If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Intrathecal and Oral Baclofen Use in Adults With Spinal Cord Injury: A Systematic Review of Efficacy in Spasticity Reduction, Functional Changes, Dosing, and Adverse Events
Corresponding author Jessica D'Amico, PhD, Kentucky Spinal Cord Injury Research Center, Frazier Rehabilitation Institute, University of Louisville, 220 Abraham Flexner Way, Louisville, KY 40202.
To examine the efficacy, dosing, and safety profiles of intrathecal and oral baclofen in treating spasticity after spinal cord injury (SCI).
Data Sources
PubMed and Cochrane Databases were searched from 1970-2018 with keywords baclofen, spinal cord injury, and efficacy.
Study Selection
The database search yielded 588 sources and 10 additional relevant publications. After removal of duplicates, 398 publications were screened.
Data Extraction
Data were extracted using the following population, intervention, comparator, outcomes, and study designs criteria: studies including adult patients with SCI with spasticity; the intervention could be oral or intrathecal administration of baclofen; selection was inclusive for control groups, surgical management, rehabilitation, and alternative pharmaceutical agents; outcomes were efficacy, dosing, and adverse events. Randomized controlled trials, observational studies, and case reports were included. Meta-analyses and systematic reviews were excluded.
Data Synthesis
A total of 98 studies were included with 1943 patients. Only 4 randomized, double-blinded, and placebo-controlled trials were reported. Thirty-nine studies examined changes in the Modified Ashworth Scale (MAS; 34 studies) and Penn Spasm scores (Penn Spasm Frequency; 19 studies), with average reductions of 1.7±1.3 and 1.6±1.4 in individuals with SCI, respectively. Of these data, a total of 6 of the 34 studies (MAS) and 2 of the 19 studies (Penn Spasm Frequency) analyzed oral baclofen. Forty-three studies addressed adverse events with muscle weakness and fatigue frequently reported.
Conclusions
Baclofen is the most commonly-prescribed antispasmodic after SCI. Surprisingly, there remains a significant lack of large, placebo-controlled, double-blinded clinical trials, with most efficacy data arising from small studies examining treatment across different etiologies. In the studies reviewed, baclofen effectively improved spasticity outcome measures, with increased efficacy through intrathecal administration. Few studies assessed how reduced neural excitability affected residual motor function and activities of daily living. A host of adverse events were reported that may negatively affect quality of life. Comparative randomized controlled trials of baclofen and alternative treatments are warranted because these have demonstrated promise in relieving spasticity with reduced adverse events and without negatively affecting residual motor function.
Spasticity is a movement disorder characterized by involuntary muscle activity such as spasms, clonus, impaired muscle coactivation patterns and increased velocity-dependent resistance to passive stretch.
Further, spasticity and the disruption of normal movement, speech, and balance often impairs function and ability to carry out activities of daily living (ADL).
In a 2018 patient-reported effect of spasticity survey, participants described stiffness associated with spasticity as the most detrimental aspect to normal function with pain in the legs reported in 57% of those with SCI.
especially transfers. Despite the significant effect of spasticity on quality of life, there are few objective measures that correlate with perceived detriment and suffering.
Spasticity after SCI develops chronically as neurons and neural circuits below the injury adapt to decreased descending inputs and movement-related sensory input. Within the spinal cord, several reflex pathways can affect sensorimotor output and changes in the excitability of these pathways occur because of decreased supraspinal modulation and plastic changes within the spinal cord itself.
At the neuronal level, the loss of descending modulation by serotonin and noradrenaline leads to both increases in the intrinsic excitability of the motor neurons themselves and disinhibition of sensory input to the motor neurons, leading to prolongation of the excitatory postsynaptic potentials.
Together, these changes contribute to an imbalance in excitatory and inhibitory inputs to the motor neurons whereby brief, innocuous sensory stimuli can trigger long-lasting involuntary muscle activity (spasms).
The most common clinical measures of spasticity are the Modified Ashworth Scale (MAS) and the Penn Spasm Frequency self-assessment. However, these clinical tools fail to entirely capture the different movement components contributing to the broader term of “spasticity,” do not correlate with neurophysiological measures of hyperexcitability,
Briefly, the MAS score assesses on a scale of 0-4 the resistance of a relaxed, single joint as it is moved throughout its full available range of motion (table 1).
