Agonistas monoaminérgicos para la lesión cerebral traumática aguda
Resumen
Antecedentes
Aunque ha habido un considerable avance en la comprensión de la sucesión de eventos que lleva a la lesión secundaria después de la lesión cerebral traumática (LCT), los esfuerzos para traducir esta comprensión en nuevos enfoques terapéuticos, denominados neuroprotectores, hasta el momento han demostrado ser desalentadores. Como alternativa, existe un interés creciente en los enfoques para mejorar la reparación cerebral después de una lesión. Los modelos animales sugieren que los agentes que mejoran la transmisión monoaminérgica (agonistas monoaminérgicos o AM), en particular las anfetaminas, promueven la recuperación motora de la lesión cerebral focal y se propone que esto podría representar un medio complementario de intervención terapéutica en la fase posterior a la lesión.
Objetivos
Evaluar la evidencia de que los AM mejoran el resultado final después de una LCT.
Métodos de búsqueda
Se hicieron búsquedas en CENTRAL; MEDLINE; EMBASE; ISI Web of Science: Science Citation Index Expanded (SCI‐EXPANDED) y Conference Proceedings Citation Index‐ Science (CPCI‐S) y www.controlled‐trials.com hasta marzo de 2009. Se estableció contacto con los autores de los ensayos no publicados conocidos en curso.
Criterios de selección
Ensayos controlados aleatorizados que comparan el uso de un AM (junto con un tratamiento de rehabilitación no farmacológico convencional) versus un tratamiento de rehabilitación no farmacológico convencional solo.
Obtención y análisis de los datos
Dos autores de la revisión examinaron de forma independiente los registros, extrajeron los datos y evaluaron la calidad de los ensayos.
Resultados principales
Aunque existe literatura clínica limitada que aborde este tema, ninguno de los estudios identificados cumple de forma completa con los criterios de inclusión para esta revisión.
Conclusiones de los autores
Actualmente, no hay evidencia suficiente para apoyar el uso sistemático de los AM para promover la recuperación después de una LCT.
Resumen en términos sencillos
¿Un grupo de medicamentos conocidos como agonistas monoaminérgicos ayudan a la recuperación cerebral después de una lesión grave?
No todo el daño cerebral sufrido después de una lesión cerebral traumática (LCT) se debe a la lesión directa que se produce en el momento del impacto. Una lesión grave puede iniciar una secuencia de eventos durante varias horas que puede conducir a un daño secundario o a la muerte del tejido cerebral. Sin embargo, la efectividad de las intervenciones llamadas "neuroprotectoras" destinadas a prevenir esta secuencia de acontecimientos o a minimizar su daño ha sido hasta ahora decepcionante.
Un enfoque alternativo podría ser el uso de fármacos conocidos como agonistas monoaminérgicos (AM), que intentan mejorar la reparación cerebral después de una lesión. Los estudios en animales han sugerido que tales fármacos promueven la reorganización cerebral y la recuperación funcional después de la lesión. La efectividad de los AM en humanos para la recuperación después de una lesión cerebral aún no se ha determinado.
Los autores de esta revisión buscaron todos los ensayos de alta calidad que investigaran la efectividad de los AM en la recuperación de pacientes con lesiones cerebrales graves de todas las edades. Ninguno de los estudios publicados identificados abordó directamente la cuestión de la investigación y, por lo tanto, no fueron elegibles para su inclusión en la revisión. Por lo tanto, los autores de la revisión concluyeron que hasta la fecha no existen estudios satisfactorios de la efectividad de los AM para los pacientes que han experimentado una LCT grave. En consecuencia, actualmente no hay evidencia suficiente para apoyar el uso sistemático de los AM para promover la recuperación de una LCT.
Los autores afirman que existe una necesidad urgente de explorar la efectividad de las intervenciones, como los AM, para mejorar la reparación cerebral después de una lesión grave. Los hallazgos de los ensayos existentes de AM requieren replicación en estudios más grandes, para involucrar a otros grupos, incluidos los pacientes con lesiones más graves y los niños.
Authors' conclusions
Background
Not all of the brain damage sustained after a traumatic brain injury (TBI) is due to direct injury occurring at the moment of impact. Severe injury sets in motion a cascade of events over several hours that can lead to secondary damage or death of brain tissue. Throughout the 1990s, much effort was directed at understanding these events, and developing neuroprotective drugs that might interrupt this cascade of secondary injury. In practice however, this has not led to major therapeutic breakthroughs. An important factor probably contributing to this disappointing conclusion is the need to intervene early within a "window of opportunity" before these processes are too far advanced.
