Abstract
Attention deficits are among the most common and persistent impairments resulting from traumatic brain injury (TBI). This study was the first to examine the effects of lisdexamfetamine dimesylate (LDX, Vyvanse) in treating TBI-related attention deficits in children. It was an extension of a previous controlled trial with adults. This was a 12-week, randomized, double-blind, placebo-controlled, dose-titration, crossover trial. In addition to weekly safety monitoring, there were assessments on a broad range of neuropsychological and behavioral measures at baseline, 6-weeks, and 12-weeks. A total of 20 carefully selected children were enrolled, ranging from 10 to 16 years of age. The sample consisted of cases with mainly mild TBI (based on the known details regarding their injuries), but they had persisting attention deficits and other post-concussion symptoms lasting from 2 to 29 months by the time of enrollment. A total of 16 children completed the trial. One of the children withdrew due to a mild anxiety reaction while on LDX. There were no other adverse effects. Positive treatment results were found on both formal testing of sustained attention and in terms of parent ratings of attention, emotional status, behavioral controls, and various aspects of executive functioning. The findings also served to highlight broader insights into the nature of attention deficits and their treatment in children with TBI.
Keywords:
traumatic brain injury;
concussion;
attention deficits;
children;
stimulant medication treatment
1. Introduction
New-onset or acquired attention deficits have been observed in both children and adults following traumatic brain injury (TBI). In a study by Levin et al. [
1], increased rates of newly diagnosed attention-deficit/hyperactivity disorder (ADHD) were found in children post-TBI (ranging from 14.5% at 12 months to 18.3% at 24 months). The rates would have been higher if selective symptoms were considered rather than requiring that the full criteria for ADHD be present. Additional studies have established a significant causal link between TBI and attention deficits in children, with prevalence rates ranging from 20–46% and with persisting deficits lasting 4–10 years or more [
2,
3]. Important moderating variables have included severity and location of injury, age at the time of injury, IQ, and psychosocial factors.
The underlying mechanisms producing attention deficits post-TBI may be conceptualized in various ways. Injury to specific areas may be involved, as consistent with models of attention components and their mediation by different regions of the brain. Mirsky et al. [
4] articulated a four-component model of attention (focus-execute, sustain, encode, shift) that has been applied widely in ADHD research. It also has been validated as applicable to children with TBI, with the underlying components affected to varying degrees depending on factors like severity of injury [
5]. The particular pattern of impairment based on this model has also been found to differ between children with TBI versus idiopathic ADHD [
6].
In a study using functional magnetic resonance imaging (fMRI), Kramer et al. [
7] examined long-range outcomes with respect to attention processing in children who sustained moderate-severe TBI in early childhood versus a group of age-matched children with orthopedic injuries. The children with TBI were found to activate similar networks of brain regions relevant to attention, albeit to a significantly greater extent in particular frontal and parietal regions compared to the controls. This may be viewed as suggesting a pattern of persistent compensatory activation in response to injury of underlying components.
Treatment of attention processes with psychostimulant medication may also be conceptualized in various ways. In an fMRI study of adults with non-TBI-related ADHD, Bush et al. [
8] found that psychostimulant medication (methylphenidate) produced increased activation in the dorsal anterior mid-cingulate cortex and dorsolateral prefrontal and parietal cortex, thereby, normalizing what ordinarily may be hypo-functioning of these regions in ADHD. Based on the Kramer et al. study noted above, it may be hypothesized that compensatory activation occurs, to some extent, naturally in TBI-related attention deficits, possibly especially in children. Stimulant medication may help to facilitate the process of compensatory activation, albeit in a more targeted or efficient fashion, such as by targeting dopamine transmission and synaptic plasticity in fronto-striatal regions [
9]. Alternatively, stimulant medication for acquired attention deficits may serve to activate secondary or backup neural circuits relevant to attention regulation. Or, rather than activating focusing or inhibitory mechanisms, per se, there may be stimulant action on general alertness and arousal. This is especially relevant in TBI in that fatigue often arises due to increased effort or exertion needed in executing normal tasks.
