Mirror neurons: action observation treatment as a tool in stroke

EUR J PHYS REHABIL MED 2010;46:517-23

Mirror neurons: action observation treatment as a tool in stroke rehabilitation

C ER O V P A Y R M IG E H DI T C ® A

M. FRANCESCHINI 1, M. AGOSTI 2, A. CANTAGALLO 3, P. SALE 1, M. MANCUSO 4, G. BUCCINO 5

M

IN

Background. The observation of actions performed by others activate in an observer the same neural structures (including mirror neurons) as when he/she actually performs the same actions. Aim. The aim of the present study was to assess whether action observation treatment may improve upper limb motor impairment in chronic stroke patients. Design. This was an observational study. Setting. Patients were recruited by three Italian Centres for Neurorehabilitation between 2006 and 2008. Population. Twenty-eight chronic stroke patients with upper limb impairment have undergone for four weeks, five days a week, a rehabilitation treatment based on observation of video-clips presenting hand daily actions, followed by the imitation of those same actions with the affected limb. Methods. Functional evaluation by means of Modified Barthel Index (MBI), Frenchay Arm Test (FAT) and Fugl Meyer (FM) was carried out twice before treatment (BT1 and BT2), at an interval of 15 days, then after treatment (AT1) and finally at a two-month follow-up (AT2). Wilcoxon Signed Rank test was applied to test differences between scores obtained from functional scales before and after treatment (BT1 vs. BT2; BT2 vs. AT1; AT1 vs. AT2). Results. In all scales, scores did not differ when comparing BT1 with BT2. Scores improved significantly in all scales at AT1 as compared to BT2 (MBI, P=0.026; FAT, P=0.005; FM, P=0.001). This improvement was still present at the two-month follow-up as testified by no score difference between AT1 and AT2. Received on January 22, 2010. Accepted for publication on March 22, 2010. Epub ahead of print on April 23, 2010.

Corresponding author: M. Franceschini, MD, Clinical and Research Director of Neurorehabilitation Units, IRCCS San Raffaele Pisana, Via della Pisana 235, 00163 Rome, Italy. E-mail: [email protected]

Vol. 46 - No. 4

1Department of Neurorehabilitation IRCCS San Raffaele-Pisana, Rome, Italy 2Department of Rehabilitation University Hospital of Parma, Parma, Italy 3Department of Neurorehabilitation University Hospital of Ferrara, Ferrara, Italy 4Section of Neurorehabilitation General Hospital of Grosseto, Grosseto, Italy 5Department of Medical Sciences Magna Graecia University, Catanzaro, Italy

Conclusion. Action Observation Treatment may become a useful strategy in the rehabilitation of stroke patients. Clinical Rehabilitation Impact. The present preliminary study suggests that stimulation of neural structures (including mirror neurons), activated when the patients actually perform the same actions as those observed could constitute a good alternative rehabilitative approach in chronic stroke patients. KEY WORDS: Action observation - Rehabilitation - Stroke.

T

he recovery of upper limb use in patients following a stroke is poor and represents a challenge for every rehabilitative team. The degree of motor impairment at the onset is, of course, a prognostic factor of poor outcome: only a range from 5% to 20% of patients presenting an upper limb paralysis at onset may improve the motor impairment over time.1 Clinical and epidemiological data reported above also indicate a substantial and intrinsic limitation of the traditional

