Botulinum toxin type A acts by presynaptic blockade of cholinergic receptors at the neuromuscular junction. It is used as an adjunct to standard therapies aimed at improving range of movement and function through the reduction of muscle tone and spasticity. There has been much in the way of investigation into the different facets of BoNT-A administration chiefly dose, duration of effect, repetition and timing of standard therapeutic intervention after injection. The long-term outcome is also under investigation and is understandably of some interest, as the period of clinically useful relaxation appears to be between two and six months following administration [4–6]. The overall aim of BoNT-A injection is to produce a reduction in tone hence allow splinting, casting or therapy for increased muscle length with the hope of improving motor function and therefore reducing, or at least delaying, the need for surgery. NICE guidelines updated in September 2014 list the indications for the use of BoNT-A in children and young people with focal spasticity of the upper limb. These include spasticity impeding fine motor function, compromising care and hygiene, causing pain, impeding tolerance of other treatments such as orthoses or causing cosmetic concerns to the patient [7]. Contraindications include severe muscle weakness, previous adverse reaction or if the patient is receiving aminoglycoside treatment.

The use of BoNT-A in the upper limb in children with cerebral palsy was the subject of a Cochrane review in 2010 [8]. They report the pooled data of seven randomized controlled trials and three unpublished trials aimed at understanding the effectiveness of intramuscular BoNT-A versus BoNT-A and occupational therapy [5, 6, 9–14]. They report a significant decrease in spasticity following BoNT-A administration, however a resultant functional gain that is sustained, was not proven. Moderate evidence exists that BoNT-A alone is not effective compared to placebo or indeed no treatment at all. High-level evidence was presented supporting the use of BoNT-A as an adjunct to concomitant occupational therapy. BoNT-A and occupational therapy was shown to be more effective than occupational therapy alone in reducing impairment, improving activity level and goal achievement based on the chosen functional outcome scoring methods. Interestingly however, no improvement in quality of life or perceived self-competence was found.

Supporting the findings of this review is a further RCT by Ferrari et al. [15] comparing BoNT-A and therapy with placebo and therapy. They found that both groups demonstrated an improvement in function with a significantly better improvement in assisted hand assessment (AHA) in the BoNT-A group. However Rameckers et al. [16] report an RCT looking at the benefit of BoNT-A and intensive therapy over intensive therapy alone. Over six months no significant differences were reported between groups with regard to functional measures. Muscle force decreased following BoNT-A injection but returned during the therapy period. The therapy group displayed significantly higher increase in force and accuracy relative to the group receiving BoNT-A [16]. Van Heest et al [17] also report a lack of functional improvement after BoNT-A injection relative to therapy alone. They again observed a significant reduction in grip strength following BoNT-A injection. In this RCT however the reduction in strength continued for six months after the final injection was administered [28, #11568].

There is no recognized, superior dose regime recommended in the current literature. Doses used in controlled trials to date range from 2 to 12 U/kg body weight per session with a maximum does of 400 units. Forearm muscles have received 0.3–2 U/kg increasing in the upper arm to 4 U/kg body weight per muscle [5]. The commonly employed concentration is between 50 and 100 U/ml [3].

Lowe in 2006 demonstrated better longer-term outcomes with a low volume, higher concentration BoNT-A injection (200 U/ml saline) [10]. However, similar results were not recreated by Kawamura et al. who specifically compared the effects of low dose BoNT-A with a high dose (double strength) BoNT-A on outcome [11]. In this study however, muscle localization was performed by palpation without stimulation or imaging. Further studies using adults also suggests the high-volume, low concentration injections achieve greater neuromuscular blockade, spasticity reduction and greater range of elbow extension than low-volume, high concentration injections [17]. With a paucity of evidence to support the use of higher doses of BoNT-A in the upper limb of patients with CP, currently there is no reason to deviate from those recommendations contained in the paediatric upper limb hypertonicity BoNT-A evidence based guidelines for intervention and aftercare published in 2010 [5].

This consensus report outlines a pragmatic and evidence based approach to calculating the dose of BoNT-A dependent on muscle groups injected and aims of treatment. Due to the small size of forearm muscles that are in much closer proximity than other areas, lower doses with less dilution are recommended if functional gains are the aim [10, 13, 18].

