Congenital Scoliosis

Congenital scoliosis is a curvature of the spine that a baby is born with. It happens because one or more of the small bones that make up the spine, called vertebrae, did not form in the usual way while the baby was still developing in the womb. This page explains, in plain language, what congenital scoliosis is, why it happens, how it is diagnosed and monitored, and the treatments that may be considered. It is written for patients, parents and families, and is based on published medical research and textbooks. It is for general information only and is not advice about any individual child.

Scoliosis simply means a sideways curve of the spine. When doctors measure a curve and it is 10 degrees or more, they call it scoliosis. The word congenital means "present from birth."

So congenital scoliosis is a sideways curve of the spine that is caused by vertebrae (the building blocks of the spine) that did not form normally before birth. These abnormally shaped bones are called congenital vertebral malformations. Because the bones themselves are unevenly shaped, the spine grows in an unbalanced way and a curve develops.

The most common type of scoliosis seen in young people is called adolescent idiopathic scoliosis. The word idiopathic means "of unknown cause." In idiopathic scoliosis, the individual vertebrae are actually normal in shape — the spine simply curves for reasons that are not fully understood. Idiopathic scoliosis affects roughly 2% to 3% of the general population, and it is more common in girls than boys among the larger curves that need treatment.

Congenital scoliosis is different in an important way: here the bones themselves are misshapen. In congenital scoliosis the malformation is present at birth, even though the curve may not become noticeable until later in childhood when the child grows. In idiopathic scoliosis the spine looks structurally normal on close inspection of each bone.

The shape of the spine is set very early in pregnancy, during roughly the first six weeks, when the pattern of the spine is first laid down in soft tissue. Once that early "mould" is formed, the cartilage and then bone follow the pattern that has already been set. If something disturbs this early process, one or more vertebrae form abnormally — and that abnormality is permanent. The bone abnormality is present at birth, but the visible curve may only appear later as the child grows and the unevenness becomes more obvious.

In most cases, no clear cause can be found, and the condition appears by chance. Research suggests that a single, isolated vertebral abnormality is usually sporadic, meaning it occurs on its own without an obvious reason and is not usually passed down through families. When multiple vertebral abnormalities are present, there is more often an inherited or genetic component. Scientists are still working to understand exactly why these malformations occur, and current understanding is discussed in more detail later on this page.

Congenital vertebral malformations — the underlying cause of congenital scoliosis — are estimated to occur in roughly 0.13 to 0.5 per 1,000 people in the general population. Because some malformations cause so little deformity that they are never noticed, the true number of people affected is not known. It is a relatively uncommon condition, but when it does cause a progressive curve, the effects can be significant, which is why careful assessment and monitoring matter.

To understand what goes wrong in congenital scoliosis, it helps to understand how the spine normally forms.

The spine is built from a column of vertebrae stacked one on top of another, with soft cushions called discs in between. During early pregnancy, the body forms a series of repeating blocks of tissue called somites, which are the early building blocks that go on to become the vertebrae. These somites bud off, one after another, in a beautifully timed and repeating process. Researchers describe this timing using what is sometimes called a "clock and wavefront" model — essentially a biological clock that signals when each new block should form as the embryo lengthens from head to tail.

This process is controlled by several groups of genes that switch on and off in waves (including genes in what scientists call the Notch, Wnt and FGF signalling pathways). When the process runs smoothly, the result is a straight, evenly built spine in which growth happens equally on both sides.

Each vertebra forms first as a soft-tissue model, then turns into cartilage, and finally hardens into bone. Importantly, in congenital problems of the spine, it is usually not a failure of the bone to harden that causes the trouble — it is a failure of the earlier soft-tissue and cartilage stages to form properly. In other words, the problem begins at the very start of the building process.

Two key steps must go right:

Formation — each vertebra must form completely. If one side of a vertebra fails to form, the bone ends up lopsided.

Segmentation — the vertebrae must separate properly from one another, so that each becomes its own independent bone with its own growth surfaces and its own disc. If two or more vertebrae fail to separate, they remain joined together.

When the timing or signalling of early spine development is disturbed, the result is a vertebra that either did not form fully (a failure of formation) or did not separate properly from its neighbours (a failure of segmentation). Sometimes both problems occur together. These are exactly the abnormalities that cause congenital scoliosis, and they are described in detail in the next sections.

Congenital vertebral malformations fall into a few main groups. Understanding the type matters, because the type strongly affects how the spine is likely to behave as the child grows.

This means that part of a vertebra did not form. If one side of a vertebra is missing or underdeveloped, the bone becomes wedge-shaped or triangular. The classic example is a hemivertebra — literally a "half vertebra." Because one side is taller than the other, the spine tips to one side, producing a curve. Failure of formation tends to be uneven (asymmetrical), which is why it commonly produces scoliosis.

This means that two or more vertebrae did not separate properly and remain fused together. If the fusion runs down only one side of the spine, that side cannot grow, while the other side keeps growing — so the spine is pulled into a curve. This one-sided fusion is called an unsegmented bar. If the whole vertebra is fused to its neighbour on both sides, it is called a block vertebra.

In many children, more than one type of abnormality is present, sometimes at the same level of the spine. These mixed patterns can be the most difficult to predict and, in certain combinations, the most likely to produce severe curves.

In the majority of children, congenital vertebral malformations appear sporadically — by chance — even within a family. This makes it difficult to identify a single cause. Because these malformations so often arise on their own, researchers increasingly suspect that epigenetic factors (influences that change how genes behave without changing the genes themselves) may play a part, alongside genetic and environmental factors.

Research has identified a number of genes involved in laying down the pattern of the spine, including genes in the Notch signalling pathway. Faults in these genes underlie certain rare inherited conditions of the spine. A panel of genes associated with vertebral patterning has been studied in patients with varied malformations, and in the Chinese Han population, variations in a gene called TBX6 have been linked to congenital scoliosis.

Scientists believe congenital malformations often result from a combination of genetic, environmental and epigenetic factors working together. Laboratory studies show that an environmental "trigger" such as low oxygen levels during pregnancy can worsen vertebral malformations in animals that already carry a genetic susceptibility — illustrating how genes and environment can interact.

