Risk factors for cerebral palsy

While we don't know what exactly causes cerebral palsy, several risk factors have been identified. Bear in mind, however, that cerebral palsy is an incredibly heterogenous disorder, so any given case of cerebral palsy could have one, some or all of these risk factors, or maybe even some novel risk factors that we haven't identified yet.

Hypoxaemia

For a long time, it was thought that hypoxaemia at the time of birth was a major risk factor for cerebral palsy. However, further studies have found that hypoxaemia is implicated in only around 8-10% of cases.

Chorioamnionitis

Chorioamnionitis, as the name suggests, is inflammation of the amnion and chorion (fetal tissues). It is usually caused by an ascending bacterial infection. Chorioamnionitis is associated with prolonged labour and an increase in the risk of cerebral palsy.

Periventricular leukomalacia (PVL)

Periventricular leukomalacia (PVL) is a cerebral lesion characterised by foci of necrosis in the white matter near the lateral ventricles. The presence of brain lesions in the white matter, which can be detected by sonography and appear prior to the onset of PVL, is one of the most important identifiable risk factors for cerebral palsy. The risk of PVL, in turn, may be increased by factors such as prematurity, asphyxia, respiratory distress and infection. For example, PVL is more likely to be seen in neonates with documented sepsis, purulent amniotic fluid, chorioamnionitis or high concentrations of IL-6. There is evidence to suggest that PVL can be initiated before birth, so prevention of cerebral palsy may need to start in utero.

Funisitis

Funisitis, which may be caused by a fetal inflammatory response, is inflammation of the connective tissue of the umbilical cord. It is typically preceded by vasculitis of the umbilical artery and/or veins. Funisitis may be caused by chorioamnionitis and is also a risk factor for cerebral palsy.

Prematurity

Premature birth, along with PVL, is one of the leading risk factors for cerebral palsy. It is estimated that over 25% of preterm deliveries are associated with subclinical intrauterine infection: the fetal (rather than maternal) systemic inflammatory response may contribute to the early onset of labour and an increased risk of cerebral palsy. One proposed mechanism is that intrauterine infection increases levels of IL-6 and other cytokines, leading to PVL and preterm labour. TNF (tumour necrosis factor) may also contribute to PVL (and thus increase the risk of CP) in the following ways:

1. Induces fetal hypotension and brain ischaemia
2. Stimulates production of tissue factors that can contirbute to the coagulation necrosis of white matter
3. Induces the release of platelet-activating factor, which damages brain cells by acting as a "membrane detergent"
4. Directly toxic to oligodendrocytes, which make up the myelin sheath of CNS neurons

Summary of the inflammatory response

I'll wrap up this post with a summary of the proposed mechanism by which an inflammatory response may contribute to cerebral palsy. Firstly, when microorganisms gain access to the fetus, mononuclear cells produce cytokines such as IL-1 and TNF, increasing blood-brain barrier permeability. Increased BBB permeability allows microbial products (as well as cytokines) to enter the brain. Microbial products then stimulate fetal microglia (basically the CNS equivalent of macrophages) to produce IL-1 and TNF. These cytokines increase astrocyte proliferation, as well as further increase TNF levels, leading to oligodendrocyte damage.

It's important to note, however, that none of this is the be-all and end-all. Most cases of intrauterine infection do not result in cerebral palsy. Yup, unfortunately we still have a lot to learn.

References

Yoon, BH, Park, C-W, Chaiworapongsa, T 2003, 'Intrauterine infection and the development of cerebral palsy', BJOG: An International Journal of Obstetrics and Gynaecology, vol. 110, supplement 20, pp. 124-127

Rehabilitation and the brain

I'm currently waiting on an appointment with a librarian to discuss and refine my systematic review, so for now it's more reading and consolidating, I guess...

Effects of rehabilitation on the brain

You've probably heard the old axiom: "if you don't use it, you lose it." In unilateral cerebral palsy (i.e. cerebral palsy only affecting one side of the body), there are delays in motor development on the affected side. If the limbs on that side are not used, they can become deconditioned, and development of cortical representation of that side of the body can be likewise suppressed. In turn, the affected limb becomes more limited in function, thus creating a vicious cycle of sorts.

Thankfully, there is some good news: motor control of the affected limb can be maintained. Maintenance may occur either via using the spared tissue of the affected hemisphere (a process known as "ipsilesional reorganisation") or via withdrawal of crossing fibres from the affected hemisphere and survival of fast-conducting ipsilateral motor projections (a process known as "contralesional reorganisation"). The type of reorganisation may have implications for treatment: subjects with ipsilesional reorganisation tend to experience an increase in motor cortex excitability following treatment, whereas subjects with contralesional reorganisation tend to experience a decrease in motor cortex excitability.

