Efficacy and Safety of Dopamine Agonists in Traumatic Brain Injury: A Systematic Review of Randomized Controlled Trials

Abstract

In the intensive care unit, dopamine agonists (DA) have been used in traumatic brain injury (TBI) patients to augment or accelerate cognitive recovery and rehabilitation. However, the efficacy and safety of DA in this population is not well established. We conducted a systematic review of randomized controlled trials (RCTs) examining the clinical efficacy and safety of DA in patients with TBI. We searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials, comparing DA to either placebo, standard treatment, or another active comparator. There was no restriction for age, date, or language of publication. Sensitivity analyses were planned to evaluate the potential effect of timing of TBI, age, drugs, and year of publication on efficacy. Among the 790 citations identified, 20 RCTs evaluating methylphenidate, amantadine, and bromocriptine were eligible. Significant clinical heterogeneity was observed between and within studies, which precluded any pooling of data. Efficacy outcomes included mainly neuropsychological measures of cognitive functioning. A total of 76 different neuropsychological tests were used, but most of them (59%) only once. Only 5 studies systematically assessed safety. No trend could be drawn from the analysis of efficacy and safety. Important sources of bias in the studies were of major concern. Considering the absence of consensus regarding clinical outcome, the lack of safety assessment, and the high risk of bias in the included trials, more research is warranted before DA can be recommended in critically ill TBI patients.

Introduction

T raumatic brain injury (TBI) is a leading cause of morbidity and mortality and is responsible for 50,000 deaths and 235,000 hospitalizations every year in the United States (Corrigan et al., 2010; Thurman and Guerrero, 1999). As many as 15–50% of patients surviving TBI will experience long-term disabilities, including neuropsychiatric sequelae such as sleep disturbances, arousal and attention disorders, cognitive impairment, depression, and aggressive behavior (Bales et al., 2009; Corrigan et al., 2010; Green et al., 2008; Fleminger, 2008; Ruttan et al., 2008).

Neurotransmitters such as dopamine influence cognitive processes, movement control, and emotion (Bales et al., 2009; Thurman and Guerrero, 1999). Central dopaminergic pathway dysfunction may contribute to some of the neuropsychiatric sequelae in patients recovering from TBI (Bales et al., 2009). In the intensive care unit, step-down units, or on the ward, dopamine agonists (DA) are anecdotally used in patients recovering from TBI in an attempt to augment vigilance or accelerate cognitive recovery and rehabilitation. However, the efficacy and safety of DA in this population is not clear (Francisco et al., 2007; Siddall, 2005; Tenovuo, 2006). The aim of the present systematic review was to determine the clinical efficacy and safety of DA in TBI patients compared to placebo, active control, or standard treatment (Moher et al., 2009).

Methods

Search strategy

Randomized controlled trials (RCTs), were identified using both electronic and manual search strategies. In June 2010, we searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Studies from the earliest accessible date. We combined terms defining the study treatment (DA) and the study population (TBI). A specific filter was applied to MEDLINE and EMBASE search strategies to identify RCTs (Lefebvre et al., 2009; Scottish Intercollegiate Guidelines Network, 2009). The bibliographies of all identified RCTs and the reviews retrieved were manually searched for additional studies. The search strategy was reviewed by a qualified librarian. The full search strategy for MEDLINE is presented in Appendix 1.

Eligibility criteria

We sought studies of TBI patients, without age restrictions, comparing the effect of the administration of a DA, including methylphenidate, bromocriptine, amantadine, levodopa, ropinirole, pramipexole, apo-morphine, cabergoline, and pergolide, to either placebo, standard treatment, or another active comparator. We included studies addressing clinical efficacy using any primary outcome measures and/or safety. There was no restriction for date or language of publication. We excluded unpublished studies, animal studies, and studies assessing dopamine agonists in any other type of brain injury. Studies of mixed populations with TBI representing more than 80% of patients, or for which data of TBI patients could be isolated, were included.

Study selection

Every study citation retrieved was evaluated by two independent reviewers (A.J.F. and S.K.), according to the previously described inclusion/exclusion criteria. Disagreements were resolved by consensus.

Data extraction and validity assessment

Each study was evaluated independently and in duplicate by two reviewers (A.J.F. and S.K., M.M.P., D.R.W., F.B., or Marie-Soleil Delisle). Any disagreement was resolved by a third independent reviewer. Information was collected using a standardized case report form that was revised after pilot testing on 5 studies by 5 different reviewers. Descriptive variables for each trial were collected, including language of publication, country of randomization, source of funding, sample size, and objectives. Information on the study design included the type of RCT, blinding, method of randomization, number of participating centers, and screening and enrolment of patients. Clinical variables including patient demographic data, past medical history, concomitant drugs permitted, timing of DA administration since the injury, description of the TBI (i.e., mechanism of trauma, severity of TBI, severity of global illness, and radiological findings), and evaluation of the pre-intervention level of functioning, cognitive function, and quality of life, were also collected if reported. Clinical outcomes included length of stay (intensive care unit [ICU] and hospital), mortality, side effects, functional scales, post-traumatic symptoms, and neuropsychological test outcomes.

Methodological quality was assessed using the rating instrument developed by the Cochrane Handbook for Systematic Reviews of Interventions (Higgins and Altman, 2008). In order to assess the carry-over effect in crossover studies, a washout period was considered appropriate when equal to at least 5 half-lives of the study drug from the last dose, when more than one dose was administered in the study. Based on clinical considerations, an a priori washout period of 2 days or more for a single-dose trial was considered suitable. In these crossover studies, we considered not reporting the results of baseline assessment after the washout period, and the inclusion of patients who dropped out after the first study period, as inappropriate (Higgins et al., 2008).

Data synthesis

Data related to study design, drug treatment, and patient demographics, were described for each study. Pre-morbid conditions were described using three categories: functional status, cognitive or psychological impairment, and quality of life, when available. Severity of TBI and illness was assessed using severity scales (Glasgow Coma Scale, Acute Physiology and Chronic Health Evaluation II, and Injury Severity Score), and radiological findings (computed tomography [CT] scan), when available. Clinical outcomes measured by neuropsychiatric questionnaires and scoring tools were categorized as either behavioral or cognitive outcomes. Cognitive outcomes were further categorized into the following domains: attention, speed of processing and reaction time, memory, and executive functioning, according to the predominant domain measured by the test (Patry and Mateer, 2006; Tsaousides, 2009). Sustained attention, focused attention, selective attention, alternating attention, and divided attention, were classified under attention. Clinical outcomes not falling into cognitive or behavioral categories were classified as “other outcomes.”

Sensitivity analyses were performed to evaluate the potential effect of the following sub-groups on efficacy: early versus late TBI, children versus adults (16 years of age or older), type of drugs, and year of publication. Year of publication was dichotomized into the following categories: before 1995, and 1995 and after, based on the increased use of computer-based neuropsychological tests after 1994. Patients were considered to have an early TBI if they were still in an acute care facility (ICU or ward). Patients recruited in rehabilitation centers or in the community were considered to have late TBI.

