Stickler syndrome

The abstracts were prepared by Jeremy Fairbank. Correspondence should be addressed to him at Nuffield Orthopaedic Centre, Oxford OX3 7LD, UK, or by email jeremy.fairbank@ndos.ox.ac.uk

DISCUSSION GROUP 1:

GENETICS OF ADOLESCENT IDIOPATHIC SCOLIOSIS

Chairman: Andrew J Carr ChM FRCS, Nuffield Orthopaedic Centre, Oxford

What is the genetic basis of adolescent idiopathic scoliosis?

A review of existing literature suggests that idiopathic scoliosis demonstrates many of the characteristics of a complex trait. It has reduced penetrance and variable expressivity. The severity of the disease within families can change and sometime generations are skipped or missed. It is also possible that more than one gene is involved in the disease. Phenocopy may occur in some families.

Papers were presented which suggested that a major gene effect may be present. This was based on an analysis of families from Novosibirsk in Russia. Separate analysis of the same data set performed in Boston and Oxford suggest a potential multi-factorial model for inheritance for the same families. The group then discussed what the likely pattern of inheritance was. It was felt that although in some families there may be a major gene affect and indeed some pedigrees demonstrate a single gene autosomal dominant pattern. The majority of families fit more into a complex trait or multi-factorial mode of inheritance. Twin studies were also reviewed. The major problem with these twin studies is with ascertainment bias. None of the studies were recruited from twin registers but from twins ascertained through Scoliosis Clinics. We believe there is scope for further investigation of heritability using twin studies based on a twin register.

The other aspect of susceptibility that was discussed was the calculation of relative risk. This is the risk of a sibling developing the condition compared to the risk in the general population matched appropriately. So far adequate relative risk calculations have not been undertaken but it should be possible to do this from existing data and this is a recommendation of the group.

Laboratory Studies

Increasing reports are becoming available of laboratory studies. The majority of these are candidate gene investigations. This involves an attempt to link a particular candidate gene to the disease in certain families. So far these investigations have proved negative and a number of candidate genes have been excluded in the families studied. The majority of candidate genes investigated have been for structural proteins including the collagens and fibrilin.

Family Collections

The Group also discussed the requirement for future family collections. It was agreed that a collaborative effort would be ideal and that the group should communicate through an international database and web site. A uniformly agreed questionnaire should be used with an accurate description of the Phenotype. Syndromal scoliosis should be excluded and a minimum lateral curvature of 10 degrees with rotation is required to confirm a positive affected status. Two main types of family are required:

The first is multiplex families with 10 or more members and the second is nuclear families with 3 or more members preferably involving two generations. Both these sets of families could be used for further candidate gene and genome wide investigations. Because different groups use different methodologies both for candidate gene and genome wide searches it is probable that the available family material should be investigated by a number of different laboratory groups to ensure that the same results are obtained. If positive results are found then it will be useful for other groups to investigate these with their families.

FAMILIES WITH CONGENITAL AND IDIOPATHIC SCOLIOSIS: EVIDENCE FOR A COMMON GENETIC ETIOLOGY

B.A. Alman, W.G. Cole

The Division of Orthopaedic Surgery, The Hospital For Sick Children and the University of Toronto, Toronto, Ontario, Canada

Adolescent idiopathic scoliosis is a type of dysplasia, as the deformity occurs with growth; while congenital scoliosis is a type of dysostosis, as the deformity is present at birth. Dysostoses and dysplasias are thought to be due to different types of genes. We found three patients with a family history that included both congenital and idiopathic scoliosis. The probands all had a congenital scoliosis involving multiple vertebral levels, and required surgical intervention for progression. There were multiple family members with congenital and idiopathic scoliosis in at least two generations. A specific mode of inheritance was not be easily identified from this small number of families.

These data raise the possibility that idiopathic and congenital scoliosis are caused by the same gene. However, our findings could also be explained by these families inheriting a gene causing idiopathic scoliosis and another gene causing congenital scoliosis. Comprehensive pedigree analysis of additional families, and DNA studies, will ultimately determine which of these is true.

SEGREGATION ANALYSIS OF IDIOPATHIC SCOLIOSIS: EVIDENCE FOR A MAJOR-GENE EFFECT

T.I. Axenovich, A.M. Zaidman, I.V. Zorkoltseva, I.L.Tregubova

Institute of Cytology and Genetics of the Siberian Division of the Russian Academy of Sciences, Novosibirsk, Russia

Idiopathic scoliosis is still assigned as a genetic disorder of unknown etiology (McKusick, 1992). The main problem is that of choosing between two models of inheritance: majorgene and multifactorial. The correct genetic prediction, which way to treat the genetic defect and which strategy in identification of disease genes depend on the choice all.

We have analysed the pedigrees of 101 probands (778 individuals with known phenotypes). The probands (80 girls and 21 boys under 18 years of age suffering from I to IV degree idiopathic scoliosis) were selected from among the patients of the Special School for Scoliosis (Novosibirsk).

We have performed a complex segregation analysis using a special version of the regressive model (Elston & George, 1989). Three mathematical components form the basis of segregation analysis: the penetrance function, the genefrequency parameters and the transmission probabilities (Elston & Stewart, 1971). This observation is in a good agreement with adolescent scoliosis inheritance model suggested by Bonaiti and co-workers (1976). It should be noted that we have only demonstrated the major-gene control of the heavily pronounced forms of idiopathic scoliosis. Slightly pronounced deformations are most likely to be genetically heterogenic. As we have now demonstrated, ignoring this fact may cause a major-gene hypothesis to be rejected, which always implies that the multifactorial control of idiopathic scoliosis is to be accepted.

ON THE GENETICS OF SPINAL ASYMMETRY AND ADOLESCENT IDIOPATHIC SCOLIOSIS. PATTERNS OF DEFORMITY AND HEREDITY

S. Berven, J.C.T. Fairbank, A.J. Carr, M. Sadovoy, I. Tregubova, A.M. Zaidman

Dept of Orthopaedic Surgery), Massachussetts General Hospital, Boston, MA

We have collected data on patterns of spinal asymmetry among extended families, and on scoliosis within first degree relations of probands with scoliosis in order to determine the pattern of inheritance of both spinal asymmetry and adolescent idiopathic scoliosis. 68 children with a diagnosis of scoliosis were chosen from the Novosibirsk Special School for Scoliosis. We gathered information on the presence of biplanar spinal asymmetry on 1544 relations, 201 first degree, 508 second degree, 630 3rd degree, and 205 fourth degree. Any relations with a known cause of their spinal asymmetry was excluded, as were any relations about whom insufficient data was available. Spinal asymmetry among extended families, and scoliosis within first degree relations both follow an inheritance pattern of a multifactorial trait. These data support the concept that there is a shared genetic aetiology for spinal asymmetry and scoliosis.

THE ROLE OF FIBRILLIN-1 AND FIBRILLIN-2 IN SEVERE SCOLIOSIS

A.H. Child, E. Park, E. Putnam, G. Brice, D.M. Milewicz St George's Hospital Medical School, London

Marfans syndrome (MFS) and congenital contractural arachnodactyly (CCA) are two phenotypically related disorders resulting from gene mutations for highly homologous proteins. fibrillin-I and fibrillin-2 respectively. Both autosomal dominant disorders have skeletal, ocular and cardiovascular manifestations. The two proteins codistribute, with fibrillin-I largely taking over the role of fibrillin-2 in late embryogenesis, especially in eye and cardiovascular system. Fibrillin-2 plays an important role in the skeletal system early in embryogenesis. CCA is much less prevalent than MFS (one in 5000 births). Types of mutation are similar in both genes. FBN-1 mutations are found throughout the 65 exons, with a cluster in exons 24-32 responsible for the most severe neonatal presentation resulting in death from cardiovascular disease in the first two years of life. Similarly, a clustering of mutations in exons 24-34 of FBN-2 have been reported in 6 out of 12 CCA patients studied. These results indicate that the rarity of CCA results from mutations in a limited region of the gene, which plays an important role in human skeletal development.

DNA LINKAGE ANALYSIS OF FAMILIAL IDIOPATHIC SCOLIOSIS

M. Inoue, S. Minami, H. Kitahara, Y. Nakata, H. Moriya Department of Orthopaedic Surgery; School of Medicine, Chiba University,, Chiba, Japan

Linkage analysis is one of the best methods for examining the association of the genetic diseases with candidate genes. It permits statistical analysis of segregation of the disease phenotype with a particular allele of a candidate gene. Disease genes causing Osteogenesis Imperfecta, Ehlers-Danlos syndrome, Stickler syndrome, Marfan syndrome, Beals syndrome, and Recklinghausen disease were considered candidate genes for idiopathic scoliosis. Segregation analysis of COL1A1, COL1A2, COL2AI, Fibrillin 1, Fibrillin 2, and NFI was performed for 3 large pedigrees consisting of more than 13 individuals. The results of segregation analysis suggested that pathogenesis of familial idiopathic scoliosis might differ from that of scoliosis with Osteogenesis Imperfecta, Ehlers-Danlos syndrome, Stickler syndrome, Marfan syndrome, Beals syndrome, and Recklinghausen disease.

