Short-term and long-term outcomes of serial robotic ...

Original article 1

Short-term and long-term outcomes of serial robotic training for improving upper limb function in chronic stroke Patrizio Salea, Federica Bovolentab, Maurizio Agostic, Pierina Clericib and Marco Franceschinia The aim of this study was to determine short-term and longterm changes in motor function in patients with chronic hemiparesis who underwent robot training and to evaluate its long-term benefit after 6 months. This was a longitudinal study with a 6-month follow-up. The 15 patients included in this study underwent the Fugl-Meyer test, the Ashworth Scale test, the Frenchay Arm test, and the Box and Block test according to the following schedule: immediately before (T1, T3) and after each treatment (T2, T4), and 6 months after T4 (T5). There were statistically significant improvements in Fugl-Meyer test between T1 and T2 and between T1 and T4; the score increased in the Ashworth Scale test for Shoulder between T1 and T3 and between T1 and T5; a statistically significant decrease was found between T1 and T2 and between T1 and T4, in the Box and Block test between T1 and T4, and also between T1 and T5. This original

Introduction Rehabilitation robotics (RR) focuses on devices or machines developed with the purpose of helping and assisting impaired individuals to recover from severe neurological disability or physical trauma (Volpe et al., 2000, 2008; Bovolenta et al., 2009, 2011; Posteraro et al., 2009, 2010; Hayward et al., 2010; Lo et al., 2010; Masiero et al., 2011; Scott and Dukelow, 2011; Sale et al., 2012a, 2012b). In western countries, neurological conditions such as stroke, multiple sclerosis, and Parkinson’s disease are a major cause of disability among the elderly population, whereas it is becoming more common in individuals of working age younger than 65 years of age (Cruz-Flores et al., 2011; Kim and Johnston, 2011). The amount of recoveries will increase, in the forthcoming years, thanks to an improved quality of hyperacute lifesaving practices and follow-up acute care, although the supply of early and long-term rehabilitation programs is significantly lagging behind with respect to a growing demand for specialized treatments (Stineman et al., 2011). Indeed, 80% of stroke patients experience long-term reduced upper limb (UL) function and manual dexterity. Because of this impairment, half of all patients could be limited in their performances of an everyday task as per activities of daily life (Kwakkel et al., 2004). The traditional motor stroke rehabilitation treatment corresponds to standardized physiotherapy and occupational therapy rehabilitation care. However, over the past few years, various studies on a new technological treatment such as RR have shown its c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0342-5282

rehabilitation treatment may contribute toward increasing upper limb motor recovery in stable chronic stroke patients. International Journal of Rehabilitation Research c 2014 Wolters Kluwer Health | Lippincott 00:000–000 Williams & Wilkins. International Journal of Rehabilitation Research 2014, 00:000–000 Keywords: robotic, stroke, upper limb a

IRCCS San Raffaele Pisana, Rome, bMedicine Rehabilitation, NOCSAE Hospital AUSL of Modena, Modena and cRehabilitation Hospital Parma, Parma, Italy Correspondence to Patrizio Sale, MD, IRCCS San Raffaele Pisana, Via Della Pisana 235, 00163 Rome, Italy Tel: + 39 065 225 2409; fax: + 39 065 225 5683; e-mail: [email protected] Received 20 March 2013 Accepted 21 August 2013

superiority in recovery from motor impairments of upper and lower extremities, with a generic improvement in activities of daily living (Fazekas et al., 2006; Pe´ter et al., 2011; Huang et al., 2012; Fazekas, 2013). Robotic devices for UL consist of either robot-driven exoskeleton orthoses or end-effectorbased solutions. Nevertheless, the mechanisms underlying the recovery process after RR still seem to be unclear. In our experience, RR for UL offers a positive result in neurorehabilitation training, particularly from severe to moderate impaired chronic stroke patients (Bovolenta et al., 2009, 2011). According to many papers, an efficient rehabilitative treatment must be intensive and specific, repetitive, functional, and motivating for the individual to allow a continuous enhancement in the process of learning, acquisition, and generalization (Fasoli et al., 2004; Van Peppen et al., 2004; Krakauer, 2006; Huang and Krakauer, 2009; Mehrholz et al., 2012). As a technology-based tool, RR devices have a number of inherent advantages over conventional, largely manually assisted therapy. Their key characteristic is the possibility to administer high-intensity, repetitive functional and also motivational physical motor therapies. This approach has been showing a general improvement, in chronic stroke patients, of motor function and motor performance (Lo et al., 2010; Posteraro et al., 2010; Bovolenta et al., 2011). Usually, RR treatment may be used, in chronic stroke patients, either as a single treatment or as an add-on to the standard poststroke multidisciplinary rehabilitation programs, although they warrant further investigation. Until now, the differences in gain to achieve a good UL motor recovery obtained by end effector and DOI: 10.1097/MRR.0000000000000036