Safety was assessed by retrieving all adverse events reported. We specifically looked for information on predetermined side effects that were clinically relevant in the evaluation of DA administration safety in TBI, in addition to the reported findings. Information regarding how safety was assessed in the individual studies was also collected. Evaluation of side effects was considered appropriate, if (1) the authors were using a prospectively validated side-effect checklist or a pre-defined objectively measured endpoint; (2) if the method of assessing specific side effects was provided; and (3) if the timing for side-effect evaluation was provided and clinically relevant (Loke et al., 2008).

Results

Study selection

The search strategy identified 790 citations. Twenty studies were included in the systematic review (Fig. 1). Demographics of the study are listed in Table 1. One study published in Chinese only provided descriptive results that were impossible to analyze with regard to the objectives of this study and was therefore rejected. Among those included, 14 evaluated methylphenidate, 4 evaluated amantadine, and 2 evaluated bromocriptine. No study evaluating the use of levodopa, apo-morphine, ropinirole, pramipexole, cabergoline, or pergolide in TBI was identified.

FIG. 1.

Flow chart of the study (RCT, randomized controlled trial; TBI, traumatic brain injury).

Table 1.

Demographics of the Included Studies

Study	Study design	Interventional arm	Comparative arm	Patients randomized (analyzed)	Children/ adults ¹	Location at randomization	Timing from injury at randomization	Description of TBI
Willmott et al., 2009	RCT Crossover	MPH 1 day/2×6 days	Placebo on alternate days×6 days	40 (40)	0/40	Rehabilitation center	Range 12–462 days Mean 68.39±77.09 days	Severe TBI (PTA, GCS)
Willmott and Ponsford, 2009	RCT Crossover	MPH 1 day/2×6 days	Placebo on alternate days×6 days	40 (40)	0/40	Rehabilitation center	12–462 days	Severe TBI (PTA, GCS)
Moein et al., 2006	RCT Parallel	MPH from day 2 until discharge	Placebo from day 2 until discharge	80 (80)	0/80	Intensive care unit	≤48 h	Mild TBI (GCS)
Kim et al., 2006	RCT Parallel	MPH×1 dose	Placebo×1 dose	18 (18)	0/18	Rehabilitation center	Mean 1.6±0.5 years MPH: 3.6±3.4 years placebo	N/A
Lee et al., 2005	RCT Parallel	MPH×4 weeks	Sertraline×4 weeks OR Placebo×4 weeks	30 (30)	0/30	Unclear	Mean years MPH 34.8±3.9; sertraline 31.9±5.8; placebo 30±6.5	Mild TBI (PTA, LOC)
Whyte et al., 2004	RCT Crossover	MPH 3 weeks/6	Placebo on alternated weeks×3weeks/6	39 (34)	0/39	Rehabilitation center and outpatients	Median: 3.2 years Range 4 months–34.2 years	Severe blunt TBI (PTA)
Alban et al., 2004	RCT Crossover	MPH 3weeks/6	Placebo on alternated weeks 3 weeks/6	35 (33–35)	0/35	Rehabilitation center and outpatients	Median: 3.2 y Range 4 months–34.2 years	Severe blunt TBI (PTA)
Mahalick et al., 1998	RCT Crossover	MPH×14 days	Placebo×14 days	14 (14)	14/0	Not mentioned/unclear	Range 1–60 months Mean 14.4±14.87 months	Mild, moderate and severe TBI (GCS)
Williams et al., 1998	RCT Crossover	MPH×4 days	placebo×4 days	10 (10)	10/0	Outpatients	Median 1.5 year Mean 2.7±3.0 years	N/A
Whyte et al., 1997	RCT Crossover	MPH 1 day/2×6 days	Placebo on alternate days×6 days	19 (9)	1/19	In- and outpatients from a rehabilitation center	Median 145 days	Severe blunt TBI (GCS)
Plenger et al., 1996	RCT Parallel	MPH 30 days	Placebo×30 days	23 (23)	0/23	Inpatients from trauma center	N/A acute phase?	Moderate to severe TBI (GCS)
Speech et al., 1993	RCT Crossover	MPH×1 week	Placebo×1 week	12 (12)	0/12	Outpatients	Median 32 months Range 14–108 months	Moderate to severe blunt TBI (PTA, LOC)
Mooney and Haas 1993	Randomized Pre-post	MPH regimen not mentioned	Placebo regimen not mentioned	38 (38)	0/38	Out patients	Mean 27.1±21.1 months	Severe blunt TBI (PTA,LOC)
Gualtieri and Evans, 1988	RCT Crossover	MPH×12 days	Placebo×12 days	15 (14)	4/15	Outpatients	Mean 46.8±41.3 months Range 5–144 months	Severe TBI (GCS)
Beers et al., 2005	RCT Parallel	AM × 12 weeks	Standard care×12 weeks	37 (variable)	37/0	Out patients	6 to – 36 months 62%≤2 years	Mild, moderate, severe blunt TBI
Meythaler et al., 2002	RCT Crossover	AM×6 weeks	Placebo×6 weeks	35 (35)	0/35	Emergency department	2–6 weeks	Severe blunt TBI (GCS)
McMahon et al., 2009	RCT Crossover	AM 4×3 weeks	Placebo×3 weeks	7 (5)	5/7	Unclear	4–10 weeks	Severe TBI (GCS) Including TBI, stroke and anoxia
Schneider et al., 1999	RCT Crossover	AM×2 weeks	Placebo×2 weeks	18 (10)	0/18	Rehabilitation center	N/A	Moderate to severe TBI (GCS)
McDowell et al., 1998	RCT Cross over	B×1 dose	Placebo×1 dose	24 (19–23)	1/24	Rehabilitation center	27 days – 300 months	Severe blunt or penetrating TBI (GCS) (1 aneurysm)
Whyte et al., 2008	RCT Crossover	B×4 weeks	Placebo×4 weeks	21 (12)	0/21	Inpatient and outpatient from clinical services	Median 3.3 years 3 months–23.5 years	Severe blunt TBI (PTA)

RCT, randomized controlled trial; MPH, methylphenidate; AM, amantadine; B, bromocriptine; PTA, post-traumatic amnesia; GCS,

Glasgow Coma Scale; LOC, loss of consciousness; TBI, traumatic brain injury; N/A, not applicable.

Adults are defined as≥16 years old.

Study characteristics

Twenty studies evaluated the administration of a DA in TBI, including a total of 480 patients (Alban et al., 2004; Beers et al., 2005; Gualtieri and Evans, 1988; Kim et al., 2006; Lee et al., 2005; Mahalick et al., 1998; McDowell et al., 1998; McMahon et al., 2009; Meythaler et al., 2002; Moein et al., 2006; Mooney and Haas, 1993; Plenger et al., 1996; Schneider et al., 1999; Speech et al., 1993; Whyte et al., 1997,2004,2008; Williams et al., 1998; Willmott and Ponsford, 2009; Willmott et al., 2009; Table 1). Dopamine agonists were compared to placebo in all but 2 studies (Beers et al., 2005; Kim et al., 2006). One these studies reported a comparison between amantadine and standard care (Beers et al., 2005). Sertraline was used as a comparator to methylphenidate in the other study (Kim et al., 2006). Of the 20 included studies, 14 (70%) used a crossover design, 5 (25%) used a parallel design, and 1 (5%) used a time series analysis. The majority of studies (80%) were single-center studies. All studies were published in English, and the majority was from the U.S.A. Other countries represented were Iran, Australia, and Korea.

