Optimization of spasticity reduction using Botulinum toxin type A injections in children with cerebral palsy. Role of motor end plate-targeted injections.

Cerebral palsy (CP) is the most common cause of physical disability in children. It is defined as a disorder of the development of movement and posture that is attributed to a non-progressive disturbance of the developing brain. In many CP patients this brain lesion causes spasticity and t he elicited increased tone leads to contractures and bony malformations. An optimal use of spasticity reduction with Botulinum toxin type A (BTX) injections, started at a young age, can prevent these complications to some extent. While m any clinical studies reported overall good results of this treatment, they also demonstrated considerable variation in outcome. This is partly due to injection variables. BTX blocks neurotransmission by inhibiting the release of Acetylcholineat the motor end plate (MEP). Animal studies already have shown thatinjecting the toxin near the MEP zone increases its paralytic effect. This was, so far, only confirmed in one human study on the biceps brachiimuscle of adults with spastic hemiplegia after acquired brain lesion (Gracies et al, 2009). Besides the lack of strong clinical evidence of theimportance of MEP targeted injections in children with CP, the clinician was confronted with the very limited information on the localization of the MEP-zones in the lower limb muscles. The overall goal of this thesis was to improve the effectiveness of lower limb treatment with intramuscular BTX injections in children with CP, by optimizing the injection location. A thorough literature search -collecting all relevant histological and anatomical studies- provided information on the exact localization of the MEP zone or the terminal nerve ramifications of most of the frequently injected lower limb muscles. After comparing these with clinical practice, it became clear that for many muscles its location was somewhat different than the currently injected areas. In the review article, o ptimal injection sites in relation to external anatomical landmarks were presented . As no informationwas found on the innervation of the psoas muscle, a cadaver dissection study was performed on 24 adult psoas muscles. With stereoscopic microscopic dissection as far as the terminal nerve ramifications, the region of intramuscular nerve endings, corresponding with the MEP zone, was identified. For both the medial hamstrings (semitendinosus, semimembranosus and gracilis muscle) as well as the psoas muscle, there was enough evidence to conclude that current popular injection techniques were not injecting the toxin at a site close to the MEP zone. To explore the clinical importance of injecting these MEP zones in children with CP, both injection techniques ( current versus MEP targeted) were compared for bothmuscle groups through the application of innovative assessments. An instrumented spasticity assessment was used to evaluate the effectof BTX in the medial hamstrings. Biomechanical (position and torque) and electrophysiological signals were measured when applying passive stretches to the medial hamstrings at different velocities. First, the sensitivity of this assessment was studied on nineteen children before and after BTX injections. The biomechanical and e lectrophysiological parameters proved to be adequately sensitive to assess the response to treatment with BTX with an average of 53% reduction in velocity-dependent root mean square electromyography (RMS-EMG) and a 47% reduction in torque. A second methodological study was set up to assess whether parameters obtained from the instrumented spasticity assessment were more sensitive than clinical scales in detecting treatment response and whether they could help explain response variability. Thirty-one children with CP (40 medial hamstring muscle groups) had a clinical and instrumented spasticity assessments of the medial hamstrings before and after BTX injection. It was concluded that the instrumented spasticity assessment showed higher responsiveness than the clinical scales. The amount of RMS-EMG was considered a promising parameter to predict treatment response. Following these methodological studies, a prospective randomized trial was set up, including 34 gracilis muscles which were injected with BTX in 27 children with CP (8.5±2.5y). Seventeen muscles were treated by proximal injections (at 25% of the length of the upper leg) and 17 muscles by MEP targeted injections (half the dosage at 30% and half at 60% of the upper leg). Clinical and instrumented spasticity assessments were performed before and after the injections. The MEP targeted injections showed a significantly better decline in pathological EMG signal compared to the conventional proximal injections, demonstrated by a higher reduction of the normalized RMS-EMG parameter. This