A study by Tramontana and associates [
10] examined the effects of lisdexamfetamine dimesylate (LDX, Vyvanse) in treating attention deficits in adults due to moderate to severe TBI. It was one of the most rigorous studies in this area and was the first controlled study using LDX with this population. Positive treatment effects were found involving various measures of sustained attention, working memory, response speed stability and endurance, and in aspects of executive functioning. No major problems with safety or tolerability were observed. Treatment was for only a six-week period (it was a 12-week, randomized, double-blind, placebo-controlled, dose-titration, crossover trial), but was enough to show impact on areas beyond narrowly defined aspects of attention. Conceivably, with more stable and regulated attention, individuals with TBI may be better able to derive benefit from other interventions, including therapies targeting other cognitive and behavioral areas affected.
The previous trial dealt with individuals with TBI ranging from 16 to 42 years of age. The present study was a similar trial but, instead, targeted children from 6 to 16 years old. Children in that age range comprise a major portion of the TBI population [
11], with most of the injuries resulting from falls, recreational activities (including sports concussions), as well as motor vehicle accidents (MVAs) and pedestrian-MVAs. Some are victims of violence. Impairments in attention can have a major adverse effect on learning and behavioral adjustment. Arguably, left untreated, these deficits in children can have an even more life-altering effect on achievement and future success than with adults.
Thus far, there have been few studies examining the use of stimulant medication in treating TBI-related attention deficits in children [
12,
13,
14,
15,
16]. One of the better controlled studies was a recent investigation by LeBlond and colleagues [
16]. Positive outcomes were reported on a range of performance and behavioral measures after four weeks of treatment on methylphenidate (MPH) vs. placebo. For the most part, however, there have been methodological problems characterizing the studies done here, including limitations in sample size and the scope of outcomes examined. The findings have tended to be weak or inconsistent. Few studies were controlled trials. Nearly all focused specifically on MPH as the stimulant treatment. The limited nature of the studies here stands in contrast to the growing body of evidence noted above documenting the prevalence and persistence of attention deficits due to TBI in children.
The present study was intended to help address this. It was a controlled clinical trial examining the effects of LDX (Vyvanse) in the treatment of TBI-related attention deficits in children (ClinicalTrials.gov, # NCT02712996, registration as a randomized controlled trial). It followed a design and methodology similar to those in the study by Tramontana et al., albeit adapted to a child sample. However, unlike the adult trial, subject selection was extended to include cases with milder TBI. Even if the attention deficits in milder cases may not prove to be chronic, it was thought that helping to enhance functioning could have a beneficial effect on overall outcomes and limit secondary problems that might otherwise arise. Another revision was the shortening of the minimum required time post-injury from six to two months. This was intended to capture child subjects earlier in the recovery process. It was also aimed at reducing the attrition of potential cases who, otherwise, might be lost to follow-up or started on other treatments.
It was predicted that, as in the previous trial with adults, positive treatment effects would be found on a range of outcomes involving attention and various behavioral and cognitive areas affected by it. Safety and tolerability were carefully examined for LDX in this clinical application. In addition, as was done in the study with adults, treatment outcomes were examined in terms of potential moderating effects involving a broad range of pre-treatment subject characteristics. More generally, the aim of the study was to gain further insights into the nature of attention deficits and their treatment in children with TBI.
2. Methods
2.1. Subject Selection
The subject sample consisted of 20 children diagnosed with TBI-related attention deficits. The specific selection criteria were as follows.
2.1.1. Inclusion Criteria
- Males and females ages 6 to 16
- TBI rated as mild/moderate/severe based on assorted factors (Glasgow Coma Scale, estimated post-traumatic amnesia, indications of intracranial injury on CT scan, etc.)
- TBI sustained 2–36 months earlier
- Considered to be neurologically stable (absence of lingering symptoms of confusion, disorientation, etc.)
