The orthopaedic side of cerebral palsy
Cerebral palsy is not one condition but a group of many similar conditions, and it therefore affects children in a variety of ways. The defining features are a non-progressive disorder of the brain that was present before the age of 2 years and that then affects muscle strength, control and development. The term cerebral palsy can be used without defining what has caused it, and sometimes the cause is not known. Sometimes it is very mild and hardly noticeable, and sometimes it is quite severe, affecting almost every part of the body. It is the commonest cause of childhood disability.
Because it affects muscle, cerebral palsy is a condition that often brings children to the orthopaedic clinic. By working with the children, the families, carers, physiotherapists and paediatricians, we can usually establish what the treatable problems are and work towards realistic goals. The broad categories of treatable problems are muscle imbalance, muscle tightness, joint deformity and pain. These problems usually overlap.
Muscle imbalance is an almost universal feature of cerebral palsy. If muscles are not balanced they can hinder each other and affect the function of a limb (and other areas such as the tongue and throat). If it is clear how a function is affected, addressing this imbalance surgically can improve it.
Muscles can be tight because they are contracting too much, too easily or at the wrong times (sometimes called spasticity, hypertonia or dystonia) or they can actually be too short (because they have not grown long enough). Tight muscles affect the joints they are crossing and therefore can affect function and cause deformity. They can also affect the way in which the bones and joints develop. Tight muscles can be painful, or they can cause pain because of their effect on the joints.
Joint deformity is therefore a result of muscle tightness. Joint deformity (contracture) can affect joint function when sitting, standing or walking. It can also affect personal care such as dressing and washing. Joint deformity can be painful, for example from a foot deformity, a tight kneecap or displaced hip.
Hip displacement is a common problem for children with cerebral palsy, mostly (but not exclusively) those children who do not have the ability to walk. Muscle imbalances affect the development of this otherwise stable ball-and-socket joint, and gradually the ball disengages from the socket. Most non-walking children will have some hip displacement. It is not painful at first, but pain or discomfort is inevitable if it is not treated. The displacement also makes the hip joint less mobile; this can affect sitting, standing, hygiene and walking. Hip displacement is one of the commonest reasons for a child with cerebral palsy to have surgery.
Orthopaedic surgery for cerebral palsy
Some aspects of cerebral palsy can be helped by orthopaedic surgery (see above). Surgery can address muscle imbalance by transferring a strong muscle to make up for a weak one. Surgery can lengthen or release a muscle that is too short or too tight. Muscle surgery can correct joint deformity, but sometimes it is useful to operate on the joint itself and/or on the bones attached to it. This might mean fusing the joint solid so it no longer causes trouble (for example the wrist or a joint in the foot). A combination of these techniques is often employed. There are a variety of techniques, and they depend on the problem and the anatomy.
Performing surgery is sometimes challenging, but the biggest challenge for any surgeon is knowing what to do and when to do it. This is particularly true for children’s orthopaedics because the conditions we treat change as the child grows. It is important therefore to have a grasp of the “natural history” of the child’s condition – in other words what is likely to happen to the child and his/her various orthopaedic conditions over time without treatment or even regardless of treatment. This is also referred to in some contexts as the prognosis, for example the prognosis for long-term independent walking. Research has taught us quite a lot about the natural history of cerebral palsy, and it is important to pass on what we know to the children, families and carers (and other health professionals) so that we can have an informed discussion.
One of the things that has been quite well researched is how a child’s early motor development can predict that child’s later motor abilities. We sometimes therefore classify children quite early on according to their motor skills – ie their motor function. The Gross Motor Function Classification System (GMFCS) is the most widely used tool for this. There are five levels: GMFCS I means normal walking, whereas GMFCS V means that the child is fully dependent.
Having agreed with all concerned what the natural history is likely to be, it is possible to discuss surgery and how it might alter the natural history, positively and also potentially negatively. Sometimes an operation might improve one aspect at the expense of another; and occasionally a treatment doesn’t work out.