It is clear that both measurements assess different aspects of spasticity, accounting for the lack of correlation between the 2 scores within the same individual.
Moreover, there is a need to develop more optimal techniques for the evaluation of spasticity in individuals with SCI. Other measures include the Tardieu Scale score, which assesses resistance to fast and slow passive mobility,
1: Slight increase in muscle tone, manifested by a catch or by minimal resistance at the end of the ROM when the affected part(s) is moved in flexion or extension
1: Mild spasms induced by stimulation
1+: Slight increase in muscle tone, manifested by a catch followed by minimal resistance throughout the remainder (less than half) of the ROM
2: Full spasms occurring less than once per hour
2: More marked increase in tone through most of the ROM, but affected part(s) easily moved
3: Spasms occurring more than once per hour
3: Considerable increase muscle in tone; passive movement difficult
Given its structure as a derivative of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA), baclofen acts as an agonist to metabotropic GABA-B receptors throughout the central nervous system to reduce the presynaptic release of excitatory neurotransmitters.
Baclofen for SCI can be administered orally or intrathecally with implantation of a metered pump. However, oral administration may be ineffective in up to 25% of patients with SCI because of baclofen's poor blood-brain barrier penetration, requiring more direct delivery via intrathecal route.
Intrathecal baclofen, introduced in the 1980s, is considered for severe spasticity, offering localized distribution to the thecal sac and lower rates of toxicity from systemic dissemination.
While substantial literature supports the use of baclofen as an effective spasmolytic, most studies analyze its use across multiple indications that may not fully appreciate mechanistic differences between pathologies and coexisting goals of recovery. For example, baclofen for spasticity after SCI reduces symptoms but also inhibits movement and functional restoration.
In this systematic review, we comprehensively investigate the efficacy in spasticity reduction, changes in function, dosing, and adverse events associated with baclofen use in the treatment of spasticity after SCI. Efficacy has been considered as reductions in spasticity scores (modified Ashworth scores, Penn Spasm scores). Recovery of motor function (motor scores, voluntary movement, standing, walking) was not often reported as an outcome; however, changes in composite scores such as the Barthel Index are reported. These scores include ADL and/or ease for caregiving as well as performance during some motor tasks.
Where available, we report level of injury and severity of injury according to the American Spinal Injury Association Impairment Scale (AIS) criteria. Quantification of efficacy, dosing, and adverse events is reported across all studies.
Methods
Data extraction
We framed the search around a Participants, Intervention, Comparison, Outcomes, Study Designs model to define the population receiving baclofen for SCI, and the study design performed to yield a comprehensive and reproducible topic search (fig 1). All studies including at least 1 patient with SCI were included in the analysis regardless of number of patients in the study with other pathologies including multiple sclerosis, stroke, cerebral palsy, and hereditary spastic paresis.
Fig 1Preferred Reporting Items for Systematic Reviews and Meta-analyses flowchart describing search and selection process of articles selected for this systematic review.
Participants, intervention, comparison, outcomes, study designs outline
Participants were adult patients aged ≥18 years using baclofen via oral or intrathecal administration for muscle spasticity secondary to SCI (acute or chronic). To include and analyze all uses of baclofen in SCI reported in the literature, studies with at least 1 patient with SCI were included, regardless of amount or focus of other pathology resulting in spasticity or additional pharmaceutical used.
Intervention was baclofen via oral or intrathecal administration for muscle spasticity, pain, or functional improvement.
Comparison between baclofen use in other pathologies (ie, multiple sclerosis, stroke, cerebral palsy), oral vs intrathecal baclofen use, with or without rehabilitation, alternative pharmaceutical agents, perioperative care regimens, or no comparison group were included in the present analysis.
Outcomes were efficacy of spasticity reduction and functional improvement using metrics such as MAS, Penn Spasm Frequency, Tardieu Scale, Fugl-Meyer, dosing of oral or intrathecal baclofen, expected or unexpected adverse events related to baclofen use, and effects on function associated with baclofen use. Clinically meaningful differences were reviewed if implicated and included in studies but not required for inclusion.
Study design was inclusive of prospective, retrospective, case reports, case series, observational studies, comparative studies, and randomized controlled trials.
Search criteria
We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines (2009) to construct the framework of the systematic review and conducted the search on November 30, 2018, using PubMed and Cochrane Databases from 1970-2018 (see fig 1).