STUDY OF TWINS AND SIB PAIRS WITH IDIOPATHIC SCOLIOSIS

M. Inoue, S. Minami, H. Kitahara, Y Nakata, H. Moriya

Department of Orthopaedic Surgery. School of Medicine, Chiba University, Chiba, Japan

Twenty-four pairs of twins, 41 individuals of which had idiopathic scoliosis with curvature of more than IO degrees, and 50 pairs of sibs, all individuals of which had idiopathic scoliosis have visited our clinic. DNA fingerprints revealed that there were 15 monozygotic and 9 dizygotic pairs of twins. Concordance for idiopathic scoliosis among monozygotic twins was 93.3%, while among dizygotic twins it was 55.6%. Monozygotic pairs tended to have greater similarity in severity of curvature than did other pairs. There were significant differences in kyphosis among monozygotes, dizygotes, and sibs (Kruskal-Wallis test; p

GENOMIC SEARCH FOR X-LINKAGE IN FAMILIAL ADOLESCENT IDIOPATHIC SCOLIOSIS

N. H. Miller, D. Schwab, P. Sponseller, E. Shugert, J. Bell, N. Maestri

Johns Hopkins Uni;versit 601 North Caroline St. Suite 5254, Baltimore, MD 21287-0882. Phone (410) 955-9553; Fax (410) 502-6816; e-mail: nhmiller@welchlink. welch.jhu.edu

Clinical observations and genetic studies have confirmed the familial nature of adolescent idiopathic scoliosis (AIS), and support the proposal that the aetiology of this disease has a major genetic component. Pedigrees were ascertained through probands(affected individuals). Individuals were determined to he affected through history, physical examination, and spinal radiographs exhibiting a minimum of I 0 degrees of lateral curvature with pedicle rotation. Fourteen pedigrees with 136 individuals were identified in which AIS is segregating in each generation with no confirmed male to male transmissions. Sixty- four individuals were deemed to be affected and seventy-two unaffected. Both clinical and genetic linkageresults do not support Xlinkage as the mode of inheritance of familial AIS in the study population.

IDENTIFICATION OF A GENE INVOLVED IN IDIOPATHIC SCOLIOSIS

Wise, D. L. Caudle, R. Barnes. J. A. Herring, H. Hobbs, M. Lovett, A.M. Bowcock

Texas Scottish Rite Hospital for Children, Dallas, TX, USA

The objective of our studies is to identify a causative agent in the pathogenesis of idiopathic scoliosis. To this end we are using a positional cloning approach to localize a gene involved in hereditary idiopathic scoliosis, HIS. To date, we have collected DNA from 13 multi-generational families segregating HIS as an apparently autosomal dominant trait. The overall penetrance of the disorder appears to be approximately 70% and the expressivity is quite variable. The sex bias in this collection of patients is 5:1 female to male, comparable to the sex bias typically reported in the literature. A genome-wide scan for an HIS-linked locus is in progress within a portion of one large family (family 9) having at least nine collected affected members and one apparent obligate carrier. Tentative linked regions identified by this approach are now being tested in additional affected individuals collected from this family as well as other collected families. This analysis should allow identification of a single locus responsible for idiopathic scoliosis in family 9 and possibly other HIS families.

IDIOPATHIC SCOLIOSIS: SOME RESULTS AND PERSPECTIVES OF STUDY

A.M. Zaidman, TI. Aksenovich, I.L. Tregubova, E.A. Dikalova, J. Fairbank

Research Institute of Traumatology and Orthopaedics, Novosibirsk, Russia

We have created a data bank of families with idiopathic scoliosis (181 families) and Scheuermann's disease (90 families). A DNA bank of some of these families has been established. Segregation analysis of the pedigrees demonstrates that genetic control of the more pronounced forms of idiopathic scoliosis (II-IV degrees) may be described by an autosomal dominant major-gene di-allele model with incomplete sex-dependent penetrance (higher in girls than in boys) of the genotypes, and major-gene effect of Scheuermann's disease with high penetrance in boys.

Several biochemical markers of idiopathic scoliosis and Scheuermann's disease have been defined. They are: increased excretion of GAG. oxiprolin, alkaline phosphatase and catabolic enzymes. We have also demonstrated high permeability of the connective tissue cell membranes to sugars.

Morphologic examination of 200 samples of spine components from different types of scoliosis were carried out. These showed that the morphogenesis of idiopathic scoliosis is based on a local (the lateral side of vertebral growth plates) disorder in the synthetic processes within chondrocytes and a matrix structure disorder.

Genetically dependent models of different types of deformities were created in mice by selection and inbreeding. Morphologic studies of these models revealed local changes in chondroblast metabolism in the region of vertebral formation in the neonatal period.

DISCUSSION GROUP 2:

PHENOTYPE AND BIOMECHANICS: CONCLUSIONS AND RECOMMENDATIONS

Ian Stokes

University of Vermont, USA

I. There is a need to describe the probability distribution of all traits which are associated with scoliosis in the normal population. Such studies would reveal the existence of directional asymmetries and would define the range of normal beyond which an individual could be classified as abnormal (e.g. greater than 2 standard deviations from the population mean), including asymmetries to right and left side, which could be different. These studies will provide the basis for the rational specification of thresholds for the referral of patients from screening programmes etc. Recognising that there are multiple morphological traits associated with idiopathic scoliosis there is a further need for comprehensive studies of the inter-relationships (correlations) between different traits. Such information would probably improve our ability to recommend referral criteria, based on multiple traits.

2. The group recognised that there is a substantial amount of information supporting the idea that mechanical forces can cause a deformity to progress. However, there is very little definite evidence of biomechanical factors which might be the original causes of scoliosis (etiological). Research is needed to understand how biomechanical forces can modify the natural history of idiopathic scoliosis, especially during growth.

3. There is evidence that both thoracic transverse plane asymmetry and spinal sagittal plane hypokyphosis are associated with early manifestation of adolescent idiopathic scoliosis.

4. Longitudinal studies should be made of patients (preferably including patients who have known risk factors, such as a family history). These studies will show whether traits associated with idiopathic scoliosis develop synchronously or whether some traits are manifested early in the development of the deformity. thereby providing evidence as to which are primary causal factors and which traits are likely secondary or consequential factors.

5. Measurement precision of morphological traits is often poor and this is a major barrier to documenting the existence of a trait and its evolution. There are multiple sources of measurement error including errors from the instrument itself but also errors due to postural and diumal effects, which generally can not be eliminated. Therefore measurement precision problems inevitably create difficulties in detecting the presence and detecting the progression of a spinal deformity.

6. The group recognised that developments in genetics and molecular biology mean that it would be desirable to identify a specific trait for which the genetic basis could be sought (candidate genes). Unfortunately, the group was not able to identify morphological or biomechanical factors for which a genetic basis might be sought with confidence.

7. The group supports the Scoliosis Research Society working paper entitled 'Review of Biomechanics in the Aetiology of Idiopathic Scoliosis'. It is in agreement with the statement that 'the most likely biomechanical mechanism for the aetiology of thoracic idiopathic scoliosis is hypokyphosis'. However, this should be interpreted as indicating that hypokyphosis is probably a risk factor, rather than being a likely unique cause of idiopathic scoliosis.

THREE-DIMENSIONAL EVALUATION OF PRE-, PER- AND POST-OPERATIVE CHANGES IN ADOLESCENT IDIOPATHIC SCOLIOSIS

S. Delonne, H. Labelle, J. Dansereau, C. Aubin, J.A. de Guise, C.H. Rivard, B. Poitras

Ecole de Technologie Superieure, Montreal, Quebec, Canada

This study was done on a cohort of 58 adolescents with idiopathic scoliosis. They were aged between II and 20 years at the time of surgery (37 CD. 19 Colorado, 2 TSRH). Pre- and post-operative 3D reconstructions (obtained from multi-planar radiographs) and per-operative measurements (using an electromagnetic digitizer) were obtained to document the 3-D changes in their spine shape. This study is a first attempt to correlate the three-dimensional changes in the shape of the spine pre-, per- and post-operatively. It indicates that prone positioning and surgical exposure are the main contributors of curve correction in the frontal plane, whereas surgical instrumentation is the key correction element in the sagittal plane. This suggests that sagittal plane correction should be the surgeon's main area of concern during instrumentation.

