This site is intended for health professionals only

At the heart of general practice since 1960

Managing scoliosis in adolescents

The term scoliosis describes a lateral deformity of the spine in which one side of the

spine appears to grow faster than the other, resulting in a characteristic spinal curvature and rib hump.

We now know that there are many causes (see table 1) of scoliosis and only a few of these patients will ever need operative intervention.


The initial assessment can be performed in the outpatient clinic without need of specialist equipment.

Most scoliotic patients suffer from idiopathic scoliosis. However, in a minority of cases, the spinal curvature is merely part of a congenital syndrome, and it is worth inspecting the child for other signs. The overlying skin should be examined for pits or naevae that might indicate a congenital spinal anomaly. The degree of each of the scoliotic curves – lumbar, thoracic and thoracolumbar – should also be noted.

The first step in assessment is to determine whether the curve is structural (fixed) or postural. Structural scoliosis represents a three-dimensional deformity. One of the most basic ways to do this is to perform an Adam's test.

In Adam's test the patient is asked to bend forwards. Since every rib is attached to a vertebra posteriorly, the rib cage soon develops a deformity of its own, and when the patient leans forward the examiner observes a striking rib hump (see figure 1). Once the initial curve becomes

well established, secondary curves may appear above and below the primary curve producing an S-shaped spine.

In other cases the curve is merely postural. A patient with a leg length discrepancy or a pelvic tilt will inevitably develop a curved spine to maintain an upright position. For this reason it is important to measure leg lengths at the time of presentation. In these patients there is nothing wrong with the actual structure of the spine and the scoliosis is described as being non-structural.

The key observation here is that the curve can be voluntarily corrected by the patient. For example, when a patient with a leg-length discrepancy sits down, his pelvis becomes level with the chair and his spine straightens.

A patient with a non-structural scoliosis does not require referral to a spinal surgeon, although they may well benefit from review in a paediatric orthopaedic clinic.

One of the biggest concerns of parents and physicians is progression. If a child has a ten-degree curve today, how severe a curve will they have in ten years' time?

Of the curves that develop during childhood and adolescence, most will stop progressing at the end of the pubertal growth spurt.

In the assessment of a child with a three-dimensional scoliotic deformity it is therefore important to remember the fourth dimension, that is, time.

Infantile idiopathic scoliosis

This condition affects children between birth and three months. It is a true

three-dimensional scoliosis and due to

the rotatory deformity there is a recognisable rib hump. The curve is typically convexto the left and is associated with plagiocephaly. Boys are affected more frequently than girls and in most cases the condition resolves spontaneously without treatment by the age of two to three years.

Adolescent idiopathic scoliosis

The majority of patients with a curve in their spine present in adolescence. Ordinarily, when we speak of ‘scoliosis' we are usually speaking of adolescent idiopathic scoliosis.

This affects children between the onset of puberty (approximately ten years) and the end of skeletal growth (20 years).

Approximately four per cent of the adolescent population are believed to be affected by idiopathic scoliosis. However, spinal deformity in children is not a reportable condition and so accurate figures are not available for its true prevalence. Very slight curves tend to be ignored by clinicians and a curve of seven degrees or more is regarded as significant.

In that sense the condition is difficult to define since, like short stature, the true prevalence depends in part on the threshold chosen for making the diagnosis.

Although girls and boys are both affected, girls tend to develop more severe curves and the male:female ratio of those receiving treatment is 1:8. The best age to examine a child for screening purposes is around the age of ten to eleven years old. The classical patient is a tall teenage girl with a thoracic curve that is convex to the right.

By definition, the cause of idiopathic scoliosis remains a mystery, but in recent years, new research has begun to offer some insight into the disease. A small proportion of idiopathic patients have obvious organic pathology when investigated by MRI. In addition, the typical adolescent scoliotic patient has other features that point to a possible aetiology.

When equivalent films are taken of the distal radius in normal and scoliotic teenagers, it can be shown that the scoliotic skeleton is about twelve to

18 months retarded relative to the norm. In addition, a taller patient is also more likely to develop scoliosis. These observations point to a multifactorial inherited aetiology for scoliosis. There is certainly no evidence of a single gene defect.

Some doctors believe that scoliosis represents a fundamental loss of body symmetry during development. Teenage girls with scoliosis may have asymmetric breasts and this problem is sometimes so severe as to require plastic surgical intervention.

As the child grows taller the curve tends to become more severe. The curve will not spontaneously resolve. However, once the patient reaches skeletal maturity (in girls this occurs about 24 months after the menarche) the curve will stop progressing. The underlying philosophy of management is therefore to watch and wait.

Those patients who appear to be developing very severe curves should be selected out and considered for surgery. For example, an influential study performed in Iowa1 suggested that thoracic curves of 30 degrees or less will not progress once the growth spurt has ended.

However, some very severe curves will continue to progress at about one degree per year during adult life, even after skeletal maturity. In order to progress, this requires an initial curve greater than

50 degrees.

