Upper limb training after stroke (ULTRA-stroke): Contrasting the clinical effects and underlying mechanisms of unilateral and bilateral training

In the Netherlands, each year more than 32,000 patients sustain a stroke (Loor
et al., 1999) and the incidence is expected to have increased by 30*45% in 2015
(Ruwaard & Kramers, 1997). About 80% of the survivors have an upper limb
paresis immediately after stroke (Nakayama et al, 1994), hampering movement of
the paretic arm and bimanual coordination (Ustinova et al., 2006).
Unfortunately, with conventional treatment programs only one third of all
stroke patients regain some dexterity within 6 months post-stroke (Dobkin et
al., 2005). However, recent studies have revealed promising results with
specific interventions aimed at arm-function improvement.
One such intervention is bilateral arm training with rhythmic auditory cues
(BATRAC), which has been shown to have beneficial effects on the paretic arm
(Whitall et al., 2000), possibly as a result of changes in contralesional
cortical networks (Luft et al., 2004).
In contrast, various controlled trials have suggested that intensive unilateral
training by constraining movements of the nonparetic arm (constrained-induced
movement therapy; CIMT) is an effective method for improving upper limb
function (Hakkenness & Keating 2005; Wolf et al., 2006). This suggests that
training may also induce beneficial changes in the affected rather than
nonaffected hemisphere and raises the question whether in BATRAC the improved
functionality of the paretic arm indeed results from exploiting
interhemispheric interactions, or merely from training with the affected arm
(cf. Luft et al., 2004).
To address this question (cf. Rose & Winstein, 2005) the proposed project
entails a randomized control trial (RCT) in which the expected merits of both
BATRAC and CIMT are compared with those of an equally intensive (i.e.,
dose-matched) conventional treatment program (DMCT), while also the effects of
BATRAC and CIMT are compared on several outcome measures. To this end,
participants are divided over three intervention groups and the effects of the
interventions are assessed (1) prior to training, (2) after 6 weeks of
training, and (3) 6 weeks after training.

The project has two principal aims: (1) to assess the relative effectiveness of
the three interventions, also as a function of patient characteristics, and (2)
to delineate the functional and neurophysiological changes that are associated
with those intervention effects. To assess the effectiveness, a range of
functional outcome measures will be determined pertaining to ADL functioning,
motor ability of the paretic arm, bimanual coordination, and peripheral motor
functioning. Besides shedding further light on the merits of bilateral versus
unilateral upper limb training in general (Stewart et al., 2006), the study
will generate specific insights into the effectiveness of distally oriented
BATRAC, aimed at improving wrist and finger extension (Kwakkel & Kollen.,
2007). In light of contrasting results and divergent perspectives on associated
mechanisms (Richards et al., 2008), the potential dependence of the
effectiveness of the interventions on neurological characteristics of stroke
survivors will also be examined. To uncover the mechanisms associated with
therapy-induced functional improvement, two kinds of analysis are included.
First, changes in three empirically identified (and functionally defined)
sources of interlimb interaction will be examined. Specifically, BATRAC is
expected to induce more improvement in these interactions than both CIMT and
DMCT. Second, MEG recordings will be analysed to identify treatment-induced
neuronal reorganizations. CIMT is expected to result primarily in changes in
ipsilesional hemisphere functioning, which may be related to restitution of its
former functionality. BATRAC, on the other hand, is expected to induce
primarily adaptations in the contralesional hemisphere (Luft et al., 2004),
which would indicate compensatory cortical reorganization in which the coupling
to the nonaffected hemisphere gains a special role in the motor control of the
paretic arm.
In short, the primary research questions are: 1) Is BATRAC or CIMT, when
compared to DMCT, more effective in terms of recovery of (unimanual and
bimanual) hand and arm function in subacute stroke patients? 2) How are the
observed changes in functionality related to changes in peripheral stiffness,
interlimb interactions and cortical inter- and intrahemispheric neural
networks?

Patients will be placed in one of three intervention groups (BATRAC, CIMT,
DMCT). Whereas the pretests (t0) are performed in the week prior to the
intervention, posttests (t1) are performed in the week after the intervention.
After the first assessment an intervention period will take place for 6
consecutive weeks, 3 times a week for 1 hour. The degrees to which changes are
sustained are examined using retention tests (t2), 6 weeks after completion of
the intervention.