2 International Journal of Rehabilitation Research 2014, Vol 00 No 00

exoskeleton robotic devices, along with the number of session of RR treatments, have not been fully demonstrated as yet. In a recent analysis, Mehrholz et al. (2012) has reported the variations between trials because of the duration and amount of training, the type of treatment, and the patients’ characteristics. The aim of our longitudinal study was to verify the effect of two consecutive RR treatments for UL, in chronic stroke patients, in the same year, and their long-term stability at follow-up (6 months after the end of the second treatment) in terms of motor function, motor performance, and dexterity.

Materials and methods This longitudinal observational study aimed to evaluate the effect of two treatments of robot therapy as a key treatment in chronic stroke patients. The study included chronic hemiparetic first-ever stroke patients only, recruited during a stable phase, at least 2 months minimum after the end of the rehabilitative treatment for an ischemic or hemorrhagic lesion (Table 1). The following inclusion criteria were established: (a) first acute event of cerebrovascular stroke; (b) conclusion of a previous rehabilitation program for UL with unsatisfactory motor recovery; and (c) discontinuation from any UL rehabilitation treatment for at least 2 months before the first examination. The following exclusion criteria were established: (a) posterior circulation infarction; (b) subarachnoid hemorrhage; and (c) patients with severe cognitive, linguistic, or perceptive impairment (Mini Mental State Examination < 24). Diagnoses in the acute phase were confirmed by means of a computed tomography scan or an MRI. The local ethics committees of the rehabilitation Table 1

Sociodemographic and clinical variables of the patients Percentiles

Variables

n

Mean

Sociodemographic variables Patient 14 Age (years) 53.2 Sex Female 5 Male 9 General clinical variables Etiology Ischemic 10 Hemorrhagic 4 Side lesion Left 6 Right 8 Disease severity Severe (FMul 0–35) 4 Moderate (FMul 36–75) 4 Low (FMul 76–115) 6 Impairment variables (at baseline T0) CMSA 3.9 Fugl-Meyer test score 74.1 Box and Block test score 12.6 Frenchay Arm test score 2.3 Ashworth Scale test score Shoulder 0.3 Elbow 2.1

SD

25th

50th (median)

75th

13.5

45.0

56.0

63.0

1.0 20.0 13.4 1.8

3.0 59.8 0.0 1.0

4.0 73.0 10.0 1.5

4.3 89.3 24.5 4.0

0.6 1.2

0.0 1.0

0.0 2.5

0.3 3.0

CMSA, Chedoke-McMaster Stroke Assessment; FMul, Fugl-Meyer upper limb.

facilities involved approved the study. All patients provided informed consent to the investigation. Treatment