Dosage regimen

Dosage did not vary much among the methylphenidate studies (0.3–0.6 mg/kg per day for adults). However, duration of treatment and follow-up period ranged from 1 day to 1 month. A similar observation can be drawn regarding the doses of amantadine that were used (200–300 mg/kg/d in adults, and 4–6 mg/kg/d in children). Duration of treatment among the studies using amantadine varied from 2 to 12 weeks. A higher variation of dosing was seen among the 2 studies using bromocriptine (2.5–10 mg/d). One study used a single dose (McDowell et al., 1998), and the other study followed patients for 4 weeks (Whyte et al., 2008).

Patient characteristics

Seven of the 20 studies included children (Beers et al., 2005; Gualtieri and Evans, 1988; Mahalick et al., 1998; McDowell et al., 1998; McMahon et al., 2009; Whyte et al., 1997; Williams et al., 1998), 3 of them exclusively (Beers et al., 2005; Mahalick et al., 1998; Williams et al., 1998). No study included patients over 65 years old, although age was not always mentioned as an exclusion criterion. Some information on baseline level of functioning and psychological and cognitive conditions was available in 12 of 20 studies (Alban et al., 2004; Beers et al., 2005; Gualtieri and Evans, 1988; Kim et al., 2006; Lee et al., 2005; McMahon et al., 2009; Meythaler et al., 2002; Mooney and Haas 1993; Whyte et al., 1997,2004,2008; Willmott and Ponsford, 2009).

The level of functioning prior to randomization was assessed using different scales, including the Disability Rating Scale (Alban et al., 2004; Whyte et al., 1997,2004,2008), the Glasgow Outcome Scale (Gualtieri, 1988; Gualtieri and Evans, 1988; Mooney and Haas, 1993), and the Rancho Los Amigos Scale (Gualtieri and Evans, 1988). Patients were described as severely disabled in 2 studies (McMahon et al., 2009; Meythaler et al., 2002), moderately disabled in 3 studies (Alban et al., 2004; Whyte et al., 1997,2004), and mildly to moderately disabled in 2 studies (Gualtieri and Evans, 1988; Mooney and Haas, 1993).

Patients were described as having significant behavioral or cognitive deficits in 3 studies (Beers et al., 2005; Meythaler et al., 2002; McMahon et al., 2009), one of them including only comatose patients (McMahon et al., 2009). Other studies reporting cognitive status included patients with intellectual quotient within normal range (Alban et al., 2004; Gualtieri and Evans, 1988; Speech et al., 1993; Whyte et al., 2004; Williams et al., 1998; Willmott and Ponsford 2009; Willmott et al., 2009), or with mild cognitive impairment (Kim et al., 2006; Lee et al., 2005). No study reported quality of life prior to randomization.

Severity of TBI

Location of patients at randomization can be found in Table 1. Of note, only 3 studies included patients in a critical care setting (intensive care unit, trauma unit, or emergency department). Heterogeneity among and within studies arises for the most part from the time elapsed from trauma to randomization, varying from less than 48 h to 34.2 years post-TBI. Information on global severity of illness at the time of injury (e.g., Abbreviated Injury Scale, Injury Severity Score, or Acute Physiology and Chronic Health Evaluation II) was not available in any of the studies. Only 7/20 (35%) studies reported CT scan findings (Alban et al., 2004; Lee et al., 2005; Mahalick et al., 1998; McDowell et al., 1998; Moein et al., 2006; Schneider et al., 1999; Whyte et al., 2004), and it was not possible to calculate a Marshall or Rotterdam score (Saatman et al., 2008). Severity of TBI was assessed using 3 different tools: loss of consciousness after injury, presence of post-traumatic amnesia, and the Glasgow Coma Scale. Nine studies (45%) included exclusively patients suffering from severe TBI (Alban et al., 2004; Gualtieri and Evans, 1988; Meythaler et al., 2002; Mooney and Haas, 1993; Whyte et al., 1997,2004,2008; Willmott and Ponsford, 2009; Willmott et al., 2009).

Studies describing the mechanism of trauma included only blunt TBI, except for 2 patients with penetrating TBI (McDowell et al., 1998), 1 aneurysm (McDowell et al., 1998), 1 stroke (McMahon et al., 2009), 1 cerebral anoxia (McMahon et al., 2009), and 1 suspected assault with unknown lesions (McDowell et al., 1998). These patients were included in the analysis since it was not possible to exclude them from the data provided.

Efficacy outcomes

Pooled estimates of outcome measures were planned, but were not calculated due to the inconsistent outcome measures used. Efficacy outcomes included mostly neuropsychological tests. Only one study used length of stay as a primary outcome, and showed both reduced hospital and intensive care unit stays with methylphenidate (Moein et al., 2006).

Neuropsychological assessment

A total of 76 different neuropsychological tests were used to assess efficacy outcomes (see Figs. 2 –7). Of these, 45 (59%) tests were only used in one study. Of note, individual results of tests were not provided in 31 out of 128 assessments (24%). The tests were categorized into cognitive- and behavioral-predominant domains, except for 4 tests (Selective Attention Task, Symbol Digit Modalities Test, Sustained Arousal and/or Attention Task 50/50 or 20/80, and Wisconsin Sorting Card Test), that were included in 2 domains.

FIG. 2.

Tests assessing executive functioning.

FIG. 3.

Tests assessing attention (WAIS-R, revised Wechsler Adult Intelligence Scale).

FIG. 4.

Tests assessing processing speed and reaction time.

FIG. 5.

Tests assessing memory.

FIG. 6.

Tests assessing mood and behavior.

FIG. 7.

Other tests (MMSE, Mini-Mental State Examination; FIM-COG, Functional Independence Measure trademark-cognitive; Wee-FIM, Functional Independence Measure for Children; GCS, Glasgow Coma Scale).

Executive functioning

Executive functioning was assessed by a total of 6 tests in 8 studies. Three tests showed statistically positive results in at least one study (Fig. 2). Only “Sustained Arousal” and “Attention tasks,” an experimental test, showed positive results in two different studies (Whyte et al., 1997,2004). These two studies evaluated methylphenidate in rehabilitating severe TBI.

Attention

A total of 21 tests were used to evaluate attention in 15 studies (Fig. 3). Of these, 10 tests (48%) yielded statistically positive results in at least one study. Two tests (Ruff 2&7 and Sustained arousal and attention tasks) had statistically significant positive results confirmed by two different studies (Mahalick et al., 1998; Whyte et al., 1997,2004; Willmott and Ponsford, 2009). All these studies evaluated methylphenidate in a heterogeneous population including mild to severe TBI. Out of the 21 tests, 9 (43%) were experimental measures.

