Background.  Many people who have had a stroke find a way to be able to walk again. This recovery is an adaptation/compensation strategy of the individuals and their therapeutic team, searching for a way to bear weight and maintain control, especially of balance. Very quickly, support/aids are used and practiced with, but what are we actually doing? Is this training to build better coordination, or is that not possible? In any case, it is not relearning, because these individuals have to compensate. Aim/Design:  Searching for the reasons why so many, even severely affected individuals, are still able to stand and walk on the affected leg, while, for example, hand/arm and foot/leg selectivity is often very severely impaired. One of these reasons is the innervation of the trunk and the large joints from the hemispheres of the brain. Because the pathways from the cortex partially cross, a connection from both hemispheres remains present, and it appears that this crossing at lower levels even approaches a 70%-30% distribution.

Result: This result therefore means that muscle patterns are also influenced by unaffected areas in the brain and that this therefore plays a role in recovery, but of course also in the control of walking and especially of the stance phase. For example, the posterior diagonal is essential to activate the affected side/hip via the unaffected shoulder, yet we often see that this is not sufficiently successful.

Discussion and conclusion: The search for an adjustment/compensation after a stroke in order to be able to walk in some way often involves providing aids and “grooving in” that walking. That is not training according to training rules, and often no activation occurs in the muscles around the affected hip joint even though this is possible. By including this activation in the training program from the start, it will also be incorporated into the walking pattern. This therefore does not mean first looking for an adjustment/compensation, or “relearning,” but immediately actively training the coordination of the affected lower trunk/hip.

Various studies show that many people are still able to learn to walk in one way or another after a stroke [1-6]. The recovery of this ability is much stronger than, for example, that of the arm and certainly the hand function. This is mainly due to the organization of the brain and the pathway of the brain to the systems that are important for walking [7-14]. Due to this organization, reduced control in the damaged hemisphere can be 'compensated' by the contribution of the other hemisphere, but the result will always be at a lower level than before. The damaged hemisphere, together with the other side, still has to ensure the highest form of selectivity, which is often no longer optimally possible, and the system has to make adjustments. These adjustments are often seen in the use of pathological stereotypical synergies and higher tone. In fact, the degrees of freedom [14] (the basis of selectivity) are limited, which means less control is required and some form of standing and walking is still possible. The control of the body to move, and in this case to walk, is then no longer completely monitored by the cortex (figure 1), but also by lower systems (Figure 2).

Figures 1 and 2: The projection of the body onto the cortex (figure 1) and it is striking how large the projections of the extremities are (hand-foot) and how small, precisely, the rest of the leg and torso are. Figure 2 shows the projection deeper in the brain, and here we see a large projection of the leg and torso as a basis for the “coarse execution” of the walking movement.

The pathways that run from the brain to the nerves downwards and vice versa, and when we look at walking, we see that those pathways to the trunk and the shoulders/hips cross 90% at the cortical level, but 10% go directly to the same side. These (cortical) brain nuclei are especially important for finer, more selective work such as balance during walking, but if we remove that balance element, control from lower centres (figure 2) should be sufficient to be able to walk with help/support within a large stable base of support. And here we see that the influence of both hemispheres is important because here the crossing part is lower, namely 70%, and the non-crossing part 30% [16]. This system thus makes it possible to practice a form of walking at a lower level, and that also gives room to train muscle patterns in a task-specific way. The outcome will always be from a lower level if the system is damaged, and often that is in the interpretation of the incoming information and the ability to respond selectively in an optimal way.

Figure 4: This figure shows the progression from the cortex projections distally. At the level of the brainstem, we see that the majority of these cortical pathways cross to the other side to continue their distally. But at that brainstem level, it is clearly visible that a portion remain on the same side and innervate distally. This concerns the motor pathways.

Diagonals

This organization, of course, has a very clear reason, and that is particularly visible in movements like walking. When we walk, we see that the arm swing is opposite to the movement of the leg but in line with the movement of the leg on the other side. [16, 6,7,8,9,10] If we look at motor patterns, then the diagonal system plays the biggest role in walking. We do need to temporarily disregard balance and see where the power comes from. The force needed to move forward over the loaded hip during the stance phase comes from the posterior diagonal running from the gluteus maximus to the opposite shoulder and through their combined effort. At the same time, the anterior diagonal is engaged in the swing phase through a contraction in that muscle chain. The muscle system therefore receives its innervation from both hemispheres.

