Quantification of skeletal, muscular and kinematics parameters in scoliosis : methodological and clinical studies

Abstract

This dissertation presents new methods derived for precise evaluation of a number of anatomical and functional variables of the complex thoracospinal deformity in scoliosis. By consensus the scoliotic angle at the frontal plane is measured on radiographs using the Cobb method. Another way of measuring this angle, the Ferguson method, although evidently more accurate than the Cobb, is not in use. The Cobb method uses only the declination of the upper and the lower vertebra of the curve to measure the degree of the curve. The Ferguson method uses the centroid of these two vertebrae and of the apical one to define the angle. In a derived new method the geometrical, i.e., the true, centre of the three vertebrae is defined on antero-posterior radiographs of the spine to delineate the scoliotic angle. Comparison of the measurements by the three methods in three groups of radiographs with different degrees of scoliosis shows that the new method gives higher accuracy and better repeatability than the two other methods.

Studies on the vertebral growth under normal and pathological conditions are based on measurements of the vertebral height at different points of interest on radiographs, since there is no method for evaluation of the vertebral volume. A new stereological method is presented which allows estimation of the vertebral volume in vivo. The volume of one isolated normal and one scoliotic vertebra was measured in a model study using frontal and lateral radiographs and one CT-scan of each. The height of the vertebrae was measured at five points on the radiographs and was expressed as weighted circumferential height. Then the area of the vertebral body was measured using a grid on the CT-scan of the vertebra and the volume was calculated using the Cavallieri method. The accuracy of the measurements with the new method was high compared with serial CT-scans of the whole vertebra. The applicability of the method was evaluated in a series of scoliotic patients who had undergone posterior spinal instrumentation and fusion. The results showed that the weighted circumferential height of the three vertebrae had increased significantly at the last follow up on average 3 years after the operation. Also the volume of the apical vertebra was found to have increased, though not significantly. It is concluded that the proposed method provides accurate evaluation of vertebral body volume changes.

The role of the intercostal muscles in the development of the thoracospinal deformity in scoliosis and of the respiratory movements of the thoracic cage is well recognised, but poorly studied because of lack of an accurate method for measurement of the surface area of the muscles and hence, indirectly, their function. In a model study the surface area of the intercostal muscles was evaluated in ultrasonographs and CT-scans by measurement of the area of their image either from the perpendiculars or by tracing. In both cases measurements of the surface area from the perpendiculars gave more accurate results than by tracing. The applicability of the method was tested in one healthy person during maximal inhalation and exhalation and no significant difference was found between the left and the right side during maximal inhalation or at full exhalation nor between full inhalation and exhalation for either side.

To investigate if quantitative whole body movement analysis is a suitable method for motor control research a model study was developed to evaluate the kinematics of different spinal segments and of shoulder-link and pelvis displacement. Fourteen girls participated in the study; eight with right convex adolescent idiopathic scoliosis and six age-matched normal controls. The subjects were asked to stand on two force plates, and seventeen passive reflective markers were attached to certain anatomical landmarks on the dorsal aspect of their body. Six surface EMG electrodes were attached to the paraspinal and hip abductor muscles. The subjects were instructed to perform standardised voluntary trunk rotation and side bending to the left and right side. Simultaneous recording of the displacement of the markers with signals from plate forms recording the ground reaction force and the muscle activity was done.

The results showed that the scoliotic group had a different way and strategy for orientation of different spinal segments displacement during voluntary trunk movements with a significant difference in the amplitude of the displacement of the centre of mass during side bending and during rotation of the trunk, compared with the control group. The resulting horizontal shear forces showed a significant difference between the two groups during side bending of the trunk, which was not the same during trunk rotation. The EMG study of the recorded activities from the paraspinal muscles and hip abductor muscles showed a different patterns and large variability between subjects and between the trials of each subject. In conclusion, this type of investigation helps to understand better the co-ordination of different body segments during voluntary movements and provides an adequate tool for investigation of spine movements and other trunk movements in normal as well as in scoliotic subjects.