Additional articles used in the references were incorporated from the references of those articles identified in the searches. We used Keyword and Medical Subject Headings of the National Library of Medicine terms for predictive outcomes to include the following terms, with numbered iterations for the 2 databases as follows:
Eligibility criteria: Current publication in English language: (1) PubMed: Baclofen AND Spinal Cord Injury: 510 articles; 91 included; (2) Ovid: Baclofen AND Spinal Cord Injury OR spinal cord injuries: 290 articles; 7 included.
Assessment of conflict of interest, funding for study and study design were assessed according to QUADAS criteria.
Exclusion criteria
Systematic reviews and meta-analyses were excluded. Exclusion criteria also included studies with nonhuman participants, pediatric population, language other than English, and those studies without full text. Combination protocols such as coadministration of morphine for treatment of spasticity were excluded. Also, studies that analyzed biomarkers with use of baclofen without clinical assessments were excluded.
Data evaluation
We followed the QUADAS tool to evaluate risk bias and result applicability of the studies according to 2003 guidelines. A total of 13 questions on the QUADAS survey that cover patient selection, index test, reference standard, and timing were addressed for each study and incorporated into the final analysis (supplemental table S2, available online only at http://www.archivespmr.org/).
Results
A total of 98 studies were included in the final analysis, and 26 of those were case reports (see supplemental table S1). There were limited studies (19/98) that were uniquely focused on participants with SCI only (fig 2). The pooled sample size for all studies was 1923 patients with SCI (see fig 2). Of these, injury level was reported in only 37% of individuals, with 393 cervical- and 321 thoracic-level injuries reported. Where reported (27% of individuals), the distribution of AIS scores included 137 patients with AIS A, 73 with AIS B, 74 with AIS C, and 56 with AIS D injuries. Chronicity of injury ranged from 1 month to 57 years for 632 patients with SCI (33% of total). When reported, the majority of participants with SCI were between 20 and 40 years old. A total of 1412 individual demographic and AIS injury characteristics were not reported (73% of total). Thirty-six studies addressed the efficacy of baclofen in reducing clinical measures of spasticity (MAS score, 591 individuals with SCI; Penn Spasm Frequency scores, 275 individuals with SCI). Seven studies reported reduced yet unspecified spasm scores after baclofen administration. A total of 8 studies including 128 participants with SCI were found to specifically discuss ability to walk or assessed function with established metrics in the study cohort. Most studies addressed adverse events or dosing; 48 studies specifically addressed adverse events associated with baclofen use, 18 of which were associated with oral dosing of baclofen only. A total of 50 studies examined intrathecal dosing (total of 621 individuals with SCI), and 26 studies examined oral baclofen dosing (total of 604 individuals with SCI), representing ∼64% of the individuals with SCI studied across all reported studies. Of 98 studies, only 6 randomized controlled trials
were identified to evaluate baclofen efficacy in participants with SCI; 4 of these studies were randomized, double-blinded, and placebo-controlled, while 1 study was a comparative trial (baclofen vs botulinum toxin vs physical therapy alone)
baclofen use, and in 2 of these studies the use of a saline injection as a placebo was only implemented during the screening process and therefore only to examine the acute effects of intrathecal baclofen administration.
Fig 2Proportion of total studies (N=98) reporting on efficacy, adverse events, dosing, and motor function following baclofen use in individuals with spasticity after SCI. Horizontal box plot illustrates median and IQR for no. of individuals with SCI across all studies when reported.
Dosing of baclofen was reported in 67 studies, representing the greatest portion of all studies included in this review (∼64 %). Most studies reported dosing ranges across participants with different pathologies. Other studies reported mean doses for participant cohorts across a study timeline or specific therapeutic doses for individual participants during observation and follow-up.
Oral baclofen dosing
Of the 67 studies, 26 studies reported oral baclofen dosing (n=604 individuals with SCI of 1205 total individuals reported) (fig 3). Eleven of these studies reported minimum and maximum dose information. The minimum oral dose of baclofen was 15 mg daily, with the maximum oral dose 266 mg daily. Lastly, where the average of stable doses were reported, the mean oral dose was ∼73 mg daily (n=223 individuals, ∼89% SCI).