Photo 1 and 2: Depending on the intensity, we see the contribution of the rear/front diagonal to the swing and swing form of the opposite arm.

These are photos of people without a disruption of the system due to brain damage to make clear which system serves as a basis for walking. But it is clearly visible here that the opposite swing between arm and leg is greater as the speed and/or step increases. 

The posterior diagonal starts in the gluteal musculature with its attachment in the thoraco-lumbar fascia and then directly to the opposite shoulder via the latissimus dorsi and many other muscles that additionally support this whole. A concentric contraction brings the opposite shoulder backward simultaneously with the hip/trunk on the other side. 

So, if there is a fixed point on the ground, the body is moved over that point. The anterior diagonal has a clearly less rigid structure because the tendon sheet at the front depends on the muscle tone in the m. rectus abdominus. We do see the connection between the m. serratus anterior and the oblique abdominal muscles, which seem to continue from one side to the other. Yet the 'connection' with the hip flexor (m. iliopsoas) is indirect and depends on 'core stability [17],' because then a point is fixed and concentric contraction of the anterior diagonal can swing the leg and opposite arm forward. So, if there is a fixed point, then movement over that point is possible and we have a step. A swing depends on the correct tension in the front diagonal, which brings the m. iliospoas to length and swings forward.   And this is often more difficult after a stroke. Heel strike.

Normal walking at a reasonable speed will typically be characterized at the start by means of the heel strike [18]. The fixed point provides the rear diagonal underneath on that side of the fixed point, with the leg positioned well forward and the gluteal muscles nicely at length, and through the tension of the rear diagonal from the opposite shoulder, the other part of the gluteal is also ensured of a fixed point. Thus, a concentric contraction of the gluteal muscle pulls the leg under the hip or, in other words, the torso over the foot. Once again, all balance necessities are disregarded, but it is true that the movement must proceed “precisely” to optimally activate the musculature (Figure 7).

Figure 7: The red line shows the posterior diagonal running from the shoulder joint to the opposite hip joint. The blue line represents the course when someone heavily leans on a cane, and then the diagonal comes out 'in front of the hip joint,' and studies have shown that the EMG activity [19] then decreases so much that muscle loss/coordination occurs. In post-stroke individuals, this will also happen, but at the same time the dominance of the adductors will increase because they can still provide extension in the leg, especially in combination with hip flexion.

The other diagonal (green line) will therefore often also not end in the shoulder joint but end in and around the shoulder blade, which often causes a retraction as a reaction in the upper torso. Standing and walking after a stroke will depend on the fixed point (the affected foot) but also on how well the diagonal can function. Because a heel strike often does not occur, weight is placed on that foot without the concentric contraction of the gluteal muscles, but instead, weight is placed and the gluteal muscles are lengthened (an eccentric contraction) together with the adductors at the back, which are further engaged due to hip flexion.  Here, we thus see the compensation that occurs when people try to walk with support using a cane. 

Figure 8: The swing phase is performed with the torso as the 'fixed point,' and because tension develops on the extended m. iliopsoas, little power is needed to swing the free leg forward. After a stroke, the continuity of the anterior diagonals is disrupted, which is often visible as the navel being pulled more toward the unaffected side. In Figure 8, we see that the diagonal actually loses its stability because the rectus is not stably positioned in the center, and, among other things, the oblique abdominal muscles lose their fixed point. Furthermore, core stability will immediately disappear, making it difficult to provide the hip flexor with a fixed point to swing the leg forward.

The red line indicates the diagonal to the swing leg on the affected side, but the blue line to the non-affected side [20], and then we know that when walking, a lot of forward movement and balance actually comes from that side, and that produces a reaction in the diagonal to the opposite shoulder, which at the front is often incomplete. To get the leg forward, we will see movements initiated by the non-affected side through an extension of the affected trunk with a rotation done by the non-affected hip, the circumduction movement [21].