All patients underwent an ambulatory rehabilitation treatment, consisting of two identical UL RR treatments, repeated in the same year 3 months apart, using the ReoGo robot device (Motorika Medical Ltd, Caesarea, Israel), a three-dimensional (3D) robot-assisted neuromuscular training system that uses high-end robotics technology. Patients recovering from stroke, traumatic brain injury, spinal cord injury, and other neurological conditions were actively engaged in an intensive, repetitive training program through a wide range of variable exercises (Bovolenta et al., 2009, 2011). A user-friendly interface allowed patients to operate the system with minimal professional involvement and a 3D display with stimulating exercises facilitated the patient’s compliance. The assessment process was designed to view the patient’s ability to perform specific exercises over time, while the system measured and displayed the patient’s progresses. The monitor showed the patient’s machine workout, on the basis of a defined and scheduled program. The treatment consisted of 20 sessions lasting 45 min each, 5 days a week, for a total period of 4 weeks, and it was repeated twice, for a total of 40 sessions. If the treatment was interrupted for less than three times consecutively, all missed sessions were made up. Our rehabilitative protocol consisted of exercises aimed at improving all main UL movements (i.e. the involved joints, with a proximal–distal progression) and its mode of execution, with progression from passive to free movement, using the forearm as a support to help perform the exercises. The robot has an arm with a platform to stabilize participants’ forearm, which moves at multidirection and multiassist levels programmed with a total-assist to a totalactive mode. In particular, the rehabilitation exercises consisted of forward thrust, horizontal abduction, forward reach 2D and 3D, horizontal reach, and functional. The training consisted of a 5–10-min warm-up of UL active movements (for shoulder, elbow, wrist, and hand) that the patient was capable of carrying out independently. Stretching of muscles (hand and finger flexors and forearm pronators) was also performed for up to 5 min for patients with increased muscle tone. After this, the patients underwent a robot training with a rehabilitative program of five different tasks that comprised the following: (a) reaching point to point (PtP) with flexion/extension of the elbow in different horizontal planes; (b) reaching PtP in various diagonal planes; (c) PtP in unaffected-affectedunaffected diagonal directions; (d) horizontal abduction reach (PtP); (e) horizontal abduction/adduction reach (PtP). The movement was repeated 200–400 times per session. Each patient underwent an UL evaluation by a trained blinded physical therapist, not involved in a rehabilitation treatment team, using the upper extremity subsection of the Fugl-Meyer (FM) Assessment Scale (Fugl-Meyer et al., 1975; Lindmark and Hamrin, 1988); the

Robotic training for improving hand function Sale et al.

3

Fig. 1

Enrollment

Assessed for eligibility (n = 35)

Excluded (n = 21) ♦ Not meeting inclusion criteria (n = 16) ♦ Declined to participate (n = 5) Eligible (n = 14)

Allocation

Allocated to first robot treatment T1 (n = 14) ♦ Received allocated intervention (n = 14 ) ♦ Did not receive allocated intervention (give reasons) (n = 0)

Allocation

Allocated to second robot treatment T3 (n = 14) ♦ Received allocated intervention (n = 14 ) ♦ Did not receive allocated intervention (give reasons) (n = 0)

Follow-up

Lost to follow-up (give reasons) T5 (n = 0) Discontinued intervention (give reasons) (n = 0)

Analysis

Analysed (n = 14) ♦ Excluded from analysis (give reasons) (n = 0)

Flowchart of the procedure. The time between T1 and T2 (first treatment) and T3 and T4 (second treatment) was 30 days, whereas it was 90 days between T2 and T3. The detailed timelines are as follows: the start of the first treatment (T1) (0 days), at the end of the first treatment (T2) (30 days from T1), 3 months after the end of the first treatment and the start of the second treatment (T3), at the end of the second treatment (T4) (150 days from T1) and 6 months after the second treatment (follow-up) (T5) (330 days from T1).

Modified Ashworth Scale for Elbow (AS-E) and for Shoulder (AS-S) (Ashworth, 1964); the Frenchay Arm test (FAT) (Heller et al., 1987); and the Box and Block test (BBT) (Mathiowetz et al., 1985). All data were collected at the beginning (T1) and at the end (T2) of the first treatment, at the start (T3) and at the end (T4) of the second treatment, and at the follow-up (T5) 6 months after the end of the second treatment. The time between T1 and T2 (first treatment) and T3 and T4 (second treatment) was 30 days, whereas it was 90 days between T2 and T3 (Fig. 1). The Chedoke-McMaster Stroke Assessment Scale was compiled to classify the patients,

according to different degrees of severity in the UL impairment, only before the first treatment (T1). At T4, patients responded to a Robot Agreement Survey designed to analyze the quality of the treatment, the quality of the result, and the expectations, assigning a score between 1 (lower agreement) and 10 (maximum agreement). During the entire duration of the study, the patients did not receive any other rehabilitation treatment. Statistical analysis

All clinical scale values were computed for each participant; moreover, the mean, median, SD, and percentile


values of all indexes were calculated. To evaluate the statistical significance of clinical outcome measures (e.g. FM, AS-E, AS-S, FAT, and BBT) among chronic patients, the changed values at pretreatment, post-treatment, and follow-up were compared using a Friedmann test for repetitive analysis of nonparametric data. Furthermore, multiple analyses (T1 vs. T2, T3, T4, and T5) were carried out using the Dunnett test. The critical limit for significance was set at P value less than 0.05.