Various exposures during pregnancy have been linked with vertebral malformations in animal studies and in some human reports. These include maternal diabetes, high body temperature (hyperthermia), alcohol use, certain anticonvulsant (anti-seizure) medicines such as valproic acid, and a number of environmental toxins. In one large series of births, the most common exposure associated with vertebral malformation was maternal diabetes. It is important to stress that, for any individual child, a specific cause usually cannot be identified, and parents should not assume they did anything wrong.

Doctors classify the abnormal bones because the type predicts how the curve is likely to behave. The main types are described below.

A hemivertebra is a "half vertebra" — a wedge-shaped extra piece of bone caused by one side of a vertebra failing to form. It sits on the outer (convex) side of the curve and acts like a wedge, tipping the spine to one side. Hemivertebrae are among the most common malformations.

Not all hemivertebrae behave the same way. A fully segmented hemivertebra has a normal disc above and below it, which means it has its own growth surfaces and can drive the curve to grow steadily. An incarcerated hemivertebra is a small, poorly formed piece of bone tucked neatly into a niche between two otherwise normal vertebrae; these tend to cause little or no curve. A semisegmented hemivertebra is partly fused to a neighbour and tends to progress more slowly. Understanding which kind is present helps doctors predict the outlook.

A wedge vertebra results from a partial failure of formation — one side of the vertebra is underdeveloped (hypoplastic) rather than completely missing. The bone is shorter on one side, so the spine tilts. Because growth on the smaller side is reduced rather than absent, wedge vertebrae tend to cause milder, more slowly progressing curves than hemivertebrae.

A block vertebra is caused by a bilateral (both-sided) failure of segmentation — two or more vertebrae are fused together on both sides. Because growth is held back fairly evenly on both sides, a block vertebra usually causes only a mild curve. In natural-history studies, block vertebrae rarely produced curves greater than about 20 degrees and seldom needed treatment.

A unilateral unsegmented bar is caused by a failure of segmentation on one side of the spine, joining several vertebrae together along that side. The bar has no growth plates and so cannot lengthen, while the opposite side of the spine keeps growing normally. This creates a strong, one-sided growth imbalance, which is why a unilateral unsegmented bar tends to produce a steadily worsening curve. It always sits on the inner (concave) side of the curve.

Some children have a combination of abnormalities. One particularly important combination is a unilateral unsegmented bar with one or more hemivertebrae on the opposite side at the same level. Here, one side cannot grow (because of the bar) while the other side has extra growth (because of the hemivertebrae). This "push-pull" combination produces the most rapidly worsening and severe curves of any type, which is why it is recognised as a distinct, high-risk pattern. Other mixed patterns can be harder to classify and to predict.

The spine forms at the same time, and in the same region of the embryo, as several other organs. Because of this shared timing, a problem affecting the developing spine can be accompanied by problems in other systems. For this reason, when congenital scoliosis is found, doctors usually check the spinal cord, kidneys and heart, and look out for other features.

In one study of 126 children with congenital spinal deformity, other organ defects were found in 55% of patients, underlining how often additional problems occur.

Hidden problems affecting the spinal cord and its coverings are common in congenital scoliosis. In the study above, intraspinal abnormalities (problems inside the spinal canal) were found in 37% of children when modern scanning (MRI) was used. These can include a tethered cord (where the spinal cord is abnormally anchored at the bottom), a fluid-filled cavity in the cord called a syrinx, and a condition called diastematomyelia, in which the spinal cord is split by a bony or fibrous spur. These abnormalities were found to be more common in children with kyphosis and with segmentation and mixed defects, and in those whose abnormal bones were in the neck or chest (cervical or thoracic) region. Importantly, the outward skin signs that sometimes hint at a cord problem were found to be unreliable, which is part of why scanning is recommended.

Because the kidneys and urinary system develop alongside the spine, kidney problems are more common in children with congenital scoliosis. In the study of 126 children, urogenital abnormalities were found in 21%. Most of these were mild and did not need treatment, but a minority did require active treatment, including, rarely, problems affecting kidney function. This is why an ultrasound scan of the kidneys is often part of the assessment.

Heart problems are also more common. In the same study, congenital heart disease was found in 26% of children. About half of these children needed treatment (medical or surgical), and the others were simply monitored. The most common heart problems found were holes in the wall between the heart's chambers (ventricular and atrial septal defects). Heart problems were more frequent in children with kyphosis and with mixed defects. Because of this, an ultrasound scan of the heart (echocardiography) is recommended, especially before any spinal surgery.

Abnormalities of the ribs are very common alongside congenital spine problems. Ribs may be missing, fused, or extra. While rib abnormalities do not by themselves cause scoliosis, they can affect the shape and growth of the chest, which matters for breathing and is discussed further under thoracic insufficiency below.

Sometimes congenital vertebral malformations are part of a recognised syndrome — a pattern of features that occur together. Examples mentioned in the research include VACTERL association (which can involve vertebral, anal, cardiac, windpipe and food-pipe, kidney and limb abnormalities), Klippel–Feil syndrome (short neck, low hairline and fused neck vertebrae), Goldenhar syndrome, Alagille syndrome, and the spondylocostal and spondylothoracic dysostoses (inherited disorders affecting many vertebrae and ribs). Identifying a syndrome can help guide which other checks a child needs.

The signs of congenital scoliosis vary widely. Some children have very little to see, while others have an obvious curve from an early age. Common features include:

Visible spinal curvature — a sideways curve of the back is the most recognisable sign, although in young children it can be subtle.

Shoulder asymmetry — one shoulder may appear higher than the other. This is especially common with curves high up in the chest (upper thoracic curves); in the natural-history research, an elevated shoulder was a frequent and distressing cosmetic concern, and a high curve could also cause the head to tilt to one side.

Trunk imbalance — the upper body may lean or "list" to one side rather than sitting squarely over the pelvis. This was a common finding with more severe lower thoracic and thoracolumbar curves and could produce significant deformity.

Rib prominence — when the spine rotates as well as curves, the ribs on one side can become more prominent, forming a "rib hump." Interestingly, in congenital curves the primary curve often has only mild rotation, but a secondary curve that develops below it can rotate severely and produce a large rib hump.

Leg length differences — when the curve reaches the lower spine, the pelvis can become tilted (pelvic obliquity), which makes one leg appear shorter than the other. In the research, some degree of pelvic tilt and apparent leg shortening developed in most children with thoracolumbar and lumbar curves.

