive Functions.
Table 3: Memory and Executive Function Testing
  • D-KEFS (2001): Tests of executive skills
  1.  
    • Trail Making
  1.  
    • Verbal and Design Fluency
  1.  
    • Word Interference
  1.  
    • Sorting
  1.  
    • Tower Test
  • WRAML2 (2003): Tests of memory
  1.  
    • Design Memory
  1.  
    • Verbal Learning
  1.  
    • Story Memory
  1.  
    • Picture Memory
  • BRIEF (2000): Observational inventory
  1.  
    • Inhibit, Shift and Emotional Control = Behavior Regulation Index
  1.  
    • Initiate, Working Memory, Plan/Organize, Organization of Materials and Monitor = Metacognition Index
  1.  
    • All subtests = General Executive Composite

Results

Side Effects Findings:

Overall, donepezil was well tolerated by the study patients.  The incidence of minor adverse effects was similar during both the treatment and placebo phases. There were no serious adverse effects.  Three subjects withdrew from the study due to adverse effects, including two who were receiving placebo at the time of discontinuation.  One patient experienced bruising while on donepezil.  There has been a 5% incidence of mild bruising with donepezil during clinical trials, which is typically mild and resolves without dosage adjustment or discontinuation.  The patient was also taking sertraline/Zoloft®, which has also been associated with bruising or prolonged bleeding in isolated case reports.  The other adverse effects seen during donepezil use, but not with placebo, included abnormal dreams and diarrhea.  A 3% incidence of abnormal dreams was reported during premarketing studies with donepezil.  The incidence of diarrhea during clinical trials has ranged from 2-8% of adult patients.  In our patients, these adverse effects were mild and did not require withdrawal from the study.

Conclusion

  1. Overall well tolerated medication with no adverse events or serious effects when compared to placebo.
  2. Three subjects discontinued due to complaints of “side effects.”
    1. Subjects withdrawing due to side effects were either not on active agent (Two subjects)

or

  1.  
    1. Subjects found to have side effects due to other medication (One subject, Zoloft® bruising).

Table 4

Symptoms

Donepezil

Placebo

Insomnia

3

2

Dizziness

2

1

Headaches

0

3

Abnormal Dreams

2

0

Depression

1

1

Fatigue

0

1

Diarrhea

2

0

No Appetite

1

1

Increased Appetite

1

2

Increased Thirst

1

0

Bruising

1

1

Muscle Cramps

1

1

Syncope

1

0

Urinary Frequency

3

1

Rash

1

1

Irritability

1

2

 

Memory Findings

  1. No global medication effect on memory skills was seen when compared to placebo phase, as measured by the WRAML 2.
  2. However, Verbal Memory Composite score on medication differed from placebo phase, demonstrating a trend towards significance (p=0.11).
    1. Active medication Composite Verbal             91.8 (average range)
    2. Placebo medication Composite Verbal           78.3 (borderline range)
    3. Arm 2: Following placebo, and on active agent, went from borderline range to the low average range.  Composite Verbal Index improved from 78.3 to 85.

Diagram 1

w

Executive Functions Findings

  1. Parent Rating (BRIEF-Weeks 1, 5, 9, 13, 17, 19)
    1. No relationship between parent report and medication or placebo phases.
      1. Parents reported improvement on placebo as well as reduced skills while on medication.
      2. “Placebo Effect by Proxy”
  2. Delis Kaplan Executive System Testing
    1. No global benefits seen while on medications.
    2. However, design fluency improvement trending toward significance.
  3. Design Fluency Composite, the continuous ability to generate novel design copy over one minute, when incorporated within a model of executive functions approached significance with a p=0.074.
  4. The selective effect on design fluency was further seen when the continuous design production task required an inhibition of visual distractions (Condition 2) which again was within the executive function model a trend p=0.079.

Discussion

Our present study is consistent with our overall research theme, which explores the use of pharmacological agents to hasten recovery of neurocognitive skills following acquired brain injury.  The current findings, while not as robust as other published findings, continues to support/suggest there are existing agents that may positively impact neuropsychological recovery.  The complementary role of science/research is to help find the rules of application for Aricept and other agents used in this population.

Our study, while under powered, also suggests that dosing and length of Aricept® use in this population needs further investigation.  The lack of statistical significance may be related to heterogeneity of the sample, length of time on Aricept® and/or dosing schedules.  The expected impact on memory return and even more importantly the possible use of this agent on subroutines of executive functioning require ongoing study and investigation.

It is a central theme of our research and work that the complementary relationship between remedial (instructional) interventions and psychopharmacological agents (biological) be further developed and explored.