Results We screened 61 patients, 14 of whom fulfilled the inclusion criteria. Nine of them were men and five were women; six individuals had a left side lesion and eight had a right side lesion. No dropouts occurred because of treatment intolerance. The sample average age was 53.21±13.5 years, with a range of 27–71 years; the average time elapsed since the acute event was 32.07±27.21 months, with a range of 6–71 months. The level of the UL impairment for each stroke patient at admission was assessed using the ‘Stage of Arm’ section of the Chedoke-McMaster Stroke Assessment Scale. Table 1 summarizes descriptive sample information, whereas Fig. 2 shows the results of the evaluations performed on all patients during the study period. The robot-assisted therapy was well accepted and tolerated by all patients, highlighted by excellent compliance, motivation, remarkable satisfaction, and the absence of dropouts because of treatment intolerance, as reported by the survey at T2 and T4. In particular, the analysis of Robot Agreement Survey values yielded 98.68±4.03 (mean and SD), with a minimum of 85 and a maximum of 100. Table 2 summarizes the result of mean ranks for Friedman two-way analysis of variance by ranks. Statistically significant improvements were found in the FM between T1 and T2 (P < 0.05) and between T1 and T4 (P < 0.01). No other statistical significance in FM was found. A statistically significant increase in the scores of AS-S was found between T1 and T3 (P < 0.01) and between T1 and T5 (P < 0.01); no other statistical significance was found. A statistically significant decrease in scores in the AS-E between T1 and T2 (P < 0.01) and between T1 and T4 (P < 0.01) was found. The BBT scores analysis showed statistically significant improvements between T1 and T4 (P < 0.01) and also between T1 and T5 (P < 0.01). Similarly, the FAT score scale showed statistically significant improvements between T1 and T2 (P < 0.01), between T1 and T4 (P < 0.05), and between T1 and T5 (P < 0.01).

Discussion Technological devices have shown a potential critical impact on the rehabilitation of patients with stroke because of their capability to produce repetitive and standard task-oriented activities that influence the relearning processes, and allow patients to be retrained while reducing the burden on the rehabilitation staff.

Fig. 2

140 120 100 q1 Min Median Max q3

80 60 40 20 0 T1 FM

T2 FM

T3 FM

T4 FM

T5 FM

6 5 4

q1 Min Median Max q3

3 2 1 0 T1 FAT

T2 FAT

T3 FAT

T4 FAT

T5 FAT

50 45 40 35 30 25 20 15 10 5 0


T1 BB

T2 BB

T3 BB

T4 BB

T5 BB

3.5 3 2.5 q1 Min Median Max q3

2 1.5 1 0.5 0 T1 AS-S T2 AS-S T3 AS-S T4 AS-S T5 AS-S 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0


T1 AS-E T2 AS-E T3 AS-E T4 AS-E T5 AS-E

The results of the clinical scores [Fugl-Meyer (FM) Assessment Scale; the Modified Ashworth Scale for Elbow (AS-E) and Modified Ashworth Scale for Shoulder (AS-S); the Frenchay Arm test (FAT); and the Box and Block test (BBT)] and the statistical significance.


Table 2 Mean ranks for Friedman two-way analysis of variance by ranks (n = 14)

Fugl-Meyer test score Box and Block test score Frenchay Arm test score Ashworth Scale test score Shoulder Elbow

T1

T2

T3

T4

T5

2.07 1.89 2.14

3.75 3.14 3.54

3.04 2.57 2.46

3.96 3.50 3.32

2.18 3.89 3.54

2.32 3.86

2.00 2.32

3.57 2.96

3.04 2.18

4.07 3.68

Comparisons (adjusted): Dunnett method (T1 group is reference): a = 0.05; Dunnett Q = 2.10; minimum significant difference = 1.26.

