The application of rehabilitation robots has grown during the last decade. While meta-analyses have shown beneficial effects of robotic interventions for some patient groups, the evidence is less in others. We established the Advanced Robotic Therapy Integrated Centers (ARTIC) network with the goal of advancing the science and clinical practice of rehabilitation robotics. The investigators hope to exploit variations in practice to learn about current clinical application and outcomes. The aim of this paper is to introduce the ARTIC network to the clinical and research community, present the initial data set and its characteristics and compare the outcome data collected so far with data from prior studies.
ARTIC is a pragmatic observational study of clinical care. The database includes patients with various neurological and gait deficits who used the driven gait orthosis Lokomat® as part of their treatment. Patient characteristics, diagnosis-specific information, and indicators of impairment severity are collected. Core clinical assessments include the 10-Meter Walk Test and the Goal Attainment Scaling. Data from each Lokomat® training session are automatically collected.
At time of analysis, the database contained data collected from 595 patients (cerebral palsy: n = 208; stroke: n = 129; spinal cord injury: n = 93; traumatic brain injury: n = 39; and various other diagnoses: n = 126). At onset, average walking speeds were slow. The training intensity increased from the first to the final therapy session and most patients achieved their goals.
The characteristics of the patients matched epidemiological data for the target populations. When patient characteristics differed from epidemiological data, this was mainly due to the selection criteria used to assess eligibility for Lokomat® training. While patients included in randomized controlled interventional trials have to fulfill many inclusion and exclusion criteria, the only selection criteria applying to patients in the ARTIC database are those required for use of the Lokomat®. We suggest that the ARTIC network offers an opportunity to investigate the clinical application and effectiveness of rehabilitation technologies for various diagnoses. Due to the standardization of assessments and the use of a common technology, this network could serve as a basis for researchers interested in specific interventional studies expanding beyond the Lokomat®.
The number of technological devices that therapists can utilize to treat people with neurological impairments has grown substantially during the last decade. Alongside this growth in clinical use, research involving robotic therapy has grown rapidly. A search in Pubmed with the terms “robot” OR “robotic*” AND “rehabilitation” revealed 2225 hits (March 2017) with research markedly increasing after 2010. Despite this increase in research activity and clinical use, the effectiveness of robot-assisted interventions in neurorehabilitation is still in debate. While in some patient populations, for example adults with stroke, meta-analyses have shown that robotic interventions for the lower and upper extremity can be beneficial [1, 2], current evidence is much less convincing in other patient groups, such as spinal cord injury (SCI), traumatic brain injury (TBI), multiple sclerosis (MS) and cerebral palsy (CP).
When comparing the effectiveness of robot-assisted gait training (RAGT) to conventional interventions of similar dosage in adult patients after SCI, it appears that neither intervention is superior [3, 4]. In other populations, such as MS, a small number of pilot studies have been conducted, and a review  concluded that the evidence for the effectiveness remained inconclusive. In adult patients with TBI, to our knowledge, there is only one randomized controlled trial that investigated the effectiveness of RAGT . While RAGT improved gait symmetry compared to manually assisted body-weight supported treadmill training, improvements in other gait parameters were not different between the interventions. In children with CP, the body of evidence is similarly small, as only two randomized trials were found [7, 8]. To the authors’ knowledge, there are no randomized controlled trials in children with other diagnoses. Studies comparing effectiveness between different patient groups are lacking.
One important factor leading to the lack of conclusive research is the relatively small number of available centers and participating patients and consequently the small statistical power of attempted studies. Multicenter collaborations are needed to achieve adequate number of participants. Several of the limitations in the evidence of the application of RAGT arise from patient selection criteria and use of different, poorly described and/or low-dosed training protocols. For example, when systematically reviewing the literature in children, we found no paper describing a training protocol on how to apply a robot for rehabilitation of gait . Most of the systematic reviews mentioned that it is extremely difficult to pool results from studies due to the large variability in treatment duration and frequency, contents of the training and inclusion criteria of the patients. For children with CP, an expert team was created to formulate goals, inclusion criteria, training parameters and recommendations on including RAGT in the clinical setting, to assist therapists who train children with CP with the Lokomat® (Hocoma AG, Volketswil, Switzerland) . Such information could be used as a first step in defining training protocols, but this information is missing for most other patient groups.
While randomized controlled trials are usually considered the “gold standard” in building solid evidence in the field of medicine, it is often difficult for rehabilitation specialists working in the clinical environment to interpret the findings with respect to the population of patients they treat on a daily basis. Randomized controlled trials require a specialized team, a controlled setting and a strict selection of patients according to well defined inclusion and exclusion criteria. These criteria often select individuals most likely to benefit based on specific parameters and lack of co-morbidities. These narrow criteria may impact the ecological validity, as results only apply to a minority of patients. This was recently investigated by Dörenkamp et al.  who reported that the majority of patients in primary care (40% at the age of 50 years and at least two-thirds of the octogenarian population ) simultaneously suffered from multiple medical problems. Further, improvements in function might be less comparable to results described in randomized controlled trials and the treatment regimens used may not be applicable to patients with multiple comorbidities.
To overcome these issues, we established the Advanced Robotic Therapy Integrated Centers (ARTIC) network to collect data from patients using RAGT in a wide variety of clinical settings. ARTIC hopes to develop guidelines for usage as well as to answer scientific questions concerning the use of RAGT. While the ARTIC network includes a general patient population, other research networks focus on a specific disorder or diagnostic group (see, for example [12, 13]). ARTIC focuses on a common technological intervention – currently the driven gait orthosis Lokomat® – and aims to gather evidence for the efficient and effective use of robotic therapy. Variation in practice among ARTIC members together with collection of common data and outcome measurements will enable the group to draw strong, generalizable conclusions. Further goals include establishing standardized treatment protocols and increasing medical and governmental acceptance of robotic therapy. The aims of this paper are to introduce the ARTIC network to the clinical and research community, present initial data on the characteristics of included patients and compare these to those known from existing epidemiological data and interventional studies.[…]
Continue —> Advanced Robotic Therapy Integrated Centers (ARTIC): an international collaboration facilitating the application of rehabilitation technologies | Journal of NeuroEngineering and Rehabilitation | Full Text