Pathology after stroke.   We see that the trunk has a very important role for the walking part and can actually almost always perform it in some way, however limited. If there is a stable fixed point that can support the body weight, limited movement over and/or “along” that fixed point leg is possible. Due to the different projections at different levels and double innervation, this is also possible after major brain damage. The system will often have to manage without heel strike, even though the foot is “fixed” in that position by an AFO (Ankle Foot Orthosis), it will often only be a creation of a fixed point. The result is that no concentric action of the gluteal is elicited, but that the leg is loaded by moving the torso over the leg from 'above,' and when that goes well, the other leg is pushed/added. This often means 'hanging' on the gluteal muscles/adductors, so an eccentric reaction and a muscle reaction elicited by a 'stretch' on these muscle groups.[22,23,24] An eccentric contraction is certainly an “economic” solution, but it will not really further improve the walking pattern; for that we need more, and that must be achieved through training of the muscle patterns and the integration of this improved coordination and power into everyday walking. Due to a pes equinus varus, this movement pattern is even more readily invoked because there is no simple alternative. [25,26] The extensive neural network, including the diagonals (front and back), also shows that the arms are now much more involved in activating the diagonals, but that there can also be consequences arising from this. The non-affected arm/shoulder is used by all people who need an aid to maintain balance to improve walking over the affected hip, and you can see that in the placement and holding of a cane.

Photo 3 and 4:  People after a stroke who usually only need a standard cane to control their balance and improve their walking. But now they use the cane to improve movement on the affected side starting from the shoulder, and this activation now begins in the opposite shoulder. We can see this from the way they hold the cane (index finger photo 4) as well as where the cane is placed. Namely at the level of the push-off of the affected leg.

That this is a perfect solution is evident from the concentric contraction of the hip extensors. This means that through the push-off with the pole, the diagonal is optimally activated and that the posterior diagonal still ends well in the hip joint, thereby maintaining and utilizing the concentric contraction.  This means that walking in this way—task-specific—with resistance will lead to training of the muscle pattern responsible for the stance phase and step length. This effect of activating the posterior diagonal from the opposite shoulder also occurs when walking with Nordic Walking poles, which are also positioned backward [27].

However, this will change if the possibilities cannot be solved with a simple stick because the balance requires support surface area and the load on the affected leg may not be optimal. That requires support on the non-affected side and at the same time causes a shift to that side, so the diagonal will no longer run optimally and therefore this results in a loss of muscle coordination and muscle.

Pay attention to the arm position on the affected side because, since the unaffected leg has to work so hard to bear weight, control balance, and facilitate the swing, the posterior diagonal will be very active and place the affected shoulder in retraction, and thus in a flexion synergy with all the consequences thereof. 

Photo 7:Swing phase of the affected leg.  In principle, the leg often remains in the same position as in the stance phase and the weight of that leg is first taken off and controlled by the non-affected leg and arm/cane. The front diagonal is not able to provide an anchor for the hip flexor, so an alternative is sought. Here, the upper torso leans mostly backward on the non-affected side, creating a stretch on the front diagonal and thereby building enough tension in it and the flexor, often in combination with the upper torso leaning sideways to the non-affected side [32] - circumduction movement.  This swing phase is therefore largely performed by the non-affected side, and improvement must come from creating more core stability and thus more cohesion in the front diagonal.

For the people in photos 3 and 4, walking in that way with some load can turn into a training of the muscle patterns and, for example, increase the stride length while walking. But the compensation strategy in photos 5, 6, and 7 will only give the person more confidence, but will not train/improve the system. The danger is even present that the possibilities that still exist due to double innervation will “disappear” due to disuse and that, due to high muscle tone for a long time, the muscle itself shortens through loss of sarcomeres, etc. [34,35,36,37].  Walking in this way is therefore not training at a better and more stable walking pattern!