Very little literature has evaluated their effectiveness and feasibility. Unfortunately, the heterogeneity in methodological procedures among the few trials available in the literature, in terms of trial design, the type of device used, and participant characteristics, is not sufficient to achieve a positive conclusion on the superiority of these procedures. Kwakkel et al. (2008) showed that UL RR is effective in enhancing function, but it is not superior to conventional physiotherapy. In a recent paper, Krebs and Volpe (2013) reported that RR is effective in reducing the impairment, but not as successful in promoting function. Several studies related to RR treatment in chronic stroke patients have shown that robotic therapy can reduce the UL impairment without any negative effect on spasticity, even if the largest randomized controlled trial on the advantage of robotic therapy, compared with conventional physiotherapy or other types of intervention, did not find any evidence. The goal and the novelty of this study is the choice of two repetitive robot-assisted treatments in chronic stroke patients in the same year, whose purpose was to verify whether a second treatment, after 3 months from the end of the first one, could improve or consolidate the gain after an RR treatment. It is common knowledge that a long-term follow-up for patients who underwent RR does not confirm the gain achieved at the end of the treatment. However, our patients showed that a repetitive treatment, in the same year, may lead to a significant gain in many proximal arm behavioral measures, on the basis of both impairment and functional-based assessments. In addition, functional gain of the BBT and FAT secondary to robot treatment, which persisted for at least 6 months after the end of the intervention, was found. This choice of UL outcome assessment is secondary to the intention of using the International Classification of Functioning, Disability and Health domain body functions and abilities to describe the patient limitation. In particular, the FM and AS investigate the body function, whereas BBT and FAT investigate the capacities (Sivan et al., 2011). Our results also show that an intensive robotassisted treatment in chronic stroke patients may lead to a significant decrease in limitation of the paretic UL, with a high compliance and a high affinity between patients and robot therapy, as well as high satisfaction for all patients.

5

In a recent study, we showed the efficacy of ReoGo robotic therapy in terms of motor improvement and compliance with the persistence of the effects after 1 month (Bovolenta et al., 2011). On the basis of this preliminary result, we designed this new study, in clinical research, with two repetitive treatments and a long-term follow-up (6 months) in the same year. Most papers have shown that the use of robot-assisted therapy in chronic patients improves motor function, arm activity, and selfperceived arm ability in patients after stroke (Hesse et al., 2003; Krebs et al., 2008; Treger et al., 2008; Posteraro et al., 2009; Lo et al., 2010; Liao et al., 2011). Few papers only, though, have shown the maintenance of these improvements during a long-term follow-up; furthermore, various systematic reviews have explored high-intensity therapy to improve functional recovery without a clear consensus on ‘the best choices’ for the clinical practice. Only a few papers showed the effect of ReoGo therapy in chronic stroke patients (Lindberg et al., 2004; Treger et al., 2008; Bovolenta et al., 2009, 2011) and our results confirm the data published, with a positive evolution of the limitation of activity (measured with FM) and functionality for all patients related to the use of robot-assisted treatment. In particular, at the end of the treatment, reductions in motor impairment and improvement in paretic UL function were found (Bovolenta et al., 2009, 2011). At the 6-month follow-up, the FM scores showed a downward trend, but were still above the average values of the initial level. In a recent paper, Lo et al. (2010) have shown an improvement in the outcome scores of motor function at the end of treatment and a decrease at the 4month follow-up. A similar result was found by Lum et al. (2002); in their work, the score of the FM showed a clear improvement in terms of gain, immediately after treatment with RR, whereas at the 6-month follow-up, there were no significant changes. As reported by Fasoli et al. (2004), the reduced performance on the FM at the follow-up could be secondary to the lack of ongoing exercise for the paretic arm after discharge from robotic therapy and it might have been difficult or impossible for patients to integrate the exercises performed during robotic treatment into their homebased programs. The management of spasticity, which can interfere with the ability to position comfortably, transfer, walk, perform activities of daily life, or maintain adequate hygiene, and that can also be painful or predispose to the development of decubitus ulcers and contractures, is a field that has not been widely investigated as yet. The results of the AS for spasticity do not become worse through robotic treatment, as reported previously in other studies (Kwakkel et al., 2008; Lo et al., 2010; Liao et al., 2011), showing temporary modifications that, at the follow-up, revert to the initial state. This result could open new frontiers in the treatment of spasticity; in particular, it could associate the robot treatment with other therapeu-