This study finding suggests there is a subset of children with severe TBI who may improve certain forms of memory with the use of Aricept®.  This may be a particularly important complement to existing remedial treatments for a most difficult area of recovery.  Also, there is the suggestion that some of the subroutines of executive functioning, behavior fluency, may also benefit from pharmacological intervention.  We examined the ability to maintain a sustained behavior response.  The literature has been very specific regarding the way psychotropic agents have an effect on various neuropsychological subroutines of the brain.  It has been pointed out routinely that agents may be very specific and focused in their effects on subroutines rather than global impact.  There has been no report of cognitive enhancing medications having a global or overriding effect on neuropsychological routines.  Instead, it appears that agents have a more particular impact on a given subroutine.  Further exploration of Aricept®’s benefits should continue to be weighed in addition to monitoring safety and side effects in this population.

Another issue that is brought into focus is the relationship between changes that may be clinically significant but not reach the “statistically significant” threshold.  As seen in this sample, results did not meet modern day “statistical significance”, but the placebo group on average was in the borderline range of performance, while the group on Aricept on average was within the normal limits.  While the differences between the two groups was less than needed for “significance” there were within the groups differences possibly pointing the way to importance.  For example, the cross over design model was helpful in this study because when the “placebo group” was switched to active agent their average score on verbal memory improved from borderline range performance to lower average range performance, but again not reaching “significance”.  While there are other influences, “practice effect” change occurred going from placebo to active agent warrants further consideration.  The improvement noted when moving to an active agent may be of even more significance given the decline in average performance of the active agent group (reducing the role of “practice effect”) that occurred in memory when study patients’ were converted to placebo and retested.

The difficulties of doing rehabilitation research with existing models for investigation have most recently received much attention (Mark Johnston, 2009).  In our study population, most samples are unavoidably heterogeneous given the nature of TBI injuries.  Also, small N trials are common and in combination with their heterogeneity, it is very difficult to find “group effects”.  For now, repeated single subject designs may find a place in our research, although the traditionalist reviewing grants and papers may not see this as strong evidence.

As an area of investigation we find the exploration of psychotropic agents’ influence on neuropsychological recovery exciting and necessary.  Instructional remedial approaches have been used for years to restore various cognitive, memory and perceptual functions. It would be an additional benefit to rehabilitation if agents that worked directly on the neurobiological substrate/networks were revealed further and clarified, and the rules of application could be scientifically determined.  This exciting area of learning about the effects and the rules of application for psychotropic agents remains a central feature of our work.

Bibliography

  • CM Smith, S. M. (1979). Physostigmine in Alzeheimer's disease. Lancet , 6(1):42.
  • DD Cardenas, M. A.-R. (1987). Oral physostigmine and impaired memory in adutls with brain injury. Brain Injury , (1):57-59.
  • Newsom M, Scheibel R, Seignovel P, Steinberg J, Troyanskaya M, Xiaoqi L, Levin H(2009) Effects of Methylphenidate on Working Memory in Traumatic Brain Injury: A Preliminary fMRI Investigation, Brain Imaging and Behavior Vol 3 (3) 298-305
  • FJ Whelan, W. M. (2000). Donepezil in the tratment of cogntiive dysfunction associated with traumatic brain injury. Annals of Clinical Psychiatry , 12(3):131-135.
  • J Beiderman, T. S. (2000). Non-stimulant treatments for ADHD. European Child & Adolescent Psychiatry , 1:151-159.
  • JP Taverni, G. S. (1998). Donepezil-mediated memory improvement in trumatic brain injury during post acute rehabilitation. Brain Injury , 12(1):77-80.
  • Laatsch, H. D. (2007). An evidence-based review of cognitive and behavioral rehabilitation treatment studies in children with acquired brain injury. J Head Trauma Rehab, vol 22 , 248-256.
  • M Shigeta, A. H. (2001). Donepezil for Alzheimer's disease: Pharacodynamic, pharmacokinetic and clnical profiles. CNS Drug Review , 7(4): 533-568.
  • M Ylyvisaker, L. T. (2007). Behavioral interventions for children and adults with behaviour disorders after TBI: A systematic review of the evidence. Brain Injury , 21(8): 769-805.
  • Mark Johnston, V. G. (2009). The Challenge of Evidence in Disability Resarch and Practice. Austin Texas: National Center for Dissemination of Disability Research.
  • Michael G Chez, B. T. (2003). Donepezil hydrochloride: a double-blind study in autistic children. Journal of Pediatric Neurology , 1(2):83-88.
  • Popper, C. (2000). Pharmacologic alternatives to psychostimulants for the treatment of attention-deficit/hyperactivity disorder. Child Adloescent Psychiatry Clin N Am , 9(3):605-646.
  • Sohlberg MM, M. K. (2000). Evaluation of Attention Processing Training and Brain Injury Educaiton in Persons with Acauired Brain Injury . Journal of Clinical and Experimental Neuropsychology , 665-676