SIMULATIONS OF A NEW ORTHOTIC APPROACH FOR THE CORRECTION OF SCOLIOTIC DEFORMITIES

C.-E Aubin, F. Poulin, H. Labelle, Y. Petit, D. Gignac, J. Dansereau

Ecole de Technologie Superieure, Montreal, Quebec, Canada

Orthoses such as Boston braces are widely used to treat spinal and thoracic deformities such as scoliosis. Even though the scoliotic curves are improved by these braces in the frontal plane (Nachemson et al., 1995), the changes are not ideal, and several undesired effects such as the reduction of sagittal curvatures or weak truncal denotations are often reported (Labelle et al., 1996). We investigate a new orthotic approach to address the question: "Can we obtain a real 3D correction with a thoraco-lumbo-sacral orthosis ?"

This biomechanical study demonstrated that brace loads can be applied in such a way to improve current brace design and correct in 3D the scoliotic deformities. The clinical validity of such approach remains to be addressed, and more simulations with other force patterns are necessary to find an even more efficient combination of brace loads to produce a real and optimal 3D correction of scoliotic deformities.

THE RELATION BETWEEN BACK SHAPE AND TORSION IN LOWER LIMB BONES MEASURED BY ULTRASOUND IN ADOLESCENT IDIOPATHIC SCOLIOSIS

S. Cavdar, R. G Burwell, A. A. Cole, A. S Kirby, F. J. Polak, R. K. Pratt, J. K. Webb, A. Moulton

Department of Anatomy, Faculty of Medicine, Marmara University, Istanbul, Turkey

This paper examines the relation between back shape in the transverse plane (back hump) and torsion in lower limbs as measured by ultrasound in two groups of patients attending the Scoliosis Clinic: (1) girls having surgery for adolescent idiopathic scoliosis (n=67, thoracic 42, thoracolumbar 22, lumbar 3, right 51, left 16; and (2) school screening referrals including boys and girls (n=95, boys 20, girls 75). Back shape was measured as angle of trunk inclinations (ATIs) at each of 10 levels by five methods namely, Scoliometer in each of three positions (standing forward bending, sitting forward bending and prone), and a commercial television-computer scanning device (ISIS) in each of two positions (standing erect and sitting erect). Femoral anteversion (FAV) and tibial torsion (TT) were measured using real-time ultrasound. The relation between the back shape variables and the torsion in the lower limb bones was examined by univariate and multivariate (canonical) correlations using Hotelling's T for multivariate analysis. The findings show that in the operative girls, left FAV (but not right FAV or either TT) is significantly associated with ATls at levels 1-10, 2-5 and 5-8 in the trunk for each of three positions using the Scoliometer and in ISIS standing. In the screening boys (but not the girls) left TT (but not right TT or either FAV) is significantly associated with ATIs in each of the Scoliometer prone and ISIS standing positions. We suggest that the findings reflect neuromuscular factors common to spine and trunk and each of a) left femur in the girls and b) left tibia in the boys. The laterality of these associations with the left lower limb is unexplained but may indicate control by the right cerebral cortex. (Supported by AO/ASIF Research Commission Project 96-W21 )

SIGNAL CHANGE ON MRI IN MULTIFIDUS MUSCLES IN ADOLESCENT IDIOPATHIC SCOLIOSIS

Y L. Chan, J. C. Y Cheng, X. Quo, A. King, J. Griffith, C. Metreweli

Departments of Diagnostic Radiology & Organ Imaging, Orthopaedics & Traumatology; Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong.

It is well established that MRI can detect evidence of acute and chronic muscle injury. (book) and changes in signal intensity on MRI have been demonstrated in paraspinal muscles after exercise. Increased resting signal intensity of the multifidus muscle was also described in patients with chronic low back pain. The objective of this study is to study the signal intensity of the multifidus muscle in AIS. 51 patients with documented AIS were recruited. 16 of the 18 patients in group I with a severe curve planned for operative treatment showed increase in signal intensity in the multifidus muscle on the concave side of the apex of the curve. 8 of 10 patients in group 2 (with abnormal SSEP) also demonstrated increased signal intensity in the concave side multifidus muscle at the apex. In group 3, of the 23 patients at their first presentation, 11 patients with mild curve (10-30 degrees). Of these, two had increased signal intensity in the multifidus muscle in the concave side. Of the 12 patients with more severe curves, 9 had increase in multifidus signal intensity on the concave side.

UPPER LIMB LENGTH ASYMMETRY IN HEALTHY CHILDREN AND IN PREOPERATIVE GIRLS WITH RIGHT THORACIC ADOLESCENT IDIOPATHIC SCOLIOSIS

A.A. Cole, R.G. Burwell, FJ. Polak, J.K. Webb

School of Biomedical Sciences, and The Centre for Spinal Studies and Surgery, Queen's Medical Centre, Nottingham, UK

This paper evaluates anthropometric upper limb data from 68 pre-operative girls with right thoracic AIS and 1422 healthy children. In healthy children, the mean directional asymmetry has a pattern, being to the right for upper arm asymmetry and to the left for forearm-with-hand asymmetry. Directional asymmetry is related to handedness with right-handed subjects showing more right asymmetry with the opposite for left-handed children. Absolute asymmetry is observed with age as the main factor determining its magnitude. There is increased absolute asymmetry in adolescents compared with prepubertal children. In preoperative AIS girls, all three components of upper limb length (total arm, upper arm and forearm-with-hand length) show significantly larger right-sided asymmetry and significantly more absolute asymmetry compared with healthy girls. There are two possibilities for the exaggerated upper limb length asymmetry in preoperative AIS girls: l.In some way secondary to a progressive scoliotic curve. 2. As a `risk factor' for the development of the spinal curvature by the increased mass of the right upper limb altering neuromuscular control during movement and hence spinal kinematics.

ANTHROPOMETRY IN ADOLESCENT IDIOPATHIC SCOLIOSIS (AIS) FINDINGS IN PREOPERATIVE GIRLS

A.A. Cole, R.G. Burwell, F.J. Polak, J.K. Webb

School of Biomedical Sciences, and The Centre for Spinal Studies and Surgery, Queen's Medical Centre, Nottingham, UK

This paper reports anthropometric data from 68 preoperative girls with right thoracic AIS (mean Cobb angle = 56) compared with data from 693 healthy school girls. In the AIS girls standing and sitting height, biacromial and biiliac widths, arm lengths, leg lengths and weight are significantly increased, AP chest diameter is significantly decreased and lateral chest diameter which is equivalent compared with controls. Allometric evaluation (assessed as the ratio of one body segment to another) shows that in the AIS girls (l) total leg lengths are proportionately longer than each of sitting height and total arm lengths: (2) sitting height and total arm length are in proportion; and (3) AP thoracic diameter is proportionately smaller than lateral thoracic diameter.

There are two possibilities for the finding that preoperative AIS girls have a larger than normal extrathoracic skeleton with a normal/diminished thorax:

1. The normal/diminished thorax is secondary to a progressive scoliotic curve in larger than normal girls

2. as a `risk factor' for the development of the spinal curvature.

LONG TERM 3-D SPINAL CHANGES AFTER C-D INSTRUMENTATION AND FUSION IN ADOLESCENT IDIOPATHIC SCOLIOSIS

S. Delonne, P. Papin, H. Labelle, J. Dansereau, J.A. de Guise, CE Aubin

Ecole de Technologie Superieure, Montreal, Quebec, Canada

We recently have reported the immediate 3-D changes in the spine after C-D instrumentation in AIS and have documented significant 3-D correction one month after surgery. Are these results stable in time and if not, what happens to the spine in 3-D? To answer these questions we have followed a cohort of 30 adolescents with AIS (27 right thoracic. 17 left lumbar curves) for an average of 2.5 years after C-D surgery. The 3-D geometry of the spine (TI-L5) was obtained using a multiplanar radiographic reconstruction technique. These results indicate that the instrumented spine continues to grow after surgery even if the 2-D height of the fused segments does not change. This observation combined with the loss of correction in the frontal plane and in the orientation of the plane of maximum deformity indicate a crankshaft phenomenon which could not be correlated with the age of the patients (x = 14.50 1.6 years). This 3-D loss of correction suggests that anterior surgery may be preferable to posterior approaches in the treatment of AIS in order to avoid long term post-op loss of correction.