tical approach to manage the spasticity in stable chronic stroke patients. The analysis of skill with the BBT shows a different behavior. The BBT assesses gross manual dexterity by counting the number of blocks that can be transported individually from one compartment of a box to another within 1 min. Higher scores are indicative of better manual dexterity (Desrosiers et al., 1994; Chen et al., 2009). In this case, the analysis of the BBT score showed good robot-related improvement and a significant gain at the long-term follow-up. The reliability, validity, and responsiveness of the BBT have been established in stroke patients (Chang et al., 2007; Hesse et al., 2008; Takahashi et al., 2008). The result of the BBT is consistent with the previous studies that indicated the highest increase in patients under robot treatment (Chang et al., 2007; Takahashi et al., 2008). The analysis of FAT showed a significant gain in this score. This result differs from Chang et al.’s (2007) papers, where the FAT and MAS scores did not show significant differences across the pretest, post-test, and retention test scores (Fazekas et al., 2007; Bernhardt et al., 2009). The evaluation of the effectiveness, feasibility, transferability, and cost/benefit ratio on the use of robotic devices in the rehabilitation programs of stroke patients has a huge impact on the organization of the National Health System. This result could be explained by the difference in the robot device and the proposed exercises. In any case, on the basis of this result, the RR treatment should be integrated with daily home exercise programs to reinforce the use of robot-trained movements to maintain acquired motor skills, resulting in a process of relearning motor activity and improve not only the activity but also function, suggesting a synergistic result that is higher than was estimated previously. Conclusion

To improve motor function, the paradigm of stroke rehabilitation strategies is currently focused on highintensity, repetitive finalized, and task-specific training, even if there is no protocol for UL rehabilitation after stroke widely accepted, whereas the treatment varies in duration, intensity, and frequency. The efficacy of RR is still under discussion, especially in European countries, whereas its cost/benefit is compared with conventional physiotherapy attentively. The use of this system in motor rehabilitation programs provides a safe and intensive treatment to patients with motor impairments after a stroke. The robotic protocol is easy and reproducible, allowing the treatment of patients with moderate to severe UL paresis. Our study examined the effect of two repetitive robot ReoGo treatments in chronic stroke patients comprehensively and its persistence after 6 months. These findings should be considered by

clinicians and researchers when deciding the appropriate treatment to recovery the body function and the activities in patients receiving stroke interventions. The first limitation of this study includes the relatively small number of patients and the inclusion in our sample of one patient at only 6 months after the acute stroke. It has already been shown that between 6 and 12 months from stroke, a spontaneous improvement could occur in the patients but the result of this patient did not influence the result of statistical calculations. The other limitation is the absence of a control group, but the long-term follow-up with a long observation can reduce this limitation. Future study with randomized controlled trials on the basis of a larger sample is required to confirm and validate the results.

Acknowledgements Conflicts of interest

There are no conflicts of interest.

References Ashworth B (1964). Preliminary trial of carisprodol in multiple sclerosis. Practitioner 192:540–542. Bernhardt J, Thuy MN, Collier JM, Legg LA (2009). Very early versus delayed mobilization after stroke. Cochrane Database Syst Rev 1:CD006187. Bovolenta F, Goldoni M, Clerici P, Agosti M, Franceschini M (2009). Robot therapy for functional recovery of the upper limbs: a pilot study on patients after stroke. J Rehabil Med 12:971–975. Bovolenta F, Sale P, Dall’Armi V, Clerici P, Franceschini M (2011). Robot-aided therapy for upper limbs in patients with stroke-related lesions. Brief report of a clinical experience. J Neuroeng Rehabil 9:8–18. Chang JJ, Tung WL, Wu WL, Huang MH, Su FC (2007). Effects of robot-aided bilateral force-induced isokinetic arm training combined with conventional rehabilitation on arm motor function in patients with chronic stroke. Arch phys Med Rehabil 88:1332–1338. Chen HM, Chen CC, Hsueh IP, Huang SL, Hsieh CL (2009). Test–retest reproducibility and smallest real difference of 5 hand function tests in patients with stroke. Neurorehabil Neural Repair 5:435–440. Cruz-Flores S, Rabinstein A, Biller J, Elkind MS, Griffith P, Gorelick PB, et al. American Heart Association Stroke Council; Council on Cardiovascular Nursing; Council on Epidemiology and Prevention; Council on Quality of Care and Outcomes Research (2011). Racial-ethnic disparities in stroke care: the American experience: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 7:2091–2116. Desrosiers J, Bravo G, He´bert R, Dutil E, Mercier L (1994). Validation of the Box and Block Test as a measure of dexterity of elderly people: reliability, validity, and norms studies. Arch Phys Med Rehabil 7:751–755. Fasoli SE, Krebs HI, Stein J, Frontera WR, Hughes R, Hogan N (2004). Robotic therapy for chronic motor impairments after stroke: follow-up results. Arch Phys Med Rehabil 85:1106–1111. Fazekas G (2013). Robotics in rehabilitation: successes and expectations. Int J Rehab Res 36:95–96. Fazekas G, Horvath M, Toth A (2006). A novel robot training system designed to supplement upper limb physiotherapy of patients with spastic hemiparesis. Int J Rehab Res 29:251–254. Fazekas G, Horvath M, Troznai T, Toth A (2007). Robot-mediated upper limb physiotherapy for patients with spastic hemiparesis: a preliminary study. J Rehabil Med 39:580–582. Franceschini M, Ceravolo MG, Agosti M, Cavallini P, Bonassi S, Dall’Armi V, et al. (2012). Clinical relevance of action observation in upper-limb stroke rehabilitation: a possible role in recovery of functional dexterity. A randomized clinical trial. Neurorehabil Neural Repair 5:456–462. Fugl-Meyer AR, Ja¨a¨sko¨ L, Leyman I, Olsson S, Steglind S (1975). The post-stroke hemiplegic patient. 1. A method for evaluation of physical performance. Scand J Rehabil Med 7:13–31. Hayward K, Barker R, Brauer S (2010). Interventions to promote upper limb recovery in stroke survivors with severe paresis: a systematic review. Disabil Rehabil 24:1973–1986.