Fig 3Baclofen dosing. (A) Intrathecal and (B) oral baclofen dosing reported across all publications. Line plots illustrate maximum (filled circles) and minimum (open circles) doses when reported. Underlying box plot illustrates no. of individuals reported across each publication. Box plots illustrate average minimum, maximum, and average baclofen doses across all publications.
Of the 67 studies, 50 studies reported intrathecal baclofen dosing (n=621 individuals with SCI of 1340 total individuals reported) (see fig 3). Minimum and maximum dose information was provided for 26 studies. To summarize, the minimum intrathecal dose reported was 60 μg over 24 hours, with the maximum dose 1668 μg over 24 hours. From the studies that reported the average or stable dose for individuals using intrathecal baclofen, the average was ∼308 μg daily (n=731 individuals, ∼35% SCI).
Efficacy
Efficacy was defined as the effectiveness of baclofen in reducing clinical measures of spasticity. The reported outcome measures are the MAS score and the Penn Spasm Frequency score (see table 1). The effects of baclofen on these clinical outcome measures were reported in 36 studies (∼38 % of total), and of these studies, only 18 studies specifically assessed the efficacy of baclofen use in individuals with spasticity after SCI. These studies represented 598 participants with SCI over the last 31 years and 365 participants with other neurologic impairments.
Changes in MAS
Seventeen studies reported on the efficacy of baclofen in reducing MAS in the population with SCI specifically (n=350 individuals) (fig 4A). Of these, 6 studies reported a reduction of 1.0±1.7 in MAS with oral baclofen use (n=260 individuals with SCI) (see fig 4A), and 11 studies reported a reduction of 2.1±1.0 in MAS with intrathecal baclofen use (n=90 individuals with SCI) (fig 4A), demonstrating increased efficacy through intrathecal administration of baclofen.
Fig 4Efficacy of baclofen. Changes in (A) Modified Ashworth scores and (B) Penn Spasm frequency scores following oral (filled circles) and intrathecal (open circles) baclofen administration. SCI-only reported values are indicated in the line plots. Averages are illustrated in the box plots. Combined average reports values obtained across individuals of different pathologies (SCI, MS, CP). No. of individuals with SCI is shown on the right axes of all plots.
Penn Spasm scores assessing frequency and severity of spasms were reported in 20 studies (∼19% of total). From these 20 studies, 275 of 464 participants were diagnosed with SCI, with spasticity arising from different etiologies comprising the remainder of the participants. Of the 20 studies, 9 studies reported findings specific to patients with SCI (see fig 4B). Of these SCI-specific values, 2 studies reported a reduction of 0.63±0. 85 with oral baclofen use (n=47 individuals with SCI). Like changes in MAS reported above, a larger reduction in Penn Spasm scores was reported with intrathecal baclofen administration (−1.9±1.5, n=83 individuals with SCI).
Efficacy outcomes from studies examining multiple pathologies
In the studies that reported combined averages across multiple pathologies, only the efficacy of intrathecal baclofen was evaluated. In these studies, baclofen effectively reduced the MAS by 2.0±0.9, with participants with SCI representing ∼53% of the individuals in this data set (n=241 SCI of 458 total individuals reported). Baclofen effectively reduced the Penn Spasm score by 2.0±1.3, with individuals with SCI representing ∼50% of the individuals in this data set (n=145 SCI of 288 total individuals reported). These results are similar to those reported in the SCI-specific literature detailed above.
Function
A total of 6 studies discuss the effects of baclofen on residual functional capacity, representing 5.7% of studies on baclofen use in SCI. In these studies it is difficult to differentiate the direct effects of baclofen on residual motor function vs the indirect effects mediated through a reduction in spasticity. Spasticity itself directly impairs muscle function, gait, and ADL. Specifically, spasticity in the lower limbs of individuals with incomplete SCI results in a loss of voluntary flexor muscle activity, while extensor spasms contribute strongly to loss of gait function.
Impact of specific symptoms of spasticity on voluntary lower limb muscle function, gait and daily activities during subacute and chronic spinal cord injury.
Stiffness affected daily living by interfering with exercise, grooming, dressing, and hygiene tasks, whereas involuntary muscle spasms and clonus were only weakly correlated to effects on daily living.
In the studies described below, motor function was assessed using mean changes in 10-m and 6-minute walk tests, functional improvement scores, or measures such as ability to carry out ADL or instrumental ADL.