Growth-related changes — because the abnormality is built into the growing spine, the deformity tends to change as the child grows, often becoming more noticeable during the rapid growth of infancy and again during the adolescent growth spurt.

Pain — children with congenital scoliosis are often not in pain, particularly when young. However, pain can occur — for example, in some patients pain developed in an over-curved (hyperlordotic) area of the lower back below the deformity.

Neurological symptoms — pressure on the spinal cord can, in some cases, cause symptoms such as weakness, numbness, changes in walking, or problems with bladder control. Subtle signs — such as a slightly smaller calf or foot on one side, a high arch, or a change in reflexes — can be early clues and are easily missed unless specifically looked for. Neurological problems are more often linked to congenital kyphosis (a forward curve) than to pure scoliosis, as discussed below.

Diagnosis combines a careful examination with a set of imaging tests. Several different scans may be needed because, as explained above, congenital scoliosis is often accompanied by problems in the spinal cord, kidneys and heart.

Assessment begins with a full history and physical examination. Doctors look at the shape and balance of the back, the level of the shoulders and pelvis, and any rib prominence. A particularly careful neurological examination is important, because subtle differences — in calf or thigh size, foot size, the arch of the foot, or reflexes — can be signs of an underlying spinal cord problem and are easily overlooked in a routine check.

Plain X-rays are the basic tool for diagnosing congenital scoliosis. They show the shape of the vertebrae and reveal the type of malformation, and they allow the curve to be measured in degrees using a standard method (the Cobb method). The same landmarks are measured each time so that the curve can be tracked accurately over the years. Special X-ray views — for example, with the child bending or lying stretched out — can show how flexible the curve is, which helps with planning.

MRI (magnetic resonance imaging) uses magnets rather than radiation to produce detailed pictures of soft tissues, including the spinal cord. Because hidden spinal cord abnormalities are common in congenital scoliosis — found in 37% of children in one study — MRI is an important investigation. Research recommends that an MRI of the whole spinal cord be obtained before any surgical correction, and in any child who has or is developing neurological signs. In young children, MRI may need to be done under sedation or a general anaesthetic to keep them still, so the timing is often planned thoughtfully rather than done in every case immediately.

CT (computed tomography) scans use X-rays to build detailed, sometimes three-dimensional, pictures of the bones. They can be especially helpful for understanding the exact shape of complex malformations and for surgical planning, because they show the bony anatomy clearly.

Because kidney abnormalities are more common in congenital scoliosis (found in around 21% of children in one study), an ultrasound scan of the kidneys is commonly performed to check that the urinary system has formed normally.

Because congenital heart disease was found in around a quarter of children with congenital spinal deformity, an echocardiogram (an ultrasound scan of the heart) is recommended, particularly before any spinal surgery. Some heart problems are picked up for the first time during this assessment. Children with kyphosis or mixed defects are at higher risk and especially warrant a heart scan.

In short, these investigations are not done to alarm families, but because finding and managing any associated problems early — particularly of the spinal cord, heart and kidneys — makes care safer and more effective.

One of the most important questions for any family is: will the curve get worse? The honest answer is that it depends on the type of malformation and where it is in the spine. Decades of research have made it possible to give families a reasonably good idea of the likely outlook.

"Natural history" means how a condition behaves if it is left untreated. The landmark study in this area followed 251 patients with untreated congenital scoliosis. It showed clearly that the prognosis is not uniformly benign. Of the patients last seen untreated after the age of ten, 36% had a curve of 40 to 60 degrees and a further 28% had a curve greater than 60 degrees. The picture was actually even more serious than these numbers suggest, because many of the worst curves had already been treated before the age of ten precisely because they were so severe. This research overturned an older belief that congenital scoliosis usually progresses slowly, if at all.

Two things matter most in predicting how a curve will behave: the type of vertebral anomaly and the site (region) of the spine where it occurs. A third important factor is the child's growth — curves tend to worsen most during periods of rapid growth.

The underlying principle is simple. Normally the spine grows evenly on both sides and stays straight. A congenital malformation upsets this balance — either by reducing growth on one side (as with an unsegmented bar) or by adding a wedge of extra growth on one side (as with a hemivertebra). The greater the growth imbalance, the worse the curve.

From most severe to least severe, the types of single malformation tend to rank as follows:

The most severe and rapidly worsening pattern is a unilateral unsegmented bar with a hemivertebra on the opposite side at the same level. This was followed in severity by a unilateral unsegmented bar alone, then two hemivertebrae on the same side, then a single hemivertebra, then a wedge vertebra, and least severe of all, a block vertebra.

Curves are not constant in their rate of worsening. In the research, curves present before the age of ten tended to deteriorate faster as the child grew, and the rate often increased sharply during the adolescent growth spurt. This is why close monitoring continues throughout the growing years.

A block vertebra holds back growth fairly evenly on both sides, so it produces only a mild curve that rarely exceeds about 20 degrees and usually needs no treatment. Likewise, an incarcerated hemivertebra (a small piece of bone neatly tucked between normal vertebrae) has no real growth potential, so curves from this cause stayed under 20 degrees with little or no progression. A hemivertebra that has fused to a neighbour also progresses more slowly.

A unilateral unsegmented bar contains no growth plates and cannot lengthen, while the opposite side keeps growing — creating a powerful, relentless growth imbalance. Curves caused by a bar that were present in infancy deteriorated very rapidly and all became very severe. The "bar plus opposite hemivertebra" combination is worse still, because one side is held back and the other side has extra growth.

To make this concrete, here is a summary of what the research found for the different patterns:

Hemivertebra progression — A single hemivertebra usually causes a slowly progressing curve. In the upper thoracic and lumbar regions it tended to worsen by about 1 degree per year before age ten, with curves often staying under 40 degrees and frequently not needing treatment. In the lower thoracic and thoracolumbar regions a single hemivertebra caused more deformity, worsening by about 2 degrees per year before age ten, and often required treatment around the adolescent growth spurt.

Fully segmented hemivertebrae — A hemivertebra with normal discs above and below has its own active growth surfaces and so tends to drive ongoing progression, unlike the incarcerated or fused types that stay quiet.

Two hemivertebrae on the same side produced a greater imbalance than one, and all such curves in the lower thoracic and thoracolumbar regions had passed 50 degrees by the age of ten; these often needed treatment before the age of five.