Heller A, Wade DT, Wood VA, Sunderland A, Hewer RL, Ward E (1987). Arm function after stroke: measurement and recovery over the first three months. J Neurol Neurosurg Psychiatry 50:714–719. Hesse S, Schmidt H, Werner C, Bardeleben A (2003). Upper and lower extremity robotic devices for rehabilitation and for studying motor control. Curr Opin Neurol 6:705–710. Hesse S, Werner C, Pohl M, Mehrholz J, Puzich U, Krebs HI (2008). Mechanical arm trainer for the treatment of the severely affected arm after a stroke: a single-blinded randomized trial in two centers. Am J Phys Med Rehabil 10:779–788. Huang S, Luo C, Ye S, Liu F, Xie B, Wang C, et al. (2012). Motor impairment evaluation for upper limb in stroke patients on the basis of a microsensor. Int J Rehabil Res 35:161–169. Huang VS, Krakauer JW (2009). Robotic neurorehabilitation: a computational motor learning perspective. J Neuroeng Rehabil 6:5. Kim AS, Johnston SC (2011). Global variation in the relative burden of stroke and ischemic heart disease. Circulation 3:314–323. Krakauer JW (2006). Motor learning: its relevance to stroke recovery and neurorehabilitation. Curr Opin Neurol 19:84–90. Krebs HI, Volpe BT (2013). Rehabilitation robotics. Handb Clin Neurol 110: 283–294. Krebs HI, Mernoff S, Fasoli SE, Hughes R, Stein J, Hogan N (2008). A comparison of functional and impairment-based robotic training in severe to moderate chronic stroke: a pilot study. NeuroRehabilitation 1:81–87. Kwakkel G, van Peppen R, Wagenaar RC, Wood Dauphinee S, Richards C, Ashburn A (2004). Effects of augmented exercise therapy time after stroke. A meta-analysis. Stroke 35:2529–2536. Kwakkel G, Kollen BJ, Krebs HI (2008). Effects of robot-assisted therapy on upper limb recovery after stroke: a systematic review. Neurorehabil Neural Repair 2:111–121. Liao WW, Wu CY, Hsieh YW, Lin KC, Chang WY (2011). Effects of robotassisted upper limb rehabilitation on daily function and real-world arm activity in patients with chronic stroke: a randomized controlled trial. Clin Rehabil 2:111–120. Lindberg P, Schmitz C, Forssberg H, Engardt M, Borg J (2004). Effects of passive-active movement training on upper limb motor function and cortical activation in chronic patients with stroke: a pilot study. J Rehabil Med 3: 117–123. Lindmark B, Hamrin E (1988). Evaluation of functional capacity after stroke as a basis for active intervention. Validation of a modified chart for motor capacity assessment. Scand J Rehabil Med 20:111–115. Lo AC, Guarino PD, Richards LG, Haselkorn JK, Wittenberg GF, Federman DG, et al. (2010). Robot-assisted therapy for long-term upper-limb impairment after stroke. N Engl J Med 19:1772–1783. Lum PS, Burgar CG, Shor PC, Majmundar M, Van der Loos M (2002). Robotassisted movement training compared with conventional therapy techniques for the rehabilitation of upper-limb motor function after stroke. Arch Phys Med Rehabil 83:952–959. Masiero S, Armani M, Rosati G (2011). Upper-limb robot-assisted therapy in rehabilitation of acute stroke patients: focused review and results of new randomized controlled trial. J Rehabil Res Dev 4:355–366. Mathiowetz V, Volland G, Kashman N, Weber K (1985). Adult norms for the Box and Block Test of manual dexterity. Am J Occup Ther 6: 386–391.