Outcomes and goal setting
How much, and whether, an operation is likely to work is usually hard to quantify. This is because there are not always objective ways of measuring what is trying to be achieved, and the child might have a different expectation from the surgeon, and even his/her family. Ideally, once a treatable problem has been identified, the magnitude of the problem should be measured. Some idea of the likely improvements that can be made is then used to help decision-making, and a clear, achievable goal can be set before surgery. After surgery, and usually a recovery and rehabilitation period, the problem can be measured again to see whether surgery has helped: this measurement is often referred to as the outcome, or the outcome score.
The problem with measuring problems is that the problem we are measuring may not be the problem that the child is experiencing. There is a natural tendency for things to be measured because they are measurable, rather than because they are important. It is inevitable therefore that some measurements are proxies for a real problem, and we have to assume that if the measurement is better the problem is better too. In practice many measurements have been validated against other outcomes to confirm that they are probably measuring the right thing.
The best outcome scores are patient-derived, or at least patient-centred. Having said that, the patient cannot be expected to score every aspect of the treatment, and there are many valuable ways in which natural history and outcome can be quantified without the patient’s opinion. What is important to the patient may not have any long-term relevance to the outcome of the condition being treated. Some patient-reported outcome measures do try to focus on the job in hand, even though they can only ask the patient to pass judgement on matters that require no medical expertise. Pain and waking at night are good examples of outcomes that can be measured; whereas how a patient or family felt about their treatment is not so easy and can be influenced by a large number of factors.
The importance of all this is to illustrate how decision-making for the surgical treatment of cerebral palsy is potentially complex: Everyone has to agree what the treatable problem is, understand the natural history, have a realistic idea of how the natural history might change, and ideally agree on how the outcome is going to be measured. The details of the surgical procedure and its potential complications and risks, the demands of rehabilitation, time off school and the inevitable pain and inconvenience of surgery, also have to be taken into account. The patient, the parents, the carers, the therapists, the paediatricians and the surgeons may all have different ideas and preconceptions that also need to be ironed out. This process is sometimes referred to as “setting realistic goals” and “managing expectations”. It is rarely as simple as “If A=B do C”.
The timing of surgery
As indicated above, conditions change with age. Physiology and psychology also change. The experiences of the child and family change too. What was an expectation at age 5 may seem totally unrealistic at age 13.
It may sometimes be tempting intervene early with an operation, but if the problem recurs with growth it may not be so easy or effective to operate a second or even third time. Other conditions do require treatment almost as soon as the diagnosis is made, and repeat surgery may be part of the treatment plan during the rest of childhood – it helps to have this plan in advance, even if just an outline.
Calf muscle lengthening
Children with cerebral palsy often have short calf muscles. This can cause deformities of the feet. It sometimes present as toe-walking, which in itself is not such a bad thing unless extreme, but it can affect the efficiency of walking, especially if it means the knee(s) cannot go straight. The calf muscles are important for straightening the knee, so although the tip-toe walking may be the most visible problem, the flexion of the knees may be more important biomechanically.
There are 3 calf muscles that operate the ankle (the others operate the toes and foot) and they can be lengthened separately in the mid-calf, or they can be lengthened together by lengthening the Achilles tendon. A cast is applied at the end of the operation and usually kept on for 6 weeks. The aims of surgery are to bring the heel closer to the floor, prevent foot deformity and improve the efficiency and balance of walking. Sometimes an AFO (ankle-foot orthosis) is prescribed afterwards, but this is not universal.
Some children go straight back up on their toes afterwards, despite the longer muscles. Some children will still be unable to straighten their knees – this is because other factors influence knee extension such as weakness of the thigh muscles and tightness of the hamstrings. If the knee won’t straighten, it is hard to bring the heel to the floor, unless the calf muscles are too long. Over-lengthening is a risk with this procedure, and it may be better to have a lot of tightness than a little weakness.