EUROPEAN JOURNAL OF PHYSICAL AND REHABILITATION MEDICINE

517

FRANCESCHINI

NEURON MIRROR

IN

M

518

bilitation of chronic stroke patients with upper limb impairment. Materials and methods Twenty-eight patients entered the study. Patients were recruited from three Italian Centres for Neurorehabilitation (General Hospital of Grosseto, A; University Hospital of Ferrara, B; University Hospital of Parma, C) between 2006 and 2008. All participants had experienced an ischemic and/or hemorrhagic stroke in the territory of the middle cerebral artery at least six months earlier. Diagnosis was confirmed by means of a computed tomography scan and/or a magnetic resonance imaging. Patients were included if they were aged between 18 and 75 years; presented with a first ever stroke; without previous neurological pathologies; had normal or corrected to normal visual and auditory acuity; no clinical evidence of neglect; mini mental state examination (MMSE) ≥24; Token test ≥23; no mood depression or apraxia; absence of clinically diagnosed total deficit of tactile sensibility. Were excluded patients whose families do not guarantee continuous access to the hospital as per protocol. Before inclusion, patients underwent neuropsychological assessment and functional evaluation. Neuropsychological assessment included: MMSE,16 the Token test 17 to rule out aphasia, the tests of De Renzi to rule out upper limb apraxia with intransitive and transitive gestures 18-20 and the Bells barrage test to rule out visual neglect or other attention disorders.21 Depression of mood was evaluated clinically according to the criteria of DSM IV revised. Functional evaluation in activities of daily living was assessed by means of the following scales: Modified Barthel Index (BI) 20 and FIM Scale (FIM).22 The motor impairment recovery was assessed by means of: Fugl-Meyer (FM),23 Ashworth Scale (AS) 24 and Frenchay Arm Test (FAT).25 Functional scales were delivered twice before treatment (BT1, BT2), at an interval of 30 days, to rule out any spontaneous improvement; then again after treatment (AT1) and at two months’ follow-up (AT2). Functional scales and motor impairment scales were delivered by a physician blinded to the study. This study and all procedures for recruitment were approved by the local Ethics Committees of the


rehabilitative approach in modifying the clinical history of upper limb motor impairment and functional ability. Independence in the Activities of Daily Living post-stroke is, however, related to numerous factors: not only the level of residual motor function of the affected limbs, but also the extent of brain lesion, the presence of anesthesia,2 cognitive disturbances 3 and mood disorders.4 Moreover, the residual motor function of the affective limb is also very important not only in the motor impairment,5 but also in the self-care activities of daily living,6, 7 such as grooming, cutting food, washing and dressing. As far as the time course of arm recovery is concerned, most of the motor improvement occurs within one-two months of stroke onset,8, 9 both in severe and mild paresis. Rehabilitation strategies, well founded on neurophysiology, but alternative to current conventional methods, are needed to improve the outcome (for a review of classical rehabilitation methods see 10). Over the last decade, basic neuroscience has greatly contributed to changing our notion of the organizational and functional properties of the motor system. It is now well accepted that observation of actions performed by others activates, in the perceiver, the same neural structures responsible for the actual execution of the same actions. Evidence in favour of this notion comes from a number of studies carried out with different neurophysiological techniques.11 Brain imaging experiments have demonstrated that during the observation of hand/arm object-directed actions there is a signal increase in the same brain regions 12, 13 also active during manipulation and action planning.14 Brain areas endowed with this mirror property are often referred to as the mirror neuron system. Thus, the mirror neuron system in humans, as in monkeys, appears to be localized in the ventral premotor cortex and in the inferior parietal lobule. Of particular relevance for the present study is the fact that during the imitation of complex actions, areas within the mirror neuron system are found to be active from the time of the observation of the model until actual execution of the actions.15 Moreover, since the mirror neuron system is bilateral, this strategy should apply regardless of the lesion site. The aim of the present study was to assess, in a clinical context, the possible use of this experience of action observation treatment approach for the reha-


December 2010

NEURON MIRROR

FRANCESCHINI

TABLE I.—Treatment planning.

Medians (error bars: 95% Cl for median) 75

65 Age (years)

To clean the table To take a clothes brush and brushing To take a soap and wash the hands To take and eat an apple To take and to change a position of a jar To take and to bring a gloss to mouth To take and to move a bottle To take and to open a box To take and to move the pot To open and closed the drawer To off the tap To open and closed the handle To brush the hairs To challenge fork To take and to bring a cup to mouth To take the toothbrush and clean teeth To take some paper and throw in the trash To take the magazine and read To take the keys and open a door To take and to try use a cut