7

Mehrholz J, Ha¨drich A, Platz T, Kugler J, Pohl M (2012). Electromechanical and robot-assisted arm training for improving generic activities of daily living, arm function, and arm muscle strength after stroke. Cochrane Database Syst Rev 13:6. Pe´ter O, Fazekas G, Zsiga K, De´nes Z (2011). Robot-mediated upper limb physiotherapy: review and recommendations for future clinical trials. Int J Rehabil Res 34:196–202. Platz T, Pinkowski C, Van Wijck F, Kim IH, Di Bella P, Johnson G (2005). Reliability and validity of arm function assessment with standardized guidelines for the Fugl-Meyer Test, Action Research Arm Test and Box and Block Test: a multicentre study. Clin Rehabil 4:404–411. Posteraro F, Mazzoleni S, Aliboni S, Cesqui B, Battaglia A, Dario P, Micera S (2009). Robot-mediated therapy for paretic upper limb of chronic patients following neurological injury. J Rehabil Med 41:976–980. Posteraro F, Mazzoleni S, Aliboni S, Cesqui B, Battaglia A, Carrozza MC, et al. (2010). Upper limb spasticity reduction following active training: a robotmediated study in patients with chronic hemiparesis. J Rehabil Med 3: 279–281. Sale P, Lombardi V, Franceschini M (2012a). Hand robotics rehabilitation: feasibility and preliminary results of a robotic treatment in patients with hemiparesis. Stroke Res Treat 2012:820931. Sale P, Franceschini M, Waldner A, Hesse S (2012b). Use of the robot assisted gait therapy in rehabilitation of patients with stroke and spinal cord injury. Eur J Phys Rehabil Med 1:111–121. Scott SH, Dukelow SP (2011). Potential of robots as next-generation technology for clinical assessment of neurological disorders and upper-limb therapy. J Rehabil Res Dev 4:335–353. Sivan M, O’Connor RJ, Makower S, Levesley M, Bhakta B (2011). Systematic review of outcome measures used in the evaluation of robot-assisted upper limb exercise in stroke. J Rehabil Med 3:181–189. Stineman MG, Xie D, Pan Q, Kurichi JE, Saliba D, Streim J (2011). Activity of daily living staging, chronic health conditions, and perceived lack of home accessibility features for elderly people living in the community. J Am Geriatr Soc 59:454–462. Takahashi CD, Der-Yeghiaian L, Le V, Motiwala RR, Cramer SC (2008). Robotbased hand motor therapy after stroke. Brain 2:425–437. Treger I, Faran S, Ring H (2008). Robot-assisted therapy for neuromuscular training of sub-acute stroke patients. A feasibility study. Eur J Phys Rehabil Med 4:431–435. Van Peppen RP, Kwakkel G, Wood-Dauphinee S, Hendriks HJ, van der Wees PJ, Dekker J (2004). The impact of physical therapy on functional outcomes after stroke: what’s the evidence? Clin Rehabil 18:833–862. Volpe BT, Krebs HI, Hogan N (2000). A novel approach to stroke rehabilitation: robot-aided sensorimotor stimulation. Neurology 54: 1938–1944. Volpe BT, Lynch D, Rykman-Berland A, Ferraro M, Galgano M, Hogan N, Krebs HI (2008). Intensive sensorimotor arm training mediated by therapist or robot improves hemiparesis in patients with chronic stroke. Neurorehabil Neural Repair 22:305–310. Wei XJ, Tong KY, Hu XL (2011). The responsiveness and correlation between Fugl-Meyer Assessment, Motor Status Scale, and the Action Research Arm Test in chronic stroke with upper-extremity rehabilitation robotic training. Int J Rehabil Res 34:349–356.