Stroke typically causes hemiparesis implicating motor deficits on one of the upper extremities [
1].
Thus, asymmetrical kinematic and kinetic functions between the paretic
and non-paretic arms frequently appear in acute and subacute patients [
2,
3], and further remain at the chronic stage of recovery [
4].
To improve these long-term motor impairments, many stroke researchers
and therapists have focused on bimanual training protocols requiring
simultaneous paretic and non-paretic arm actions because of potential
benefits on paretic arm functions through inter-limb coupling processes [
4,
5].
A theoretical basis underlying the bimanual training protocols assumes
that repetitive bimanual actions may facilitate balancing cortical
excitation and inhibition patterns between more-affected and
less-affected hemispheres contributing to motor recovery post stroke [
6,
7,
8].
Interestingly,
prior studies evidenced that a simple bimanual condition resulted in
transient functional improvements in the paretic arm movement control.
Harris-Love and colleagues reported that the paretic arm revealed an
increase in peak velocity and acceleration during bimanual reaching
tasks as compared to the unimanual conditions [
9,
10].
Similarly, Rose and Winstein found higher peak velocity in the paretic
arm and lower peak velocity in the non-paretic arm during bimanual
aiming tasks than those during unimanual aiming tasks [
11].
These findings indicated that the paretic arms revealed better
kinematic functions while performing bimanual actions at maximal speeds.
On the other hand, when individuals post stroke executed bimanual
movements at their preferred speeds (i.e., submaximal movement
condition), no motor improvements in the paretic arm were observed.
Moreover, the non-paretic arm adapted movements closely aligned with the
paretic arm in the bimanual condition, and no functional changes
occurred in the paretic arm [
12,
13,
14].
Finally, no changes in a reach-grasp-lift-release performance at
preferred-speed across paretic and non-paretic arms between unimanual
and bimanual conditions [
15].
Similar
to these kinematic findings, individuals post stroke showed better
kinetic functions in their paretic arm during the maximal bimanual
isometric force production tasks. For example, the bimanual conditions
transiently increased maximal isometric handgrip force for the paretic
arm [
16] and maximal isometric wrist and fingers extension force [
17]
than those for the unimanual conditions. However, at the submaximal
targeted force levels (e.g., 20–80% of maximum voluntary contraction:
MVC) the paretic arm revealed no changes in mean handgrip force between
the bimanual and unimanual conditions [
16,
18].
Taken together, these kinematic and kinetic findings commonly support a
proposition that bimanual training protocols requiring high intensity
levels (e.g., maximal movement speed or force production) may be more
effective on functional recovery of the paretic arm.
Importantly,
during submaximal isometric force control tasks, the ability to maintain
forces near targeted force levels with minimal variations is an
additional crucial indicator estimating individual’s progress toward
stroke motor recovery [
19,
20,
21].
Indeed, individuals post stroke improved their isometric force control
capabilities after completing rehabilitation protocols without an
increase in force production [
19,
20].
These findings lead to a possibility that the bimanual conditions
change the force control capabilities of the paretic arm without
altering force outputs as compared to the unimanual condition. Recently,
Kang and Cauraugh [
17]
found that the paretic arm produced greater maximal force during simple
bimanual wrist and fingers extension task than that during unimanual
condition. However, during submaximal force production tasks, how the
unimanual and bimanual conditions alter force control capabilities in
the paretic arm is still unclear.
Traditional bimanual training
protocols included homologous movements of the paretic and non-paretic
arms because of the possibility that more motor improvements may occur
after task-related training protocols typically requiring dynamic
movements [
22].
However, given that muscle weaknesses post stroke are highly related to
deficits in activities of daily living, many stroke researchers
additionally focused on various types of resistance training for
restoring muscle strength in the paretic arm [
23].
Specifically, several studies used isometric resistance training on the
unimanual paretic arm, and reported improvements in muscle force [
24,
25] and motor control ability [
26].
Moreover, Saunders and colleagues found positive effects of an
isometric resistance training protocol on reducing blood pressure [
22].
Presumably, isometric resistance training can be an additional
treatment protocol for optimizing stroke motor recovery, and further
applying specific intensity (e.g., either submaximal or maximal) and
contraction type (e.g., either unimanual or bimanual) may facilitate
these beneficial effects on the paretic arm functions [
8,
23].
Thus, investigating transient bimanual effects on paretic and
non-paretic arm force control across submaximal and maximal targeted
levels may provide useful information for developing potential bimanual
training protocols based on isometric force production paradigms.
Beyond
force production changes across submaximal and maximal force control
tasks, quantifying the variability of force production within a trial is
crucial for estimating an individual’s force control capabilities [
21].
Specifically, force variability using either standard deviation or
coefficient of variation (CV) is a conventional measurement to quantify
noise of motor outputs so that greater variability increased instability
of the motor system implicating impaired force control capabilities [
27].
However, given that a motor system may acquire stability by solving
environmental and biomechanical limitations as a nonlinear process, more
force variability does not necessarily lead instability of the motor
outputs interfering with task performance [
27,
28].
Further, the force regularity, a temporal structure of variability
(e.g., sample entropy: SampEn), is a nonlinear outcome measure
indicating the adaptability of motor outputs. Although a certain level
of force variability exists in the intact motor system, less force
regularity denotes more adaptability contributing to better force
control performance. However, these patterns may not appear in the
paretic arm because of the impaired motor system post stroke. Indeed,
individuals post stroke revealed more force variability with increased
force regularity during isometric force control as compared to
age-matched healthy controls [
19,
29,
30].
Taken together, measuring both variability and regularity of force
production is necessary to further elaborate altered force control
capabilities across unimanual and bimanual conditions post stroke.
Thus,
the current study examined force control capabilities in chronic stroke
patients across unimanual and bimanual conditions to provide additional
information regarding the beneficial effects of bimanual contraction on
restoring paretic arm functions. Post stroke individuals performed
isometric force control tasks with wrist and fingers extension at
submaximal targeted force levels (i.e., 5, 25, and 50% of MVC) and a
maximal level (i.e., maximal sustained force production) with their
unimanual arms (i.e., paretic vs. non-paretic arms) and both arms
simultaneously. We selected the three targeted force levels because many
activities of daily living require submaximal force generation with
5–50% of the maximum efforts [
31], and paretic arm functions potentially varied with these submaximal ranges [
19,
21,
32].
Further, our force control outcome measures included force production,
variability, and regularity. We hypothesized that bimanual conditions
would increase force production and reduce force variability and
regularity in the paretic arm during submaximal and maximal force
control tasks when compared to the unimanual conditions.