In terms of concentration the smaller the volume injected, the lower the diffusion within and beyond the muscle [19, 20]. If the treatment is to improve function, it is recommended that the calculated dosage be distributed in a small amount of normal saline in the muscle at one to two injection sites, depending on the age and size of the child. At the forearm, a volume exceeding 0.5 ml is likely to diffuse into neighbouring muscles. A preparation of 100 U/ml saline dilution is therefore recommended dependent on the product used. In children with larger muscles or if the aim of intervention is to improve appearance, tolerance of orthoses, or facilitate care, then a higher volume would be appropriate such as 100 units in 2 mls.

The therapeutic effect of BoNT-A lasts approximately 2–6 months. There is a risk of antibody formation with BoNT-A use and in order to minimise this, intervals between injections should exceed 3 months [21, 22]. In practice it seems sensible to prolong intervals between injections as long as clinically justifiable without jeopardising treatment. This approach will help to allow adaptation to reduce tone and avoid excessive weakness through repeated injections.

No study was found investigating the effect of repeat BoNT-A injections against single injections. Olesch et al. performed an RCT into repeat BoNT-A injections and therapy vs therapy alone. They report progressively reduced spasticity and improved parental perception of performance in the BoNT-A group [23]. They included 22 children without fixed contractures. Three BoNT-A injections were administered no more frequent than every 16 weeks. They prescribed 6 weeks of intensive therapy (twice weekly sessions) followed by routine therapy until the next injection. This study was underpowered however and the only significant difference between the groups was seen in areas where the assessors were not blinded.

In a recent RCT, Lidman et al. compared repeat BoNT-A injections and therapy with therapy alone on hand function. Their injection protocol included BoNT-A administered 2 weeks prior to 8 weeks of therapy. Thereafter parents went back to normal therapy with a second injection at 6 months. They report improvements in ROM and goal performance in both groups. The BoNT-A group demonstrated a significantly better AHA. They included all functional levels; however there were clinical differences at baseline that may have affected functional outcomes. This was a small study (20 participants) with no power calculation and only partial blinding [24].

An investigation into multi-sessional intramuscular injections of BoNT-A performed at weeks eight and 20 following initial injection to muscles still exhibiting marked spasticity was performed by Koman et al. [25]. This prospective, double-blind RCT concluded that after these three injections, children receiving BoNT-A displayed a clinically meaningful improvement in function at 26 weeks compared to those who received placebo. Their cohort included children with varying levels of functional impairment some of whom were not surgical candidates. Follow-up was to 26 weeks but grip strength was not assessed.

Using a similar protocol with injections at initiation, 12 and 24 weeks followed by a supervised therapy regime and continued home therapy; Van Heest et al. did not reproduce these findings. They performed a final assessment 6 months after last injection and report no significant improvements in measurements of bodily impairment, activity limitations or participation restrictions in the BoNT-A group or those treated with ongoing therapy alone at 12 months. Furthermore, they report a significant reduction in grip strength in the BoNT-A group up to 6 months following last injection.

There is great variation in therapy regimes following BoNT-A injection in the literature. To date there is little high quality evidence that any is superior. Hoare et al. report an RCT into whether modified constraint-induced movement therapy (mCIMT) leads to superior gains compared with bimanual occupational therapy (BOT) in young children with unilateral CP following BoNT-A injections. They conclude that there is no benefit to mCIMT despite increased intensity of the home programme over standard BOT [2]. Most regimes utilise static splinting between therapy sessions to maintain any improvement in spasticity achieved.

The 2010 Cochrane review reported excessive grip weakness as the most common adverse event. Several studies report the difference in strength reduction in the smaller muscles of the upper limb compared to the larger muscles of the lower limb. BoNT-A use in the forearm can result in a 90 % reduction in the amplitude of the EMG after electrical stimulation of the corresponding nerve compared to 20 % in the gastrocnemius muscle [26, 27]. The impact of BoNT-A on grip strength has been investigated by Van Heest et al. They report that the reduction in grip and pinch is not only present immediately after administration but persists for at least 6 months after last injection [28]. Other less commonly reported complaints include nausea, vomiting, flu like symptoms, coughing, soreness at injection site, respiratory infections and headache.

There are still many unanswered questions when it comes to the use of BoNT-A in children with CP, not least of all the fundamental question as to whether it is beneficial to functional outcome at all. Key areas for future research include: