Prescribing Patterns of Amantadine During Pediatric Inpatient Rehabilitation After Traumatic Brain Injury: A Multicentered Retrospective Review From the Pediatric Brain Injury Consortium

Introduction:

Traumatic brain injury (TBI) is the leading cause of admission to pediatric inpatient rehabilitation units.¹ Youth with TBI who are admitted to inpatient rehabilitation have variable levels of cognitive functioning, spanning from unresponsive wakefulness syndrome (UWS, also known as vegetative state) to acute cognitive impairment (ACI, representing resolution of post-traumatic amnesia [PTA] and a state where remaining cognitive dysfunction is expected to improve more slowly over time).² Children and young adults admitted to inpatient rehabilitation after TBI have persisting cognitive deficits limiting functional independence at discharge,² and the vast majority continue to demonstrate cognitive changes affecting academic function years post-injury.³ Deficits in specific neuropsychological domains such as attention/executive functioning,^4,5 learning/memory,⁶ and processing speed⁴ are common after pediatric TBI and broadly impact long-term outcomes.^7,8

TBI is associated with an increase in dopamine release acutely; however, long-term, TBI-related alterations to dopaminergic systems result in a hypodopaminergic state.⁹ Dysfunction in the dopaminergic system has been suggested as underlying cognitive impairments following TBI, and the dopaminergic system has long been a target for pharmacological intervention.¹⁰ Amantadine hydrochloride is a dopamine agonist and an N-methyl-D-aspartate (NMDA) antagonist approved by the Food and Drug Administration (FDA) for treatment and prophylaxis of influenza A in children and adults, as well as for treatment of extrapyramidal symptoms and Parkinsonism in adults.¹¹

For adults with TBI, administration of amantadine at a total daily dose of 200 to 400 mg (divided across two doses per day) is a Clinical Practice Guideline for patients in UWS or a minimally conscious state (MCS) 4–16 weeks post-injury based on a randomized controlled trial by Giaicino et al.^12,13 Meythaler et al. conducted a placebo-controlled pilot trial of 35 adult patients spanning a broader range of cognitive states in the weeks to months following moderate to severe TBI; they reported a trend for more rapid functional recovery in patients treated with a fixed dose of amantadine 200 mg per day compared to placebo.¹⁴ In limited studies, amantadine has also shown promise as a treatment for post-TBI agitation¹⁵ and for symptoms of mild TBI. Taken together, these studies suggest that amantadine may be useful for alleviating cognitive/behavioral dysfunction across a wide range of adult patients with TBI.

In terms of pediatric studies, two prospective^16,17 and two retrospective studies^18,19 have examined the efficacy of amantadine in children with TBI in primarily UWS or MCS in inpatient rehabilitation settings. However, due to limited sample sizes and/or lack of comparison groups, these studies have produced inconclusive findings regarding efficacy.²⁰ In a retrospective study of adolescents with mild TBI, results showed greater magnitude of improvements in patients treated with amantadine compared to those who were not.²¹ Across all of the available pediatric amantadine studies, results suggest amantadine is well tolerated in children. Vargus-Adams and colleagues (2010) found that among seven children with DoC (UWS/MCS) treated during inpatient rehabilitation, only one child experienced nausea/vomiting.²² Similarly, in the retrospective study evaluating 54 pediatric inpatients with TBI, Green et al. (2004) found that only 9% experienced adverse effects including nausea/vomiting, behavioral changes (e.g. aggression, hallucinations, delusions) and tremor.¹⁹