Studies involving combinations of constraint-induced movement therapy (CIMT), neurodevelopmental treatment (NDT), occupational therapy (OT), intensive motor training and/or training camps have shown that the brain can undergo reorganisation following such rehabilitative activities. In the affected hemisphere, the primary motor cortex and/or primary somatosensory cortex may undergo greater activation during active motor tasks (evidence is less consistent for passive motor tasks). However, these effects only take place if such tasks are done with the affected hand. Other functional effects also take place in the affected hemisphere: for instance, there is an increase in somatosensory evoked magnetic fields (SEFs) elicited by tactile stimulation. In patients with ipsilesional reorganisation (but not contralateral reorganisation), SEF latency may decrease and motor evoked potentials (MEPs) may increase in the primary motor cortex. Aside from functional changes, there are also structural changes in the affected hemisphere: the primary motor cortex, primary somatosensory cortex and hippocampus have all been found to increase as a result of rehabilitation.

The unaffected hemisphere also undergoes some changes. While there are no clear functional changes at a population level (some changes have been seen at a single-subject level), there are structural changes: both the primary motor cortex and hippocampus increase in volume.

Important factors in neurorehabilitation

Nielsen et al. (2015) identified many important factors in neurorehabilitation programs. They are as follows:

Active patient participation

It is important that patients participate actively in treatment, as sensory feedback via passive manipulation is not the same as via active manipulation. When muscles are contracted actively, gamma motor neurons are activated, causing greater deviations in the activity of muscle spindle afferents as opposed to passive movement. Furthermore, active contraction increases the load on the muscle tendon, which causes activation of Golgi tendon organ afferents. Joint afferents may result in different types of feedback depending on whether movement is active or passive.

I'm not done yet! During active manipulation, spinal motoneurones can influence spinal neural networks via Renshaw cell inhibition. Active manipulation is also a factor in influencing presynaptic inhibition of sensory input to spinal networks. Perception of sensory stimuli tends to be reduced when performing an active movement at the same time. (For example, when you move your arm, unless you are focusing on it, you generally don't tend to notice that your arm is moving. My understanding is that's because your brain is expecting that arm to move, so it blocks out a degree of sensory input pertaining to movement.)

What about those who are paralysed? There is hope that mental training and imagination may activate the same areas of the brain as voluntary movements. Current studies, however, are unconvincing. Mental training tends to be more effective in people who practise such techniques regularly, such as elite athletes. Other alternatives are virtual reality therapy and mirror therapy.

Physical aids only when necessary

Nielsen et al. recommend against the use of physical aids unless absolutely necessary, as a) time spent using a physical aid (e.g. a wheelchair) takes away from opportunities to practice a skill (e.g. walking), and b) external sensory feedback (e.g. training in a robotic device) may disrupt learning. There are, however, newer devices that allow a patient to contribute to the movement to an ever-increasing extent.

Challenges that support learning

It is important to keep the patient challenged and learning new things. If a patient learns something once, they get a short-lived increased representation of muscles in the corresponding brain areas. If training is repeated, they gain more long-lasting changes in cortical representation. However, if they keep repeating the same thing over and over, performance increases by a lesser amount each time, and cortical representation may even decrease. Cortical expansion only occurs when difficulty increases, so it is important to keep up the challenge. It might be helpful to have goals in mind to work towards.

Patient responsibility

It takes many hours to learn new skills, so it is important that patients have ways to train at home.

Every day, as long as possible

One study showed that no plastic changes occurred after training for 15 minutes, so blocks of 20 minutes or more may be necessary. On the other hand, another study suggested that multiple short bouts might be just as effective as one long bout.

Motivation and reward

When dopamine is released by the "reward system," there is increased consolidation of the motor program in the motor cortex and basal ganglia. There are also increased influences on the prefrontal areas, which facilitate the decision to perform similar behaviours in order to get that sweet, sweet dopamine. The effect of rewards and feedback are most optimal if given immediately.

Optimise acquisition and retention

Acquisition and retention can be optimised by optimising the practice structure. In this area, there is a bit of a trade-off between increased performance in the practice session (which can lead to increased motivation) and increased performance in the long-term. Practising a single skill during constant conditions or practising multiple skills in a "blocked" format (i.e. practise skill A and only skill A for x minutes, and then skill B and only skill B for y minutes) can increase performance short-term, as well as motivation. On the other hand, varying the conditions and using random or interleaved practise is better for long-term retention.