The hamstrings are the muscles that bend the knees, and also extend the hips. However, gravity does a pretty good job of bending the knees without the hamstrings; and the main muscle that extends the hips is the biggest muscle in the body – gluteus maximus. Therefore the hamstrings are often seen as expendible.
If they are tight or contracting at the wrong time, they can prevent the other thigh muscles (the quadriceps, or “quads”) from extending the knees straight. Walking without straightening the knees (known as “crouch gait”) is particularly inefficient: it can be exhausting and even painful – try it! It is harder than walking on tip-toe. Younger children may even prefer to walk on their knees than on their feet if they can’t get their knees straight.
There are therefore often good reasons for lengthening or releasing the hamstring muscles with surgery. Some surgeons use casts or splints (such as gaiters) after this surgery, but it is not my practice to do so. I have reached this conclusion after working with other surgeons and physiotherapists and following children up after hamstring surgery.
Hip displacement is very common for children who do not have the ability to walk. 90% of the most severely affected children (GMFCS V) have displaced hips, i.e. when the ball is more than 30% outside the socket on x-ray. Many of these children will need hip surgery to prevent or treat pain and deformity. Some walking children also need surgery for hip displacement, but this is less common.
Most surgeons rely on a surveillance system to detect children who might need surgery. This is usually coordinated by the community physiotherapist or community paediatrician, and involves a clinical examination and an x-ray of the hips every year or so, depending on the GMFCS level (see above). Cerebral palsy registries in Sweden and Victoria (Australia) have demonstrated that this is an effective way of detecting hip displacement early and reducing the magnitude of surgery that children undergo. A registry has been introduced in Scotland, and a similar registry is being planned in England, pending NHS funding. Without a registry, hip surveillance can be a bit hit and miss, so some children are still presenting quite late to the orthopaedic service.We do now know (from the Victorian registry) that nearly all displaced hips become painful, and that it is more effective to prevent the pain than to treat it once it has started. Surgery is the only treatment proven to correct hip displacement reliably.
It is my recommendation that surgery for hip displacement should, with a few exceptions, always include a femoral osteotomy. This means surgically dividing the upper femur (thigh bone) rotating it in the appropriate planes, and fixing it in its new position with a metal plate and screws. Muscle releases are added as needed. Sometimes, the acetabulum (hip socket) needs to be re-aligned as well. This means gently cutting the pelvic bone and usually inserting a piece of bone from the femur to hold the acetabulum in its new position; occasionally additional fixation is needed with wires or screws. There is usually no need to enter the hip joint itself these days, but severely displaced hips do, in my opinion, sometimes benefit from exploration during the operation in order to inspect the condition of the joint and ensure that obstacles to ball-and-socket alignment are removed.
This looks like a lot of surgery, and it is. It is a major undertaking for the child and family. If hip displacement can be detected early, the magnitude of the surgery can be reduced. Nevertheless it is still a big operation. Many children with cerebral palsy have complex medical conditions that might affect their recovery after surgery. This is why a team of specialists is needed to make sure the experience is as safe, effective and as positive as it can be. This includes paediatricians, physiotherapists, anaesthetists and nurses; and some children spend time on the intensive care unit after surgery. It is important that major surgery, such as hip surgery, is undertaken in a hospital that has these facilities and the expertise to make the recovery as quick and as safe as possible.
Orthopaedic surgery for cerebral palsy used to have a bad reputation, but this was probably because children were referred too late for the surgery to be effective. Ineffective surgery discourages referral to the surgeon, and so the problem continued. These days, orthopaedic surgeons are usually involved much earlier on, often well before surgery is needed.The orthopaedic surgeon is part of a wider multidisciplinary team. By working together, the right goals can be set, and appropriate solutions can be found.
205-209 Great Portland Street
Great Ormond Street
164-178 Cromwell Rd