Currently, empirical evidence on optimal dosing of amantadine in pediatric TBI does not exist. FDA labelling of amantadine for children is 4.4 to 8.8 mg/kg/day (not to exceed 150mg per day) divided into two daily doses for children aged 1–9 years and 200 mg daily (in divided doses) for ages 10 and older.²³ McMahon et al. evaluated a small cohort of children with TBI and other forms of acquired brain injury who were in UWS/MCS using amantadine dosed at 4mg/kg/day (maximum 300 mg/day) escalating to 6mg/kg/day (maximum 400 mg/day).¹⁷ Pharmacokinetic analyses from that cohort demonstrated that the two children with the clearest positive responses to amantadine had the highest serum concentrations of amantadine.²² These two children also had the highest weights, thus, received the highest total doses.²² This led Vargus-Adams et al. to recommend that, in the absence of adverse effects, clinicians consider increasing dosing of amantadine beyond 6 mg/kg/day in younger/smaller children in order to improve clinical response.²² Similarly, in a prospective study of 43 children (5–19 years) with autism and hyperactivity, impulsivity, and aggression, amantadine dosing at 5 mg/kg/day divided twice daily (breakfast and afternoon) was “too well tolerated” with no drop-outs from the treatment arm and no statistically significant differences in side effects when compared to placebo.²⁴ King et al. reported that one “non-responder” was converted to a “responder” with an open-label dosage increase.²⁴ Similar to the Vargus-Adams study,²² this finding from King et al. suggests that doses higher than 5–6 mg/kg/day may be required to yield clinical benefit.²⁴

The lack of consistent practice guidelines, including optimal dosing and limited efficacy research, for amantadine in children with TBI, coupled with the limited number of small and heterogeneous studies (e.g., wide range of age, injury severity, time since injury) suggests the potential for variability in clinical practice. The goal of this study is to describe amantadine prescribing practices across the 8 inpatient rehabilitation units which comprise the Pediatric Brain Injury Consortium in order to inform clinical use and clinical trial design. We hypothesized that 1) amantadine is prescribed more frequently to patients in UWS or MCS than children with higher levels of cognitive functioning, 2) a wide range of amantadine doses are used in clinical practice, and 3) adverse effects are similar to those previously reported.

Methods:

Participants:

After obtaining individual Institutional Review Board approval from 8 sites participating in the Pediatric Brain Injury Consortium (PBIC) (2 – Northeast, 2 – South, 3 – Midwest, 1 – West), the medical records of 234 patients aged 2 months to 21 years admitted to inpatient rehabilitation with a primary diagnosis of TBI were reviewed. The cohort and data extraction process have been previously reported.² Briefly, at each site, data were reviewed from admissions spanning a 1-year period; specific dates of the 1-year block reviewed varied by site, with all 1-year time blocks falling between June 2017 and July 2019. Data were de-identified prior to sharing with Blythedale Children’s Hospital, which served as the coordinating center for the study. One site had 96 admissions during this time period, and the others ranged from 11–36 admissions. Demographic and injury-related information collected include gender (female, male, or other), age at injury, weight, mechanism of injury, time from injury to rehabilitation admission [time to admission (TTA)] and length of inpatient rehabilitation stay (LOS).

Amantadine Dosing and Adverse Effects:

Patients who had received amantadine during the inpatient rehabilitation admission were identified by review of medication records as well as other clinical documentation (i.e. admission, progress, and discharge notes). For patients who had received amantadine, each recorded dose of amantadine was reported. For each recorded dose, days since injury of first and last recorded use were provided. Some, but not all, sites had access to and included information regarding amantadine use (initiation and titration) during the acute care admission. Due to incomplete data regarding acute care use, amantadine doses are referenced as lowest and highest “recorded” doses rather than using more definitive terms such as “initial” dose. For the purpose of describing lowest recorded doses, only patients who either had a) first recorded use of amantadine after admission to rehabilitation or b) escalation of amantadine dose during inpatient rehabilitation were included (n=40).

Data on adverse effects associated with amantadine were collected as a free text response. Medical record reviews for adverse effects were completed by a clinician (physiatrist or neuropsychologist, 4 sites) or by a research assistant or medical student with training and data review performed by a clinician (4 sites). Neuropsychologists consulted with medical providers as needed to verify information regarding adverse effects.