Improvements in walking
One study of 3 participants reports improvements in spatiotemporal gait parameters in 2 individuals after baclofen administration,
Effect of the intrathecal baclofen screening test on the spatiotemporal gait motion parameters of patients with cervical spinal cord injuries who exhibited diffuse spasticity: a report of three cases.
specifically a ∼13-second reduction in walking time, a reduction in the total step count (∼7 steps), and an increase in step length of ∼3.5 cm during the 10-m walking test.
Effect of the intrathecal baclofen screening test on the spatiotemporal gait motion parameters of patients with cervical spinal cord injuries who exhibited diffuse spasticity: a report of three cases.
Effect of the intrathecal baclofen screening test on the spatiotemporal gait motion parameters of patients with cervical spinal cord injuries who exhibited diffuse spasticity: a report of three cases.
Another study of 12 individuals with incomplete SCI and varying degrees of walking abilities examined the effects of baclofen, cyproheptadine, and clonidine on walking function.
More specifically, oral baclofen did not improve walking ability despite reducing spasticity, and in some individuals treadmill speed of walking actually increased during the washout period after baclofen administration.
After completion of this study, 7 individuals continued drug treatment, with only 1 individual continuing baclofen therapy in combination with cyproheptadine.
In line with previous findings, a 2016 study noted no functional improvements in walking or ambulation for 6 individuals receiving intrathecal baclofen with mean doses of 600 µg per day.
The functional improvement measure contains 18 items, each rated 1-7, where 7 involves 100% independence, assessing ability to carry out ADL in several domains to determine disability and likely necessity of care dependence.
In a separate study, an improvement of 23.25 in the Barthel Index scale after 6 weeks of oral baclofen use for 75 patients with mean age of 36 years was reported.
Improvements in instrumental ADL scores for 1 participant in a report of 2 participants noted improvements in dressing, bathing, clipping toenails, and transfer measures such as moving without supervision and driving.
A total of 43 studies (∼41 % of all studies) discussed expected or unexpected adverse events associated with baclofen. Of the 43 studies, 15 were associated with oral dosing of baclofen only. The most common complication was muscle weakness reported in 15 studies and at least 107 individuals with SCI (56 oral), followed by fatigue, which was reported in 20 studies with at least 69 individuals with SCI (12 oral). Other frequently reported adverse events were nausea (∼18 participants [6 oral]), dizziness (∼16 participants [9 oral]), altered mental status or hallucination (∼13 participants), overdose (∼15 participants [1 oral]), erectile dysfunction (∼13 participants), withdrawal (∼9 participants), hypotension (∼13 participants [1 oral]), infection (∼5 participants), respiratory depression (∼5 participants), and bloating and constipation (∼6 participants) (Fig 5).
Fig 5Adverse events reported after oral and intrathecal baclofen administration. Bar chart (left axis) illustrates no. of individuals reporting adverse events, with the no. of publications reported on the right axis (line plot).
While several studies have reported on baclofen's efficacy in managing spasticity, only 6 of these studies were randomized controlled trials conducted in patients with SCI.
In 1993, Coffey et al conducted one of the largest and most robust trials in which 93 patients (59 SCI, 31 multiple sclerosis, 3 other pathologies) were enrolled in a randomized, double-blind, placebo-controlled trial comparing the acute effects of test injections of varying doses of baclofen (50, 75, or 100µg) or placebo to determine whether individuals qualified for implantation of an intrathecal baclofen pump. During these screening assessments, no patients responded to the placebo injection; however, in 94.6% of participants, baclofen injections reduced MAS and spasm scores by at least 2 points.
Of these patients who passed the screening assessments, 75 individuals (47 SCI, 27 MS, 1 other) underwent pump implantation to examine long-term administration of intrathecal baclofen; however, this portion of the trial was not placebo-controlled. During the 19-month follow-up period, long-term intrathecal administration of baclofen reduced the MAS score by an average of 2.1 for patients with SCI, with a concomitant average reduction in muscle spasm score of 2.2.
The effective daily dose of intrathecal baclofen increased over the duration of the study from 187 to 405 μg/d on average, with doses typically higher in individuals with SCI (increased from 196μg/d to 462μg/d, on average).