Unsegmented bars caused rapid progression — for example, around 5 degrees per year for lower thoracic curves and 6 degrees per year for thoracolumbar curves before age ten, increasing further after age ten.

Combined anomalies (bar plus contralateral hemivertebra) were the worst, with thoracolumbar curves of this type worsening by more than 10 degrees per year.

Thoracic versus lumbar curves — For every type of malformation, the outlook was generally worst in the thoracolumbar region (the junction of the chest and lower back), only slightly less severe in the lower thoracic region, less severe again in the lumbar region, and most favourable in the upper thoracic region. The earlier a curve appears — especially in the first year of life — the more it signals a strong growth imbalance and a worse outlook.

Because congenital curves can change as a child grows, regular review is a central part of care.

The whole purpose of monitoring is to catch a worsening curve early — while it is still small and easier to manage — rather than waiting until it has become severe and harder to treat safely. The research strongly emphasises identifying curves that are at risk of progressing and planning treatment before severe deformity develops.

Since progression is closely tied to growth, doctors keep a close eye on a child during the periods of fastest growth: infancy and the adolescent growth spurt. These are the times when curves are most likely to worsen quickly.

X-rays taken at intervals allow the curve to be measured and compared over time. By always measuring from the same landmarks, doctors can detect even small increases in the curve. The rate of change — how many degrees per year — is often more telling than any single measurement.

At each visit, clinicians watch for an increasing curve, any change in the balance of the trunk or the level of the shoulders or pelvis, the development of a secondary curve, and any new neurological signs. Spotting these changes early is what allows timely, less drastic treatment.

Not every child with congenital scoliosis needs surgery. The right approach depends on the type of malformation, the size and location of the curve, the child's age, and how the curve is behaving over time.

For curves that are mild and not progressing — such as those caused by a block vertebra or an incarcerated hemivertebra — careful observation with regular review may be all that is needed. Many of these curves stay small and never require active treatment.

In the context of congenital deformity caused by abnormal bones, exercises do not straighten the spine, because the curve comes from the shape of the bones themselves. In the natural-history research on congenital kyphosis, exercises of any type, frequency or duration were of no value in correcting the deformity. Physiotherapy may still have a role in general wellbeing and strength, but families should understand that it cannot correct a congenital curve.

A brace is a supportive external jacket sometimes used in scoliosis. For idiopathic scoliosis, full-time bracing can be effective at holding a flexible curve while a child finishes growing.

For congenital curves, the evidence is much less encouraging. Because a congenital curve is driven by abnormally shaped, often rigid bones, a brace cannot correct it. In the congenital kyphosis research, attempts to treat children with a brace failed to produce significant improvement and, at best, only slowed the rate of progression. A brace may still occasionally have a supporting role — for example, helping to manage a long, flexible secondary curve that develops above or below the rigid congenital segment, or as part of after-surgery care — but it is not a way to fix the congenital curve itself.

Non-surgical management — chiefly observation — is most appropriate for curves that are mild, in low-risk locations, caused by low-risk malformations (such as block vertebrae or incarcerated hemivertebrae), and shown to be stable over time. The key is regular monitoring, so that if a curve does start to progress, treatment can be considered without delay.

Surgery is considered when the balance of evidence suggests that leaving the curve alone carries a greater risk than operating. The natural-history research is sobering here: in untreated children, curve progression was the rule for the high-risk patterns, and roughly three quarters of children with congenital scoliosis show some progression, with about half eventually needing surgery. Situations that may prompt consideration of surgery include:

Progressive curves — a curve that is documented to be worsening, particularly one caused by a high-risk malformation, is the most common reason to consider surgery. The research repeatedly stresses treating progressive curves early, while they are still relatively small.

Significant deformity — a curve that has already become large, or is producing a visible deformity that affects the child, may warrant surgery.

Growth-related concerns — because the malformation is built into a growing spine, doctors weigh up how much growth remains and how the curve is likely to behave during the growth spurt.

Trunk imbalance — when the upper body leans to one side and a compensating curve cannot keep the trunk balanced, surgery may be needed to restore balance.

Thoracic insufficiency — concerns about the growth of the chest and its effect on the lungs (explained in the next section) can be a reason to intervene, particularly in very young children.

Neurological concerns — pressure on the spinal cord, or signs that the cord is at risk, can make surgery necessary, sometimes urgently.

There are several surgical strategies. The choice depends on the type and site of the malformation, the size of the curve, the child's age and remaining growth, and whether the chest and lungs need room to develop. The broad goals of all of them are to stop the curve worsening, to correct it where possible, and — in young children — to do so while protecting future growth of the spine and chest.

This operation removes the wedge-shaped half-vertebra that is driving the curve and joins the vertebrae above and below together. Because it removes the cause of the curve directly, it can produce good correction, and it is now regarded as the treatment of choice for a progressive curve caused by a single hemivertebra. It can be performed by more than one surgical route, and is discussed in detail in its own section below.

Goals: remove the offending bone and correct the curve at its source. Advantages: directly addresses the cause; can give good, lasting correction, especially when done early while the curve is still small and flexible. Limitations: it is major surgery with the usual surgical risks, including a risk to the spinal cord. Typical indications: a single hemivertebra producing a progressive or significant curve, often in a younger child.

"In situ" means "in place." This operation joins the affected vertebrae together to stop that segment of spine from growing and curving further, without trying to straighten it. The aim is to halt progression rather than to correct the existing curve. Goals: stabilise the spine and prevent further worsening. Advantages: relatively straightforward; an effective way to stop progression. Limitations: it does not correct the curve already present, and it stops growth in the fused segment. Typical indications: a progressing curve where the main aim is to prevent it getting worse.

In very young children, fusing a long section of spine would stop it growing and could harm the development of the chest and lungs. Growth-friendly techniques aim to control the curve while still allowing the spine and chest to grow. They generally use implants that can be lengthened as the child grows. Goals: control the curve while preserving growth of the spine and chest. Advantages: protects lung and trunk development in young children. Limitations: usually requires repeated procedures over time and carries a higher overall rate of complications. Typical indications: young children with progressive deformity where preserving growth matters.