The interventions are applied by physiotherapists and/or occupational
therapists working at the RCA. Where possible, interventions in groups is
preferred (no more than 3 patients per group).
The BATRAC group receives 60-minute sessions, 3 days a week for 6 consecutive
weeks. Treatment will be applied in 5-minute movement periods interspersed with
5-minute rest periods. The lower arms are fixated and the subject performs
flexion and extension movements about the wrist in the horizontal plane, paced
by an auditory metronome. The tempo of the auditory cues depends on the
severity of upper limb deficit and is selected individually. Over the course of
training the tempo is adjusted in response to improvement in task performance.
The movements are performed in phase (simultaneous flexion or extension) and
antiphase (flexion of one wrist coincides with extension of the other), where
maximal flexion/extension should coincide with the auditory cue. Changes
between the two modes are included to avoid loss of motivation.
The CIMT group receives 60 minutes of task-oriented training of the upper
extremity, aimed at improving dexterity of the paretic arm (i.e., shaping), 3
days a week for 6 consecutive weeks. Task difficulty is increased progressively
using behavioural techniques of shaping and successive approximation (Taub et
al., 1999). In addition, the Padded Safety Mitt (Samson Preston # 6727; Sammons
Preston, Inc, Bolingbrook, IL, USA) is applied to immobilize the non-paretic
arm for at least six waking hours each working day to prevent from the
non-paretic arm taking over tasks of the paretic arm. The content and duration
of the CIMT therapy as well as the shaping exercises are recorded in a
patients* log reflecting the progress in reaching treatment goals.
Dose-matched control treatment (DMCT) consists of exercise therapy based on
existing guidelines for exercise therapy as shown by Dutch Society of
Occupational Therapy (NVE) and Royal Dutch Society of Physical Therapy (KNGF)
(van Peppen et al, 2004). Therapy will be applied 60 minutes per treatment
session, 3 days a week for 6 consecutive weeks and will not contain elements of
the other two therapies. The content and duration of the sessions is recorded
in patient logs.

We are aware of the burden on patients of intervention and (clinical)
measurements. However, evidence is compelling that the functionality of the
impaired paretic arm may be improved by using specific therapeutic
interventions. Next to the opportunity to evaluate specific interventions for
the paretic arm, we expect to find answers to fundamental research questions
that are vital to our understanding of functional recovery after paresis of the
upper limb. The proposed BATRACT-CIMT trial goes beyond testing the
effectiveness of the interventions, by extending the analyses to the mechanisms
of exercise-induced neuroplasticity that are associated with functional
improvement (e.g., the importance of cortical reorganization in the affected
and non-affected hemispheres and cerebellum for regaining dexterity).

Clinical assessments
Assessments of included patients for the trial are performed by questionnaires
and (functional) tasks. There are no invasive measurements. Care is taken that
the total time to perform all the required assessments is no more than 1 hour.

Peripheral stiffness assessment
Peripheral stiffness assessment involves the use of a haptic robot
(Wristalyzer, Moog FCS INC) and EMG. The Wristalyzer is a one-degree-of-freedom
wrist perturbator for neurological research and diagnostics. Surface EMG is
measured for additional validation. Both methods are non-invasive and do not
have any side effects. The assessment will take about 40 minutes (including
mounting and demounting).

Interlimb interactions
At the Faculty of Human Movement Sciences at the Vrije Universiteit an
apparatus is developed for the assessment of interlimb interactions. The
patient*s forearms will be fixated in a neutral position, whereas the hands,
supported by horizontally movable manipulanda can be actively or passively
moved around the wrist. The movement trajectory is determined by the patient*s
own abilities. The method is not invasive and assessment is separated in two
sessions, each no longer than 1 hour.

Brain dynamics
The neural effects of BATRAC and CIMT are assessed with MEG. With MEG, the
brain can be observed *in action* without any of the risks associated with
other imaging techniques. Surface EMG is used to monitor task performance. All
patients perform simple unimanual and bimanual isometric force production tasks
with their fingers. The assessment takes no longer than 1 hour.