- Persistent (>2 months) problems with focused or sustained attention
- Problems with attention/concentration rated as among the most prominent cognitive changes
- Accompanying features may include diminished arousal/speed/stamina and/or hyperactivity/impulsivity symptoms
2.1.2. Exclusion Criteria
- Cases with primarily penetrating head trauma
- Pre-injury history of diagnosed ADHD
- Pre-injury history of other neurodevelopmental disorders including intellectual disabilities, major communication disorders, or autism spectrum disorder
- Unstable or serious psychiatric conditions, such as psychotic symptoms. (Concurrent problems with depression, anxiety, or post-traumatic stress disorder may be present, but are judged as stable and as not requiring pharmacologic treatment.)
- Treatment with psychotropic medication(s), including stimulants, within the past six weeks, but eligible thereafter for inclusion in the trial
- Lifetime history of stimulant abuse or dependence. Other (non-stimulant) substance abuse within the past six months.
- Tics or other contra-indications for psychostimulant use including cardiovascular disease, uncontrolled hypertension or hyperthyroidism, glaucoma, agitation, and use of a MAO inhibitor within the past six weeks. Pregnancy was also an exclusion for girls of childbearing age.
- Estimated IQ < 70
- Sensory and/or motor impairment(s) seriously limiting testing options
- Neurological conditions including uncontrolled epilepsy, degenerative disorders, brain tumor, or primary stroke
- Physical condition affecting decreased arousal, activity level, or stamina including uncontrolled hypothyroidism, severe or symptomatic anemia, autoimmune or metabolic disorders, untreated moderate/severe sleep apnea, etc.
2.1.3. Recruitment Process
Children were recruited from hospitals and clinics at Vanderbilt University Medical Center. An initial review of medical records served to narrow the pool in terms of age, indications of TBI, and any contraindications. All study methods, including recruitment procedures, were approved by the Vanderbilt University Institutional Review Board (IRB) for Human Subjects Research (Project IRB# 151965, 11 February 2016).
2.1.4. Screening Assessments/Enrollment
Parents of cases meeting the initial criteria from record review were contacted via letter informing them of the study and asking them to consider participating in further screening to determine eligibility. Parents were then contacted by phone to further inform them of the study and to ask permission to participate in a brief telephone screening (10–15 min) to get basic information concerning eligibility. A set script was followed, which focused mainly on whether there were cognitive problems involving attention/concentration and to determine if there were disqualifying conditions.
Next, potentially appropriate children and their parent(s) were invited to come in person for an in-depth determination of eligibility. Each underwent a semi-structured interview by the project neuropsychologist/principal investigator (MGT) to obtain more detailed information about the TBI, post-concussion symptoms, persisting problems with attention and related areas, presence of any co-morbid psychiatric conditions, and clarification of premorbid history. Additionally, rating forms were used in eliciting detailed information about current cognitive and behavioral status. This included behavior ratings on the Conners-3 assessing ADHD and related areas. Separate forms were completed by parent and child. Selection required a T-score of 65 or higher (+1.5 SD) on one or more of the following subscales: Inattention, Hyperactivity/Impulsivity, ADHD Inattentive Symptoms, and ADHD Hyperactive-Impulsive Symptoms. They also completed a Post-TBI Symptom Questionnaire, which further delved into mental functioning. It was a 40-item inventory spanning a total of eight cognitive domains (alertness/attention, orientation, perception, communication, mental control, thinking, memory/new learning, and specific skills). Each item was rated on a 0-3 scale of severity, with a score of 2 and higher considered as indicating a significant complaint. Subject selection required that, categorically, attention problems were rated as among the most troubling cognitive symptoms persisting since the TBI. Each case was also screened for the necessary minimum IQ of 70. That was estimated with the Vocabulary subtest of the Wechsler Intelligence Scale for Children-Fifth Edition (WISC-V) and required a scaled score of 4 or higher (M = 10, SD = 3).
Lastly, each candidate underwent a physical exam and review of medical and psychiatric history by a board-certified child and adolescent psychiatrist on the investigative team (EW). Enrollment was contingent on verifying the absence of any contra-indications for psychostimulant use, as noted above. Female patients of child-bearing age also had to have a negative urine pregnancy test. Parents were provided with information to help them guide their daughters on avoiding pregnancy and what actions should be taken if they were to become pregnant while in the study.