Cognitive State:

Patients’ cognitive states were documented for the time points of admission to and discharge from inpatient rehabilitation as well as at weekly intervals during the inpatient rehabilitation stay. At the time of data extraction, patients were classified into one of four categories at admission and discharge: UWS, MCS, PTA, or ACI. Method for determining cognitive state varied by site, as has been previously reported,² and involved review of behavioral descriptions of function and available data from administrations of the Coma Recovery Scale-Revised²⁵ and age-appropriate measures of PTA (e.g. Children’s Orientation and Amnesia Test).²⁶

Evaluation tools and behavioral manifestations of disorders of consciousness have only recently been developed and described for children younger than 5 years of age.^27,28 Similarly, criteria/assessment tools do not currently exist for post-traumatic confusional state in children, with criteria/tools for determining emergence from PTA only existing for children 3 years or older. In light of these limitations, in the current study, cognitive state was collapsed into a binary variable of Disorder of Consciousness (DoC, representing children in UWS or MCS) versus Confusional or Conscious State (CCS, representing children classified as being in PTA or ACI). This binary classification system mirrors prior work in pediatric TBI examining children in UWS/MCS in comparison to those who emerged from MCS.²⁹

Statistical analyses:

Data were maintained in a local REDCap database at the data coordinating site. Descriptive statistics were performed for dosing information and for description of the population; median and interquartile range (IQR) were used for continuous demographic variables and frequency (%) for discrete variables. Independent samples t-tests were used to examine differences between patients who received amantadine and those who did not with respect to age at injury, weight, TTA, and LOS; chi-squared or Fisher’s exact tests were performed to examine between-group differences in gender and cognitive state at admission and at discharge from rehabilitation. Exploratory analyses examined differences in demographic factors based on Cognitive State at admission. Data were analyzed using IBM SPSS Statistics v27.0.

Results:

Patient characteristics

The total cohort comprised 234 patients ranging in age from 2 months to 21 years old; 61% were male (Table 1). Most patients (78%) were in CCS at the time of admission. The most frequent mechanisms of injury involved motor vehicles (70%); the next most common mechanism of injury was assault/abusive head trauma (14%).

Table 1.

Demographics of the patient population comparing those who received amantadine compared to those who did not.

TABLE 1	Total (n=234)	Amantadine (n=49)	No Amantadine (n=185)	p-value
Median Age in Years (Q1-Q3)	11.6 (5.0–15.3)	12.4 (6.10–15.4)	11.1 (4.2–15.3)	0.206
Gender	88 F, 145 M, 1 O	21 F, 28 M	67 F, 117 M, 1 O
Median Acute LOS (TTA) (Q1-Q3)	16 (10–29)	27 (12–39)	16 (9–26)	0.003
Median Rehab LOS (Q1-Q3)	23 (14–39)	40 (28–58)	20 (13–31.25)	<0.001
State at Admission	---	---	---	<0.001
DoC	51 (22%)	23 (47%)	28 (15%)	---
CCS	183 (78%)	26 (53%)	157 (85%)	---
State at Discharge	---	---	---	0.007
DoC	19 (8%)	9 (18%)	10 (5%)	---
CCS	215 (92%)	40 (82%)	175 (95%)	---

M – Male, F – Female, O – Other, LOS – Acute care Length of Stay, TTA – time from injury to rehabilitation admission, DoC – Disorder of Consciousness, CCS – Confusional or Conscious State

Of the 234 patients, 49 (21%) received amantadine. Comparisons between patients who did and did not receive amantadine are summarized in Table 1. Across the whole cohort, patients who received amantadine had a longer TTA (median 27 vs 16 days, p=0.003) and longer inpatient rehabilitation LOS (40 vs 21 days, p<.001). There was a significant between-group difference in the distribution of Cognitive States at admission and discharge. Forty-five percent of patients admitted to inpatient rehabilitation in a DoC were treated with amantadine while 14% of children admitted with higher levels of functioning received amantadine. The group treated with amantadine had a higher percentage of children in DoC at admission (p<.001) and discharge (p=.007) compared to the group not treated with amantadine.