Training. Very often this is explained as a form of relearning, and the well-known forms of relearning (explicit and implicit) come into play. However, the principle is to regain control of walking from the combination of the extra support that is offered and the remaining capacity that is still present. In the beginning, this will involve searching for the best solution (a form of explicit/implicit learning), but if the combination is correct, it will become automatic. Photos 5, 6, and 7 show such an image, and “strangely” most of the attention goes to the aids on the non-affected hand/arm and affected leg. Furthermore, we see that the distance and speed increase until a ceiling is quickly reached in this group, and the “training” brings minimal change there. And actually, it is simple, because to improve walking the muscle patterns need to be activated and improved in terms of coordination and power. [38,39,40]

So, in order for the training to be successful, it must be targeted (task-specific [41,42,43]) and there must be an increase in coordination and power that is also transferred to the running pattern. To achieve this, training must be done to induce muscle fatigue (the stimulus for coordination/power improvement) and its application with many repetitions with variation in the different forms of running. Therefore, we need to determine what the 1 R.M. is of the muscle pattern (for example, stance phase) and we know that 75% of that with at least 10 repetitions, three times in a row, and at least twice a week should lead to an improvement in coordination and power [44,45,46,47]. 

The integration of this skill is achieved by task-specific training, preferably with resistance so that the energy demand is higher and thus muscle fatigue occurs. In individuals with a stroke, there is always the question of whether the damaged brain system is capable of providing this coordination, and precisely because of the double innervation during walking, it is possible, but it does require targeted training. That which is not used must, if possible, be activated, and that which actually inhibits selectivity (pathological synergies and pathological tone) must be controlled because the activation of the posterior diagonal is at “risk” and therefore inhibited.

Photo 8, 9 and Picture 1: Training form in lying position of the posterior diagonal starts in the leg/foot that is on the bench. Photo 9 shows the 1 R.M. [47] determined and the weight on the free leg ensures activation and concentric contraction in the other leg and posterior diagonal. It is about the reaction-activation of the affected side and in photo 8 we see without resistance that there is no control in the hip and a plantar flexion in the ankle. And Picture 1 shows that the lift of the buttocks cannot be maintained.

No activation means no concentric contraction of the muscles around the hip of the leg on the bench and therefore no control. Often caused by the dominance of a pathological synergy with a pathological tone. We see plantar flexion in the ankle but also sliding of the leg in extension and 'collapsing' of the tentacle (Picture 1) as signs of the dominance of this pathological synergy/tone. Specifically, this means that this training will not work, because there is no activation.

Picture 2:  Start of an activation in the position of the musculature around the hip, as the starting point of the posterior diagonal towards the opposite shoulder. It is very important that the degrees of freedom under the hip, in the knee, and foot are limited. In the knee, this is done with a posterior splint and in the foot, for example, with a bandaging technique, because this now makes it possible to properly put weight on the affected leg. To keep balance completely under control, we use an adjustable height mobile bench, which provides a large support surface.

The support -for- will mainly be on the non-affected elbow, so the activation will be optimized through hands-on technique. The height -picture 2- of the bench is partly determined by that activation, and now it will also occur earlier because the extensors are under stretch. Essential for that activation is a concentric reaction of the hip muscles, especially extensors/abductors, and for that a good alignment of the diagonals is needed and a lift of the other leg. And start by lifting and moving backward against resistance; this will often provide activation but also immediately give a first impression of 1.R.M. of the muscle pattern that ensures good weight-bearing on the affected leg.

Walking backwards is then the next step, important because with a splint a backward swing phase is easier and because it often provides quicker extension of the trunk. But certainly, also sideways along a bench is a perfect activation of the lower part of the posterior diagonal, and keep testing with hands-on to ensure that the activation takes place, but also a test of the 1R.M. by providing resistance to the swinging leg (non-affected), and we can set up training according to the training rules. Walking forward is possible over a long distance with a mobile bench, and the muscle patterns can be trained optimally against resistance. A 1 R.M. can be determined, pushing a bench can be such a load, so a training program can be realized, which improves the coordination and power of the muscle pattern. Pushing provides activation of the anterior and posterior diagonals (core stability [50]) and therefore also affects the swing phase as well as the stance phase, even with proper extension of the calf. It is important that the activation of the hip extensors remains optimal and that a progression is achieved through the task-specific resistance training program. From the beginning, focus must therefore be placed on the activation of the hip extensors, which must always be monitored optimally, because as soon as the compensation / adaptation walking technique is applied daily, it will require much more training to improve it. Stabilization of the knee and foot is necessary and thus directly affects the way/quality of walking; the posterior splint directly affects the swing phase, but an AFO (Ankle Foot Orthosis [48,49]) can also have an influence. Unfortunately, with the standard carbon AFO, where no movement is allowed, we see that in some people this leads from the beginning to walking with an apparent equinus and, with prolonged use, to a shortening of the calf musculature [25,26].