Upon meeting all eligibility requirements, each candidate completed an approved informed consent process (separate consent/assent procedures were used for parent and child). Financial incentives and reimbursement of travel expenses were offered for participation in the study.
Recruitment was continued until the enrollment goal of 20 cases was met (which took about two years). Thirty-four candidates completed a screening visit. Of the 14 who were not enrolled, eight did not meet the full selection criteria. The other six cases declined to participate for one reason or another (scheduling demands, reluctance to pursue medication treatment, etc.).
2.2. Study Design
As in the initial study, this was a randomized, double-blind, placebo-controlled, dose-titration, crossover trial. Following enrollment, each case was randomly assigned to one of two treatment sequences, alternating on whether stimulant treatment or placebo came first. Each phase was six weeks long, resulting in a total duration of 12 weeks. Comprehensive neurobehavioral assessments were performed at baseline, six weeks, and 12 weeks. Streamlined behavior ratings along with safety monitoring and medication/placebo dispensing were done during weekly visits.
2.3. Medication Trial
2.3.1. Source
Medication was supplied by Shire Pharmaceuticals GmbH, which is the manufacturer of Vyvanse and the funding source for this investigator-initiated research trial (IIR-USA-000881). The Vanderbilt Investigational Drug Service (IDS) repackaged the active medication to provide placebo and drug capsules identical in size (the smallest available) and appearance that remained the same throughout the trial. The IDS performed medication blinding and distribution to the study staff for dispensing.
2.3.2. Protocol
Enrolled children entered a pre-determined randomization scheme as designed by the IDS. They received LDX (Vyvanse) 20–70 mg or placebo for six weeks. At the end of six weeks (day 43 after treatment initiation), each subject was switched from the current agent to the alternative one. Based on manufacturer’s guidelines, no taper or washout period was deemed necessary either in switching from active drug to placebo or at termination of treatment.
2.3.3. Titration
All subjects in the LDX treatment phase of the protocol began dosing at 20 mg po on study day 1 and continued that for week 1. (The usual starting dose is 30 mg for children 6 years of age and older with idiopathic ADHD, but a more conservative starting point was chosen given the off-label use in the present trial.) If tolerated without indication of medication sensitivity (such as mild increases in anxiety, insomnia, weight loss, etc.), the dose was increased to 30 mg for week 2. Thereafter, if tolerated, it was increased to 50 mg for week 3 and again for week 4 to a maximum dosage of 70 mg. Increments were scaled back to a rate of 10 mg weekly at any point if there was concern about possible medication sensitivity. Subjects remained at the maximum tolerated dose for the remainder of the trial unless they met safety endpoints for withdrawal (see below) or requested to exit the study. Cases with certain medical conditions (e.g., those with known or suspected renal dysfunction) were not to be advanced to a daily dose beyond 50 mg. If a subject tolerated lower dosing(s) but reported tolerability problems after a dose increase, the dose was titrated downward to the prior tolerated dose level. Dosing adjustments were allowed up to the start of week 6 if necessary. The same titration schedule and guidelines were applied during both the drug and placebo phases of the trial.
2.3.4. Weekly Monitoring
Once started, all cases underwent weekly (+/−3 days) clinical monitoring, drug trial implementation, and safety and compliance assessments by the project medical staff. Safety monitoring included assessment of any self-reported or parent-reported adverse events (AEs), assessments of blood pressure, heart rate, and weight, as well as psychiatric symptom assessment. There were pre-defined safety endpoints, based on both medical and psychiatric AEs, that served as withdrawal criteria if met.
2.3.5. De-Blinding
The study investigators and subjects were blinded with respect to drug/placebo status. The IDS provided this information to the principal investigator or medical personnel in the case of a patient’s medical emergency. If blinding had to be broken for this reason, the subject was to exit the study.
