Scoliosis is a 3-D deformation of the spine:​ the majority of cases occur in adolescence and are of unknown etiology. This is called adolescent idiopathic scoliosis (AIS). Since the cause of AIS is not fully understood, its management is still controversial. Orthotic treatment which is commonly accepted intervention has been applied for more than half a century, however, the knowledge of underlying mechanisms and responses of scoliotic curves are still not fully complete. Moreover, it has some inevitable drawbacks that may affect the treatment outcome.

This thesis consists of three parts. The first part aimed to understand more about the mechanisms and responses of scoliotic curves to the orthotic intervention and the other two parts targeted to study new interventions for AIS. These areas are 1) evaluation on the effectiveness and biomechanics of spinal orthoses in the treatment of adolescent idiopathic scoliosis;​ 2) effectiveness of audio-biofeedback in postural training for adolescent idiopathic scoliosis patients;​ and 3) *+ effects of the deviation of spatial image on the posture of adolescent idiopathic scoliosis patients. In the first part, thirty-three female patients with progressive adolescent idiopathic scoliosis (AIS) were selected and managed with spinal orthoses. The responses of the spinal deformities and the control mechanism of the orthoses were investigated prospectively for 20 months. Before completing the data collection period, three defaulted and four underwent surgical treatment due to curv e deterioration. The curve control rate was 87% (26 out of 33). The full data set obtained from the remaining twenty-six patients were analyzed. The values of standing AP Cobb’s angle, apical vertebral rotation, lumbar lordosis as well as thoracic kyphosis showed significant reduction (p < 0.05), however, the angle of trunk inclination and trunk listing did not. The values of those reducible parameters reached their lowest values within the first 12 months of orthotic treatment and then the values gradually increased but they were still significantly below the pre-brace values. The mean pressure of the pressure pads was found to be 53.2 ± 13.3 mmHg while the mean tension of the straps was 26.8 ± 5.2 N.

For the second part, the possibility of using learned control of physiological responses in control of progressive AIS was investigated. A postural training device (electronic orthosis) was applied as a prototype or experimental model for automatic portable treatment system, incorporating the principle of instrumented learning for extended and continuous training o f medically desirable response. Sixteen AIS patients with progressing or high-risk curves were selected under a strict protocol and fitted with the devices. In the first 18 months of application, three patients were defaulted and four showed curve progression > 10° (two changed to rigid spinal orthoses and two underwent surgery). The curves for the nine patients were kept under control (within ±5° of Cobb’s angle) and five of them have reached skeletal maturity and terminated the application. The remaining four patients were still using the devices till skeletal maturity or curve progression. The results were promising, as the curve control rate was 69%, which seemed lower than that rendered by the rigid spinal orthoses (87% for the first study). However, in this study a long-lasting active spinal control could be achieved through the patient’s own spinal muscles, instead of the temporary passive spinal control via the forces exerted by the rigid orthosis on the patient’s body as in the first study. Nevertheless, before the postural training device to become a treatment modality, a long-term study for more AIS patients was necessary. This project is on-going in the Duchess of Kent Children's Hospital, Sandy Bay, Hong Kong.

The other element of the study was a preliminary investigation to detect the postural changes and body sway of four AIS patients with similar deformities under different spatial images. Two pairs of low-power prisms (Fresnel prisms) with 5 dioptre and 10 dioptre were used. In the experiment, the apices of the prismatic lenses were orientated randomly at every 22.5° from 0° to 360° for testing changes. It was found that the low-powered prisms at specific orientations could cause postural changes of AIS patients that might be used for controlling their scoliotic curves. However, apart from this laboratory test, a longitudinal study was necessary in order to investigate the long-term effect of the prisms at different powers and orientations (under both static and dynamic situations) on the patient’s posture, spinal muscular activities, vision, eye-hand coordination, psychological state and other daily activities before it becomes an alternative treatment modality for AIS patients.

On the assumption that AIS is at least partly related to muscular weakness. There might be a point or a level at which muscular control information from the three studied interventions can be thought to have common mechanism (biofeedback) and interactions. Experiments such as combining the rigid orthosis and postural training device or postural training device and prisms or prisms and rigid orthosis or all three interventions together can be attempted to enhance the effects. It is worthy of investigating their long-term effects.

The management of AIS should be developed along the clinical pathway of early intervention with reliable indicators/predictors, patient’s active participation, dynamic control mechanism, holistic psychological and psychosocial considerations, and effective and long-lasting outcome.