Picture 3 and Photo 9: Picture 3 gives an image of how to secure a foot in the ankle.  Photo 9 gives an image of the effect of a carbon AFO in the shoe and the limitation that arises at the end of the stance phase and the compensations that this causes.

The posture that the person displays gives a signal that they want to move the unaffected foot forward but are being limited. Yet, they try to reach their “old” step length and we see: knee hyperextension, outward rotation of the affected hip, or better, a counter-rotation on the other side, and trunk reaction. This means that a poor walking pattern is being “imposed,” but also that the calf muscle is not, with every step, being brought to length, with the consequence that this can lead to calf muscle shortening due to loss of sarcomeres. [51,52] 

Extra training, besides gait training with hands-on correction, is necessary to improve coordination and power to such an extent that they can be used in that walking pattern. That means the intensity of the training must be optimal, but also that walking throughout the day should not differ too much, because otherwise integration will only become more difficult. This also means that other caregivers must have the knowledge and skills to properly guide this. For example, to be able to use what has been trained, it remains essential that the end of the rear diagonal ends in the correct position in the hip joint, and walking with an aid to the side creates a risk that this will not be achieved. Especially at the beginning of rehabilitation, this can quickly occur due to fatigue, and it is important to include this element in the training.

Photo 10 and 11:  Training of the stance phase through task-specific resistance therapy. By applying resistance against the swinging leg, you can determine 1 R.M. and thus set up training according to the rules, and an improvement in coordination and power can be achieved.

By consistently incorporating this section into the treatment, there will be an improvement in the ability to enhance the stance phase, and the activation of the hip musculature will become easier, but this does require that the training rules are also followed. So per session 3 sets of 10 repetitions at 75% of 1 R.M. [41,42,43], leading to muscle fatigue, and twice in a row. And at least twice a week, but preferably three times a week, with an increase in resistance to consistently reach muscle fatigue. The variation of the resistance is very large because a small change in direction already produces a different muscle response. Important (Photo 12) keep checking whether the activation of the muscles is optimal because leaning on a chair back can always create a diagonal that does not optimally engage the hip joint. Also realize that the front diagonal is involved because resistance is applied against the swinging leg, and it is therefore important that the opposite shoulder moves forward (protraction), which also partially determines the choice of how much percentage of 1 R.M. can be given as resistance and thus also the number of repetitions. This is necessary to achieve the training effect but also to prevent that, due to too high resistance, the person has to rely on pathological synergies with pathological tone, because then the activation of the hip muscles is inhibited.

Picture 4: Training of the hip muscles is essential for the walking pattern after a stroke and certainly has potential. Due to the bilateral innervation from the brain, the trunk and the major joints are controlled “doubly,” and of course, this is reduced after an injury. What applies to starting a standing and/or walking training also applies to training and activation of the main muscle patterns, which means that where there is no control, we must create it. Picture 4 shows these in the form of a posterior splint that secures the knee, a bandage around the ankle, and support on a table to the side and a therapist in front and on the side. Now it is possible to provide resistance and determine 1 R.M. and carry out the training program, whereby the variation can be very large for the entire hip region.

It is remarkable that the activation of the hip region/lower trunk as part of the posterior diagonal works best when someone is optimally standing on the affected leg.

Photo 12: Weight on one leg requires an adjustment in the lower trunk. In other words, the diagonal must fall in or preferably just beyond the hip joint. This requires not only selectivity but also a great perceptual ability, because this is and remains a balance moment, yet still within limits.  Especially in people after a stroke, this perceptual control is often severely impaired, and one of the reasons that a movement is made or dared sideways, which means that the activation of the muscle patterns is lost with all the associated consequences.