POSSIBLE ROLE OF MUSCLE SPINDLE IN THE AETIOLOGY OF IDIOPATHIC PROGRESSIVE SCOLIOSIS

K. Dobosiewicz

Department of Rehabilitation Silesian Medical Academy, Katowice, Poland

There has been a growing interest in the role of muscle spindle in the aetiology of idiopathic scoliosis. Information conveyed from the muscle spindle afferents plays a major role in the reflex regulation of muscle tone. The intrafusal fibres of the muscle spindle are arranged in parallel with extrafusal fibres, so that stretching of a muscle causes the spindle to discharge, while contraction of the extrafusal fibres tends to unload the spindle. Muscle spindle is supplied by an independent group of motor axons called fusimotor or gamma axons. Fusimotor axons are of two types-dynamic and static. The dynamic gamma axons increase the sensitivity of the spindle primary endings to changes in muscle length. Static gamma axons reduce spindle afferent sensitivity to changes in the length, but increase afferent firing at a constant length. The aim of the present investigations was to determine which parts of the muscle spindle can be particulary involved in the aetiology of idiopathic progressive scoliosis. The study included 91 healthy children /66 girls and 25 boys/ and 554 children with scoliosis. The latter were divided into fast /204 girls and 23 boys/ and slow- progressive /295 girls and 32 boys/ groups. In fast progressive scoliosis the most commonly observed curve patterns were double major and right thoracic curves. The analysis of muscle spindle function was based on EMG measurements of the unloading reflex in paraspinal muscles in children with fast and slow progressive sco- liosis as well as in healthy controls. The method used to elicit the unloading reflex was published in European Spine Journal, 1997:6;93-97.

EVALUATION OF VERTEBRAL AND RIB ROTATION USING REAL-TIME ULTRASOUND

A S Kirby, R G Burwell, A A Cole, R K Pratt, J K Webb, A Moulton

The Centre for Spinal Studies and Surgery and the School of Biological Sciences, Queens Medical Centre. Nottingham, UK

We have previously reported the development of a realtime ultrasound method to measure each of segmental vertebral laminal rotation (LR) and rib rotation (RR) in 20 scoliosis school screening referrals (Kirby et al 1997). The equipment included an Aloka SSD 500 portable machine with a 3.5mhz wide field of view (172 mm) linear array transducer. The patients lay prone on a couch with the forehead supported (mean Cobb angle 19 deg., range 0-37 deg., thoracic 5, thoracolumbar 5, lumbar 6, double 3, straight spine 1, girls 17, boys 3). Measurements of each of LR and RR were obtained twice by one observer (ASK). After walking around the room the child lay prone again and measurements of Ir and rr were repeated again twice. Pedicle and spinous process vertebral rotation (VR) were measured using the Perdriolle (t5-14) and Bunnell (tl-4) templates respectively by one observer (RGB). A statistically significant difference was revealed by level but not by measure or position. The 95% confidence limits for LR at each of the 18 levels (TI-SI) ranged from + 1.7o to +/3.1 deg.; and RR from +//- 1.8 deg. to +/- 2.8 deg.. Comparison of ultrasound LR against x-ray VR shows statistically ignificant correlations at each of T7-14, the scoliosis curves being mainly thoracolumbar and lumbar. VR on ap standing spinal radiographs is greater than ultrasound LR measured in the prone position. Summated ultrasound LR at TI-12 correlates significantly with summated ultrasound RR (p

GAIT ANALYSIS IN PATIENTS WITH SCOLIOSIS

I.A. Kramers-de Quervain, R. Muller, D. Grob, E. Stussi Spine Unit SchuIthess Klinik, Lengghalde 2, CH-8008 Zurich, Switzerland

A cerebral asymmetry or a developmental abnormality in the central nervous system have been implicated as possible causes of idiopathic scoliosis. The goal of this study was to identify asymmetries during level walking in scoliotic subjects which may be related to a neurological dysfunction or to the spinal deformity itself.

Subjects: Nine scoliotic females (median age 14.4 years) with a left lumbar and a right thoracic component of their curve were included in this study.

Gait analysis: consisted of 3D kinematic- (VICON) and kinetic- (Kissler Force Plates) analysis combined with dynamic electromyography of major lower extremity and paraspinal muscle groups. The traditionally recorded gait parameters such as time-/ distance parameters, verticalanterior-posterior and medio-lateral ground reaction forces, as well as the sagittal plane joint angles did not demonstrate a significant or clinically relevant asymmetry between the two sides in this group of young scoliotic subjects. However. a significant and marked asymmetry was seen as a torsional shift of the upper trunk in relation to the symmetrically rotating pelvis. This motion pattern was reflected by an asymmetry of the free external rotational moment and impulse. Further studies are needed to investigate if this behaviour is solely an expression of the structural deformity or if it could enhance the progression of the torsional deformity.

LOWER EXTREMITY DYSMETRIA, SCOLIOSIS AND LOW BACK PAIN

A. Manganiello L.D. Radiology - Via Savonarola 175 - 35137 Padova Italy

Diagnostic imaging shows that during growth lower extremity dysmetria, even of the order 3-4 mm, causes various scoliotic deviations. These are differentiated by the metameric site at which the compensatory counterdeviation of the spine starts, and by load distribution. Curves which are initially of mild degree, non-structural and mobile may subsequently diminish, remain stationary or progress and become structural. The course of curves is above all affected by the load distribution at the lumbosacral level since. if it is asymmetric. it may produce: wedging of the disc L4L5 and/or LS-SI; wedging of vertebral body L4 and/or L5. Pelvis torsion with asymmetric drop of the sacrum. These changes, by modifying the orientation of the resting surface of the sacrum and the basal vertebrae, cause spine deviation in the frontal plane on the side of the greater load. There will be either a progression or a reduction of the curve according to whether the lateral deviation due to asymmetric overload sets up an acting synergism or a compensation. If, however, the load distribution at the lumbosacral level is symmetric, the curve will usually be stationary. Nevertheless it is not possible to predict the course of the scollosis because of the shifting of the load due to the appearance of further compensatory curves, as well as posture change if not stabilised. The scoliosis is the fixing of a scioliotic attitude.

Furthermore, in dysmetric adults the asymmetric overload at the lumbosacral level may produce premature degenerative disease of the disc-somatic and the interapophyseal joints L4-L5 and/or L5-SI, as well as subluxation of the sacroiliac 'joint with anterior rotation of the corresponding ilium. These alterations, involving structures supplied with nociceptors, may give rise to low back pain of mechanical origin. The pain. with or without sciatica, more often affects the side of the lesser load if caused by hemiary pathology, otherwise it almost always affects the side of the greater load. The most frequent causes of sciatica are discal herniation and degenerative stenosis of the central canal, of the lateral recess and of the neural foramen

ANTEFLECTION STOP AND IDIOPATHIC SCOLIOSIS

H.-H. Matthias

Muenster/Germany

The flattening of the physiological curvature of the spine plays an important role in the origin of the deformation of the spine.

We asked whether a stop of anteflection in the spine plays a role in the origin of the idiopathic scoliosis (AIS). We measured kyphosis using a miniaturized DEBRUNNER Kyphometer. This device allowed us to measure the segmental mobility. By "Anteflexion Stop" we understand the diminution of mobility from habitual posture to ventroflexion in the sagittal plane.

We confirm that the function of the scoliotic-spine in the sagittal plane is anteflexion (TOMASCHEWSKI 1994).

In this study we examine. whether there is a connection between a stop of anteflection and the development of scoliosis (Measurement with the BUNNELL scoliometer).

We discriminated between a non-pathological-group (under 5 degrees) and a pathological-group-with greater than 5 degrees. We examined a total of 623 girls with a mean age of 12 years.

The result was that there was no differencebetween the pathological and the non pathological group . This is underlined with tabels and graphics.The significance of these findings for the judgement of the pathogenesis of AIS is discussed in detail.

CAN A DYNAMIC ORTHOSIS CHANGE NEUROMUSCULAR DEVELOPMENT IN IDIOPATHIC SCOLIOSIS?

G. Nijenbanning, A. Burleigh-Jacobs. A. Veldhuizen, R. Jacobs

Institute for Biomedical Technology, University of Twente, Enschede, The Netherlands

In erect standing, weighting of proprioceptive. visual and vestibular sensory information is important to obtain a reference frame of the body vertical alignment. Recently, several papers have suggested the idea that idiopathic scoliosis (ISc) results from disrupted sensory processing and/or cerebral asymmetry (Herman et al 1985: Goldberg et al 1995). The prospect of a neuromuscular component to ISc therefore has questioned the effectiveness of using a Thoraco-Lumbar-Sacral-Orthosis (Tl SO) for the treatment of scoliosis. In this presentation. we introduce a newly developed dynamic orthosis which may have a positive influence on the sensory contributions from axial trunk muscles.