These included hypo- and hypertension, seizure, depression, drowsiness, temporary weakness, and respiratory depression. Thirty-one device-related complications occurred across 28 individuals, requiring a secondary surgical or invasive procedure (did not report diagnosis group allocation).
This study did not report effects on motor function or ADL outcomes. In 1989, Penn et al performed a randomized double-blinded, crossover study of intrathecal baclofen use in 20 patients, 10 of whom had SCI and the other 10 were patients with multiple sclerosis.
All enrolled patients had previously failed maximum oral dosage of baclofen. During the blinded placebo-controlled crossover portion of the study, MAS score decreased by 2.8 points, and the spasm score decreased by 2.9 points across the entire study group after baclofen but not saline infusion.
During long-term follow-up of these individuals in an open trial of intrathecal baclofen (∼19 months, not placebo-controlled), the reductions in MAS and spasm scores were maintained.
Device-related complications such as dislodgement, pump failure, and pain at the implantation site were reported; however, no drug-related adverse events such as confusion or drowsiness were noted.
Lastly, it was reported that during maintenance treatment, many patients reported improvements in ADL such as self-care, transferring, sleeping, and urinary continence.
In a 2018 study, Kumru et al performed a randomized, double-blinded, placebo-controlled trial examining the acute effects of a single bolus of intrathecal baclofen vs saline on neuropathic pain unrelated to spasticity in 11 individuals with SCI. Although not a primary outcome, MAS was significantly reduced in the intrathecal baclofen group (pre: 2.0±1.3 vs post: 0.0±0.1) but not in the placebo group.
To date, 3 randomized controlled trials have examined the efficacy of oral baclofen administration. In one of the largest studies to date, 336 patients with SCI were randomized to 3 different treatment groups: physical therapy alone, physical therapy with botulinum toxin type A, or physical therapy with baclofen in a 1:1:1 ratio.
Outcome measures included MAS scores, Barthel Index, disability assessments, and modified Medical Research Council (mMRC) scores. MAS scores were significantly improved in the baclofen and botulinum toxin type A groups at the 2-week time point and beyond.
It took 4 weeks for spasticity to significantly improve in the physical therapy group; however, at the 6-week time point, there were no significant differences in improvements of MAS scores between the baclofen and physical therapy only groups.
Again, improvements in the Barthel Index occurred only at the 6-week time point in the physical therapy only group. The Barthel Index assesses independence across 10 different activities, with a total assigned score from 0 (total dependence) to 100 (complete independence).
Because this index is a combined measure of independence across ease of caregiving, ADL, and motor tasks (eg, ambulation, stair climbing), it is difficult to determine which specific tasks were improved with baclofen administration. In line with these findings, improvements in disability assessment scores and mMRC scores occurred earlier (2-week and 4-week time points, respectively). While baclofen was associated with improvements in mMRC scores, those receiving physical therapy alone failed to demonstrate an improvement in these scores.
In the baclofen group, MAS score decreased by 1.79 points (3.34±0.05 to 1.55±0.05), with improvements in muscle strength (mMRC scale; 1.31±0.05 to 2.79±0.03) and functional outcomes measured by the Barthel index (36.05±1.41 to 59.31±1.32).
Similar improvements in spasticity and muscle strength were obtained after oral administration of tolperisone after 6 weeks; however, a greater improvement in the Barthel Index was observed (38.53±1.41 to 73.35±1.32).
Adverse events were reported with increased incidence after baclofen administration compared with tolperisone (36 vs 14 adverse events), with asthenia cited as the most frequently reported adverse effect.
In 2011, Nance et al conducted a multiple-dose, randomized, double-blinded placebo-controlled trial examining the efficacy of arbaclofen placarbil, a prodrug of the R-isomer of baclofen that allows for an extended-release oral formulation. Thirty-seven individuals with spasticity due to SCI were enrolled. Ten individuals were randomized to the 10-mg arbaclofen placarbil/placebo group, 13 patients to the 20-mg arbaclofen placarbil/placebo group, and 14 patients to the 30-mg arbaclofen placarbil/placebo group.
Compared with placebo, the 20- and 30-mg arbaclofen placarbil doses significantly reduced MAS scores and patient-reported weekly average spasticity severity measures.
No differences were noted in secondary outcome measures of muscle strength, night-time awakening due to muscle spasms, pain severity, or sleep quality.