difference could not be demonstrated using the clinical scale. It was concluded that BTX injection in the gracilis muscle at the sites with a high concentration of MEPs resulted in improved spasticity reduction in children with CP. Further, we demonstrated that different injection protocols could be compared sensitively and objectively using the instrumented spasticity assessment that integrates biomechanical and electrophysiological measures. The ultimate goal is to optimize motor function and thus to understand the influence of spasticity and tone reduction treatment on functional activities, such as gait. Therefore, a study was set up to search for functional mar ke rs of spasticit y of the gastrocnemius and hamstring muscles during gait. Because spasticity is a velocity dependent feature, it has been suggested that signs of spasticity during gait may be highlighted by increasing the walking velocity. Gait parameters (kinematic, kineti c and E MG pa ra met er s, muscle length and muscle lengthening velocity MLV) of 17 typical developing ( TD) children ( 10.46±2 .36y) and 53 patients dia gnose d with spastic CP (9.8±3.0y ) were collected during a 3D gait analysis at different walking velocities (normal, fast and as fast as possible without running) and compared at twosimilar non-dimensional velocities, estimated by a linear regression model. A number of gastrocnemius and hamstrings related parameters could be considered as functional markers for spasticity, due to significantly different difference scores (between slow and fast walking velocity) between CP and TD. The spastic gastrocnemius muscle, while walking at high velocity, was characterized by a higher ankle angular velocity, planta r fle x ion mom e nt and powe r a bsorption during loading response. Additionally, this muscle demonstrated an increa s e d EMG signal during stance phase. The increa s e d wa lki ng v e loci t y a f fe c ted the spastic ha mst ring s a t the leve l of the hip a nd kne e joi nt s a t mid- stanc e by a delayed max im u m kne e e x tension mom e nt and b y an incr ea sed hip e x tension mom e nt and power generation. The hamstrings also presented with a lower MLV during swin g pha se. To evaluate both injection techniques for the psoas muscle, a quantitative evaluation using muscle volume assessment by digital magnetic resonance imagination (MRI) segmentation was done. The temporary chemical denervation caused by BTX injections leads to muscle atrophy. MRI sensitively identifies these changes inmuscle volume as was confirmed by a good intra-class correlation (0.988) and within-subject coefficient of variation of 3.506% in our study. Inseven spastic diplegic children, the MEP targeting versus a widely usedmore distal injection technique were compared. Five patients received two different injection techniques randomly applied to both psoas musclesand in two patients a bilateral MEP targeting technique was used. MRI images of the psoas were taken before, two months and -in three patients-six months after the injections. The average injection volume two months after the injection (in relation to pre-injection volume) for the nineMEP targeted muscles was 79,5% versus 107.8% for the five distal injected psoas muscles. This difference was statistically significant. In all five asymmetric injected patients, the MEP targeted psoas had an averageof 27% (range 9-37%) larger volume reduction than the more distal injected psoas muscle. This atrophy remained even six months after the treatment. We therefore concluded that i njections in the MEP zone of the muscle, which is the more proximal part of the psoas muscle, caused muscleatrophy -as a demonstration of the effect of the toxin-, in contrary tomore distal injections were this atrophy was not observed. The newly developed assessment tools (the instrumented spasticity assessment and the digital MRI segmentation muscle volume assessment) proved to be reliable and valid to compare different BTX injection protocols. The results from the gracilis and psoas study have shown that BTX injections atthe sites with high MEP concentrations, have an improved efficacy compared to injections more distant from these MEPs. It is therefore reasonable to state that all BTX injections preferably should be given close to the MEP zone(s) of the injected skeletal muscles. The effect on functionof the child with CP when using these more efficient MEP targeted BTX injections will be further explored by studying the effect on the functional spasticity markers during gait. Future studies comparing different dosage and dilution protocols injected at these MEP zones, documented by the sensitive instrumented spasticity assessment and muscle volume measurement, can further improve the treatment efficacy. This can eventually lead to the use of lower dosages thus decreasing economic costs and the risk of side effects.status: publishe