Growing rods are one type of growth-friendly system. Rods are anchored to the spine above and below the curve and are periodically lengthened as the child grows, controlling the curve while allowing continued growth. The underlying principle of using anchors above and below a curve connected by rods, with controlled lengthening, is well established in spinal surgery. Goals and advantages are as for growth-friendly surgery. Limitations: each lengthening has traditionally required a small operation, meaning a series of repeated procedures while the child grows. (Techniques in this fast-moving area continue to develop, and the specifics of any device are best discussed directly with the surgical team.)

A definitive (final) spinal fusion permanently joins the affected vertebrae together, correcting the curve as much as is safe and then holding it in the corrected position while the bone heals into a solid block. Modern fusions use metal implants — screws and rods — to correct and hold the spine. Goals: lasting correction and stabilisation. Advantages: a durable, final solution once growth is complete or when the situation requires it. Limitations: it permanently stops growth and movement in the fused segment, so in young children it is balanced against the need to preserve growth. Typical indications: older children and adolescents, or situations where a permanent solution is appropriate.

A note on implants: a great deal of research has gone into the design of the screws and rods used in spinal surgery. The stiffness of a rod depends on its material (titanium, cobalt chrome and stainless steel differ in stiffness) and its thickness, and surgeons shape and contour the rods carefully to restore the spine's natural front-to-back profile as well as correcting the sideways curve. The aim of modern surgery is not only to straighten the spine when viewed from behind, but also to restore good overall balance when viewed from the side.

Because hemivertebra is one of the most common causes of congenital scoliosis, the operation to remove it deserves a closer look. Removing a hemivertebra is now regarded as the standard treatment of choice for a progressive curve caused by a single hemivertebra.

The surgeon removes the wedge-shaped half-vertebra that is tipping the spine, along with its disc material, and then brings the healthy vertebrae above and below together, fixing them with small implants — usually over just one spinal segment, an approach called monosegmental instrumentation. Removing the wedge takes away the cause of the curve and allows the spine to be straightened at that point.

There are two main ways to reach and remove a hemivertebra. The older, "classical" method approaches the spine from both the front and the back in the same operation — a combined anterior and posterior approach. A newer method removes the hemivertebra working only through the back — a single posterior approach.

A comparative study of children who had a hemivertebra removed for congenital scoliosis looked closely at how these two routes compared. Both approaches straightened the spine to a similar degree — on average about 55% to 59% correction of the main curve — and both gave good correction of the compensating curves above and below as well as restoring side-to-side and front-to-back balance. The difference lay mainly in the recovery and the complications. Children operated through the back alone tended to spend less time on a breathing machine after surgery, less time in intensive care (about one day versus three), and fewer days in hospital (around twelve versus nineteen), and they had markedly fewer general complications. In that study, none of the children operated through the back alone had a surgery-related general complication, compared with around 38% of those operated through the combined front-and-back route, where complications such as fluid around the lungs, a sluggish bowel, or partial lung collapse occurred (all of which settled with treatment). The need for a blood transfusion was broadly similar between the two routes.

Because of these advantages, the single posterior approach is increasingly favoured where it is suitable. The right choice for any individual child still depends on the anatomy of the malformation and is made by the surgical team.

It may be recommended when a single hemivertebra is causing a curve that is progressing or is already producing significant deformity or imbalance. Because it tackles the problem at its source, it can achieve good correction over a relatively short segment of spine.

A recurring theme across the research is that earlier treatment of a progressive congenital curve, while it is still small and flexible, tends to give better and safer results than waiting until the curve is large and rigid. The landmark natural-history study showed that untreated high-risk curves frequently reach severe sizes and that the rate of worsening tends to climb during growth — which is exactly why surgeons aim to operate before a curve becomes severe. Operating early can mean a smaller, single-segment operation and a better-balanced spine. The exact timing for any child depends on many factors and is a decision made carefully by the surgical team with the family.

When performed for the right child, hemivertebra removal can correct the curve well and stop it progressing. In the comparative study above, the corrected position was generally well maintained, the spine healed solidly in every child (no failed fusions were seen in either group), and balance was restored both side-to-side and front-to-back. As with all spinal surgery, the results depend on the individual situation, and the operation carries the general risks described later on this page, including a small risk to the spinal cord.

Research on the surgical treatment of congenital scoliosis, including hemivertebra resection in young children, has reported a wide range of general (non-neurological) complication rates between studies — from a few per cent up to around 40% or more — with the combined front-and-back approach tending to carry the higher rates and the single posterior approach the lower. Serious neurological complications are uncommon; reported rates vary across mostly older or smaller surgical series (higher after the combined approach than after the posterior-only approach), and many such problems are temporary rather than permanent, while large registry reviews put the rate of a new neurological problem at around 1%. Failure of the bone to fuse solidly (a pseudarthrosis) has been reported in roughly 9% to 14% of children after the combined approach in some series and is a recognised cause of later loss of correction; modern monosegmental techniques, combined with routine monitoring of the spinal cord during surgery, have helped reduce these problems, and recent series report solid fusion in all or nearly all children. Studies comparing the two approaches have examined not only the correction achieved but also the children's recovery and quality of life afterwards. Overall, the available evidence indicates that, in carefully selected children, surgery can stop progression and improve the shape and balance of the spine, while carrying real but generally manageable risks.

Thoracic insufficiency syndrome describes a situation in which the chest (thorax) is unable to support normal breathing or the normal growth of the lungs. In other words, the chest is too small or too stiff to do its job properly.

Congenital scoliosis, especially when combined with abnormal ribs, can deform and restrict the chest. If this happens in a very young child, it can interfere with the growth of the lungs at a critical time. This is one of the main reasons that, in young children, surgeons try so hard to preserve and encourage growth of the spine and chest rather than simply fusing the spine.

Abnormal ribs — fused, missing or extra ribs — are common in congenital spine conditions. When several ribs are fused together, the chest cannot expand properly. Certain inherited conditions that affect many vertebrae and ribs, such as spondylothoracic dysostosis, are specifically associated with a short, stiff chest and reduced lung volume, and thoracic insufficiency is a recognised feature.

The lungs do most of their growing in early childhood. A chest that cannot expand or grow restricts this development, which can have lasting effects on breathing. This is why preserving chest growth — sometimes with growth-friendly spinal surgery or specialised chest-expanding surgery — is such an important goal in young children with congenital scoliosis and rib involvement.