Consolidation

Failure to consolidate a new skill is most likely if the patient learns another task or is subjected to competing external stimuli (i.e. stimuli that activate the same neural circuits) within 3-4 hours of learning. It has been suggested that it may help to schedule the timing of sessions so that patients can sleep afterwards. This way, patients won't be affected by external stimuli during the critical 3-4 hour period. Furthermore, certain phases of sleep may influence consolidation of new skills.

Dopamine, as well as being important in motivation, is also essential for the late phase of long-term potentiation. It is related to consolidation and structural network changes. As such, maybe it might be important to increase dopamine levels, through methods such as aerobic exercise towards the end of a session. (Amphetamines also increase levels of monoamines such as dopamine, thus increasing consolidation, but please don't try this at home.)

Focus on paresis (muscular weakness), not spasticity

Many interventions focus on spasticity (muscle stiffness), but this focus may be misplaced. The abnormal stretch reflexes that are the hallmark of spasticity only occur in the resting state, not during movement. Also, it has been suggested that spasticity may be an adaptive change aimed at maintaining functional output despite diminished descending drive. Nielsen et al. suggest teaching patients how to optimally integrate sensory feedback, rather than focus on eliminating spasticity.

Other influences on learning and memory

Since the brain lives off glucose, increasing sugar intake prior to learning may be helpful. (Of course, you shouldn't overdo it, and you need to be careful if the patient has diabetes or some other dietary requirement.) Another supplement that might be important is DHA (docosahexaenoic acid), which is a component of the neural membrane. Yet another important component is exercise.

Individualisation

We're all individuals, so rehabilitation needs to be individualised :)

Neurorehabilitation still has effects late in life

Old brains still maintain considerable plastic potential: motor learning and consolidation are only slightly slower as compared to younger brains. Therefore, even older patients can stand to benefit from neurorehabilitation. In a similar vein, while the 3-4 months following an injury is the optimal period for rehabilitation, rehabilitation can still occur afterwards. The main take-home point from this is that patients can't use "I'm too old"/"It's been too long, there's no hope for me any more" as excuses for not doing their exercises :D

Virtual reality training

As my project will (hopefully) incorporate elements of virtual reality, I decided to have a look at some virtual reality stuff as well. Virtual reality can provide rich sensory environments for mass practice of skills. The ability to make simple alterations in graphics and sound effects can influence attention levels, which is especially important in rehabilitation of children.

One of the easiest components to implement in virtual reality is vision, so let's start there. Many motor, premotor and parietal neurons are modulated by visual information, and movement errors in the visual domain can influence motor control areas during motor learning. Interacting with a virtual representation of hands can recruit the angular gyrus, precuneus and extrastriate body area, which are all regions of the brain involved in attribution of agency (subjective awareness of initiating, executing and controlling one's own actions). Furthermore, when actions are observed, motor evoked potentials (MEPs) may increase in magnitude, and there may also be influences in corticocortical interactions in the motor and premotor areas. Visually-stimulated motion also has an effect on postural responses.

A component of virtual reality that is perhaps more difficult to implement is haptic (tactile) feedback. Tactile feedback can assist with advanced skill learning, and forces that augment errors are more effective in teaching desired movements.

Virtual reality treatments have been found to alter brain activation in at least one child with cerebral palsy. Before a virtual reality treatment, the child had predominantly bilateral activation of sensorimotor cortices. After the treatment, the bilateral activation disappeared, leaving only contralateral activation. These changes were associated with an increased ability of the child to perform reaching, dressing and self-feeding tasks.

References

Adamovich, S 2009, 'Sensorimotor training in virtual reality: A review', NeuroRehabilitation, vol. 25, no. 1, pp. 29-44

Inguaggiato, E, Sgandurra, G, Perazza, S, Guzzetta, A, Cioni, G, Sale, A 2013, 'Brain reorganization following intervention in children with congenital hemiplegia: A systematic review', Neural Plasticity, vol. 2013.

Nielsen, JB, Willerslev-Olsen, M, Christiansen, L, Lundbye-Jensen, J, Lorentzen, JR 2015, 'Science-based neurorehabilitation: Recommendations for neurorehabilitation from basic science', Journal of Motor Behaviour, vol. 47, no. 1, pp. 7-17

Motor Control and Motivation

Since my systematic review will be looking at how gaming and/or virtual reality impact motor control and motivation, I did a bit of reading on both motor control and motivation today. Hopefully the stuff below is actually accurate and is not just me misinterpreting the articles I've read. (As with previous posts, I've put references at the bottom so you can read the original stuff if you don't trust me.)