GROWTH AND TRUNK ASYMMETRY IN A COHORT OF YOUNG ADULTS

M. Nissinen, lt. Heliovaara, J. Seitsamo, M. Pousa Laakso Hospital, National Health Institute, Occupational Health Institute, the Orthopaedic Hospital of the Invalid Foundation, Helsinki, Finland

Growth and trunk asymmetry of a total of 1060 children (515 girls, 545 boys) were followed up annually from the average age of 10.8 to 13.8 years to find out whether trunk asymmetry or growth profile has any prognostic value in the development of idiopathic scoliosis. Of the children 865 (80.70,t) participated In the final examination. These 855 children were invited to a re-examination at the average adult age of 21,9 years. Six participants had died during the follow-up time of II years and 44 did not have a permanent addresses. Of the 803 subjects invited, 430 ( 208 women and 222 men, 53.5% of those invited. 40.6% of the original cohort) were examined by one the authors The physical examination consisted of height. sitting height. weight, total arm length and log length inequality determinations. Trunk asymmetry was measured by forward bending Only 16.5ok of the adults were found to be symmetric (hump 0.2 mm) in the forward bending test, 53.5Ck had a hump of 3-7 mm and 30% had a hump > 8 mm at 21.9 years. Humps of 8 mm or more were as prevalent among women as men at the adult age (16.3% vs. 14.7%, P0.62). The prevalence of major trunk asymmetry -> I 0 mm was the same in both women and men in contrast to the findings at the prepubertal age of 10.8 years in this cohort . The fact that major trunk asymmetry is more prevalent among girls at the mean age of peak height velocity" (I 1.8 years) might explain why severe scoliosis Is more prevalent in girls as the rapidly growing spine could be considered more unstable. The boys experience their peak growth at a more skeletally developed stage and therefore are less scoliotic. Mild scoliosis may be assumed to be as prevalent in adult women as men on the basis this study. For ethical reasons the subjects were not X-rayed at the final examination.

EFFECT OF STRAP FASTENING IN BRACE TREATMENT OF ADOLESCENT IDIOPATHIC SCOLIOSIS: A PRELIMINARY STUDY

Y. Petit, C.-E Aubin. J. Dansereau, D. Gignac, J. Joncas, H. Labelle

Ecole de Technologie Superieure Montreal, Quebec, Canada

The treatment of progressive adolescent idiopathic scoliosis IAIS) with thoraco-lumbo-sacral orthosis (TLSO) is widely used. However. the relationship between the strap tension adjustment and the pressure applied by the brace on the patient torso as well as the 3D correction obtained by this treatment is not well understood. The objective of this study is to evaluate the effect of fastening the straps on pressure distribution.

This study was conducted on 7 patients (13+2 year old) with AIS presenting right thoracic or right thoracic and left lumbar curves and treated by Boston brace. A flexible mat with 192 thin pressure sensors was put under the patient's brace and strain gauges were installed along the thoracic and pelvic straps. The straps were in turn tightened at 20N and 60N and the patients were asked to take two positions normally taken during the day (standing and sitting). Each time, the pressure between the brace and the patient torso as well as the tension on the straps were recorded during 30 seconds.

This preliminary study indicates a different relationship between the strap tension and the pressure generated by the Boston brace from one position to another. However, this study should be continued on a larger number of patients in order to confirm these observations and to clearly identify these relationships as well as to evaluate the biomechanical effectiveness of brace treatments.

THE ROLE OF THE RIB CAGE IN INFANTILE IDIOPATHIC SCOLIOSIS (IIS)

R.K. Pratt, R.G. Burwell, J.K. Webb

The Centre for Spinal Studies and Surgery and the School of Biomedical Sciences, Queen's Medical Centre Nottingham, England

The ribs may cause scoliosis by transmission of abnormal muscle forces to the spine (Stromeyer 1836, Taylor 1904, Wojcik 1990), by unbalanced load transmission (Pal 1991) or by asymmetric rib growth (Sevastik 1984). What is the evidence for these mechanisms in ItS?

The X-ray films were reviewed of 13 patients with IIS treated by Luque trolley and convex epiphysiodesis who had 5 year follow-up. Spinal curvature (Cobb angle), vertebral tilt and rotation and the angle the ribs at the apex of the curve made with the TlSI line were measured. Spearman rank correlation coefficients and multiple linear regression analysis were used.

The findings are as follows:

1. Pre-operative concave rib angle predicts both the change in spinal curvature due to surgery (p= 0.003) and the spinal curvature at 5 year follow-up (p= 0.038).

2. During follow-up, the concave and convex apical ribs tend to move together (p= 0.027, r 0.58). The direction of movement of the convex apical rib during follow-up correlates negatively with the direction of movement of both convex (p= 0.019) and concave apical ribs (p=0.031) with surgery.

3. Apical rib angle changes were not associated with changes in spine length (TISI) at surgery (concave p= 0.31; convex p= 0.20) or during follow-up (concave p= 0.148: convex p= 0.886).

It is suggested that the concave apical rib angle indicates the extent to which the rib cage allows surgical correction of spinal curvature. Surgery forces a change in rib cage configuration which reverses during follow-up. Consideration of changes in spine length does not reveal evidence for the action of a muscle tether on the apical ribs.

In conclusion, the rib cage appears to act as a buttress to the spine, stabilising it against both deforming and correcting forces.

SYSTEM OF SCREENING AND PRIMARY DIAGNOSIS OF SPINE DEFORMITIES IN CHILDREN IN AMBULATORY-POLYCLINIC CONDITIONS

M.A. Sadovoy, N.G. Fomichev, TN. Sadovaya, VN. Samadsky

Novosibirsk Research Institute of Traumatology and Orthopaedics, 17, Frunze str., 630091, Novosibirsk RUSSIA

A computer system for estimation of spine deformities based on original principles of obtaining information was developed at Novosibirsk Research Institute of Traumatology and Orthopaedics. The introduction of the optical topography method opened new methods of description and registration of the spine status, diagnosing even insignificant spine asymmetry with a low risk of false negative results, simplified interpretation and provided possibility of examination of a large number of children during a short time.

We noted that clinico-radiological and topographic characteristics describing dorsal body surface both in normal and pathological cases have high degree of dependence on sex. age and constitutional peculiarities of a child which must be considered in determination of presence or absence of spine deformity. There are stable correlations between similar clinical, radiological and topographic characteristics of three plane measurements of dorsal body surface which provides the possibility to calculate radiographic characteristics using topographic ones on the basis of linear regression equations, without radiographic examinations of majority of children. Algorithms of analysis and interpretation of average meanings and deviation limits of the average meanings of seven basic topographic characteristics consisting of the following blocks: topographic tests, screening tests and primary diagnosis, allow to carry out general estimation of the degree of spine pathology expression on the basis of working classification of three plane assessment of spine deformity in children. Efficiency analysis of application of topographic screening and diagnostic test demonstrated its high specificity and sensitivity, which allow to use this method in wide clinical practice.

Optimal structure of regional children's spine diagnostic centre is represented by two main divisions: diagnostic and therapeutic. Diagnostic division provides screening examination of up to 100 children and consultation of up to 20 children per day. Therapeutic division provides courses of therapeutic exercises, massage and physiotherapy procedures for 30 patients a day. The centre has the possibility to carry out screening examination of 25 000 children, deep diagnostic examination of 2 000 children, and therapeutic courses for 750 children per year.

COMPUTER OPTICAL TOPOGRAPH FOR DIAGNOSTICS OF SPINAL DEFORMITIES

M.A.Sadovoy, VN.Samadsky, N.G. Fomichev. Novosibirsk Research Institute of Traumatology and Orthopaedics, 17, Frunze str, 630091, Novosibirsk, RUSSIA

A new automatic system (TODP) for 3-D back surface measurement to detect of the trunk deformity in scoliosis has been developed on a fringe projection technique and personal computer based image processing system. In comparison with ISIS and JENOPTIC formetric, the bestknown automatic systems for 3-D surface measurement, TODP has better accuracy, space resolution and more speedy processing. It differ from these systems in the method of image data acquisition. Instead of direct measurement of the projected line coordinate, the TODP uses spatial phase-detection method, that provides measurement of surface shape in the all image pixels without interpolation.

This system provides interactive graphical and quantitative shape analysis of back surface with evaluation of the series of topographic data, that correlate with traditional radiological and clinical parameters.

The TODP may be applied both for mass screening of children for early detection of spinal deformity, and for following treatment monitoring of patients with scoliosis and kyphosis.