Reaction time

Processing speed and reaction time was evaluated in 8 studies, using a total of 8 different tests (Lee et al., 2005; Plenger et al., 1996; Speech et al., 1993; Whyte et al., 1997, 2004, 2008; Williams et al., 1998; Willmott and Ponsford, 2009; Fig. 4). Five of them were experimental. Positive results were shown by 3 tests. Choice reaction time task, a test used in all 8 studies, yielded positive results in 50% of subjects (Lee et al., 2005; Whyte et al., 1997; Willmott and Ponsford, 2009). Studies showing positive results evaluated methylphenidate in mild and severe TBI, with duration of treatment varying from 6 days to 4 weeks.

Memory

Memory was assessed by only 4 studies (Gualtieri and Evans, 1988; Plenger et al., 1996; Schneider et al., 1999; Speech et al., 1993), using a total of 4 tests (Fig. 5). All tests are published. Only 2 studies report individual test results, both not showing any statistically significant difference.

Mood and behavior

A total of 21 tests were used to evaluate mood or behavior (Fig. 6). No trend can be drawn from the analysis of the results. A total of 6 studies showed positive results on a behavioral measure (Beers et al., 2005; Lee et al., 2005; McDowell et al., 1998; Mooney and Haas, 1993; Whyte et al., 1997,2004), but none of these positive results were replicated in another study.

Other measures

A total of 20 tests could not be classified as either a cognitive domain measure or a behavior measure, and are thus classified as other measures (Fig. 7). Of these, the Woodcock-Johnson battery, measuring general intellectual ability, specific cognitive abilities, scholastic aptitude, oral language, and achievement, showed positive results in one study (Lee et al., 2005). The Disability Rating Scale, which measures general functional changes over the course of recovery from TBI, yielded positive results in one study (Plenger et al., 1996), which were not corroborated by a second (Meythaler et al., 2002).

In summary, no clear beneficial effect could be drawn in any of the predefined categories of outcome (see Figs. 2 –7). Predetermined sensitivity analyses did not provide additional information.

Safety outcomes

Twelve of 20 studies reported evaluation of adverse events (Alban et al., 2004; Beers et al., 2005; Lee et al., 2005; McMahon et al., 2009; Meythaler et al., 2002; Moein et al., 2006; Mooney and Haas 1993; Plenger et al., 1996; Schneider et al., 1999; Whyte et al., 2008; Williams et al., 1998; Willmott et al., 2009). However, only 5 studies (Alban et al., 2004; McMahon et al., 2009; Plenger et al., 1996; Williams et al., 1998; Willmott et al., 2009) systematically assessed safety using pre-defined objective measures or a tool for at least one side effect.

Two RCTs evaluating methylphenidate reported results of side effects in a second publication as a primary outcome: Whyte and associates (2004) in Alban and associates (2004), and Willmott and Ponsford (2009) in Willmott and associates (2009). Although both studies mentioned a statistically significant increase in blood pressure and heart rate, their clinical significance was not described. Among the 10 remaining studies using methylphenidate, 5 reported the evaluation of side effects (Table 2). No study reported systematic measures of aggressive behavior, agitation, or delirium. All four studies using amantadine reported side effects. Most side effects were reported as minor, and resolved spontaneously upon discontinuation of the study drug.

Table 2.

Evaluation of Safety and Side Effects

	Bromocriptine			Amantadine			Methylphenidate
Side effects	Number of events	Studies	Comments	Number of events	Studies	Comments	Number of events	Studies	Comments
Rash/itching	3	Whyte et al., 2008		3	Schneider et al., 1999; Beers et al., 2005	1 patient withdrawn because of rash (Schneider et al., 1999)	5	Alban et al., 2004
Seizures	N/A	None		1	Meythaler et al., 2002; Beers et al., 2005		2	Moein et al., 2006
Delirium/confusion/hallucinations	0	Whyte et al., 2008		N/A	None		4	Alban et al., 2004
Hemodynamic changes	2	Whyte et al., 2008	1 patient withdrawn because of decreased blood pressure (Whyte et al., 2008)	N/A	None		?	Lee et al., 2005; Alban et al., 2004; Moein et al., 2006	Number of events missing (Lee et al., 2005) Statistically more prevalent than with placebo (p<0.05) (Moein et al., 2006; Alban et al., 2004)
Dizziness/lightheadedness	2	Whyte et al., 2008	1 patient withdrawn because of lightheadedness and 1 patient withdrawn because of dizziness (Whyte et al., 2008)	N/A	None		4	Alban et al., 2004
Diarrhea/constipation	4	Whyte et al., 2008		?	Beers et al., 2005	Number of events missing	2	Alban et al., 2004	Number of events missing (Lee et al., 2005)
Nausea/vomiting	2	Whyte et al., 2008	1 patient withdrawn because of vomiting (Whyte et al., 2008)	1?	Beers et al., 2005; McMahon et al., 2009	Number of events missing	2 ?	Alban et al., 2004; Lee et al 2005	Number of events missing (Lee et al., 2005)
Headache	N/A	None		N/A	None		?	Plenger et al., 1996	Number of events missing
Anxiety	1	Whyte et al., 2008		N/A	None		7	Alban et al., 2004
Tremors	2	Whyte et al., 2008		N/A	None		4	Alban et al., 2004
Sleep disorder/drowsiness	1	Whyte et al., 2008		?	Numbers of events missing (Beers et al., 2005) No difference found on sleep questionnaire (McMahon et al., 2009)		12	Alban et al., 2004	Statistically more prevalent than with placebo (p<0.05) (Alban et al., 2004) Number of events missing (Plenger et al., 1996; Willmott et al., 2009)
Others	1 sexual disability	Whyte et al., 2008		N/A	None		4 Irritability; 11 reduction in appetite; 2 difficulty in urinating; 2 blurred vision; 6 dry mouth; 2 weakness; 1 sexual disability	Alban et al., 2004; Willmott et al., 2009	Number of events for irritability missing (Willmott et al., 2009) Irritability and reduction in appetite statistically more prevalent than with placebo

?=total number of events impossible to calculate since data were not provided by at least one study.

All side effects were predetermined except for side effects listed in the “others” section.

Itching and tremors were the most common side effects reported among patients on bromocriptine. However, the incidence did not differ significantly from placebo.

Five patients were withdrawn from the study because of either probable or possible side effects (Beers et al., 2005; Schneider et al., 1999; Whyte et al., 2008). These included vomiting, dizziness, and elevated blood pressure, light-headedness and behavioral agitation, and rash and seizure.