Are there approaches other than, or additional to, trying to interrupt the immediate injury‐related cascade of events? After the injury has been established, partial recovery of function occurs; can this be enhanced in any way?
The processes underlying the restoration of function after injury are poorly understood, but probably include: (i) full recovery of reversibly injured tissue, (ii) spared neural pathways taking over functions of destroyed networks, (iii) compensation through behavioural adaptation, and (iv) internal changes within neural networks caused by functional adaptations (changes in synaptic plasticity) brought about by external stimulation (Levin 2000).
It has been suggested that drugs that promote monoaminergic neurotransmission may improve the functional recovery of the brain after an acquired injury has become established. There is specific experimental evidence for the efficacy of dexamphetamine in improving motor recoveries following focal brain injury in rats (Feeney 1997; Gladstone 2000). Indeed, in some animal models, not only has enhanced recovery of function been demonstrated, but also restoration of function that would otherwise have been permanently lost (Goldstein 2000). Beneficial effects on outcome are reported in some clinical studies in the context of stroke (Crisostomo 1988; Walker‐Batson 1995) but not others (Sonde 2001). The use of amphetamines in stroke is the subject of a separate Cochrane systematic review (Martinsson 2001).
The mechanisms of these effects are unclear. Hypotheses need to accommodate observations (in the rat) that recovery can occur within hours of administration; that short term administration leads to long acting benefits; and that behaviourally relevant training appears necessary in combination with the drug (Gladstone 2000). There appears to be a synergistic effect of dexamphetamine with therapy (re‐learning of movement) in animal studies (Feeney 1982). In animal experiments, benefits are seen even if drug therapy is commenced some weeks after injury (Feeney 1998). There are, however, some animal data showing deleterious effects of amphetamine therapy under certain conditions (Boyeson 1991).
It has been suggested that these drugs are contributing to "metabolic normalisation" of dysfunctional, but still viable, neurons in the vicinity of the injury (Sutton 2000). However, immediate positive effects of amphetamines on experience‐dependent learning have also been demonstrated in healthy human volunteers where presumably "normal" metabolic conditions pertain (Butefisch 2002). Increased expression of gene products and proteins associated with plasticity have been demonstrated (Stroemer 1998; Wang 1994). Amphetamines are believed to increase levels of noradrenaline (NA) in the brain by enhancing NA release. Another drug, methylphenidate, blocks NA re‐uptake. L‐threotops, a synthetic NA precursor, is also being evaluated in stroke (Miyai 2000).
A separate group of drugs within the class of MAs are dopaminergic agonists (Bales 2009) of which the most widely studied has been amantadine, a drug also used in the treatment of Parkinson's disease.
Objectives
To evaluate the evidence that MAs improve final outcome after TBI.
Two separate hypotheses will be tested:
-
that MA therapy improves rate of recovery of orientation as assessed by the Galveston orientation and amnesia test (Levin 1979) or similar instruments.
-
that MA therapy affects final (late) global outcome as assessed by Glasgow outcome score and Disability Rating Scale (DRS). Other indices of cognitive, functional independence or gross motor outcome will be included if reported.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled studies of any MA therapy (together with conventional non‐pharmacological rehabilitative therapy) versus conventional non‐pharmacological rehabilitative therapy alone. The search strategy did not exclude trials without placebo control. The effects of excluding trials without placebo control if any are identified in due course (none have been at time of writing) will be the subject of a subsequent sensitivity analysis.
Types of participants
Patients without prior significant neurological disability sustaining acute onset severe coma (Glasgow coma score on admission less than or equal to 8 or comparable index of severity) of unambiguously traumatic cause.
Types of interventions
Use of any MA (to include amantadine, dextroamphetamine, metamphetamine or methylphenidate) by any dose, duration, time post injury or route.
Types of outcome measures
Primary outcomes
-
all‐cause mortality at the end of the follow‐up period.
-
severe disability at the end of the follow‐up period. If Glasgow outcome score (GOS) data were presented, severe disability was defined as a GOS outcome of "severe disability" or "persistent vegetative state". If other outcome scoring systems were used, these were mapped to the GOS.