At the moment that this activation does not occur due to perceptual and/or motor reasons, it cannot be expected that it will be incorporated despite training. The only solution is to ensure that the situation changes (becomes safe) and that this activation does occur, because then there will also be training/stimulation of the input (perception) alone but also through the muscle contractions. [53,54] And then it is important to create a situation where someone can handle that, and picture 4 provides a fine solution. By stabilizing the knee, many people will dare to stand on the leg and can therefore train their lower trunk muscle patterns but also use them while walking, although the swing phase is less graceful. The opposite, trying to keep all elements under control with too few fixed points, often results in a situation where the activation of the muscle patterns in the affected lower trunk does not come across at all and is therefore never incorporated into the walking pattern. In that case, stabilizing the knee and foot so that this activation can take place and be trained optimally is the best option to further improve walking.

Photo 13, 14, 15 and 16: The search of a man affected by a stroke for a safe way to walk with a cane. This is practiced intensively under supervision, but he does not seem to like to evoke and train an activation of the hip muscle patterns. The placement of the cane and the way he holds it indicate that he is trying everything possible to activate the diagonal towards the affected leg through the upper body. Of course, there is a double problem here: due to perception, there is no optimal control over the affected leg and therefore more stabilization needs to be provided, which must also be 'felt' because then activation can occur.

The red line indicates the sideways movement and the green line the rear diagonal, and then it is fairly far in both compared to normal.

Photo 17, 18, 19, 20 and 21: At the moment the gentleman tries to practice his walking pattern without an extra aid on the unaffected side, it becomes clear that his perception of the affected side is such that he can hardly dare/can hardly fully load the affected side. The consequence on the affected side is that activation of the hip muscles cannot take place, and photo 21 shows an increase in retraction in the affected shoulder, which affects the selectivity of the entire trunk. The photos also show that a lot is being demanded, and that the moment the unaffected foot lifts, this leads to lower activation and higher pathological tone.

Photo 22, 23, 24 and 25: Through an AFO with the ability to move at the ankle, it becomes clear that not only a stabilization of the ankle is achieved, but also of the knee. However, it is still insufficient when he walks to include the activation of the hip extensors and thus to incorporate the training of these muscle patterns into his walking pattern. Training of this muscle patterns should therefore be incorporated into the walking pattern, which means always creating optimal stabilization but also training other ways of walking with support, and not choosing lateral support too quickly as this often 'inhibits' activation.

Other walking training

The additional training of the lower torso to enable optimal movement over the affected hip coordinatively and with sufficient power, but that must be incorporated back into walking.  At the moment when walking with a lot of lateral support, incorporation will be almost impossible, so it is important not to move too quickly to that form. Especially not if this person is increasingly able to walk longer distances and is better off than choosing another 'walking support' and additional stabilization so that the acquired abilities are also used.  Often it is also still possible to provide an extra task-specific resistance.

Photo 26, 27 and 28: Walking behind a high chair backrest. Photo 26 and 27 show that now the diagonal runs perfectly and this can be seen in a clearly improved walking pattern. But in photo 28, the twisting of the hip is still present.

No matter how severe the damage in the brain is, the double innervation provides an opportunity to mainly improve the movement over the affected hip, but that requires training of these muscle patterns and finding a solution to incorporate it into the walking pattern. It is precisely this incorporation that is essential, because if this does not succeed, what has been achieved through training will disappear again.

Recovery of walking after a stroke actually has a reasonably good prognosis due to the double innervation of the trunk and large joints [55]. The process these individuals have to start is a search for the best control and at the same time the best and safest way to get somewhere else. This therefore does not mean 'learning' but searching through adaptation and compensation to find that stabilization, and unfortunately, aids are often used, which increase the support base and stabilize, enable loading, and control balance.  Practicing a lot and regularly is then the therapy strategy to automate this adaptability. This is not training according to the training rules and certainly not an activation of the hip extensors of the affected hip; the opposite will happen because these muscle patterns will deteriorate faster in terms of strength and coordination [19]. It is precisely a treatment aimed at activating the hip muscles by optimally stabilizing the affected leg, and not seeking lateral support base enlargement, that will trigger this activation and, through proper training, lead to a positive change in the walking pattern with possibilities for improving balance as well. This training is actually best done in water [56], both the task-specific resistance training of the muscle patterns and the incorporation of these possibilities into the walking pattern.