^{3.3 ORTHOTIC TREATMENT}

^{3.3.1 BRACE TREATMENT FOR IDIOPATHIC SCOLIOSIS}

Charter 3, section 3.3.1  Literature review of scoliosis bracing, Master degree by thesis, Wood, 2003.

Disagreement in aetiology and natural history of scoliosis further complicates the debate of whether spinal bracing is effective in changing the natural history of progressive scoliosis.  Although Dickson (1984) claimed that there is no evidence which demonstrates conclusively that wearing a brace affects the natural history of scoliosis, there is a general agreement that the TLSO can prevent progression in adolescents with idiopathic scoliosis  (Carr et al., 1980; Weiss and Werkmann, 1996; Rigo, 1999a; Wood and Rigo, 1999; Rigo and Wood, 2001a).  Scoliosis is a three-dimensional deformity, and thus correction must be considered not only in the coronal plane, but also in the sagittal and transversal planes.  Winter et al., (1975) as well as Willers et al., (1993), have reported that deformity in the sagittal and transversal planes has an important effect on pulmonary function and cosmesis.  Although the TLSO Boston brace endeavours to correct rotation, Emans et al., (1986); Peltonen et al., (1988)  and Olafsson et al., (1995)  have reported that it does not provide any permanent change in rotation.  Additionally, Labelle et al., (1996) and Aubin et al., (1996, 1997) claimed that the TLSO Boston brace produces a thoracic flat back and reduces the sagittal diameter of the thorax.  Rigo (1999a) presented at the meeting of the International Research Society of Spinal Deformities in Burlington, Vermont, that the hypercorrection Chêneau brace system seems to obtain sagittal plane normalisation as well as correction of rotation and lateral curvature in idiopathic scoliosis.

I) TLSO BOSTON BRACE STUDIES

Willers et al., (1993) presented the long-term effect  of TLSO Boston brace treatment of the Cobb angle, vertebral rotation, rib hump and translation of the apical vertebra, in idiopathic scoliosis.  Computed tomography measurements were completed before the start of treatment with the TLSO Boston brace and subsequently after bracing during an 8.5 years mean follow-up in 25 patients with idiopathic scoliosis.  At follow-up, the pre-treatment Cobb angle, vertebral rotation, rib hump and the translation were not significantly decreased.  As a result, this study demonstrated that the TLSO Boston brace does not improve, however prevents progression of the Cobb angle, vertebral rotation, rib hump and translation in idiopathic scoliosis.  Willers et al., (1993) claimed that the reduction of the sagittal diameter was noteworthy and may be of importance for cosmesis and pulmonary function.

Katz et al., (1997) presented a study of 319 patients with adolescent idiopathic scoliosis treated with either the TLSO Boston brace or a Charleston bending brace.  The results found that the TLSO Boston brace is more effective than the Charleston bending brace, both in preventing lateral curve progression and in avoiding the need for surgery.  These finding were most notable for patients with curves of 36 to 45 degrees Cobb angle in whom 83% of those treated with a Charleston bending brace had curve progression of more than 5 degrees, compared with 43% of those treated with the TLSO Boston brace.  As a result, Katz et al., (1997) recommend the TLSO Boston brace and that the Charleston bending brace should be considered only in the treatment of smaller single thoracolumbar or single lumbar curves.

Goldberg et al., (1993)  reported two groups of 32 girls with adolescent idiopathic scoliosis, one group was treated during late onset of idiopathic scoliosis with the TLSO Boston brace and the second group was untreated.  The groups were based on curve size, location, and age at diagnosis, furthermore, all were Risser sign 0 at diagnosis.  Goldberg et al., (1993) found that there was no statistically significant difference between the groups on any parameter of curve progression (Cobb angle and vertebral column rotation).  Therefore, doubts were raised about the efficacy of spinal orthoses in modifying the natural history of late-onset idiopathic scoliosis and removes the ethical problems inherent in a prospective trail in which the only treatment permitted to the control group is surgery.

There are numerous studies on the effectiveness of braces in preventing the progression of the deformity, by taking the Cobb angle as the evaluated parameter, and occasionally the axial rotation angle as well (Mellencamp et al., 1977; Hopf and Heine, 1985; Liljenqvist et al., 1998) but the majority are not conclusive for several reasons.  These reasons are related to the design of the retrospective studies, the heterogeneity of the specimens, including males, females, juvenile and adolescent scoliosis, with initial Cobb angles and also with a very variable initial bone age.  The conclusive study on the effectiveness of the brace in preventing the progression of the Cobb angle is the study directed by Nachemson and Peterson, (1995).  This is a prospective, controlled study, in which, patients were observed and placed in the control group, or treated by electro-stimulation or with a TLSO Boston-type brace.  Although, TLSO Boston brace appears effective in preventing lateral curve progression, it does not necessarily mean 3D correction.

II) MILWAUKEE BRACE COMPARED WITH THE TLSO BOSTON BRACE

 

Long-term studies of both the Milwaukee and TLSO Boston brace have demonstrated that the main effect of orthotic treatment is to produce a curve that is only a few degrees better than that of the original deformity (Edmondson and Morris, 1977; Mellencamp et al., 1977).  Therefore, it is assumed that bracing prevents deterioration but does not convert major deformities into normal physiological shapes.