THE POLYMORPHISM OF IDIOPATHIC SCOLIOSIS

J. A.Sevastik

Department of Orthopaedic Surgery; Huddinge University Hospital, Karolinska Institute, Stockholm, Sweden According to the SRS in 1969, idiopathic scoliosis is classified into three groups related to the age of the patient at the start of the deformity and in another seven groups with regard to the pattern of the curve . From a clinical point of view this sorting of the cases of IS is simple, convenient and sufficient. However, besides these rude differences regarding the clinical manifestations of the deformity there are also other known disparities between the three main forms of IS related to racial, geographical, hereditary, gender and other factors as well as to the pattern and the orientation of the curve. Moreover, differences in some somatometric variables related to the pattern of the curve have been reported in girls with adolescent IS. These known variations indicate the polymorphism of IS, which includes an aggregation of pathologic conditions, that besides the thoracospinal deformity have nothing else in common but their uknown etiology, i.e., the factor or factors causing the deformity, pathogenesis, i.e. the mode of origin of the morbid process and pathomechanism, i.e. the sequence of events in the evolution of the structural and functional changes that result from the pathological process. Therefore, a reapraisal of the classification of the idiopathic scolioses taking into consideration not only the clinical manifestations of the deformity but also the known particular characteristcs of each form seems to be a crucial prerequisite to promote research on issues related with the etiology (-ies) , the pathogenesis (-es) and the pathomechanism (-s) of the idiopathic scoliosis conglomerate and hence to create new methods for treatment based on pathophysiological concepts.

THE ROLE OF THE RIBS IN THE PATHOGENESIS OF IDIOPATHIC SCOLIOSIS

J.A. Sevastik

Department of Orthopaedic Surgery, Huddinge University Hospital, Karolinska, Institute,Stockholm, Sweden

The main results of a series of experimental and clinical investigations have shown: a) that rib osteotomy/shortening in growing rabbits induces scoliosis which was ascribed to stimulation of the longitudinal rib growth because of the fracture and led to the hypothesis that the spinal deformity in IS can be triggered by asymmetric longitudinal rib growth, b) that in 5 of 6 deceased women with right convex thoracic IS, the left ribs were somewhat longer than the right ones c) that the left breast in scoliotic, but not in normal, girls was significantly more vascularized than that of the right one, c) that there is a slight vertebral rotation towards the right predisposing the normal spine to rotate to the same direction, d) that in growing rabbits resection of 4 intercostal nerves, including the sympathetic fibres, leads to increased vascularity of the denervated hemithorax, increased osteogenetic activity at the costochondral junction, increased longitudinal rib growth and progressive scoliosis concave to the side of denervation,with decreased kyphosis and vertebral rotation to the convexity, e) that similar results were obtained by forced mechanical lengthening of one rib in rabbits and f) that in girls with early IS the 3-D structural vertebral changes appear simultaneously and not in any particular plane. The presented set of facts when analysed in relation to one another, lend strong support to a new, the thoracospinal theory of the pathogenesis of IS according to which overgrowth of the left ribs, due to hypervascularity of the ipsilateral anterior hemithorax, alters the equilibrium of forces controlling the alignment of the normal spine, as it is predisposed to rotate to the right, and triggers the thoracospinal deformity in IS. The theory better than any other hypothesis explains the mode of origin of at least the most common form of IS, with location on the thoracic spine, concavity to the left, apex at the T7-T9 level, vertebral rotation to the right, decreased kyphosis and almost exclusive affection of adolescent girls.

ARE RIB OPERATIONS A PHYSIOLOGICAL APPROACH FOR TREATMENT OF EARLY SCOLIOSIS?

Xiong, J. A. Sevastik

Department of Neurosurgery; Chen Zhou First Peoples Hospital, Chen Zhou City, Hunan, R. China

In conection with earlier anthropometric and experimental studies. related to the pathogenesis of IS, attempts also have been made to correct the spinal deformity by interventions on the ribs. Thus, right convex scoliosis induced in growing rabbits by resection of 3 intercostal nerves on the left side was corrected by mechanical elongation of one rib on the side of the convexity. In still another experimental study, in rabbits in which scoliosis had been induced by the same procedure resection of three intercostal nerves on the side of the convexity resulted in reduction or stop of the progression of the spinal deformity. Moreover, a 7-year-old girl with progressive thoracic IS and a Cobb angle of 46has recently been operated by 2 cm shortening of three ribs on the side of the concavity. One year after the operation the Cobb angle was reduced to 28-. The relation between thoracotomy or other operations on the thoracic cage and the developement of scoliosis is well documented . Rib resection or costodesis on either the concavity or the convexity of scoliotic curves has been used by several earlier authors either isolated or in combination with other therapeutic interventions. However, the results of these clinical trials in patients of different ages and with curves of varying etiology, degree and pattern have been largely disappointing . In the light of these reports and of the results of our studies it seems probable that operations on the ribs as shortening or resection on the side of the concavity or lengthening of ribs on the convexity of the curve if used under well defined indications may consist a new approach in the treatment of early thoracic curves. The vast implications of new, easy-to-perform and harmless operations based on physiological concepts on the overall treatment of early progressing curved need not to be underlined.

SCOLIOSIS AND RHYTHMIC GYMNASTICS

P. Tanchev, A. Djerov, A. Parushev, D. Dikov University Hospital of Orthopaedics "Gorna Bania", 56, N. Petkov St., 1614 Sofia, Bulgaria. Phone: 00359 2 55 01 42; Fax: 00359 2 55 30 63

In a survey of 100 girls, aged 11-15, having been trained actively in rhythmic gymnastics at least for a period of 5 years, we found a fivefold higher incidence of adolescent idiopathic scoliosis (AIS). These data were compared to the results from a screening of 4800 students of the same age, carried out in Sofia by the same examining team in 1996.

A detailed familial and past diseases anamnesis was registered in all the probands. General physical and meticulous back examinations were carried out, and X-rays were taken in all the AIS suspected cases. Special attention was paid to the presence of generalised joint hyperlaxity, which had been usually the main criterion for the primary selection of 5-year-old-girls to start training in rhythmic gymnastics. Eating regimen, weight-height relations, growth and maturing development (delayed menarche or secondary amenorrhea) were assessed too. Intensity, duration and specific moments of the training process were carefully analysed, and a significant asymmetric loading and an extreme physical overloading of the body were found.

This study supports the multifactorial origin of AIS, and strongly suggests the important role of the following aetiologic triad in disturbing the growth of vertebrae and intervertebral discs: generalised joint hyperlaxity as a hereditary characteristic; delayed maturing caused by physical, dietary and psychic stresses; and persistent asymmetrical overloading of the growing spine. These findings in girls, practising rhythmic gymnastics, generate a hypothesis which needs to be juxtaposed on the characteristics of nongymnast patients with AIS.

MECHANICAL MODULATION OF SPINAL GROWTH: CONTRIBUTION TO THE PROGRESSION OF SCOLIOSIS

I.A.F Stokes

University of Vermont. Dept. of Orthopaedics and Rehabilitation, Burlington VT 05405-0084, (iSA

In the present empirical approach to management of progressive scoliosis, it is generally accepted that the progression is at least partly due to mechanical modulation of spinal growth. Roaf (JBJS 42B:40-59,1960) described a "vicious circle" in which spinal asymmetry (lateral curvature) leads to asymmetrical loading of the vertebral physes and hence to asymmetrical growth (according to the Hueter-Volkmann principle). The rationale for brace treatment is based on this premise, and the failure of posterior spinal fusion in very young patients to arrest growth and progression (the crankshaft phenomenon) is testimony to the difficulty of controlling growth.

My own research is directed at several aspects of this 'vicious cycle' hypothesis. Since it is not well known how much of the scoliosis deformity is in the discs and how much is in the vertebrae, nor how this ratio differs by anatomical level of the curve, by diagnosis, and how it evolves with progression of the deformity, we have studied radiographs and found that the wedging is greater in vertebrae than discs at the apex of a thoracic curve, and the converse is true in idiopathic lumbar curves. The wedging ratio between discs and vertebrae did not change significantly with progression, contradicting the idea that the deformity originates predominantly in one of these two structures. There was no evidence of the cause of the scoliosis (idiopathic versus neuromuscular) determining the distribution of deformity between discs and vertebrae.

In order to quantify the gains in the positive feedback loop of the 'vicious cycle' we have to quantify the relation ship between the degree of scoliosis curvature and the degree of asymmetrical loading of the vertebral physes and discs. This has been addressed by means of a biomechanical analysis, since it is not possible to measure details of spinal loading directly. Also. we need to quantify the relationship between the loading of the spine and its growth. A rat tail model with vertebrae either axially loaded in compression or distraction, or asymmetrically loaded (curved tail model) have been used for this purpose.

COSMESIS IN ADOLESCENT IDIOPATHIC SCOLIOSIS

T. N. Theologis, R. J. Jefferson, A. H. R. W. Simpson, A. R. Turner-Smith, J. Cq T. Fairbank

Nuffield Orthopaedic Centre NHS Trust. Oxford

This abstract can be found at J Bone Joint Surg [Br] 1992; 74-B, Suppl I 9S.