Motor Control

Unlike robots, which send the same commands to their motors, human motor commands are highly variable and context-dependent. For example, we might need to adjust to factors such as fatigue, pain, carrying heavy objects, and so on. One model suggests that we have two systems of information for estimating our body state and adjusting accordingly: firstly, our brain can predict what should happen, and secondly, our sensory system reports what is actually happening. These two systems may be integrated and their relative importance optimally weighted in order to help us predict our body state and improve our motor control.

But why have two systems? you may ask. What's wrong with just having one and sticking to it? The long and short of it is that it's a trade-off between speed and accuracy. Sensory feedback is important because it tells us what is actually happening. However, there are delays inherent in feedback, which can lead to instability (see here). Using our brain to predict sensory feedback can mostly eliminate such instability. The cerebellum can predict the state of our limbs from the history of motor commands. It can also use an efferent copy of a motor command to predict its consequences and correct said motor command, if necessary.

In another hypothesis of motor adaptation, reflex pathways can in themselves act as a "teaching signal" for the brain. That is, to my understanding, our brain can "learn" the corrected response. However, experiments suggest that motor adaptation is due to error signals (visuomotor cues etc.) rather than due to error corrections (i.e. reflex responses).

Motor Rehabilitation

The motor control systems described above and the resulting "predictive control" are important in adapting our movements to make them seamless and efficient. However, in cerebral palsy, there are deficits in movement execution, movement representation and movement planning. Such deficits may result in poor motor imagery (i.e. the ability to imagine the movements that you need to carry out), as the neural networks for motor imagery and predictive control have been found to overlap.

In motor rehabilitation, it is important that patients can use feedback to compare executed versus the intended actions. If the ability to do so is compromised, even sheer repetition of skills may not result in any improvement. In cerebral palsy, sensory information may be compromised, which in turn leads to issues in implementation of error correction and predictive control. As such, motor rehabilitation should take into account methods of augmented (multisensory extrinsic) feedback and techniques that cue attentional focus.

Augmented feedback provides feedback above and beyond naturally-occurring intrinsic feedback. Augmented feedback can include knowledge of results (information about the outcome of a movement, for example percentage of successes), knowledge of performance (information about the manner in which the movement was performed and its form) and concurrent augmented feedback (real-time feedback that may be visual, kinaesthetic or auditory). Concurrent augmented feedback can aid in the development of coordination, leading to enhanced stability.

In attentional training, external cues encourage the patient to focus on the effects of the movement, rather than the movement itself. Such an external focus allows the patient to enlist rapid control processes, including the ability to implement adjustments. An internal focus, on the other hand, might encourage a focus on the self and on self-evaluation, which might interfere with the unconscious flow of skilled performance. Attentional training has been found to increase performance on retention and transfer tasks, as actions are most efficient when planned according to intended outcomes.

Motivation

Tatla et al. (2013) defined motivation as "an energy and drive function that causes an individual to move toward satisfying specific needs and general goals in a persistent manner." It is thought to be a critical modulator of functional plasticity, leading to improved motor and functional outcomes. A systematic review by Tatla et al. (2013) found that combining a motivating intervention with therapy instruction resulted in a greater level of biofeedback as compared to therapy alone. The type of motivating intervention also mattered: virtual reality resulted in more biofeedback than watching a DVD. If games were used as a motivator, the type of game was also important. However, some studies found a possible decline in intensity and initiative to play over time.

Another interesting point that was brought up in the Tatla et al. (2013) systematic review is that children with cerebral palsy have lower levels of motivation as compared to their typically developing peers. Important factors that correlate with motivation include self-efficacy and competence. Improvements in motor ability, self-care, communication and socialisation may help to increase motivation in this population.

References

Shadmehr, R, Smith, MA, Krakauer, JW 2010, 'Error Correction, Sensory Prediction, and Adaptation in Motor Control', Annual Review of Neuroscience, vol. 33, pp. 89-108.

Tatla, SK, Sauve, K, Virji-Babul, N, Holsti, L, Butler, C, Loos, HFM 2013, 'Evidence for outcomes of motivational rehabilitation interventions for children and adolescents with cerebral palsy: an American Academy for Cerebral Palsy and Developmental Medicine systematic review', Developmental Medicine and Child Neurology, vol. 55, no. 7, pp. 593-601.

