The levels of evidence (class I-IV) supporting the recommendations and ratings of recommendations (A-C) are defined at the end of the "Major Recommendations" field.
Rehabilitation of Aphasia
The conclusions of the Cochrane review of aphasia rehabilitation after stroke are not compatible with grade A recommendation for aphasia therapy. There is however considerable evidence from class II and III studies, as well as from rigorous single-case studies indicating its probable effectiveness (grade B recommendation). There is clearly a need for further investigations in the field. In particular, the evidence of effectiveness of pragmatic-conversational therapy after traumatic brain injury (TBI) is based on a limited number of studies on small samples and is in need of confirmation.
Rehabilitation of Unilateral Spatial Neglect
Several methods of neglect rehabilitation were investigated in level I or II studies. The present evidence confers level A recommendation to visual scanning training and to visuo-spatio-motor training, and level B recommendation to the combined training of visual scanning, reading, copying and figure description; to trunk orientation; to neck vibration; and to forced use of left eye. The use of prism goggles obtains the same level of recommendation for transient effect and level C for long-term effect if used over longer periods. Level B recommendation exists for video feedback; and level B–C for training of sustained attention and alertness. Level C of recommendation is valid for transient effects because of caloric or galvanic vestibular stimulations as well as transcutaneous electrical stimulation of neck muscles. Visual cueing with kinetic stimuli and the use of computers in neglect rehabilitation remain controversial.
Rehabilitation of Attention Disorders
During the acute period of recovery and inpatient rehabilitation, evidence is insufficient to distinguish the effects of specific attention training from spontaneous recovery or more general cognitive interventions for patients with moderate-to-severe TBI and stroke. Therefore, specific interventions for attention during the period of acute recovery are not recommended. On the contrary, the availability of class I evidence for attention training in the post-acute phase after TBI is compatible with a grade A recommendation.
Rehabilitation of Memory
One group of researchers (using a different rating system from the one used here) recommended compensatory memory training for subjects with mild memory impairments as a practice standard. These authors point out that independence in daily function, active involvement in identifying the memory problem to be treated and the capability and motivation to continue active and independent strategy use strongly contribute to effective memory remediation. Based on the currently available evidence the Task Force members judge the use of memory strategies without electronic aid as possibly effective (level C) although it remains unclear to what degree the benefit depends on the severity of the memory impairment. Specific learning strategies such as errorless learning are supported by a series of class III studies and are thus rated as probably effective (level B). However, some studies suggest that the efficacy of a specific learning technique may depend on the task used, whether implicit or explicit memory is implicated, and the severity of the memory impairment. Two class III studies supported by several class IV studies have shown possible efficacy (level C recommendation) of non-electronic external memory aids such as diary or notebook keeping. Electronic external memory devices such as computers, paging systems or portable voice organizers have been shown to be effective in several class III studies and are thus recommended as probably effective (level B) aids for improving TBI or stroke patients' everyday activities. The use of virtual environments has shown positive effects on verbal, visual and spatial learning in stroke and TBI patient in two class III studies. A direct comparison of performing learning and memory training in virtual environments versus non-virtual environments is still lacking and no recommendation can be made as to the specificity of the technique. Currently, memory training in virtual environments is rated as possibly effective (level C).
Despite the many studies investigating memory rehabilitation, the problems raised in previous reports concerning the heterogeneity of the population studied (in terms of age, aetiology and type of brain damage, severity of brain-damage, severity of functional impairments, time post-onset) and the subsequent difficulty of interpreting the results are still valid. It is conceivable that the type and intensity of training has different effects depending on the neural circuits damaged, the functional impairment profile, the age and gender of the patient, the time post-injury, the education level of the patient, and other external factors (such as social and vocational situation). The number of variables involved makes generalization across individuals difficult and favours training programmes tailored to the individual circumstances. No specific recommendations are made for different diagnostic groups or stages of severity. There is still a lack of studies that directly compare patients with different aetiologies (e.g. stroke versus TBI), type and severity of brain damage, age, gender, or stage of recovery.
Rehabilitation of Apraxia
There is grade A evidence for the effectiveness of apraxia treatment with compensatory strategies. Treatment should focus on functional activities, which are structured and practised using errorless learning approaches. As transfer of training is difficult to achieve, training should focus on specific activities in a specific context close to the normal routines of the patients. Recovery of apraxia should not be the goal for rehabilitation. Further studies of treatment interventions are needed, which also address if the treatment effects generalize to non-trained activities and situations.
Rehabilitation of Acalculia
Overall, the available evidence suggests that rehabilitation procedures used to treat selected variants of disorders of number processing and calculation (DNPC) were successful (level C rating). Notably, significant improvements were observed even in severely impaired and chronic patients. Several caveats need to be mentioned in this context. At present, little is known about the prognosis and spontaneous recovery of DNPC, thus, the effects of different interventions in the early stages of numerical disorders may be difficult to evaluate. Moreover, different underlying neurological disorders (e.g. stroke, dementia, and trauma) have only partly been compared as to their specific effects on DNPC. Furthermore, it has not been studied in detail how impairments of attention or executive functions influence the rehabilitation process of DNPC.
General Recommendations
In the guideline developers' opinion, there is enough overall evidence to award a grade A, B, or C recommendation to some forms of cognitive rehabilitation in patients with neuropsychological deficits in the post-acute stage after a focal brain lesion (stroke, TBI). This general conclusion is based on a limited number of randomised controlled trials (RCTs), and is supported by a considerable amount of evidence coming from class II, III, and IV studies. In particular, the use of a rigorous single-case methodology has been considered by the present reviewers as a source of acceptable evidence in this specific field, in which the application of the randomised controlled trial methodology is difficult for a number of reasons, related to the lack of consensus on the target of treatment, the methodology of the intervention, and the assessment of the outcomes.
Definitions:
Evidence Classification Scheme for a Therapeutic Intervention
Class I: An adequately powered prospective, randomized, controlled clinical trial with masked outcome assessment in a representative population or an adequately powered systematic review of prospective randomized controlled clinical trials with masked outcome assessment in representative populations. The following are required:
- Randomization concealment
- Primary outcome(s) is/are clearly defined
- Exclusion/inclusion criteria are clearly defined
- Adequate accounting for dropouts and crossovers with numbers sufficiently low to have minimal potential for bias
- Relevant baseline characteristics are presented and substantially equivalent among treatment groups or there is appropriate statistical adjustment for differences
Class II: Prospective matched-group cohort study in a representative population with masked outcome assessment that meets a–e above or a randomized, controlled trial in a representative population that lacks one criteria a–e
Class III: All other controlled trials (including well-defined natural history controls or patients serving as own controls) in a representative population, where outcome assessment is independent of patient treatment
Class IV: Evidence from uncontrolled studies, case series, case reports, or expert opinion
Rating of Recommendations
Level A rating (established as effective, ineffective, or harmful) requires at least one convincing class I study or at least two consistent, convincing class II studies.
Level B rating (probably effective, ineffective, or harmful) requires at least one convincing class II study or overwhelming class III evidence.
Level C rating (possibly effective, ineffective, or harmful) requires at least two convincing class III studies.