Even minor scoliotic curves can be distressing for the teenage patient. Some of the more severe ones can be quite alarming and it is tempting to suggest that severe scoliosis ought to lead to cardiovascular and respiratory compromise. However, such crippling complications are rare, but a child afflicted by scoliosis before the age of eight is at considerable risk of impaired pulmonary development, and these children are also believed to be at risk of premature death in adult life.2

A curve of over 30 degrees can be shown to cause significant VQ abnormalities in the adult. These patients often require formal lung function tests before corrective surgery.

Some teenage girls with scoliosis are under the impression that they will be unable to carry children and that any children they have will be badly affected by scoliosis. Broadly speaking this is not true. The risk of a scoliotic mother producing a scoliotic child is less than

1 per cent.

A patient presenting with a relatively minor curve at the end of her adolescent growth spurt can usually be reassured. However, a well-established curve in a younger child is more alarming and such patients need to be followed up regularly to check for progression.

Thoracic curves that are convex to the left often reflect serious underlying pathology and in the 1990s, suspicious cases such as this were investigated by MRI. Nowadays, as MRI becomes more accessible, many spinal surgeons request MRI as a matter of routine in childhood scoliosis.

As mentioned above, scoliosis may be associated with other pathologies. There are recognised warning signs regarding this. Cafe au lait spots point to neurofibromatosis. A small hairy patch on the base of the spine points to spina bifida. A thoracic curve in an adolescent that is convex to the left or a curve in a boy should provoke concern.

Remember that adolescent scoliosis is not normally painful and that children in particular do not normally invent the symptom of back pain.3 The pain of osteoid osteoma is characteristically worse at night and relieved by aspirin. Any evidence of altered neurology or

decompensation requires urgent referral.

Radiological investigations

X-rays of the whole spine are usually taken on a special cassette with the child facing away from the X-ray source to reduce the radiation dose to the breasts. There is widespread concern that many children with scoliosis accumulate very heavy

X-ray packets, which in turn reflects a high dose of radiation.

Some units have attempted to evade this problem by using colour photography of the torso illuminated by transverse beams of lights. Unfortunately, the topographical representation the produces is not a complete substitute for radiographs, and in some units the move to topographical pictures has been abandoned.


Bracing has the advantage that it attempts to treat a serious condition without the potential risks associated with spinal surgery. In recent times this policy has been more enthusiastically adopted in the US than in the UK.

In some cases, the apex of a thoracic curve may be relatively low. For example, when the apex of the curve is at, or below, T8, the child can be treated in an

under-arm brace such as the Boston brace. In contrast, a curve that affects a higher part of the spine requires a more obtrusive brace such as the Milwaukee brace, which actually immobilises the neck.

There is still controversy as to whether the wearing of a brace will make any difference to final outcome in scoliosis. In the course of this debate, one thing is certain: a brace that is issued and not worn is of no use at all. Carefully blinded studies in the UK have suggested that persuading a modern adolescent to wear a brace can be very difficult.

In the first instance it is reasonable to refer a child with scoliosis to a local orthopaedic surgeon. The majority of such referrals will result in reassurance for the patient and their families, but progressive adolescent idiopathic scoliosis has an incidence of about two per 1,000 children and in these patients surgery may be necessary. The DGH consultant will then refer the patient to a regional spinal specialist with an interest in scoliosis.

In October 2001 there were 30 spinal surgeons in the UK and Republic of Ireland undertaking the management of spinal deformity.4 The most appropriate consultant to treat a scoliotic patient would be an active member of the British Scoliosis Society. Some consultants only perform ten operative procedures per year while others perform over a 100.

There is a significant degree of risk associated with spinal surgery of this kind. In particular there is the risk of significant neurological injury sustained at the time of surgery, including paraplegia. This has been estimated to be 0.2 per cent to

0.5 per cent.

For this reason many surgeons have questioned whether corrective surgery should be performed at all given that scoliosis is in some respects a cosmetic deformity. Many units try to avoid the risk of neurological injury by performing continuous EMG studies throughout the procedure, thus providing early warning

of any neurological damage long before the patient can be woken up and assessed clinically.

The procedures themselves are tailored to the individual case. Operations involve either anterior or posterior fusion, with or without instrumentation. Most implants used are made of stainless steel, although titanium is sometimes preferred as this facilitates post-op MRI scanning.

The operative techniques used for this kind of surgery are under continuous improvement and there is much interest in thoraco-laparscopy to perform anterior body fusion.

Once the spine has been fused, the range of movement is significantly reduced. The rib hump will be reduced but not obliterated.


1 Weinstein SL, Zavala DC et al. Idiopathic scoliosis. Long-term follow-up and prognosis in untreated patients. J Bone Joint Surg Am 1981;63A:702–12

2 Siegler D, Zorab PA. The influence of lung volume on gas transfer in scoliosis Br J Dis Chest 1982;76:44–50

3 George H Thompson. Back Pain in Children Chapter 22 Rosemont, Illinois. US:American Academy of Orthopaedic Surgeons;1994

4 The Management of Spinal Deformity in the UK, Guide to Practice

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say