Secondary outcomes
-
duration of post‐traumatic amnesia as assessed by the Galveston orientation and amnesia test (Levin 1979), the analogous children's orientation and amnesia test (Ewing‐Cobbs 1990) or comparable indices.
-
incidence of adverse effects.
Search methods for identification of studies
Electronic searches
We searched the following electronic databases:
-
Cochrane Injuries Group's Specialised Register (searched 2 April 2009);
-
CENTRAL (The Cochrane Library 2009, Issue 1);
-
MEDLINE 1950 to March (week 3) 2009;
-
EMBASE 1980 to March 2009 (week 13);
-
ISI Web of Science: Science Citation Index Expanded (SCI‐EXPANDED) 1970 to March 2009;
-
Conference Proceedings Citation Index‐Science (CPCI‐S) 1990 to March 2009;
-
PubMed (last 6 months; searched March 31 2009).
The searches were based on the strategy described in Appendix 1, adapted as appropriate to the specification of each database.
Searching other resources
We also contacted the authors of published trials and other researchers to try to identify further published and unpublished studies.
Data collection and analysis
Selection of studies
Two authors independently checked titles and abstracts of potentially relevant articles found by the database searches. In cases of doubt we obtained the full text of the article.
Data extraction and management
Two authors extracted data independently. Where possible we contacted the authors of trials to provide missing data.
Assessment of risk of bias in included studies
We documented the methodological quality of studies using the following criteria:
-
baseline comparison of experimental groups;
-
explicit diagnostic criteria;
-
allocation concealment;
-
completeness of follow‐up;
-
blind outcome assessment and blind administration of the drug.
Allocation concealment was scored using the scale devised by Schulz (Schulz 1995).
-
Adequate allocation ‐ central randomisation; numbered or coded bottles or containers; drugs prepared by the pharmacy; serially numbered, opaque, sealed envelopes; or other description that contained elements convincing of concealment).
-
Unclear ‐ trials in which the authors either did not report an allocation concealment approach at all or reported an approach that did not fall into one of the other categories.
-
Inadequate ‐ trials in which concealment was inadequate (such as alternation or reference to case record numbers or to dates of birth).
Data synthesis
The results were expressed as Peto's odds ratio (OR) and 95% confidence intervals (CI) for dichotomous outcomes.
Results
Description of studies
Included studies
No study met every inclusion criterion: the most problematic inclusion criterion in this regard was the "severe injury" criterion prospectively operationalised as a Glasgow Coma Score (GCS) <= 8.
Excluded studies
Three studies came close to meeting the inclusion criteria but were ultimately excluded.
Meythaler 2002
The study by Meythaler 2002 limited study entry only to a GCS of 10 or less (not <= 8). This study assessed the effect of amantadine on neurorecovery following TBI in 35 participants. The methodology used was a double‐blind, placebo controlled, cross‐over design trial. Participants were enrolled when directly admitted to the emergency department with a TBI following a motor vehicle crash. Amantadine or placebo was delivered for six weeks in a cross‐over trial with no washout period. All patients were recruited within six weeks from injury. Several outcome variables were measured, primarily the Disability Rating Scale (DRS), the Mini Mental State Exam (MMSE), the Glasgow Outcome Scale (GOS), Galveston Orientation and Amnesia Test (GOAT), and the FIM cognitive Score (FIM‐Cog). Comparison of scores at the end of the first six week period showed a significantly lower (better) DRS in the active treatment group. However DRS scores in the group receiving amantadine first were by chance lower (i.e. less disabled) than in the group receiving placebo first. The change in DRS score between the two groups at the end of six weeks was not statistically significant.
Mini Mental State Examination (MMSE) scores were comparable in the two groups at treatment outset: there was a trend for a greater (better) absolute MMSE and gain in MMSE over the first six weeks in the group receiving active treatment for the first six weeks, however this did not reach statistical significance. There was no significant improvement in Glasgow Outcome Scale (GOS) or FIM Cognitive score (FIM‐Cog) among the group receiving amantadine during the first six weeks. Due to the crossover nature of the study it is very difficult to assess whether use of amantadine affects final level of recovery over a longer period of follow‐up.