Risk of bias

The analysis of risk of bias revealed that the majority of studies did not provide enough information on sequence generation and allocation concealment to assess their appropriateness (Table 3). All but one study (Mooney and Haas, 1993) were double-blinded, but the efficacy of randomization was rarely assessed. Of note, in 2 studies, the majority of patients or caregivers correctly guessed their attrition (Speech et al., 1993; Willmott and Ponsford, 2009). None of the studies had a published version of their protocol. Therefore, the selective reporting item could not be evaluated. Other sources of bias could be found in the majority of studies, and are described in Table 3. Based on the Cochrane Handbook for Systematic Reviews, crossover design was considered appropriate if the randomized population included only stable patients (Higgins et al., 2008). Based on this criterion, crossover design was considered inappropriate in all studies. The most common bias retrieved in parallel design studies was selection bias. A lack of information did not allow comparison of groups to determine the efficacy of randomization.

Table 3.

Risk of Bias Assessment

	Risk of bias assessment items
Study (drug)	Sequence generation	Allocation	Blinding	Outcome data	Selective reporting	Other threats to validity
Willmott et al., 2009 (MPH)	L	L	H 75% of patients guessed correctly their attrition	L	U	H Crossover design less suitable for patients in lower range of timing post TBICarry-over effect possible
Moein et al., 2006 (MPH)	H Based on hospital record number	H Based on hospital record number	L	L	U	H Impossible to assess similarity of groupsPossible confusion bias affecting length of stay
Kim et al., 2006 (MPH)	L	L	L	L	U	H Impossible to assess if groups are similar
Lee et al., 2005 (MPH)	U	U	U	H Missing information on excluded patients	U	L Impossible to assess similarity of groupsChoice of tool and comparator (depression)
Whyte et al., 2004 (MPH)	U	U	L	H Participants that can't perform a task or have schedule conflicts are withdrawn	U	H Crossover design less suitable for unstable patientsRisk of carry-overLow external validity
Mahalick et al.,1998 (MPH)	U	U	L	L	U	H Crossover less suitable for unstable patientsRisk of carry over
Williams et al., 1998 (MPH)	U	U	L	H Several measures not completed by all subjects	U	H Crossover less suitable for unstable patientsLow external validity
Gualtieri and Evans, 1988 (MPH)	U	U	L	L	H Reporting results for responders only	H Crossover less suitable for unstable patientsRisk of carry-over
Mooney and Haas, 1993 (MPH)	U	U	H Simple blind	L	U	H Impossible to assess similarity of groups
Speech et al., 1993 (MPH)	U	U	H 12 patients and 10 observers correctly guessed attrition	U	U	H Crossover less suitable for unstable patientsRisk of carry over
Whyte et al., 2008 (B)	U	U	L	H Patients randomized not analyzed due to serious adverse eventsPatients consented but dropped or withdrawn	U	H 1/3 of patients did not crossoverLow external validityCrossover less suitable for unstable patientsRisk of carry-over
McDowell et al., 1998 (B)	U	U	L	H Number of patients varies with test	U	H Crossover less suitable for unstable patientsRisk of carry over
McMahon et al., 2009 (AM)	U	U	L	H 2 patients withdrawn from analysis	U	H Crossover less suitable for unstable patients
Beers et al., 2005 (AM)	L	U	H Parents aware of allocation	H Patients lost to follow up and their allocation is unknownResults on outcome provided only for 13 patients	U	H Standard treatment is not describedImpossible to assess if groups are similar
Meythaler et al., 2002 (AM)	U	U	U	L	H Number of patients screened in unknown	H Crossover design unsuitable for acute patientsRisk of carry-over
Schneider et al., 1999 (AM)	U	U	L	H Missing data from 8 subjects who did not complete the study	U	H Timing from injury unknown

U, uncertain risk of bias; H, high risk of bias; L, low risk of bias; MPH, methylphenidate; B, bromocriptine; AM, amantadine.

Discussion

In this systematic review, we identified 20 RCTs evaluating the effect of DA on different functional outcome measures. However, considering the important clinical inconsistencies, a meta-analysis of the included studies could not be performed. As an effect, we could not draw meaningful conclusions regarding the efficacy of dopamine agonists in the patient recovering from TBI. The main factors contributing to our inability to pool data were the lack of consistency concerning patient severity of trauma, timing from injury, pre-randomization psychological and cognitive condition, and most importantly, the wide range of clinical outcome measures used in the different studies. While other systematic reviews of pharmacological therapies including dopamine agonists exist (Fleminger et al., 2006; Sivan et al., 2010; Writer and Schillerstrom, 2009), this is the first to focus on the use of these drugs specifically in patients suffering from TBI, and to evaluate both cognitive and behavioral outcomes in addition to other meaningful clinical outcomes from an acute care point of view. Methodological limitations of other reviews include the use of a unique database, inclusion of non-randomized trials, an incomplete search strategy description, language restriction, and lack of rigorous evaluation of bias (Sivan et al., 2010; Writer and Schillerstrom, 2009).

The impressive number of different tests used to assess neuropsychological outcomes in the included studies may be due to different reasons, such as the lack of a gold standard, the broad range of year of publication, the familiarity with a particular battery of tests available, and the resources available to researchers in different settings. Many studies used non-validated or experimental tools to assess outcomes. Moreover, clinical interpretation of statistically significant results was often difficult due to a lack of effect size and reliable change index reporting, the use of composite measures, and the lack of reproducibility of the neuropsychological tests among the patients. Choice of outcome measures and their analysis in TBI is still a matter of debate (Bagiella et al., 2010; Maas et al., 2010). Heterogeneity in outcome measures precludes any comparison between studies, and consequently from drawing conclusions regarding the efficacy of DA in TBI. Different groups have proposed various validated outcome combinations, including quality of life, neurobehavioral and neuropsychological status, and global functional outcome, but none has reached the status of a gold standard (Shukla et al., 2011). Future studies should aim at evaluating outcomes that are well described in the literature, validated, appropriate for international use, have well established normative data, are applicable to a broad spectrum of TBI, and are easy to administer (Wilde et al., 2010). Consequently, the Traumatic Brain Injury Outcome Workgroup has proposed selected outcomes evaluating different domains according to those recommendations in both adults and pediatric populations (McCauley et al., 2011; Wilde et al., 2010). Perhaps these recommendations should guide future studies regarding the efficacy of DA in TBI.

The poor methodological quality of the included studies was another concerning finding. The majority of studies used a crossover design, which may be appropriate when high inter-patient heterogeneity is likely, such as in a TBI population. However, patient condition should be stable in order to avoid the natural evolution of the condition to account for the benefit or harm. Evolution and brain recuperation from a TBI may continue to occur up to several years after injury and is not predictable (Lammi et al., 2005). A crossover design also mandates a washout period between treatments, which has to be long enough to avoid a carry-over effect. It has been shown that dopamine receptors can downregulate during treatment and upregulate following cessation of therapy in patients suffering from Parkinson's disease (Thobois et al., 2004). In the studies identified, the washout period was often not long enough to avoid a carry-over effect. The use of a parallel design, which would overcome the carry-over problem, is limited by the heterogeneity of the TBI population, which poses a methodological challenge. Similarity of the study groups seen in this population is often impossible to achieve unless large sample sizes are used, and pre-specified covariate adjustments are made (Maas et al., 2004,2010).