SCOLIOSIS PROGRESSION AND BACK SHAPE

T. N. Theologis, A. R. Turner-Smith, J. C. T. Fairbank Nuffield Orthopaedic Centre NHS Trust, Oxford

This abstract can be found at J Bone Joint Surg [Br 1994;76-B(Supp 1):7

DIFFERENCES IN SAGITTAL THORACOLUMBAR CURVE DEVELOPMENT IN GIRLS AND BOYS AROUND PUBERTY

D. Uyttendaele, L. De Wilde

Department of Orthopaedic Surgery, University Hospital, De Pintelaan 185, B9000 GENT, BELGIUM

Biomechanical factors probably play an important role in the pathogenesis of adolescent idiopathic scoliosis and an explanation for the development of lateral deviation and rotation of the spine. characteristic of scoliosis, can be found in physical laws (Euler's theory). The sagittal dorsolumbar curve distribution of the spine may play an important role in the development of idiopathic scoliosis, a longer so-called declive segment increasing the risk for rotational deformity under the influence of postural forces. The purpose of this study was to look for differences in the sagittal dorsolumbar curve distribution in healthy boys and girls around puberty which might explain the preponderance of girls over boys in subjects affected by adolescent idiopathic scoliosis.

Using Moire topography, the spinal contour was studied in 292 girls (aged 9-16) and 191 boys (aged ll-18). Our results showed no differences between boys and girls in the youngest age group studied. However there is a different evolution around the maximal growth spurt. The length and inclination of the proximal proclive segment increase only slightly in girls but considerably in boys, and remains constant after the maximal growth spurt. This difference is statistically significant (p

The length and the inclination of the distal proclive segment increase in girls in all age groups studied. In boys these parameters remain almost constant. The results of this study corroborate the idea that girls are more prone to the development of idiopathic scoliosis because their spines with a relatively longer declive segment and a shorter and less inclined proximal proclive segment neutralise less easy rotation inducing forces than does the spine in boys.

SPINAL MOBILITY: CHANGES DURING GROWTH WITH STRUCTURAL SCOLIOSIS

S. Viola

Buda Children's Hospital, 1277 Budapest 23. POB 14. Hungary

A number of cases studies were followed for 10 years. Children suffering structural scoliosis were separated from a normal group. Spinal mobility and posture were compared with the mobility of normal healthy boys and girls aged 5-14 years in the recent longitudinal study. The range of motion was not physiologic in scoliosis: rotation, flexion thoracic hypermobility with decreased lumbar flexion ability was found. This study was started in 1980, involving 1200 5 years old children chosen from various nurseries. Measurement of spinal mobility and posture were taken. The same examination was repeated at ages 10 and 14 years in a longitudinal study. In addition the spinal movement of 45 suffers have been tested in a cross sectional study for controlling our results we have got in the longitudinal study. Lateral bending of spine with Howmedica protractor, thoracolumbar rotation with inclinometer, flexion ability of the spine with modified Shober method, and positional sagittal curves were measured. The comparison of reference and abnormal curves was carried out by statistical analysis. Both of spine rotation and flexion ability in structural scoliosis happened significantly higher (p=0.01), in addition the lumbar flexion ability and thoracic kyphosis were significantly lower (p(0.01) in scoliosis compared with the normal values of the healthy children in the longitudinal study even before the diagnosis. There were established in our control cross-sectional study.

DISCUSSION GROUP 3:

CENTRAL CONTROL

Chairmen - Marc Asher (Kansas City, USA) and John Swaddle (Bristol UK/Chicago, USA) Five sessions were held of this discussion group: the first reviewed the relevant S.R.S. literature summaries. In the second, third and fourth sessions 17 papers were submitted and discussed. A separate discussion took place on the role of developmental asymmetry.

In the last session possible lines of further research were discussed. No single hypothesis or group of hypotheses were proposed. However, any hypothesis that is proposed must accommodated these possibilities.

A sub clinical cell membrane defect in platelet/muscle that might be under genetic control.

Mounting evidence that the mid line hindbrain and/or mid brain is the likely location of the neuro-anatomical defect if one can ever be identified or the origin of a hormonal defect.

Any theory must accommodate the differences between stable and progressive scoliosis and the behaviour of scoliosis in boys and girls.

Further areas of promising central control research would seem to be:

Cell membrane defects.

Survey of diseases involving the mid line, mid and hind brain for scoliosis.

Refined MRI study and analysis, for instance radiologists may be undercalling tonsillar ectopia.

Further development of animal models of scoliosis.

Better understanding of hormonal interactions, particularly growth hormone, gonadotrophins, melatonin. and calmodulin.

The hypothesis proposed by developmental stability theory is that patients with large fluctuating asymmetries in skeletal characters will be more likely to develop a spinal deformity. This hypothesis is eminently testable through a detailed longitudinal analysis of morphological asymmetries. There were two principle suggestions for further research in asymmetry:

1) It is imperative to define the phenotypic appearance of the normal spine. This could be achieved by three-dimensional imaging and computer-graphic reconstruction of spinal segments in non-scoliotic patients. Developmental stability is the deviation from the norm, hence we.need to know the norm before we can fully assess this deviation. 2) There is a need for a detailed longitudinal study of spinal growth and development in scoliotic and non-scoliotic patients. For this to be achieved, researchers will need to focus their efforts on high risk patient categories such as familial scoliotic groups. If we can assess the ontogeny of asymmetry in spinal and skeletal characters we will be able to assess whether subtle skeletal asymmetries could be used as indicators of whether a patient is likely to develop a scoliosis.

THE PHENOMENON OF SCOLIOSIS PRODUCTION FOLLOWING PINEALECTOMY IN A YOUNG CHICKEN

K.M. Bagnall. Moreau, L. Raso, J. Mahood, X. Wang, J. Zhao, M. Beuerlein

Departments of Cell Biology and Anatomy, and Surgery, University. of Alberta, Edmonton, Alberta, Canada. T6G 2H7

Recently pinealectomy in young chickens has been shown consistently to produce scoliosis. Several studies, including one of our own, have shown that this scoliosis has many characteristics similar to those seen in patients with adolescent idiopathic scoliosis (AIS). Consequently this might be a valuable animal model for the study of AIS. There is no appropriate animal model currently available and an understanding of the mechanism behind the phenomenon has the potential to provide new insights into AIS and guide the way towards novel and innovative treatment methods. In a series of experiments we have explored the possibility that it is an artefact of the delicate surgery that is responsible for the production of the scoliosis but are, now convinced that some consequence, related to the actual removal of the pineal gland is the underlying cause. The main product of the pineal gland is melatonin and a logical assumption might be that low levels of melatonin are responsible for the production of the scoliosis. Indeed, some authors have reported success in treating the scoliosis with melatonin therapy. However, our own studies are in conflict with these results. For example, all our pinealectomised chickens have zero levels of melatonin and yet not all (45%) of the chickens develop scoliosis. Similarly, we, determined a physiologically appropriate dose of melatonin that would replicate the circadian rhythm of serum melatonin levels and applied it to young, pinealectomised chickens but with no effect. This melatonin therapy did not prevent the development of scoliosis nor did it affect ft progression of any scoliosis that had been allowed to develop prior to therapy administration. Our results and doubts about melatonin's. involvement in this phenomenon are also supported by findings of normal serum melatonin levels in patients with AIS although we appreciate that any differences might be recognisable only during the early stages of curve development. Our attention is currently being diverted from melatonin towards other products of the pineal gland and towards other substances such as growth hormone and calmodulin both of which have connections with melatonin. We recognise that there is a large phylogenetic gulf between avians and mammals but we believe that an understanding of the mechanism behind the production of scoliosis following pinealectomy in the young chicken will make significant contributions to the understanding of AIS in humans.

SCOLIOSIS, HORIZONTAL GAZE PALSY AND GENITAL DYSGENESIS IN THREE SIBLINGS IN TWO GENERATIONS

H. Berk, T. Berk, M. Bozkurt

Dokuz Eylul University; School of Medicine, Departments of Orthopaedics and Traumatology and Ophthalmology, 35340 Inciralti, Iznmir, Turkey

Adolescent Idiopathic Scoliosis has been shown to be related with postural disequilibrium, vestibular abnormality, defective proprioceptive functions and ocular motor disturbances. Scoliosis in association with horizontal gaze palsy (HGP) have been reported in literature. We report three new cases with scoliosis, HGP and genital dysgenesis. These three patients have two distinct differences from previously reported 39 patients. They present two generations of same stigmata of scoliosis and horizontal gaze palsy. Two female patients have genital dysgenesis that has not been described previously.