Schneider 1999
This randomised controlled trial of amantadine has also been excluded on GCS severity threshold grounds. No GCS threshold was specified in the protocol. Seven out of 10 included patients had a GCS of <= 8; one had a GCS > 8 and GCS was unknown for two. It was not possible to identify and report separately on the GCS <= 8 subgroup. The study used a crossover design with two week treatment periods (active or placebo) separated by a two week washout period. The outcomes measured were standardised neuropsychological testing and the Neurobehavioural Rating Scale. Of 18 enrolled patients only 10 completed the study, others withdrew consent for assessment or violated protocol.
Plenger 1996
In this trial the participants were less severely impaired at the time of study than the original intended study population of this review. This was because (i) entry criteria were not limited to severe injury (GCS < 8) and (ii) use of methylphenidate was deferred until a certain degree of spontaneous recovery had occurred (as reflected in recovery of the Galveston orientation and amnesia test to > 65). Although concealment and blinding were adequate (grade A, Schulz 1995) and treatment and placebo groups were well matched, there were significant methodological weaknesses. These related primarily to marked patient drop out due to transfers to other rehabilitation facilities. Of 80 eligible patients only 23 were enrolled, the primary reason for non‐participation being "rehabilitation readiness" of the remainder and their transfer to other facilities. Of the 23 enrolled, 12 patients reached 30‐day follow‐up (six active, six placebo) and only nine patients reached 90‐day follow‐up (five active, four placebo). Differential attrition resulted in a significantly higher initial GCS in the treatment group reaching 30‐day follow‐up relative to the placebo group, due to the presence of three "complicated mild" patients in the latter group, although these patients were retrospectively excluded in the published analyses. The study originally used a Wilcoxon rank sum test analysis reporting significant improvements in DRS at 30‐days (P < 0.007) but not at 90‐days (P = 0.12). There were no deaths in this study and thus no data are available on the effects of methylphenidate on mortality.
The high and progressive loss to follow‐up is a potential source of bias, since by implication loss to follow‐up occurred among those patients recovering to "rehabilitation readiness", although the published data suggests that this loss occurred to a comparable extent in each group. There was an excess of "complicated mild" patients (not requiring intensive care but with positive findings on CT scan) in the treatment group at 30‐day follow‐up who were omitted from the published analysis because of the potential for biasing in favour of a treatment effect. These concerns have led to the study being excluded for the purposes of this review.
Risk of bias in included studies
No included studies.
Effects of interventions
No included studies.
Discussion
The Plenger study discussed above (Plenger 1996) was reported as supporting the use of methylphenidate to promote rate of recovery after TBI. The non‐parametric analysis used in the original paper appeared to show significantly lower disability in the treatment group at 30 days, although the effect was lost by 90 days. These conclusions are, however, severely affected by the selective loss to follow‐up of less disabled survivors, as they were transferred to other centres.
Several excluded studies looked specifically at the use of methylphenidate in improving attention after head injury: a role more related to the widespread use of methylphenidate in attention deficit disorders. The role of methylphenidate in promoting motor recovery in more disabled survivors has not been addressed. A key priority of head injury rehabilitation research is the evaluation in clinical settings of approaches that animal data suggest may promote neural plasticity. One excluded study (Kaelin 1996) was a multiple‐baseline (A‐A‐B‐A) design study of the effect of methylphenidate on measures of attention. Disability rating scale (DRS) changes were also reported. Improvements in DRS were noted both spontaneously during the first seven‐day interval (A‐A) and the second seven‐day interval, during which methylphenidate was administered (A‐B). Friedman two‐way ANOVA by ranks was used to compare average weekly change in DRS score and suggested that the mean improvement in DRS on treatment (27%) approached a significant difference (P < 0.06) from the A‐A change in baseline DRS score (15%). Speech et al (Speech 1993) used a placebo‐controlled cross‐over study to look at attentional endpoints and a measure of social adjustment (the Katz Rating Scale) following the use of methylphenidate late in the chronic recovery phase after head injury (minimum 14 months). No significant effects were reported on attention or social function. However, animal data on plasticity and reorganisation after brain injury would support the suggestion that this may be too late after injury to expect a beneficial effect.
Overall completeness and applicability of evidence
The US National Library of Medicine lists a trial (Blum 2002; NCT00035139) whose results have not yet been published although recruitment has now closed. Primary endpoints relate to the effects of eight weeks of methylphenidate on rate of recovery of cognitive, memory, and attentional skills in children with TBI. This study appears to be intervening late after injury.
Another paediatric pilot study (Philips 2002) of potential relevance to this review has closed. Data analysis is ongoing at the time of writing (June 2009).