Comparisons of patients who did and did not receive amantadine, stratified by Cognitive State at admission, are summarized in Table 2. The patients admitted to inpatient rehabilitation in DoC who did not receive amantadine were significantly younger (median 3.0 years) than the patients who were in DoC and treated with amantadine (median 11.6 years, p=.008). The finding of longer TTA in the group that received amantadine was only observed for patients admitted to inpatient rehabilitation in CCS, while longer rehabilitation LOS was observed in patients treated with amantadine who were admitted to inpatient rehabilitation in either DoC or CCS.

Table 2.

Subgroup analyses between groups who received amantadine and those who did not based on Conscious State at admission to inpatient rehabilitation facilities.

Table 2	Admission State
	DoC (UWS or MCS)
	Amantadine (n=23)	No Amantadine (n=28)	p-value
Median Age at Injury (Q1-Q3)	11.6 (5.4, 15.2)	3.0 (0.4, 10.3)	0.008
Gender	11 F, 12 M	15 F, 13 M
Median Acute LOS (Q1-Q3)	29 (14.5–37.5)	22 (14–31.8)	0.346
Median Rehab LOS (Q1-Q3)	47 (38.5–62)	31 (25–47.5)	0.027
State at Discharge	---	---
DoC	9 (39.1%)	10 (35.7%)	---
CCS	14 (60.9%)	18 (64.3%)	---
	CCS (PTA or ACI)
	Amantadine (n=26)	No Amantadine (n=157)	p-value
Median Age at Injury (Q1-Q3)	12.8 (8.70–15.5)	12.0 (5.9,15.6)	0.393
Gender	10 F, 16 M	52 F, 104 M, 1 O
Median Acute LOS (Q1-Q3)	26.5 (10.8–43.3)	14.0 (9.0–23.0)	0.012
Median Rehab LOS (Q1-Q3)	35.5 (23–53.3)	15.5 (10–22.3)	<0.001
State at Discharge	---	---
CCS	26 (100%)	157 (100%)	---

DoC – Disorder of Consciousness, UWS – Unresponsive Wakefulness Syndrome, MCS – Minimally Conscious State, CCS – Confusional or Conscious State, PTA – Post-traumatic Amnesia, ACI – Acute Cognitive Impairment

Amantadine dosing characteristics

Amantadine dosing information is summarized in Table 3 and Figures 1 and 2; as described in Methods, data for lowest doses represents 40 patients while all other dosing data represent all 49 patients who received amantadine. The youngest child treated with amantadine was 0.9 years old at injury and also had the lowest weight (9.6 kg). Twenty patients (40%) received amantadine prior to or at the time of admission to inpatient rehabilitation. For all patients who received amantadine, Cognitive State at first reported use of amantadine was the same as Cognitive State at admission to inpatient rehabilitation (Table 1). At the time of discharge from inpatient rehabilitation, median duration of reported amantadine use was 24 days (range 1–225 days). Seventeen patients (35%) treated with amantadine were taking the medication on the day of discharge from inpatient rehabilitation.

Table 3.

Dosing and adverse effect data of amantadine in the study population.

TABLE 3	Median (IQR)	Range
Lowest Dose in mg/kg/day	3.6 (2.2–4.7)	0.7–8.2
Highest Dose in mg/kg/day	5.2 (3.4–7.1)	1.1–13.5
Adverse Effects	N (%)	Associated doses mg/kg/day (mg/day)
Total	8 (16)
-Nausea/Abdominal Discomfort	3 (6)	1.3 (100); 4.6 (100), 8.1 (400)
-Agitation	3 (6)	0.7 (20), 2.3 (150), 8.2 (400)
-Dizziness	1 (2)	2.7 (150)
-Diarrhea	1 (2)	13.5 (400)

Figure 1.

