The DDB has been evaluated with use of different outcome measures, including radiographic parameters, such as the Cobb angle [
36] and the Perdriolle angle of rotation, [
37], cosmesis, evaluated by changes in the Angle Trunk Inclination (ATI), and the quality of life, as assessed with the Brace Questionnaire (BrQ), [
38,
39].
Cobb angle and Perdriolle angle of rotation
In 1995, Valavanis et al reported an initial correction rate of 49.54%. At two years' follow-up, the correction rate was 44.1% [
4]. In 2003, Grivas et al reported an overall improvement rate of 35.7%, while the deformity had remained stable in another 46.42% of their patients. The scoliotic deformity increased in 17.83% of patients in that report [
5].
A more detailed analysis of the 28 scoliotics treated conservatively with the DDB in the series of Grivas et al follows:
Seventeen children had double curves (double curve group), four had thoracic curves (thoracic group), and seven had thoracolumbar curve (thoracolumbar group).
In the double curve group, 17 children (15 girls, 2 boys; mean age, 12.3 years; range, 8 - 17) had right thoracic and left lumbar curves with a mean thoracic Cobb angle of 23.2°(range, 10°- 35°) and a mean lumbar Cobb angle of 21.2°(range, 8°- 30°). Using the Perdriolle method, rotation was measured at a mean value of 6.9°(range, 3°- 25°) for the thoracic curve and 7.8°(range, 4°- 15°) for the lumbar curve. The mean follow-up was 28 months (range, 14 - 84).
In the thoracic group, 4 children (all girls; mean age, 13.8 years; range 12 - 15) had right thoracic curves with a mean thoracic Cobb angle of 25°(range, 22°- 31°) and a mean apical vertebral rotation of 6.8°(range, 3°- 10°). The mean follow-up was 8 months (range, 6°- 18°).
In the thoracolumbar group, 7 children (6 girls, 1 boy; mean age, 13.5 years; range, 12 - 17) had thoracolumbar curves with a mean Cobb angle of 24°(range, 20°- 38°) and a mean apical vertebral rotation of 10°(range, 4°- 30°). The mean follow-up was 9 months (range, 6 - 21), [
5].
The results presented pertain to the Cobb angle and Perdriolle readings, as measured on follow-up radiographs obtained without the brace.
At final follow-up of the double curve group, 6 curves improved, 7 curves remained stable and 4 curves increased. In five children, curve improvement involved both thoracic and lumbar Cobb angles, whereas in one child only the thoracic Cobb angle improved. In the four children with increase of their scoliosis, this was noted on both the thoracic and the lumbar curves. Concerning rotation at final follow-up, the mean value for the thoracic curves at the apical vertebra was 6.2°(range, 0°- 22°) for thoracic and 5°(range, 0°- 12°) for lumbar curves.
In the thoracic group, the Cobb angle remained unchanged in two curves and improved in another two. The mean rotation at follow-up was 5.4°(range, 2°- 10°).
In the thoracolumbar group, two curves improved, four remained stable and one increased. At final follow-up, the mean rotation measured 8.6°(range, 4°- 20°).
The overall results, including all three (double, thoracic and thoracolumbar) groups, indicate that 10 curves out of 28 (35%) improved, 13 (46%) remained stable and five (18%) became worse, as assessed by measuring the Cobb angle. One (4.3%) patient underwent surgical correction. Four out of five curves that increased were double (right thoracic and left lumbar).
Rotation (measured with use of a Perdriolle template at the apical vertebrae in patients out of the brace) improved only in the lumbar component of the double curves (p < 0.004) and remained unaffected (statistically non-significant change) in the thoracic component of double and in single curves.
By comparison, the Cobb angle of the same children was measured while in-brace. Nineteen (67.8%) curves improved, seven (25%) remained stable and two curves (7.1%) increased. When children with double curves were wearing the brace, the mean apical vertebral rotation measured 5.8°and 3°for the thoracic and lumbar curves, respectively. An additional de-rotating effect for the lumbar component was thus clearly observed. For further details on this study, the reader is referred to Table one of the publication by Grivas et al (2003) [
5].