Since the first report of the use of botulinum toxin A (BoNT-A) in the management of children with cerebral palsy (CP) in this journal almost 20 years ago, the drug has been licensed in many countries. It is now widely used and considered by many to be a ‘standard of care’. With more widespread use has grown an appreciation of both its benefits and risks, leading to an urge for caution. Fortunately, serious adverse events and deaths following injection of BoNT-A are rare and preventable so long as risk factors which reside within the child are recognized and appropriate dose and administration guidelines are observed.
However, a more insidious and silent problem may be post-injection muscle atrophy, reported by Williams et al. In contrast to the alarming reports from animal studies, which include substantial reductions in muscle mass and force generation, the results in children with CP are quite different and broadly reassuring. Williams et al. report a study of 15 children, aged 5 to 11 years with spastic diplegia, at Gross Motor Function Classification System levels I and II, who received injections of BoNT-A to the gastrocnemius muscles in both lower limbs, and in some patients to the medial hamstring muscles when clinically indicated. Muscle volumes were computed from magnetic resonance imaging (MRI) and Mimics software at 2 weeks prior to injection and 5 weeks post-injection. Although the volume of the gastrocsoleus muscle group remained unchanged, there was a 5% reduction in gastrocnemius muscle volume and a 4% increase in soleus volume following BoNT-A injection. In addition, significant quadriceps hypertrophy was also found, interestingly only in the group who did not receive injection to the hamstring muscles. The atrophy of the gastrocnemius is much smaller than that reported in rabbit quadriceps and is consistent with the known effects of the toxin. The hypertrophy of the soleus and the quadriceps is a novel finding and may represent a compensatory strategy secondary to muscle and/or nervous system plasticity. Importantly, there was no evidence of impaired function in the injected children.
What does this mean for the practicing clinician? Firstly, with appropriate injection techniques, atrophy of the target muscle is reasonably limited and may to some degree be compensated for by hypertrophy in neighbouring muscles. However, given that spastic diplegia is associated with a 22% smaller medial gastrocnemius muscle volume as part of the natural history, any additional atrophy would be a matter for concern. Furthermore, new evidence has found medial gastrocnemius growth rate to be slower in children with CP compared with typically developing peers, with a further 60% reduction following BoNT-A injection.
Obtaining serial MRI is difficult in younger children because of compliance issues and expense. The authors and the children in the study should be congratulated on completing this demanding protocol. As always, more studies, larger numbers, and longer-term follow-up hold the key to improved understanding of the muscle responses in growing children to injections of this widely used therapy. How much BoNT-A and how frequent is not well established.
In spastic diplegia, the two-joint gastrocnemius is more implicated in gait dysfunction than the one-joint soleus. Preserving soleal strength and moment generating capacity is critical to long-term walking ability and in preventing crouch gait. In older children with fixed contractures, the most common intervention is a gastrocnemius recession in which the soleus is not lengthened and its function is preserved. It is fascinating that the results of this study suggest that similar effects can be achieved by selective injection of the gastrocnemius, that is atrophy of the gastrocnemius and preservation of the soleus. Non-ionizing imaging techniques, including both three-dimensional ultrasound and MRI, hold great promise in the understanding of muscle biomechanics and the response to intervention.[4, 5]
Finally, it is important to consider the broader context of gait management in children with spastic diplegia in the first decade of life. Prior to the introduction of BoNT-A, gastrocsoleus lengthening was often utilized in very young children with spastic diplegia. This resulted in an alarming degree of progressive crouch gait which in many cases resulted in more functional disability and required more complex surgical reconstruction than the original equinus gait problem. With the introduction of BoNT-A injections to the gastrocsoleus, followed by multilevel surgery at an appropriate age, the prevalence of crouch gait has been dramatically reduced in our centre.