Assessment of safety was not consistent between studies, and a lack of rigor in the methodology of side-effect evaluation was considered a major limitation. Although some studies did use a questionnaire or a standardized tool to assess side effects, they often relied on self-reporting of symptoms. The studies also often failed to look prospectively for potentially concerning side effects in a TBI population. Neurological and psychiatric adverse reactions such as seizures, delirium, hallucinations, and insomnia should be targeted as key safety outcomes, since they are clinically relevant in the TBI population, and are associated with the use of dopamine agonists in other human populations and animal models (Barone et al., 1992; Wood, 2010). A lack of information regarding diagnosis of these side effects was also an issue.

Despite the results of our review, the dopaminergic system is an obvious target for research of cognitive impairments in TBI. Dopamine receptors are known to be present in many areas of the brain important to cognition, and are often damaged post-TBI (the hippocampus, striatum, and frontal cortex; Bales et al., 2009). The role of dopamine receptors in cognitive domains such as attention, working memory, executive functioning, and speed of processing is still a matter of debate (Nieoullon, 2002). However, some observations in humans showing impairment in visuospatial, executive, and verbal fluency tests related to induced parkinsonism lead one to think that dopamine participates in cognition (Burruss et al., 2000; Nieoullon, 2002; Stern et al., 1990). Radioimaging studies in patients with Parkinson's disease, schizophrenia, and attention-deficit/hyperactivity disorder provides supporting evidence (Bruck et al., 2001; Duchesne et al., 2002; Heilman et al., 1991; Tanaka, 2006).

The influence of dopaminergic drugs on neuronal plasticity, a concept of brain adaptation to injury over time, is also a concerning feature (Ampuero et al., 2010; Larson and Zollman, 2010). Dysfunction of the molecular processes influencing neuronal plasticity in the hippocampus may be a potential cause for temporal and spatial memory impairment seen in TBI (Ampuero et al., 2010; Falo et al., 2006). The role for dopamine pathways in this process is unclear, but some data suggest that a modification of plasticity occurs with the exposure of various neurotransmitter agonists and antagonists, including dopamine (Bales et al., 2009; Meintzschel and Ziemann, 2006). For example, benzodiazepines and haloperidol have been incriminated as possible causes of adverse effects on neuronal plasticity (Ampuero et al., 2010; Falo et al., 2006). Proper safety evaluation of dopamine agonists in TBI is therefore warranted.

Conclusion

This systematic review is the first to evaluate the effect of the administration of dopamine agonists on different domains of cognition, behavior, and safety in TBI patients. Because of the significant clinical heterogeneity in the outcome measures used in the different studies, pooled analyses could not be performed, and consequently clear conclusions about the efficacy or safety of dopamine agonists in TBI patients are difficult to draw. Despite the low incidence of adverse events reported, most studies did not systematically screen for them. In summary, our systematic review does not support the use of dopamine agonists as part of the treatment regimen for improving functional outcome measures in either the acute or sub-acute phases of care in TBI patients. Future studies should first aim at standardizing and validating a neuropsychological battery of tests in TBI. A clinical trial of dopamine agonists in TBI could then be designed using standardized outcome measures with clear clinical significance, including a validated neuropsychological battery of tests. It should also use a parallel design in a large sample with predefined covariates, and should systematically screen for potential side effects.

Footnotes

Acknowledgments

The authors would like to thank Ms. Marie-Soleil Delisle, Ms. Marie Frenette, and Ms. Jodi Peters for their help with this study.

Author Disclosure Statement

No competing financial interests exist.

Appendix 1.

Search Strategy for MEDLINE

1	head trauma.mp.	9919
2	head injury.mp.	36681
3	head injuries.mp.	11177
4	brain trauma.mp.	2421
5	“brain injuries”.mp. or expBrain Injuries/	105383
6	cerebral trauma.mp.	998
7	cerebral injury.mp.	2592
8	cerebral injuries.mp.	666
9	“craniocerebral trauma”.mp. or exp Craniocerebral Trauma/	211373
10	craniocerebral injury.mp.	920
11	craniocerebral injuries.mp.	1008
12	brain injury.mp.	81937
13	Traumatic brain injury.mp.	24215
14	Traumatic brain injuries.mp.	1275
15	1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14	234953
16	exp Dopamine Agonists/	134451
17	dopamine agonist$.tw.	11285
18	bromocriptine.tw.	12227
19	amantadine.tw.	5300
20	levodopa.tw.	14134
21	methylphenidate.tw.	8292
22	apomorphine.tw.	15887
23	ropinirole.tw.	1058
24	pramipexole.tw.	1367
25	pergolide.tw.	2025
26	cabergoline.tw.	1591
27	exp Dopamine Agents/	298400
28	or/16-27	312593
29	15 and 28	3630
30	Randomized controlled trial.pt.	572709
31	Controlled clinical trial.pt.	160808
32	randomized.ab.	548100
33	placebo.ab.	338077
34	Drug therapy.fs.	1393362
35	randomly.ab.	359540
36	group.ab.	2580745
37	trial.ab.	506749
38	37 or 36 or 35 or 34 or 33 or 32 or 31 or 30	4626546
39	38 and 29	925
40	remove duplicates from 39	790

References

Alban

J.P.

, Hopson

M.M.

, Ly

, Whyte

2004. Effect of methylphenidate on vital signs and adverse effects in adults with traumatic brain injury. Am. J. Phys. Med. Rehabil., 83:131–137.

Ampuero

, Rubio

F.J.

, Falcon

, Sandoval

, Diaz-Veliz

, Gonzalez

R.E.

, Earle

, Dagnino-Subiabre

, Aboitiz

, Orrego

, Wyneken

2010. Chronic fluoxetine treatment induces structural plasticity and selective changes in glutamate receptor subunits in the rat cerebral cortex. Neuroscience, 169:98–108.

Bagiella

, Novack

T.A.

, Ansel

, Diaz-Arrastia

, Dikmen

, Hart

, Temkin

2010. Measuring outcome in traumatic brain injury treatment trials: recommendations from the traumatic brain injury clinical trials network. J. Head Trauma Rehabil., 25:375–382.

Bales

J.W.

, Wagner

A.K.

, Kline

A.E.

, Dixon

C.E.

2009. Persistent cognitive dysfunction after traumatic brain injury: A dopamine hypothesis. Neurosci. Biobehav. Rev., 33:981–1003.

Barone

, Palma

, de Bartolomeis

, Cicarelli

, Campanella

1992. Dopaminergic regulation of epileptic activity. Neurochem. Int., 20,Suppl:245S–249S.

Beers

S.R.

, Skold

, Dixon

C.E.

, Adelson

P.D.

2005. Neurobehavioral effects of amantadine after pediatric traumatic brain injury: a preliminary report. J. Head Trauma Rehabil., 20:450–463.

Bruck

, Portin

, Lindell

, Laihinen

, Bergman

, Haaparanta

, Solin

, Rinne

J.O.

2001. Positron emission tomography shows that impaired frontal lobe functioning in Parkinson's disease is related to dopaminergic hypofunction in the caudate nucleus. Neurosci. Lett., 311:81–84.

Burruss

J.W.

, Hurley

R.A.

, Taber

K.H.

, Rauch

R.A.

, Norton

R.E.

, Hayman

L.A.

2000. Functional neuroanatomy of the frontal lobe circuits. Radiology, 214:227–230.

Corrigan

J.D.

, Selassie

A.W.

, Orman

J.A.

2010. The epidemiology of traumatic brain injury. J. Head Trauma Rehabil., 25:72–80.

10.

Duchesne

, Soucy

J.P.

, Masson

, Chouinard

, Bedard

M.A.

2002. Cognitive deficits and striatal dopaminergic denervation in Parkinson's disease: a single photon emission computed tomography study using 123iodine-beta-CIT in patients on and off levodopa. Clin. Neuropharmacol., 25:216–224.

11.

Falo

M.C.

, Fillmore

H.L.

, Reeves

T.M.

, Phillips

L.L.

2006. Matrix metalloproteinase-3 expression profile differentiates adaptive and maladaptive synaptic plasticity induced by traumatic brain injury. J. Neurosci. Res., 84:768–781.

12.

Fleminger

, Greenwook

R.J.

, Oliver

D.L.

2006. Pharmacological management for agitation and aggression in people with acquired brain injury. Cochrane Database Syst. Rev., 18:CD003299.

13.

Fleminger

2008. Long-term psychiatric disorders after traumatic brain injury. Eur. J. Anaesthesiol. Suppl., 42:123–130.

14.

Francisco

G.E.

, Walker

W.C.

, Zasler

N.D.

, Bouffard

M.H.

2007. Pharmacological management of neurobehavioural sequelae of traumatic brain injury: a survey of current physiatric practice. Brain Inj., 21:1007–1014.

15.

Green

R.E.

, Colella

, Hebert

D.A.

, Bayley

, Kang

H.S.

, Till

, Monette

2008. Prediction of return to productivity after severe traumatic brain injury: Investigations of optimal neuropsychological tests and timing of assessment. Arch. Phys. Med. Rehabil., 89:S51–S60.

16.

Gualtieri

C.T.

, Evans

R.W.

1988. Stimulant treatment for the neurobehavioural sequelae of traumatic brain injury. Brain Inj., 2:273–290.

17.

Gualtieri

C.T.

1988. Pharmacotherapy and the neurobehavioural sequelae of traumatic brain injury. Brain Inj., 2:101–129.

18.

Heilman

K.M.

, Voeller

K.K.

, Nadeau

S.E.

1991. A possible pathophysiologic substrate of attention deficit hyperactivity disorder. J. Child Neurol., 6,Suppl:S76–S81.

19.

Higgins

, Altman

2008. Assessing risk of bias in included studies. Cochrane Handbook for Systematic Reviews of Interventions. Higgins

J.P.T.

, Green

The Cochrane Collaboration, Version 5.0.1. www.cochrane-handbook.org.

20.

Higgins

, Deeks

, Altman

2008. Special topics in statistics. Cochrane Handbook for Systematic Reviews of Interventions. Higgins

J.P.T.

, Green

The Cochrane Collaboration, Version 5.0.1. www.cochrane-handbook.org.

21.

Kim

Y.H.

, Ko

M.H.

, Na

S.Y.

, Park

S.H.

, Kim

K.W.

2006. Effects of single-dose methylphenidate on cognitive performance in patients with traumatic brain injury: a double-blind placebo-controlled study. Clin. Rehabil., 20:24–30.

22.

Lammi

M.H.

, Smith

V.H.

, Tate

R.L.

, Taylor

C.M.

2005. The minimally conscious state and recovery potential: a follow-up study 2 to 5 years after traumatic brain injury. Arch. Phys. Med. Rehabil., 86:746–754.

23.

Larson

E.B.

, Zollman

F.S.

2010. The effect of sleep medications on cognitive recovery from traumatic brain injury. J. Head Trauma Rehabil., 25:61–67.

24.

Lee

, Kim

S.W.

, Kim

J.M.

, Shin

I.S.

, Yang

S.J.

, Yoon

J.S.

2005. Comparing effects of methylphenidate, sertraline and placebo on neuropsychiatric sequelae in patients with traumatic brain injury. Hum. Psychopharmacol., 20:97–104.

25.

Lefebvre

, Manheimer

, Glanville

2009. Searching for studies. Cochrane Handbook for Systematic Reviews of Interventions. Higgins

J.P.T.

, Green

The Cochrane Collaboration, Version 5.0.2. www.cochrane-handbook.org.

26.

Loke

, Price

, Herxheimer

2008. Adverse effects. Cochrane Handbook for Systematic Review of Interventions. Higgins

J.P.T.

, Green

The Cochrane Collaboration, Version 5.0.1. www.cochrane-handbook.org.

27.

Maas

A.I.

, Marmarou

, Murray

G.D.

, Steyerberg

E.W.

2004. Clinical trials in traumatic brain injury: current problems and future solutions. Acta Neurochir. Suppl., 89:113–118.

28.

Maas

A.I.

, Steyerberg

E.W.

, Marmarou

, McHugh

G.S.

, Lingsma

H.F.

, Butcher

, Lu

, Weir

, Roozenbeek

, Murray

G.D.

2010. IMPACT recommendations for improving the design and analysis of clinical trials in moderate to severe traumatic brain injury. Neurotherapeutics, 7:127–134.

29.

Mahalick

D.M.

, Carmel

P.W.

, Greenberg

J.P.

, Molofsky

, Brown

J.A.

, Heary

R.F.

, Marks

, Zampella

, Hodosh

, von der Schmidt

3rd . 1998. Psychopharmacologic treatment of acquired attention disorders in children with brain injury. Pediatr. Neurosurg., 29:121–126.

30.

McCauley

S.R.

, Wilde

E.A.

, Anderson

V.A.

, Bedell

, Beers

S.R.

, Campbell

T.F.

, Chapman

S.B.

, Ewing-Cobbs

, Gerring

, Gioia

G.A.

, Levin

, Michaud

, Prasad

M.R.

, Swaine

, Turkstra

L.A.

, Wade

S.L.

, Yeates

K.O.

2011. Recommendations for the use of common outcome measures in pediatric traumatic brain injury research. J. Neurotrauma[Epub ahead of print].

31.

McDowell

, Whyte

, D'Esposito

1998. Differential effect of a dopaminergic agonist on prefrontal function in traumatic brain injury patients. Brain, 121,Pt. 6:1155–1164.

32.

McMahon

M.A.

, Vargus-Adams

J.N.

, Michaud

L.J.

, Bean

2009. Effects of amantadine in children with impaired consciousness caused by acquired brain injury: a pilot study. Am. J. Phys. Med. Rehabil., 88:525–532.

33.

Meintzschel

, Ziemann

2006. Modification of practice-dependent plasticity in human motor cortex by neuromodulators. Cereb. Cortex, 16:1106–1115.

34.

Meythaler

J.M.

, Brunner

R.C.