At least 1 treatment-related adverse event was reported in 75.7% of individuals receiving arbaclofen placarbil and in 48.6% of individuals receiving placebo.
Across all groups, the most commonly reported adverse events were urinary tract infection, pain, insomnia, and nasopharyngitis, with pain and insomnia uniquely associated with administration of arbaclofen placarbil.
The present review of 98 studies is a comprehensive review of baclofen use specific to SCI and demonstrates (1) there is a lack of pivotal trials with robust study method to examine efficacy of baclofen administration in the population with SCI; (2) efficacy as defined by reduction in MAS and Penn Spasm score were consistently reported after both oral (short-term) and intrathecal (short- and long-term) administration; (3) there is limited evidence for functional motor improvement with oral or intrathecal administration; and (4) the most prevalent reports are related to drug-related adverse events and dosing. Most trials examined baclofen use across different pathologic indications, including SCI, multiple sclerosis, and stroke. Only 6 randomized controlled trials investigated the effects of baclofen on MAS score and spasm measures with variation in specificity for SCI, route of administration, and dosing. While baclofen is widely used clinically to control spasticity after SCI, the lack of robust quantitative efficacy measures and the limited number of randomized, double blinded, placebo-controlled clinical trials should be considered and addressed. Additionally, a host of adverse events are associated with its use that may significantly affect quality of life, affect residual motor function, and lead to additional invasive operations.
However, minimal clinically important differences may at least be considered greater than the minimal detectable difference to determine clinical improvement.
Semiquantitative ordinal rating scales such as the MAS or Penn Spasm Scale may be more difficult to determine a universal clinical improvement for patients with SCI. The MAS score is often used clinically to evaluate clinical outcome of intervention or decide treatment in patients with spasticity.
Based on an analysis of patients with stroke, the Rehabilitation Institute of Chicago has reported 1 point in the MAS scale suggests a clinically meaningful improvement.
Alibiglou et al compared quantitative neuromechanical assessments of reflex stiffness for patients with spasticity from stroke and found no correlation between muscle stiffness and torque with Ashworth score.
Fleuren et al also demonstrate no association between flexion electromyographic activity parameters and MAS score. Another study contradicts the reliability of the MAS on evaluation of spasticity that showed poor and moderate interrater reliability from the intraclass correlations coefficient. Further, they recommend interpreting scores with great caution and advise to stop using MAS score for spasticity assessments.
more studies are needed to systematically examine whether baclofen remains the most promising treatment in individuals with SCI. Specifically, the mechanisms driving the development of spasticity after SCI are not likely to be the same across different pathologies. Therefore, it is surprising that most studies group together patients across different pathologies. Second, the reliability, validity, and responsiveness of the tools available to assess the efficacy of baclofen on spasticity and function remains unexplored if not inconclusive. Promising results have been demonstrated in studies comparing baclofen use with other pharmacologic and nonpharmacologic treatments.
Additionally, inclusion of quality of life metrics may benefit the current analysis as other outcome measures that indicate severity of burden associated with drug-related adverse events.
Pharmacologic alternatives
Like baclofen, most pharmacologic treatments are associated with their own adverse event profiles; however, some studies have indicated that while baclofen has detrimental effects on motor function, other drugs may improve function. Specifically, in 78 individuals with SCI, tizanidine (and clonidine), a noradrenergic α2 receptor agonist, effectively reduced MAS scores compared with placebo, with no change in residual walking function and ADL.
Tizanidine may be used cautiously in patients with kidney or hepatic dysfunction given its kidney clearance and risk of hepatotoxicity, requiring monitoring of creatinine and aminotransferase levels.
Additionally, tizanidine is known to prolong the QT interval and may interact with other commonly prescribed medications in this patient population with similar QT-prolonging effects such as fluoroquinolones.
In 25 participants with SCI, the off-label administration of cyproheptadine, a nonspecific serotonergic (5HT2) and noradrenergic (α1) receptor blocker, was as effective as baclofen and clonidine in reducing excessive muscle activity.
; however, no quantitative evidence supports substantial clinical effects. Botulinum toxin injections also provide localized relief from spasticity symptoms by blocking the release of acetylcholine at the neuromuscular junction with repeated injections required because of the waning effects after 2-4 months.