After completion of the full trial, individual participants were provided information from the IDS, indicating the order and dosing of treatment in their case. This allowed them and their parents the option of sharing their subjective experiences with their primary care provider, including any perceived benefits from LDX, in consideration of possibly pursuing further treatment on their own. However, the blinding of project staff with respect to treatment order was maintained for all cases throughout the study until completion of the final study subject.
2.3.6. Note
Concomitant medications not listed in the exclusion criteria were permitted. No medications were changed or held for the purposes of entering the research study. If a child was started on a new medication by their medical provider, and that medication was on the list of excluded medications, the patient was to exit the study. Inquiry as to possible medication changes/additions were specifically assessed as part of the monitoring of safety and compliance in weekly visits with the study staff.
2.4. Neurobehavioral Assessments
All cases received a one-time assessment at baseline on the following measures. These were used as covariates or component measures facilitating interpretation on other tests.
- Abbreviated Wechsler Intelligence Scale for Children, Fifth Edition (WISC-V, general intelligence)
- Wisconsin Cart Sorting Test (WCST, set maintenance/shifting, executive functioning)
- Finger Oscillation (fine-motor speed/persistence)
The following are repeatable measures that were administered at baseline, 6 weeks (+/−3 days), and 12 weeks (+/−3 days):
- Conner’s Continuous Performance Test (CPT, sustained attention, delay, response modulation)
- Stroop Color-Word Test- Children’s Version (set maintenance/shifting, regulation of competing response tendencies)
- Letter & Animal Word Fluency, WISC-V Coding (processing speed/mental control)
- Woodcock-Johnson Understanding Directions (listening comprehension, following spoken instructions)
- WISC-V Digit Span (working memory)
- Wide Range Assessment of Memory and Learning-2 (WRAML): Verbal Learning and Design Memory subtests (short-term auditory-verbal memory and visual memory)
- Conners-3 Parent and Self-Report Forms (ratings of ADHD symptoms and related areas). Short-form versions of these were obtained during weekly visits.
- Child Behavior Checklist (CBCL, parent ratings of more general behavioral and emotional problems)
- Children’s Depression Inventory-2 (CDI, self-report of depression symptoms)
- Children’s Manifest Anxiety Scale-2 (RCMAS, self-report of anxiety symptoms)
- Behavior Rating Inventory of Executive Functioning-2 (BRIEF)—Parent and Self-Report Forms
Note
The above provided a broad-based assessment of cognition and behavioral/emotional status, as well as focusing on attention-related areas. It incorporated measures assessing the four components in Musky’s model of attention: focus-execute, sustain, encode, and shift [
4]. Descriptions and normative data for many of the tests can be found in a Compendium of Neuropsychological Tests [
17].
2.5. Data Analyses
The double blind, crossover design allowed for the assessment of both within-subjects and between-subjects contrasts. The primary analyses consisted of multiple paired-samples t-tests comparing LDX versus placebo on each of the neurobehavioral dependent measures. There was no power analysis based on sample size. Nor was there a formal correction applied for the multiple comparisons performed. As an initial study of its kind, the objective was to not limit sensitivity in detecting possible treatment effects. A p-value equal to, or less than, 0.05 was considered statistically significant. All tests were two-tailed. All analyses were performed using the Statistical Programs for the Social Sciences (IBM,2020,SPSS StatisticSubscription1.0.0.1327.Retrieved from
https://www.ibm.com/products/spss-statistics/).
Possible order effects (depending on whether drug treatment came before or after placebo) were examined through a separate analysis of variance (ANOVA) for each dependent measure using a two-factor model (treatment, order, and treatment x order interaction).
There were also applications of analysis of covariance (ANCOVA) to determine possible mediating or moderating effects of pre-treatment variables on treatment outcomes (demographics, injury variables, IQ and other cognitive factors, behavioral symptoms, and other features).
Additionally, safety data were examined based on weekly visits over the course of the trial. Comparisons were made on LDX versus placebo for indices such as weight, blood pressure, and heart rate, side effects (insomnia, decreased appetite, etc.), or significant adverse events, if any. Each of these were examined using repeated measures ANOVA.
www.telegraph.co.uk