All spinal surgery carries risks, and it is important that families have a realistic, balanced picture. Discussing the possible risks is not meant to frighten, but to allow genuinely informed decisions. The main risks include the following.

Infection — wound infections (both superficial and deeper) are among the more common complications of spinal deformity surgery and may require antibiotics or further treatment.

Blood loss — spinal surgery can involve significant blood loss, and steps are taken during the operation to manage and replace blood as needed. In hemivertebra removal, around half to two-thirds of children need a blood transfusion, whichever surgical route is used.

Neurological injury — the most serious risk is injury to the spinal cord or nerves, which could cause weakness, numbness or, rarely, paralysis. Fortunately this is uncommon: in a very large review of pediatric scoliosis surgery, the rate of a new neurological problem was around 1.1% for congenital scoliosis, and the risk of death was very low (around 0.3%). When nerve problems do occur after hemivertebra removal they are often temporary — for example, a brief weakness or altered sensation in one leg that recovers over a few weeks. Surgeons take great care to protect the spinal cord throughout the operation, and routine spinal-cord monitoring during surgery is now standard practice.

Implant complications — the screws, rods or other implants can occasionally move out of position, fail, or need adjustment, sometimes requiring further surgery.

Failure of correction — sometimes the spine does not heal solidly. A failure of the bone to fuse is called a pseudarthrosis (a "false joint"). This was a particular problem historically; in older congenital kyphosis surgery, pseudarthrosis was common after a posterior (back-only) fusion, and combining front and back surgery greatly reduced this risk. After hemivertebra removal, some older series reported pseudarthrosis in around 9% to 14% of children, and it was the main reason a corrected curve later lost some of its correction; with modern single-segment techniques, recent series report solid fusion in all or nearly all children. A pseudarthrosis can allow the deformity to progress and may require revision surgery.

Revision surgery — for the reasons above, some children need one or more further operations. This is especially true of growth-friendly surgery, which by design involves repeated procedures and has a higher overall complication rate.

Growth-related considerations — fusing a segment of spine stops it growing. In young children this must be weighed carefully against the need to preserve growth of the spine and chest. Reassuringly, in congenital kyphosis the research noted that any tendency to develop a slight reverse curve after early fusion was actually beneficial, and that some shortening of the trunk from early fusion was a fair trade for preventing a much worse deformity.

It is worth putting these risks in context. The broad message from outcomes research across spinal deformity is that, despite real risks, surgery in carefully selected patients has the potential to produce meaningful improvement — but families should have appropriate expectations of both the benefits and the risks.

Recovery varies with the type of operation, the child's age, and the individual situation, so the following is a general guide only.

Hospital stay — children stay in hospital for a period after surgery to manage pain, monitor healing and begin gentle activity. The length of stay depends partly on the operation: after hemivertebra removal through the back alone, a hospital stay of around two weeks is typical, whereas the combined front-and-back operation, which involves entering the chest, tends to need closer early supervision and a longer stay.

Pain management — pain is expected after spinal surgery and is managed with medicines and supportive care. It improves steadily as healing progresses.

Immobilisation and bracing — after hemivertebra removal, children are often placed in a protective cast for a period (commonly around six weeks), followed by a removable thoraco-lumbar brace for several further months while the fusion becomes solid, with gradual return to upright activity. The exact regimen is set by the surgical team.

Return to school — children gradually return to normal activities, including school, as they recover. The timing depends on the operation and the child, and is guided by the surgical team.

Activity restrictions — there is usually a period during which certain activities are limited to protect the healing spine. Spinal fusions take many months, not weeks, to become solid, so a period of protected activity is normal. Modern implants have made after-care less restrictive than it once was, but details are set by the surgical team.

Follow-up — long-term follow-up is essential. Children are reviewed regularly to check that the fusion is solid, that the implants are in good position, and that the spine remains balanced as the child continues to grow.

The aim of treating congenital scoliosis is not just to change an X-ray, but to give the child the best possible long-term spinal health, function and quality of life.

Curve control — the central goal of surgery is to stop the curve worsening and, where possible, to improve it. Achieving a balanced, stable spine is more important than any single measurement on an X-ray.

Growth — in young children, treatment aims to control the curve while still allowing the spine and chest to grow as much as possible. In older children and adolescents, a definitive fusion provides a durable, final correction once growth is largely complete.

Function — the goal of modern deformity surgery is to maintain and ideally improve function. Restoring good overall balance — both side-to-side and front-to-back — helps the spine work efficiently and comfortably over a lifetime.

Quality of life — across spinal deformity as a whole, the large majority of published studies show that, in carefully selected patients, surgical treatment has the potential to produce meaningful improvements in quality of life. At the same time, the evidence specifically for congenital scoliosis outcomes is more limited than for other types, and researchers continue to call for further high-quality study, including longer follow-up of children operated on at a young age.

Long-term spinal health — keeping the spine balanced, both when viewed from behind and from the side, is increasingly recognised as important for long-term back health. Although much of this evidence comes from idiopathic scoliosis, the underlying principle — that good overall spinal balance matters for long-term comfort and function — applies broadly.

1. What is congenital scoliosis? It is a sideways curve of the spine caused by one or more vertebrae that did not form normally before birth. The abnormal bones cause the spine to grow unevenly and curve.

2. How is congenital scoliosis different from ordinary (idiopathic) scoliosis? In congenital scoliosis the vertebrae themselves are misshapen and the problem is present at birth. In idiopathic scoliosis the bones are normal in shape and the cause is unknown.

3. What causes congenital scoliosis? It is caused by a failure of vertebrae to form completely (failure of formation) or to separate properly (failure of segmentation), or both. These problems arise very early in pregnancy. In most children, no specific cause can be identified.

4. Did I do something during pregnancy to cause this? In the great majority of cases no specific cause can be found, and the malformation occurs by chance. Parents should not assume they did anything wrong.

5. Is congenital scoliosis inherited? Usually not. A single, isolated malformation is generally sporadic and not passed down through families. Multiple malformations more often have an inherited component.

6. How common is it? Congenital vertebral malformations are estimated to occur in roughly 0.13 to 0.5 per 1,000 people, though the true figure is uncertain because mild cases may never be noticed.

7. What is a hemivertebra? A hemivertebra is a "half vertebra" — a wedge-shaped extra piece of bone caused by one side of a vertebra failing to form. It tips the spine to one side.