Wilson, Peter 2014, 'Developmental cognitive neuroscience perspective on motor rehabilitation: The case for virtual reality-augmented therapy', International Journal of Child Health and Human Development, vol. 7, no. 4, pp. 341-348.

Current Treatments for Cerebral Palsy

Yesterday I had a look through some papers to figure out what treatments are currently available for cerebral palsy. I found that there were a LOT of treatments, but unfortunately we still have a long way to go with regards to efficacy. Here's a list of treatments that have been used on patients with CP:

• Physiotherapy: focuses on gross motor skills, functional mobility, etc.
• Occupational therapy: daily skills like feeding, dressing, etc. Can enhance the outcomes following botulinum toxin injections, which I'll talk about later.
• Neurodevelopmental treatment (NDT): treatment that aims to normalise muscle tone and inhibit abnormal reflexes.
• Conductive education (CE): an educational and task-oriented approach given to groups of children.
• Therapeutic exercises: includes passive stretching, static weight-bearing exercises, strength training and fitness training.
• Electrical stimulation
• Constraint-induced therapy: restraining the less affected arm. Seems promising.
• Orthoses (external devices to modify structural and functional characteristics of the musculoskeletal system): seem to have short-term effects, but the long-term effects are still unclear.
• Oral medications: include benzodiazepines, baclofen, sodium dantrolene, tizanidine, alpha2-adrenergic agonists, gabapentin and tigabine. Many act via decreased excitation via glutamate, increased inhibition via GABA, or both. The most useful anti-spasticity agent is diazepam (Valium, a benzodiazepine), but it may cause drowsiness. Baclofen (a structural analogue of GABA) can help in spasticity related to spinal cord dysfunction. Sodium dantrolene, the only oral medication that affects the muscles and not the brain (if I remember correctly), is infrequently used due to liver toxicity.
• Intramuscular medications: generally work by causing neuromuscular blockade. Intramuscular phenol and alcohol are neurolytic (block nerves by injuring them). Botulinum toxin type A is a frequently used chemodenervation agent that prevents acetylcholine release at the neuromuscular junction. Effects of botulinum toxin last 8-12 weeks, but effects can last for a longer time if combined with another treatment (orthoses, physiotherapy, etc.). Even though botulinum toxin has limited long-term effects, it can reduce the need for complex surgery.
• Intrathecal (into spinal cord) baclofen.
• Hippotherapy (horse riding)
• Hyperbaric oxygen therapy: lacks evidence of efficacy. Furthermore, it is not risk-free: patients have had seizures following hyperbaric oxygen therapy.
• Adeli suit treatment: use of suits that provide resistance to movement).
• Acupuncture

References

Papavasiliou, AS 2009, ‘Management of motor problems in cerebral palsy: A critical update for the clinician’, European Journal of Paediatric Neurology, vol. 13, no. 5, pp. 387-396.

Principles of Serious Games

Honours isn't exactly busy at the moment- I'll be diving into a systematic review soon, but I can't start yet until my team has nutted out our search terms and so forth. For now I've been mainly floundering around reading articles while trying to figure out what to do with myself. One of the articles that my supervisor sent to me a while back was about principles of effective design of "serious games," or games that are played for a primary purpose other than simply entertainment. This will become relevant to me later, as my Honours project will be looking at virtual reality gaming and robotics in the treatment of cerebral palsy.

So what are the main principles of effective gaming, according to the paper that my supervisor sent me? They are:

• Storylines and characters- help to create a sense of immersion, which in turn leads to more positive outcomes.
• Short-, medium- and long-term goals.
• Continuous feedback and rewards- preferably of the type that targets intrinsic motivation. (Avoid too much negative feedback as this may serve to decrease motivation and learning potential. Rehabilitation games are meant to target things that patients aren't good at, so negative feedback might just add to the frustration.)
• Individualised difficulty levels. One way of doing this is by using adaptive progressions, in which the level of difficulty is adapted to the player's in-the-moment game performance. Usually the overall performance level is maintained between 75% and 85% for optimal results.
• Provision of choice- increases motivation and learning.

Addition of multiplayer elements was also suggested as another way to increase motivation as well as social skills (particularly pertinent given that this paper is about serious gaming in autism).

References

Whyte, EM, Smyth, JM & Scherf, KS 2015, ‘Designing serious game interventions for individuals with autism’, Journal of Autism and Developmental Disorders, vol. 45, no. 12, pp. 3820-3831.