Lowest recorded doses by age (A) and weight (B) and highest recorded doses by age (C) and weight (D). Sites are represented by marker type.

The median lowest recorded dose was 3.5 mg/kg/day. Range of lowest recorded doses was 0.7–8.2 mg/kg/day, representing a 12-fold difference in lowest doses. The site with most variability in lowest recorded dose had a range from 1.5–8.2 mg/kg/day, reflecting a 5.5-fold difference. The median highest recorded dose of amantadine was 5.2 mg/kg/day (range of 1.1–13.5 mg/kg/day) reflecting a 12-fold difference. Five sites reported children treated with doses greater than 6mg/kg/day for a total of 15 children (31% of those receiving amantadine). Three sites reported doses greater than 8mg/kg/day for a total of 7 children (14%). Maximum total daily dose reported was 400mg.

Eight (16%) patients were reported to experience adverse effects attributed to amantadine (16.3%; Table 2). The most common adverse reactions were nausea/abdominal discomfort (n=3) and agitation (n=3). Adverse effects were reported for doses ranging from 0.7 mg/kg/day to 13.5 mg/kg/day. Of 15 children who received a dose of amantadine >6 mg/kg/day, three experienced adverse effects. The highest reported dose for which an adverse effect was not reported was 10.1 mg/kg/day. For five of the eight children who experienced adverse effects, there was a reported lower dose at which no adverse effects were experienced. Two children who experienced nausea or abdominal discomfort were treated at the dose associated with the side effect for more than one week; one child with agitation and one child with dizziness continued treatment with amantadine albeit at a previously tolerated lower dose. For the other four children with adverse effects, amantadine was discontinued. No serious adverse events were reported.

Discussion:

This study evaluated the patient characteristics and dosing practices related to amantadine use for patients with TBI admitted to pediatric inpatient rehabilitation units at 8 different institutions in the United States. Approximately one in five patients with TBI was treated with amantadine. As hypothesized, patients in DoC were more likely to receive amantadine though some patients with higher levels of cognitive function were also treated with the medication. There was a wide range of doses used. Even at higher weight-based doses than previously reported in this population, side effects reported were similar to those in the literature.^13,19,22 Approximately one third of patients treated with amantadine during hospitalization continued to receive amantadine at the time of discharge from inpatient rehabilitation.

Patients who received amantadine had more severe injuries and/or greater medical complexity, as reflected in their longer acute care and rehabilitation lengths of stay. Nearly one-half of children who were admitted to inpatient rehabilitation in DoC received amantadine, compared to 1 in 7 children admitted at a higher level of cognitive functioning. This is consistent with a single site retrospective evaluation of amantadine use from nearly two decades ago from a pediatric unit not included in the current work; in that study, children treated with amantadine had a lower starting Ranchos Los Amigos level (median 3) compared to children who did not receive amantadine (median 4).¹⁹ It is important to highlight that amantadine is not used exclusively in children in DoC, though children with higher levels of cognitive functioning and shorter rehabilitation lengths of stay are currently less likely to be treated with amantadine. In light of data suggesting benefits across the severity spectrum of TBI,^14,20 additional studies are needed to understand whether amantadine may be beneficial even for children who are rapidly recovering and able to quickly transition to the home and outpatient settings.

Patients with DoC who were treated with amantadine were significantly older (median=12 years) than patients in DoC who did not receive amantadine (median=3 years). Although amantadine is FDA approved to treat influenza in patients beginning at 1 year of age, the current data suggest that providers may be less comfortable prescribing amantadine to young children after TBI. Prior studies of children with severe brain injury have been limited to children at least 3,¹⁹ 5,¹⁷ or 8^16,18 years of age, which may also influence clinical use. Young children represent a cohort of interest for future study which may be facilitated by use of measures designed to assess cognitive functioning in young children, such as the Coma Recovery Scale for Pediatrics³⁰ and the Cognitive and Linguistic Scale.^28,31 Ensuring inclusion of young children in future studies would limit disparities in care based on age.³²