8. What is an unsegmented bar? It is a bony bridge joining several vertebrae together along one side of the spine. Because that side cannot grow while the other side does, it produces a steadily worsening curve.

9. What is a block vertebra? It is two or more vertebrae fused together on both sides. It usually causes only a mild curve because growth is held back fairly evenly on both sides.

10. Which type of malformation is most serious? A unilateral unsegmented bar with a hemivertebra on the opposite side at the same level has the worst outlook, because one side is held back while the other has extra growth.

11. Will my child's curve get worse? It depends on the type of malformation and its position. Some curves stay stable, while others — particularly those caused by bars or combined anomalies in the thoracolumbar region — can worsen rapidly. Overall, around three quarters of children show some progression, and about half eventually need surgery. Regular monitoring is how this is tracked.

12. When does congenital scoliosis get worse fastest? Curves tend to worsen most during the periods of rapid growth: infancy and the adolescent growth spurt.

13. Does the position of the curve matter? Yes. For most malformation types, curves at the thoracolumbar junction have the worst outlook, lower thoracic curves are next, and upper thoracic curves tend to be the most favourable.

14. Can congenital scoliosis cause other health problems? Yes. Because the spine forms alongside other organs, children may also have spinal cord, kidney or heart abnormalities, which is why these are checked.

15. Why does my child need an MRI scan? MRI shows the spinal cord in detail. Hidden spinal cord problems are found in around a third of children with congenital scoliosis, and MRI is recommended before surgery and if there are any neurological signs.

16. Why does my child need a heart scan? Congenital heart disease is found in around a quarter of children with congenital spinal deformity, so an echocardiogram is recommended, especially before surgery.

17. Why does my child need a kidney scan? Kidney and urinary problems are more common in congenital scoliosis (around one in five children), so an ultrasound checks the urinary system.

18. Are there often warning signs of a spinal cord problem on the skin? Not reliably. Research has found that outward skin signs are unreliable indicators of an underlying cord problem, which is part of why scanning is used.

19. Can a brace cure congenital scoliosis? No. Because the curve comes from abnormally shaped bones, a brace cannot correct it. In congenital kyphosis, bracing at best only slowed progression.

20. Can exercises or physiotherapy straighten the spine? No. Exercises cannot correct a congenital curve caused by abnormal bones, although they may help general strength and wellbeing.

21. Does every child with congenital scoliosis need surgery? No. Mild, stable curves — such as those from block vertebrae or incarcerated hemivertebrae — can often simply be observed with regular review. Across all children with congenital scoliosis, roughly half eventually need surgery.

22. When is surgery considered? Surgery is considered for curves that are progressing, that are causing significant deformity or imbalance, that threaten the chest and lungs, or that put the spinal cord at risk.

23. Why operate early rather than wait? A consistent message from research is that treating a progressive curve early, while it is small and flexible, tends to give better and safer results than waiting until it is large and rigid.

24. What is hemivertebra excision? It is an operation to remove the wedge-shaped half-vertebra causing the curve, then join the healthy vertebrae above and below — usually over a single spinal segment. It tackles the problem at its source and is the treatment of choice for a progressive curve due to a single hemivertebra.

25. Is hemivertebra removal done from the front, the back, or both? It can be done either through a combined front-and-back approach (the older method) or through the back alone (a newer method). Both straighten the spine to a similar degree, but the back-only approach generally means fewer general complications, a shorter time in intensive care, and a shorter hospital stay, so it is increasingly preferred where suitable. The surgical team decides which route best fits a child's anatomy.

26. What is in situ fusion? It is an operation that joins the affected vertebrae together to stop further curving, without trying to straighten the existing curve.

27. What is growth-friendly surgery? It is a range of techniques used in young children that control the curve while still allowing the spine and chest to grow, usually with implants that can be lengthened over time.

28. What are growing rods? Growing rods are implants anchored above and below the curve that are periodically lengthened as the child grows, controlling the curve while allowing growth. Traditionally each lengthening has required a small operation.

29. What is a definitive spinal fusion? It is a permanent joining of the affected vertebrae, correcting the curve as much as is safe and holding it while the bone heals into a solid block. It permanently stops growth in that segment.

30. What is thoracic insufficiency syndrome? It is when the chest is too small or stiff to support normal breathing and lung growth. Congenital scoliosis with abnormal ribs can cause it, especially in young children.

31. What is the most serious risk of surgery? The most serious risk is injury to the spinal cord or nerves, which is uncommon — around 1% for a new neurological problem in congenital scoliosis surgery in a large review, and when nerve problems do occur after hemivertebra removal they are often temporary.

32. What is a pseudarthrosis? A pseudarthrosis is a failure of the spine to fuse solidly — a "false joint." It can allow the deformity to progress and may need further surgery. It was reported in around 9% to 14% of children in some older hemivertebra-removal series, but modern single-segment techniques report solid fusion in almost all children.

33. How common are complications after congenital scoliosis surgery? Reported rates vary widely between studies. For hemivertebra removal, general (non-neurological) complications tend to be lower after the back-only approach and higher after the combined front-and-back approach. Serious neurological complications are uncommon.

34. Will my child need more than one operation? Possibly. Growth-friendly surgery in particular involves repeated procedures by design, and some children need revision surgery for other reasons.

35. Is it harmful to fuse the spine in a young child? Fusion stops growth in that segment, so in young children it is balanced carefully against preserving growth. When only one or two segments are fused, as in hemivertebra removal, the loss of growth is small. In congenital kyphosis, the research found early fusion worthwhile, with any small loss of trunk height being a fair trade for preventing severe deformity.

36. Can congenital scoliosis cause paralysis? It is uncommon, and is more associated with congenital kyphosis (a forward curve), where pressure on the cord can develop, particularly with curves high in the chest. This is one reason these curves are monitored and treated carefully.

37. What happens at follow-up appointments? The curve is measured on X-rays and compared over time, and the doctor checks the balance of the trunk, the shoulders and pelvis, and looks for any new neurological signs.

38. Can children with congenital scoliosis lead normal, active lives? Many children do very well, especially when curves are mild or treated effectively. Outcomes depend on the type and severity of the malformation and any associated conditions.

39. What is the long-term outlook? The outlook depends heavily on the type and site of the malformation and on early, appropriate management. In carefully selected children, treatment can stop progression and improve the shape and balance of the spine, supporting good long-term function and quality of life.