III) TLSO BRACES COMPARED WITH NATURAL HISTORY

Miller et al., (1984) studied 255 female patients with initial curvature measuring 15-30 degrees and who ranged in age from 8 to 17 years.  These patients were divided into two closely matched groups with 144 patients treated with various types of TLSO bracing and 111 patients going without active treatment.  The results after a mean period of 1.9 years, suggested that bracing reduced the overall probability of progression when compared with the untreated group.

IV) TLSO COUPLING EFFECT

Aubin et al., (1996, 1997) reported that orthoses are widely used to treat scoliotic deformities of the trunk, but the way the corrective forces are transmitted from the thorax to the spine remains poorly understood, and several undesired effects such as the reduction of sagittal curvatures or weak derotations are often reported.  A biomechanically measurable element model of the trunk was used to investigate the hypothesis that a coupling mechanism exists between the scoliotic spine and rib cage, which may explain incomplete and unexpected results obtained by orthotic treatments.  Forces were applied to the model on the rib hump and lateral side of thorax.  These biomechanical simulations demonstrated the existence of coupled motions between the spine and rib cage subjected to orthotic loads.  Aubin et al., (1996, 1997) showed that reduction of physiological sagittal curvatures (up to 30%) are possibly related to anterior orthotic loads applied on the rib hump.  These loads also contributed to increase lateral shift of the spine (up to 7 mm) as well as scoliotic frontal curvatures (up to 4 degrees).  Based on these results, another approach was proposed and this consisted of applying loads laterally on the convex side as well as on the anterior thorax opposite to the rib hump, with a system that mechanically constrains the backward movement of the posterior rib hump.  This biomechanical model was simulated on four scoliotic patients presenting thoracic curves between 22 and 54 degrees to evaluate its practicability.  It was found that derotation of the trunk (7 to 13 degrees) and reduction of frontal curvatures could be done without reducing physiological sagittal curvature.  More simulations on different scoliotic configurations are necessary to find the most optimal combination of forces to produce a real 3D correction of scoliotic deformities.

Willers et al., (1993) demonstrated an undesirable effect of a reduction in the sagittal diameter of the thorax caused by the TLSO Boston brace.  Labelle et al., (1996) and Aubin et al., (1996, 1997) reported that the TLSO Boston brace produces a lordotic effect (hypokyphosis) in the thoracic region as a result of the coupling of the spine and the ribs of the costal gibbus.  This is caused by the forces acting from dorsal to the ventral aspect.  The authors Labelle et al., (1996) and Aubin et al., (1996, 1997) have recently proposed a modification of the correction principles, which consists of applying loads laterally on the convex side as well as on the anterior thorax opposite to the rib hump, with a system that mechanically constrains the backward movement of the posterior rib hump.  Indeed, these proposed modifications are not very different from those originally proposed by Chêneau, (1990, 1994, 1996a). The correction principles originally proposed by Chêneau in 1979 facilitate correction as a result of the location and size of the forces, as well as the expansion rooms on the opposite side of the convexities, which permit derotation, coronal plane correction and sagittal normalisation.

V) CHÊNEAU BRACE STUDIES

A study by Oberthaler et al., (1985) was conducted on 115 patients with idiopathic scoliosis treated with either the TLSO Boston brace or the Chêneau brace.  The results found that there was excellent Cobb angle correction of the deformity in both braces.  The TLSO Boston brace seems to be better for lumbar and thoracic curves whereas the Chêneau brace lends itself more for thoracic curves or when more than one primary curve is present.  It was also claimed that the TLSO Boston brace was effective in treatment of mild hyperkyphosis.

Von Deimling et al., (1995) compared long-term influence of idiopathic scoliosis in 47 patients who wore either the Milwaukee brace or the Chêneau brace with an average follow-up of 7.8 years.  It was reported that Chêneau brace had significantly better results.  There was a Cobb angle correction of 62% and 38% the patients who wore the Chêneau and Milwaukee braces respectively.  The initial correction of the Cobb angle was 35% and 47% of the pre-treatment value in the Chêneau and Milwaukee braces.

Rigo (1999c) reported a retrospective study of 105 patients (mean age of 12.5 years) with progressive idiopathic scoliosis who were treated with the Chêneau brace.  Of this group, 44 patients had been wearing other braces from other clinics before the start of treatment with the Chêneau brace.  All of these 44 patients presented curve progression even while wearing their previous brace.  The major Cobb and torsion angles had a mean of 37 degrees and 17 degrees respectively at the start of Chêneau brace treatment.  The major Cobb and torsion angles had a mean primary correction of 31% and 22% respectively.  In the group of patients with end results (n=37) the mean initial major Cobb and torsion angles were  36.4 and 16.9 degrees respectively and at follow-up they were 34.1 and 15.7 degrees.  The results show high initial Cobb angles at the start of treatment and a low primary correction.  The final Cobb angle at a 2-year follow-up, showed a tendency of a loss of correction, without reaching significance.  Rigo (1999c) claimed that the Chêneau brace could effectively prevent the progression of the Cobb and torsion angles, even in cases of bad prognosis. In agreement with other authors, these results show better end results with a low initial Cobb angle and high primary correction.  The primary correction is less than those of Hopf and Heine, (1985) as well as Liljenqvist et al., (1998) however this could be due to the higher initial Cobb angle and poor effect of previous treatments of lower quality braces producing more rigid curves.