Bracing for adolescent idiopathic scoliosis: evidence and guidance

Adolescent idiopathic scoliosis (AIS) is a three-dimensional deformity of the spine that begins at puberty without a known cause. It is diagnosed when the Cobb angle on X-ray is 10° or more, usually with vertebral rotation and an unbalanced trunk.

Epidemiology. About 80% of scoliosis cases are idiopathic, and AIS is the most common form at 2–3%. A 2026 global meta-analysis in Spine (data 1975–2025) found that clinically significant AIS, a Cobb angle of 20° or more, has a prevalence of 0.28%. It is more common in girls than in boys.

Psychological burden. A 2026 systematic review in Health Expectations pooled 19 qualitative studies from 10 countries (323 participants). Stigma, prejudice and lack of psychological support made the experience of children and carers heavier. A 2025 physiotherapy review noted that more than 30% of AIS patients under observation or in a brace had clinically significant psychological distress.

About 4 in 10 children with AIS have curve progression if untreated. The meta-analysis of Di Felice et al. (2018) in the American Journal of Physical Medicine & Rehabilitation estimated progression at 42% in AIS and 49% in a mixed juvenile–adolescent group.

This is why intervention must happen while the child is still growing: it is the only period in which a brace can change the course of the disease.

What happens in adulthood without treatment? The 50-year follow-up of Weinstein et al. (JAMA 2003) of 117 untreated people with late-onset idiopathic scoliosis versus 62 controls: survival was similar to the general population (about 0.55 versus 0.57); 61% had chronic back pain versus 35% of controls, most of it only mild to moderate; 22% had shortness of breath in daily activities versus 15%, with a higher risk when the Cobb angle exceeded 80° with a thoracic apex. The authors concluded that untreated adults remain productive and functional. This puts the goal in proportion: treatment at puberty aims to stop the curve reaching a severe level, and decisions must be individualised for each child.

Treatment decisions rest on three main parameters: curve size, skeletal maturity and rate of progression.

A note on error and X-rays. Because measurement error is about 5°, changes smaller than that should not be taken as real improvement or worsening. The authors of the 2026 network meta-analysis also stress that follow-up X-rays should not be repeated at less than 6-month intervals without a clinical indication, to limit cumulative radiation.

Bracing is the only non-surgical treatment for scoliosis proven in a multicentre randomised trial to reduce the rate of surgery.

BrAIST was a multicentre trial in the US and Canada of 242 children with a Cobb angle of 20°–40°, Risser 0–2, aged 10 to under 15, who wore a rigid TLSO or were observed only. The outcome was whether the child reached skeletal maturity with a Cobb angle below 50°.

Temperature sensors inside the brace recorded the real hours worn. Hours worn were closely linked to the success rate (P < 0.001).

In the trial children were told to wear the brace at least 18 hours a day, but actual wear depended on the curve. The review of Karavidas (2019) also pooled studies showing that those wearing more than 90% of the prescribed hours had an 11% progression rate, versus 56% among the non-adherent.

To show what “in-brace correction” means, we use a sample case from the clinic: a 15-year-old (born 2007) with a single C-shaped lumbar curve, apex L1, Cobb angle 38.65° (film of 2 August 2022). Treatment was with two individualised braces (day and night), periodic review at each stage and exercise. The patient adhered very well to the wearing schedule.

How to read it: in-brace correction is the Cobb angle measured on a film taken while wearing the brace, compared with the angle out of the brace. The difference between the two films is the corrective force the brace produces. A negative number means the curve is pushed past straight to the opposite side.

Reading the numbers: out of the brace the curve fell from 38.65° to 10°, about 74%. On the film of 24 October 2023, compared with 10° out of the brace, the day brace brought the curve to about 0° and the night brace over-corrected it to a negative value. At the review of 28 June 2024 both braces were still “negative”, and the patient had completed the final stage of weaning off the day brace. At that review the software measured two angles (upper and lower curve), so each row has two numbers.

Why does in-brace correction matter? The best-evidence synthesis of van den Bogaart et al. (Eur Spine J 2019; 26 studies) concluded that there is strong evidence that a lack of initial in-brace correction is associated with treatment failure, and moderate evidence that wear time is associated with both failure and success; the authors call for research on a minimum correction threshold. The Boston-brace study of Emans et al. (Spine 1986; 295 patients, best in-brace correction about 50% on average) found that greater in-brace correction predicted better correction later. Katz et al. (JBJS Am 2010; 100 completers) recorded no progression in 82% of those wearing the brace more than 12 hours a day, versus 31% of those wearing it under 7 hours. The 2016 SOSORT guidelines also call for periodic X-rays to check the brace’s effect.

Note: this is an association, not proof of cause and effect; one child’s result also depends on age, skeletal maturity, curve pattern and adherence (this case adhered very well), so it does not represent every child. In-brace correction should also not be chased at any cost: a good brace balances corrective force, comfort and the ability to wear it for enough hours.

Most corrective scoliosis braces today are thoraco-lumbo-sacral orthoses (TLSO). In 2022 SOSORT, together with SRS, ISPO and POSNA, developed a new brace classification system, endorsed by ESPRM.

Night bracing: a new option with randomised evidence. The CONTRAIS trial was carried out in 6 public hospitals in Sweden from 2013 to 2018 with 135 children with moderate AIS. The night-brace-plus-physical-activity group had a 76% success rate. This was higher than the specific-exercise group (58%) and the physical-activity-only group (53%). A secondary analysis found in-brace correction of the main curve of 48% with the night brace and 51% with the full-time brace; the two had equivalent effectiveness.

How is the brace made? According to SOSORT, every stage from prescription and manufacture to checking, adjustment and follow-up must strictly follow brace-management standards for each individual brace. Effectiveness depends on the correction achieved when worn, so a skilled orthotic engineer is the key factor.

Scoliosis-specific exercises give a small but real benefit, and work best combined with a brace, not replacing it.

PSSE (Physiotherapeutic Scoliosis-Specific Exercises), including the Schroth method, focus on three-dimensional self-correction, posture control, rotational breathing and neuromuscular retraining. A randomised trial in Turkey showed that the group doing three-dimensional Schroth exercises for 6 months had better Cobb angle results than the observation-only group.

Reading the numbers correctly. A 3–5° reduction sits close to the measurement error of the Cobb angle (±5°). Exercise should therefore be seen as a supportive therapy for posture control and for giving the child an active role, rather than a stand-alone way of straightening the curve. Clinical guidelines recommend that physiotherapy complement the brace and not replace it for moderate curves. SOSORT also recommends specific spinal mobilisation exercises to prepare for bracing. The evidence for exercise has improved a lot: the first Cochrane review (2013) found only 2 studies with 154 people at very low certainty, whereas the 2024 update has 13 randomised trials.

The international societies agree that a child who is still growing, with a curve of about 20°–25° or more and a risk of progression, should be braced.

SOSORT 2016 guidance (published 2018):

SRS criteria for bracing studies: age 10 or more, Risser 0–2, Cobb angle 25°–40°, premenarchal or under 1 year after the first period. This is also the group of patients who benefit most clearly from bracing in the trials.

This is a reference framework. The treatment decision needs a specialist to individualise it by age, skeletal maturity, curve pattern, risk of progression and the family's wishes.

The most important thing parents can do is detect early and help their child wear the brace for enough hours.

A brace is made for each child's own body and curve, its in-brace correction checked and adjusted as the child grows.