45

35

Token test

36

M

Vol. 46 - No. 4

32

* P=0.300

31

MMSE

IN

We chose 20 daily activities implying the use of an object to be used during action observation treatment (Table I). These actions, performed both by an actor and an actress, were recorded from different perspectives, to make the video clips more interesting and to sustain the attention of patients during the rehabilitation sessions. Each action (e.g., having a coffee) was subdivided into 3 or 4 constituent motor acts. In the case of having a coffee, for instance, the subdivisions were: reaching for the cup, grasping it, bringing the cup to the mouth and drinking. Each motor act was presented for three minutes so that the total duration of each video-clip was 9-12 minutes. Figure 1 shows frames taken from the video clip used during the treatment. Patients underwent a four-week rehabilitation program, five days a week. Treatment took place in a quiet room, with the patient sitting comfortably in front of a computer screen, in the presence of an expert physiotherapist with at least three-year experience. During each rehabilitation session only one action out of twenty was trained. Over sessions, the different actions were presented in a fixed order, according to an increasing level of complexity as judged by the experimenter. Treatment consisted of an

34

30

Hospitals in which the study took place. All patients gave informed consent before participating. Treatment procedure

* P=0.448

55


1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

29

* P=0.213

27

Centres

A

B

C

Figure 1.—Median values for variables by Centres are shown. P values relative to comparisons are also reported. Error bars: 95% CI for median. MMSE: mini mental state examination;*the probability for the Kruskal-Wallis one-way analysis of variance by ranks.

observation task and an imitation task, respectively. At the beginning, patients were required to carefully observe each motor act shown in the video clip for three minutes; later, they were asked to imitate the motor act, soon after observation, for two minutes with the affected limb. During this phase patients were provided with the same objects used in the video clip, in order to make the task as valid as possible in a real-life context. It is worthy of note that patients were advised prior to imitation that the quality of their execution was not the goal of the rehabilitation treatment. In this way, we hoped to facilitate spontaneous use of the affected limb, even if not fluid or elegant. During the session, the physiotherapist simply reminded patients to pay attention to the video clips and encouraged them during the imitation task.


519

FRANCESCHINI

NEURON MIRROR

TABLE II.—Socio-demographic and clinical variables of the patients. Percentile Variables

N.

%

Mean

SD 25th

50th median

54.0

61.0

70.0

75th

Socio-demographic variables Subject Center A Center B Center C Age, years Gender Female Male

28 16 4 8

57.1 14.3 28.6 58.5

Etiology Ischemic Clinical syndromes LACI PACI TACI Hemorrhagic Side lesion Left Right Day time from ictus Impairment variables MMSE Token test Bells test (left) Bells test (right)

32.1 67.9

21

75.0

5 2 6 7

23.8 9.5 28.6 25.0

19 9

67.9 32.1


General clinical variables

9 19

14.0

817.6

606.5

427.0

619.5

1165.0

28.4 32.4 16.8 17.0

2.5 4.1 0.7 0.0

27.0 31.0 17.0 17.0

29.0 34.0 17.0 17.0

30.0 35.5 17.0 17.0

SD: standard deviations; MMSE: mini mental state examination.

IN

Since video clips were presented twice during each rehabilitation session, each session lasted about 40 minutes in total.

M

Statistical analysis

Kruskal Wallis test and Fisher-Freeman-Halton exact test were applied to assess differences in the three groups of patients for quantitative and qualitative variables, respectively. Sex, age, type of lesion, time interval from acute event (in days) and scores of the neuropsychological assessment were the independent variables. Wilcoxon Signed Ranks test was applied to assess differences between scores obtained from functional and motor impairment scales before and after treatment (BT1 vs. BT2; BT2 vs. AT1; AT1 vs. AT2).

520

Results

Demographic and clinical data of all patients are reported in Table II. There were no statistically significant differences among the three groups as far as demographic (age) and clinical parameters (Token test and MMSE) were concerned (Figure 1). Otherwise the time interval from acute event was shorter for Center B (Table II). Functional and motor impairment evaluation scores are reported in Figure 2. No difference was found when comparing BT1 with BT2. In contrast, when comparing BT2 with AT1 a significant difference was observed in all scales taken into consideration. In fact, BI median value moved from 16.0 to 17.0 (P=0.026). Median value of FM shifted from 38.5 to 49.0 (P=0.001), the Ashworth Scale decreased from 7.5 to


December 2010

NEURON MIRROR

FRANCESCHINI

5.5 (P