A large degree of variability was observed in dosing both within and across sites. Variability in lowest recorded dose was of particular interest as this more closely approximates starting dose which does not account for individual patient clinical response or tolerance, whereas these factors may contribute to highest recorded dose. The greatest range of weight-based dosing was observed in children with weights below 40–50 kg. As no total daily dose exceeded 400mg, the upper limit recommended in the Practice Guideline for adults with TBI,¹² the oldest children had less variability in weight-based dosing as they approached this upper dosing limit. The level of variability between the lowest and highest dose across different sites highlights the need for a more standardized dosing regimen. Even within sites, there was a high level of variability between some of the highest doses observed. This level variability alone could represent the difference between obtaining an anticipated clinical response, having no detectable clinical response, or causing an adverse event. Furthermore, without clear pediatric-specific guidelines for initiating amantadine, patients who may have similar injuries may or may not receive amantadine or may receive the drug at different times or in different doses, which may alter rates of clinical recovery. In order to perform a large-scale, multicentered efficacy study with amantadine in pediatric patients with TBI, more information on an optimal standardized dose and patient population are needed.

The rate of adverse effects reported to be associated with amantadine in this cohort (16%) was slightly higher compared to a previous retrospective study of patients receiving inpatient rehabilitation after TBI (9%),²⁰ though the types of effects were similar. Lack of weight-based dosing information from the prior cohort limits ability to compare the cohorts in this regard.¹⁹ The overall rate of adverse effects mirrored a small prospective study in children with varying etiologies of acquired brain injury who were all in DoC at the time of amantadine initiation (14%)²² and the randomized controlled trial in adults with TBI and DoC (in which a range of prospectively-tracked individual adverse effects were noted to occur in up to 21% of the amantadine cohort, with no significant differences between amantadine and placebo cohorts).¹³ In the current study, three patients who experienced adverse effects received the highest total daily dose (400 mg), but patients experienced adverse effects across the full range of reported doses (0.7 to 13.5 mg/kg/day). Differences in serum concentration of amantadine, which are not routinely evaluated in clinical use, may underlie some of this variability in tolerance. This was highlighted in the work of Vargas-Adams et al. in which the sole child who experienced an adverse effect had the highest serum concentration of amantadine.²² Even when pediatric patients receive a standard weight-based dose of amantadine, there is large (up to two-fold) between-patient difference in plasma concentration not well predicted by body weight or size.²² This underscores the challenges associated with understanding optimal dosing in clinical practice. Future prospective studies are needed which evaluate clinical responses as they relate to drug concentration.

The retrospective nature of data collection poses a risk for missing or misinterpreted clinical data, particularly related to assignment of cognitive state and identification and description of adverse effects of amantadine. For some patients, amantadine was started prior to admitting to the inpatient rehabilitation unit, and full information was not available regarding dosing and duration of use. This incomplete information regarding dosing, demographic bias in children who received amantadine, and lack of standardized outcomes and assessment intervals across the sites prevented exploration of impact of amantadine on functional recovery. Furthermore, adverse effects were collected as a free-text response and obtained solely from clinical documentation relying on clinician attribution and documentation as well as detection of this information during the chart review. There was no standardized procedure for review of records across sites. These factors may contribute to both underestimation and overestimation of adverse effects related to use of amantadine.

Conclusion:

This study highlights the high level of variability in dosing of amantadine among patients admitted to pediatric inpatient rehabilitation units after TBI. Amantadine is a frequently prescribed medication, particularly among older children and young adults in UWS or MCS, which is believed to be safe and typically well-tolerated with adequate renal clearance. Future prospective dose escalation studies, which also include young children and incorporate clinically meaningful outcome measures, will help bridge the gaps in the current knowledge and improve patient outcomes.