The spinal curve in standing and sitting postures in children with idiopathic scoliosis

A sample of convenience of children with moderate idiopathic scoliosis without bracing or surgery was studied. The sample consisted of 19 children, aged 9 to 16 years, with mean Cobb angle of 24 degrees. The spinal configurations and paraspinal muscle activity in several commonly assumed postures were examined.

To determine how the apex angles, verticality of spine, and muscle activity vary with the assumed posture and whether the location and the number of spinal curves affect these variables.

It has been suggested that the configuration of the spine in commonly assumed postures can affect the spinal curve in scoliosis because of gravitational bending moments. There is, however, a paucity of data obtained in subjects in sitting postures that school-age children assume daily for prolonged periods. Absence of bilateral symmetry in pressure during sitting has been described, but its effect on the spinal apex angle has not been investigated.

Infrared-emitting markers, whose three-dimensional positions could be tracked by a pair of cameras, were affixed to the spine. The natural postures studied were relaxed standing, relaxed sitting, erect sitting, and writing while seated. Electromyographic activity in muscles close to the spinal apexes was recorded bilaterally for each test posture.

Subjects leaned laterally and, in general, anteriorly, in all test postures, rather than placing the C7 vertebra vertically above S1. The direction of lean and the change in the spinal apex angle from standing to sitting varied depending on whether the spinal curve was single or double, thoracic or lumbar. Subjects with single curves, whether thoracic or lumbar, tended to lean laterally toward the convexity of their curve apex--that is, the lean was in a direction that reduced the apex angle. Subjects with double curves (thoracic and lumbar), in all postures except relaxed sitting, tended to lean toward the convexity of the lumbar curve, thereby reducing the lumbar apex angle and exacerbating the thoracic angle. Most subjects' apex angles were smaller in relaxed or erect sitting than in relaxed standing. Electromyographic activity was in general greater on the convex side of the curve, with greatest activity in erect sitting.

The findings indicate that in self-selected postures the gravitational effect of leaning and the muscle activity in paraspinal muscles may serve to reduce the apex angle. Thus, a fully upright, centered posture may not be best for correction of every patient's spinal curve.