Despite potential pharmacologic alternatives to baclofen, depression of central nervous system excitability cannot be avoided. Therefore, treatment of spasticity through or in combination with nonpharmacologic methods may remain a more suitable alternative when possible.
Nonpharmacologic alternatives
Nonpharmacologic alternatives to treating spasticity after SCI vary based on residual motor capacity of the individual. The first line of physical therapy treatment applicable for most individuals aims to maintain joint range of motion through stretching exercises and passive movements.
These techniques have effectively reduced measurements of spasticity; however, results are inconsistent regarding improvements in motor function because of reduced spasticity. While the above techniques may be considered “passive,” alternatives involving voluntary contractions of the muscle for individuals with residual motor capacity have shown promise in treating spasticity. Locomotor training reduces ankle clonus, flexor, and extensor spasms as well as cocontraction while improving strength, walking speed, and self-reported mobility.
Comparison of the effects of body-weight-supported treadmill training and tilt-table standing on spasticity in individuals with chronic spinal cord injury.
Electrical stimulation of the nervous system has also been used to manage spasticity with no detriment to motor function. Peripheral nerve stimulation of different intensities has been demonstrated to reduce spasticity
and sometimes to a similar extent as baclofen, even at intensities of stimulation that do not produce muscle contractions (transcutaneous electrical nerve stimulations at 100Hz).
More sophisticated stimulation techniques that are not common practice in the clinic have also demonstrated promise in reducing spasticity after SCI. Noninvasive brain stimulation paradigms using transcranial magnetic stimulation delivered at different frequencies has effectively reduced spasticity measures and in some cases increased neurophysiological measurements of excitability in the short-term.
Spinal cord stimulation, transcutaneously and through implanted epidural electrodes similarly improved spasticity measures while improving residual motor function; however, results have been variable.
In more recent work, stimulation from 50-100 Hz, delivered through implanted electrodes, improved passive stretch resistance and reduced the use of antispastic medication.
More so, noninvasive transcutaneous stimulation of the spinal cord at 50 Hz reduced MAS, clonus, and spasms, with clinically meaningful improvements lasting up to 2 hours post intervention.
More promising, the applicability of transcutaneous spinal cord stimulation as a home-based therapy was explored with progressive improvements in therapeutic effects.
; however, the stimulation parameters used for the management of spasticity (50Hz) are not the same as those implemented for recovery of locomotion (typically ∼30Hz). Finally, most of the above studies focused on improvements in spasticity and effects on residual motor function in individuals with incomplete SCI; therefore, further examination of the potential applicability of these techniques to manage spasticity in individuals with severe SCI are warranted.
Study limitations
Limitations in the present review stem from the variability in study design, clinical assessments, and diagnostic indications across studies on baclofen use for spasticity in traumatic SCI. Heterogeneity of investigation designs (eg, retrospective, prospective, case series) with differing durations of treatment and modes of administration (intrathecal or oral) and a limited randomized controlled trial cohort may confound generalizability of the current findings. Additionally, many studies compare baclofen use across different pathologies and indications such as SCI and stroke that further reduces generalizability of claims made. Common assessments used in measurement of spasticity, including the Modified Ashworth score and Penn Spasm score have interrater reliability concerns that also confound validity of measurements in spasm and interfere with structured study comparisons of results and conclusions in a review or meta-analysis.
Conclusions
Current literature would benefit from improved study methods evaluating the clinical efficacy of baclofen in patients with SCI. Moderate improvements in spasticity is evident with baclofen; however, negative effects on residual motor function are not adequately addressed in the literature. Future prospective, double blinded, randomized controlled trials are necessary in patients with SCI and should compare baclofen with alternative treatment methods.
Impact of specific symptoms of spasticity on voluntary lower limb muscle function, gait and daily activities during subacute and chronic spinal cord injury.
Effect of the intrathecal baclofen screening test on the spatiotemporal gait motion parameters of patients with cervical spinal cord injuries who exhibited diffuse spasticity: a report of three cases.
Comparison of the effects of body-weight-supported treadmill training and tilt-table standing on spasticity in individuals with chronic spinal cord injury.
00461-0&id=gr1.jpg){kind=link}
00461-0&id=gr2.jpg){kind=link}
00461-0&id=gr3.jpg){kind=link}
00461-0&id=gr4.jpg){kind=link}
00461-0&id=gr5.jpg){kind=link}