Several important studies underpin our understanding of congenital scoliosis.

Landmark natural history studies. The most influential study of how congenital scoliosis behaves over time followed 251 untreated patients and showed that the condition is often progressive and far from benign. It established that the rate of worsening and the eventual severity depend on both the type of malformation and the region of the spine, and it ranked the malformation types from most to least severe. This work made it possible to anticipate which curves are at high risk and to plan early treatment — a major shift from the older belief that congenital scoliosis usually progresses slowly.

Natural history of congenital kyphosis. A separate landmark study of 130 patients with congenital kyphosis (a forward, rather than sideways, curve) documented that, without treatment, progression of the curve was the rule, averaging around 7 degrees per year and reaching a maximum during the adolescent growth spurt. It also showed that bracing and exercises were ineffective, that early posterior fusion gave the best results, and that paralysis could occur — particularly with high thoracic curves caused by failure of formation. These findings continue to influence how forward curves are managed.

Associated abnormalities. Research assessing 126 children with congenital spinal deformity demonstrated how often other organs are involved: intraspinal abnormalities in 37%, other organ defects in 55%, congenital heart disease in 26%, and urogenital abnormalities in 21%. This evidence is the basis for the modern recommendation that children undergo MRI and heart and kidney assessment as part of their work-up.

Hemivertebra resection: comparing surgical approaches. A comparative study of children who underwent hemivertebra removal for congenital scoliosis examined two surgical routes — the classical combined front-and-back approach and a newer back-only (single posterior) approach. Both achieved similar correction of the main and compensating curves (around 55% to 59%), but the back-only approach was associated with significantly fewer general complications, less time on a breathing machine, shorter intensive-care and hospital stays, and faster recovery, with comparable, low rates of pseudarthrosis and of (mostly temporary) neurological problems. This kind of evidence supports the trend towards single-segment, posterior-only hemivertebra removal where the anatomy allows.

Genetics and embryology. Research into the genetics of congenital and idiopathic scoliosis has clarified how the spine forms through the precisely timed budding of somites, controlled by genes in the Notch, Wnt and FGF pathways, and has identified genes and exposures linked to vertebral malformation. It also highlights that most malformations occur sporadically and likely result from a combination of genetic, environmental and epigenetic factors.

Surgical outcomes and complications. Reviews of surgical treatment of pediatric spinal deformity report a wide range of complication rates for congenital scoliosis surgery, with low rates of serious neurological injury (around 1%) and death (well under 1%) in large registries. The same body of work stresses that, while surgery carries real risks, it has the potential in selected patients to improve quality of life — and that the evidence base specifically for congenital scoliosis remains limited.

Surgical techniques. Research on the surgical management of scoliosis has advanced the design of implants (screws and rods of differing materials and stiffness), the techniques for correcting the spine in three dimensions, and the importance of restoring not just side-to-side but also front-to-back balance. Strategies for severe, rigid curves — such as gradual traction to improve flexibility before surgery — have also been studied.

Despite this progress, important uncertainties remain. The number of high-quality studies specifically reporting the outcomes of surgery for congenital scoliosis is limited, and more research is needed, including longer follow-up of children operated on at a young age. The exact genetic and environmental causes of most malformations are still being unravelled, and for any individual child a precise cause usually cannot be found. The best timing and choice of operation must be tailored to each child, and the ideal balance between controlling a curve and preserving growth in very young children continues to be refined.

Looking ahead, researchers anticipate that improvements in genetic technology and in classifying patients by their specific malformation will lead to a better understanding of why these conditions occur and, ultimately, to more targeted prevention and treatment. In spinal surgery more broadly, a major hope for the future is the ability to guide and modulate spinal growth so that healthy growth and function can be preserved while reducing the need for fusion.

Congenital scoliosis is a sideways curve of the spine caused by vertebrae that did not form normally before birth; the abnormal bones are present from birth even if the curve appears later.

It is fundamentally different from idiopathic scoliosis, in which the bones are normal in shape and the cause is unknown.

The malformations fall into two main groups — failure of formation (such as a hemivertebra) and failure of segmentation (such as an unsegmented bar) — and some children have both.

In most children the condition occurs sporadically, by chance, and parents should not assume they caused it.

How a curve behaves depends on the type of malformation and the region of the spine; this can be predicted reasonably well.

The highest-risk pattern is a unilateral unsegmented bar with a hemivertebra on the opposite side; block vertebrae and incarcerated hemivertebrae are the lowest risk.

Curves at the thoracolumbar junction tend to have the worst outlook; upper thoracic curves the most favourable.

Congenital scoliosis is often progressive — and frequently more serious than once believed — so early diagnosis and regular monitoring are essential. Around three quarters of children show some progression and about half eventually need surgery.

Curves worsen fastest during the rapid growth of infancy and the adolescent growth spurt.

Children are routinely checked for associated spinal cord, kidney and heart problems, which are common; MRI and heart and kidney scans are part of the assessment.

Braces and exercises cannot correct a congenital curve, because it is driven by the shape of the bones.

Surgery is considered for progressive curves, significant deformity, trunk imbalance, threats to chest and lung growth, or risk to the spinal cord.

Treating a progressive curve early, while it is small and flexible, generally gives better and safer results than waiting.

For a single hemivertebra, removing the offending bone (hemivertebra excision) is the treatment of choice; it can be done through a combined front-and-back approach or through the back alone, with similar correction but fewer complications and a faster recovery from the back-only route.

Other surgical options range from in situ fusion to growth-friendly techniques such as growing rods that preserve growth in young children, and definitive fusion in older children.

Surgery carries real but generally manageable risks, and in carefully selected children it can stop progression, improve the shape and balance of the spine, and support good long-term function and quality of life.

This page is for general information only and is based on published medical research and textbooks. It is not a substitute for individual medical advice. Every child with congenital scoliosis is different, and decisions about investigation and treatment should always be made together with a qualified spinal surgeon who can assess the individual situation.

• —NICE — Low Back Pain and Sciatica in Over 16s (NG59)
• —NHS — Spinal Conditions
• —British Association of Spine Surgeons (BASS)

This information is for general patient education and is not a substitute for an individual consultation. Always seek personalised advice from a qualified spinal specialist.