, Johnson

, Novack

T.A.

2002. Amantadine to improve neurorecovery in traumatic brain injury-associated diffuse axonal injury: a pilot double-blind randomized trial. J. Head Trauma Rehabil., 17:300–313.

35.

Moein

, Khalili

H.A.

, Keramatian

2006. Effect of methylphenidate on ICU and hospital length of stay in patients with severe and moderate traumatic brain injury. Clin. Neurol. Neurosurg., 108:539–542.

36.

Moher

, Liberati

, Tetzlaff

, Altman

D.G.

2009. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ, 339:b2535.

37.

Mooney

G.F.

, Haas

L.J.

1993. Effect of methylphenidate on brain injury-related anger. Arch. Phys. Med. Rehabil., 74:153–160.

38.

Nieoullon

2002. Dopamine and the regulation of cognition and attention. Prog. Neurobiol., 67:53–83.

39.

Patry

B.N.

, Mateer

C.A.

2006. Neuropsychological assessment of moderate to severe traumatic brain injury. Psychological Knowledge in Court. Young

, Nicholson

, Kane

A.W.

Springer: New York, 352–377.

40.

Plenger

P.M.

, Dixon

C.E.

, Castillo

R.M.

, Frankowski

R.F.

, Yablon

S.A.

, Levin

H.S.

1996. Subacute methylphenidate treatment for moderate to moderately severe traumatic brain injury: a preliminary double-blind placebo-controlled study. Arch. Phys. Med. Rehabil., 77:536–540.

41.

Ruttan

, Martin

, Liu

, Colella

, Green

R.E.

2008. Long-term cognitive outcome in moderate to severe traumatic brain injury: A meta-analysis examining timed and untimed tests at 1 and 4.5 or more years after injury. Arch. Phys. Med. Rehabil., 89:S69–S76.

42.

Saatman

K.E.

, Duhaime

A.C.

, Bullock

, Maas

A.I.

, Valadka

, Manley

G.T.

2008. Classification of traumatic brain injury for targeted therapies. J. Neurotrauma, 25:719–738.

43.

Schneider

W.N.

, Drew-Cates

, Wong

T.M.

, Dombovy

M.L.

1999. Cognitive and behavioural efficacy of amantadine in acute traumatic brain injury: an initial double-blind placebo-controlled study. Brain Inj., 13:863–872.

44.

Scottish Intercollegiate Guidelines Network. Search filters, in: Scottish Intercollegiate Guidelines Network. November 2009. http://www.sign.ac.uk/methodology/filters.html#random.

45.

Shukla

, Devi

B.I.

, Agrawal

2011. Outcome measures for traumatic brain injury. Clin. Neurol. Neurosurg., 113:435–441.

46.

Siddall

O.M.

2005. Use of methylphenidate in traumatic brain injury. Ann. Pharmacother., 39:1309–1313.

47.

Sivan

, Neumann

, Kent

, Stroud

, Bhakta

B.B.

2010. Pharmacotherapy for treatment of attention deficits after non-progressive acquired brain injury. A systematic review. Clin. Rehabil., 24:110–121.

48.

Speech

T.J.

, Rao

S.M.

, Osmon

D.C.

, Sperry

L.T.

1993. A double-blind controlled study of methylphenidate treatment in closed head injury. Brain Inj., 7:333–338.

49.

Stern

, Tetrud

J.W.

, Martin

W.R.

, Kutner

S.J.

, Langston

J.W.

1990. Cognitive change following MPTP exposure. Neurology, 40:261–264.

50.

Tanaka

2006. Dopaminergic control of working memory and its relevance to schizophrenia: a circuit dynamics perspective. Neuroscience, 139:153–171.

51.

Tenovuo

2006. Pharmacological enhancement of cognitive and behavioral deficits after traumatic brain injury. Curr. Opin. Neurol., 19:528–533.

52.

Thobois

, Vingerhoets

, Fraix

, Xie-Brustolin

, Mollion

, Costes

, Mertens

, Benabid

A.L.

, Pollak

, Broussolle

2004. Role of dopaminergic treatment in dopamine receptor down-regulation in advanced Parkinson disease: a positron emission tomographic study. Arch. Neurol., 61:1705–1709.

53.

Thurman

, Guerrero

1999. Trends in hospitalization associated with traumatic brain injury. JAMA, 282:954–957.

54.

Tsaousides

, Gordon

W.A.

2009. Cognitive rehabilitation following traumatic brain injury: Assessment to treatment. Mt. Sinai J. Med., 76:173–181.

55.

Whyte

, Hart

, Schuster

, Fleming

, Polansky

, Coslett

H.B.

1997. Effects of methylphenidate on attentional function after traumatic brain injury. A randomized, placebo-controlled trial. Am. J. Phys. Med. Rehabil., 76:440–450.

56.

Whyte

, Hart

, Vaccaro

, Grieb-Neff

, Risser

, Polansky

, Coslett

H.B.

2004. Effects of methylphenidate on attention deficits after traumatic brain injury: a multidimensional, randomized, controlled trial. Am. J. Phys. Med. Rehabil., 83:401–420.

57.

Whyte

, Vaccaro

, Grieb-Neff

, Hart

, Polansky

, Coslett

H.B.

2008. The effects of bromocriptine on attention deficits after traumatic brain injury: a placebo-controlled pilot study. Am. J. Phys. Med. Rehabil., 87:85–99.

58.

Wilde

E.A.

, Whiteneck

G.G.

, Bogner

, Bushnik

, Cifu

D.X.

, Dikmen

, French

, Giacino

J.T.

, Hart

, Malec

J.F.

, Millis

S.R.

, Novack

T.A.

, Sherer

, Tulsky

D.S.

, Vanderploeg

R.D.

, von Steinbuechel

2010. Recommendations for the use of common outcome measures in traumatic brain injury research. Arch. Phys. Med. Rehabil., 91:1650–1660.

59.

Williams

S.E.

, Ris

M.D.

, Ayyangar

, Schefft

B.K.

, Berch

1998. Recovery in pediatric brain injury: is psychostimulant medication beneficial? J. Head Trauma Rehabil., 13:73–81.

60.

Willmott

, Ponsford

2009. Efficacy of methylphenidate in the rehabilitation of attention following traumatic brain injury: a randomised, crossover, double blind, placebo controlled inpatient trial. J. Neurol. Neurosurg. Psychiatry, 80:552–557.

61.

Willmott

, Ponsford

, Olver

, Ponsford

2009. Safety of methylphenidate following traumatic brain injury: impact on vital signs and side-effects during inpatient rehabilitation. J. Rehabil. Med., 41:585–587.

62.

Wood

L.D.

2010. Clinical review and treatment of select adverse effects of dopamine receptor agonists in Parkinson's disease. Drugs Aging, 27:295–310.

63.

Writer

B.W.

, Schillerstrom

J.E.

2009. Psychopharmacological treatment for cognitive impairment in survivors of traumatic brain injury: a critical review. J. Neuropsychiatry Clin. Neurosci., 21:362–370.