These patients clearly represent a distinct syndrome.

HOW DOES SKELETAL GROWTH AFFECT THE ONSET AND PROGRESSION OF ADOLESCENT IDIOPATHIC SCOLIOSIS? THE RED QUEEN CONCEPT APPLIED TO TRUNK GROWTH AND 3D BALANCE. A SPECULATION

R. G. Burwell, P H. Dangerfield

Skeletal growth, in contributing to scoliosis pathogenesis, is generally thought of as acting directly on the deforming spine. It seems likely that there is a second, hitherto unsuspected, mechanism which involves the changing relation between the increase of skeletal mass during growth and the developing nervous system of the child. The pattern of gait changes during growth. There is evidence that the early intense learning process for automatic movements such as gait involves: (I) neuronal plasticity at synapses for which neurotrophic factors are needed and which are present in sensory nerves; and (2) the cerebellum as a proprioceptive mismatch corrector (PMC) for central pattern generators (CPGs) and lower motor neurons. To act as a PMC the cerebellum is in receipt of input from both CPGs and the periphery . Growth increases skeletal size which, in the legs and hips, will generate increased forces in trunk muscles during gait accompanied in some individuals with asymmetries. In order to balance these changing and asymmetric forces in the lower trunk, the nervous system must continually produce appropriate and changing muscular contractions in the upper trunk during physical activities. Such information from the lower trunk will arise from the periphery (muscle spindles, tendons and joint capsules) and enter the spinal cord through sensory nerves leading to centres (1) cerebellum and (2) cerebral cortex with corrective action through the central pattern generators controlling upper trunk muscles. The motor output to the upper trunk must keep pace with the changing sensory input from the lower trunk - and the two must 'run' together like the Red Queen and Alice with the forest (Alice through the looking glass) or fail. This biological view provides another example of the `Red Queen Concept'. It is suggested that any unilateral impairment of the `proprioceptive spinocerebellar loop' involving the periphery, sensory nerves, spinal cord, brain stem, cerebellum or cerebral cortex, leads to the cerebellum as a PMC driving the trunk muscles inappropriately (imbalance) and so to a secondary spinal growth torsion, namely idiopathic scoliosis. Other possibilities include unilateral impairment of the `proprioceptive spinocortical loop', or loops involving the vestibular apparatus and/or the eyes.

A MULTIFACTORIAL CONCEPT OF THE CAUSATION OF IDIOPATHIC SCOLIOSIS (IS) R.G. Burwell, PH. Dangerfield

Currently there is no generally accepted theory for the cause of IS so treatment is pragmatic and not founded on aetiopathogenetic knowledge. Recent MRI evidence shows that 20-25% of children less than 12 years of age with putative JIS have some detectable neurological abnormality including Chiari malformation associated with a syrinx or an astrocytoma in the cervical cord In adolescent idiopathic scoliosis a neurological deficit relating to the afferent side of the reflex postural control system has been detected which may operate only under dynamic conditions.

Scoliosis curve progression in our view is determined by how much the immature vertebrae remodel under the eccentric loading created by neurological imbalance in the trunk in the presence of gravitational forces. Such imbalance is mediated through locomotor central pattem generators (CPGs) in the spinal cord. The concept involves an initial developmental disturbance in the peripheral and/or central nervous system - possibly at different sites in different subjects - causing asymmetric function of locomotor CPGs which control trunk movements in 3D. The resultant asymmetry of trunk muscle function together with gravity cause a gradual alteration of spinal growth as a secondary growth-induced torsion of the spine termed scoliosis with inevitable anterior spinal overgrowth. This torsion is a final common pathway of deformity from different aetiologies in muscle, nerve or central nervous system.

ADOLESCENT IDIOPATHIC SCOLIOSIS - THE CORRELATION OF CURVE SEVERITY, SEP AND MRI FINDINGS

J. C. Y Cheng, Y. L. Chan, X. Guo. H. L. Sher

Dept Orthopaedics & Traumatology, The Chinese University of Hong Kong

Disruption of the postural reflex system was found not uncommonly in patients with adolescent idiopathic scoliosis. Results of our recent SEP study in 147 AIS patients showed functional deficiency of somatosensory pathway in 11.57% of patients. The correlation with the clinical severity, the progression and the underlying neuropathological basis are not clear. The aim of the study was to correlate the disturbance of somatosensory function with MRI studies in AIS patients with different curve severity. Subjects included 157 new adolescent patients presenting with idiopathic scoliosis who was first seen and has not undergone any treatment before. They were subdivided into 2 groups : Group I included 135 adolescents with minor curve of Cobb's angle between 20-40 degrees and Group II with 22 patients with severe curve of Cobb angle more than 40 degrees . A control group (Group Ill) of 30 normal adolescents with no scoliosis were also included in the study. Posterior tibial nerve-somatosensory evoked potentials (PTN-SEPs) and Median nerve-somatosensory evoked potentials (MN-SEPs) were used for neurological functional evaluation, and MRI of the brain and whole spine for detecting structural deformities in the spinal cord and brain stem. This study demonstrated a significantly higher correlation of SEP and MRI abnormalities with clinical severity of the curve in AIS. The further high correlation of SEP and MRI points to a neuroanatomical basis for the disturbance of somatosensory function, which may play an important role in the etiology and progression of AIS.

FLUCTUATING AND DIRECTIONAL ASYMMETRY IN IDIOPATHIC SCOLIOSIS

PH. Dangerfield, D. Scutt, S. Ashton, J.T. Manning, J.C. Dorgan

The University of Liverpool, Liverpool, UK. Entail: spine92@liv.ac.uk

Asymmetry is a well known physical feature of scoliosis, expressed as the linear asymmetry of the spinal curvature found in the vertebral column, and the directional asymmerry (DA) of the limbs. Fluctuating asymmetry (FA) is a deviation from bilateral symmetry in normally bilaterally symmetrical morphological traits are normally distributed in a population and in which signed differences between the right and left sides have a mean value of zero. An important feature of FA is that it gives a composite reflection estimate of the well-being of an individual in their environment. Developmental stability within a species may be measured in two ways, namely by fluctuating asymmetry where there is a small random deviation from normal and by phenodeviants, where there is a marked deviation from normal developmental traits. The marked deviation from normal development found in idiopathic scoliosis, particularly in growth, is an example of the latter. The present study has examined the relationship between scoliosis and FA. A sample of patients with idiopathic scoliosis, attending the Royal Liverpool Children's Hospital, was measured for both DA and FA using an anthropometer and vernier caliper. Upper limb symmetry was recorded by measuring both limbs. 8 traits of FA were recorded with true FA being identified in all 8 traits. The results indicate that marked directional asymmetry is present in cases of adolescent scoliosis. Additionally, for FA, the relationship between the degree of spinal curvature and the FA is a positive one with a more severe curvature being associated with a higher degree of FA. This effect was very significant when FA was summated across the traits. These initial findings indicate that a major phenodeviant in the form of idiopathic scoliosis is associated with a general developmental instability, a finding which has important aetiological consequences for scoliosis. The directional asymmetry identified is an important component of the overall developmental problem of scoliosis.

QUANTITATIVE ANALYSIS OF THE VOLUNTARY TRUNK MOVEMENTS AND THE GROUND REACTION FORCES FOR EVALUATION OF SPINAL DEFORMITIES

K. M. Diab. H. Hirschfeld, J. A. Sevastik Dept. of Orthopaedic Surgery, Huddinge University Hospital, Karolinska Institute, Stockholm, Sweden Performance of trunk movements require an interaction between different trunk segments and the control of body equilibrium. Understanding of the global movement characteristics of the normal and deformed spine is important for improving the evaluation and the treatment of a spinal deformity. Voluntary trunk movements such as side bending and trunk rotation to the left and to the right side have been studied in a group of eight girls with AIS. The obtained results were compared with the values from a control group of six girls of similar age. Different body kinematics have been recorded with 2 CCD cameras and a digital motion analysis system (ELITE). Simultaneously, the ground reaction forces (GRF) were recorded from 2 AMTI force plates. Nineteen passive reflecting markers were attached to anatomical landmarks on the dorsal aspect of the subject who was standing on the two force plates. Off lines the centre of the mass (COM) and the centre of pressure (COP) were computed. Segment angles of the upper and lower thoracic and the lumbar spine segments, as well as the projection between these segments were calculated vs the X-axis (medial-lateral) during side bending. Furthermore, the shoulder and the pelvic segment angles were computed and analysed vs the Z-axis (anterior-posterior) during trunk rotation. The results showed that in the controls during side bending to the right, the movement direction of the head and COM was consistently to the right; in AIS the COM moved to the right and to the left in a inconsistent manner. The controls had a distinct loading and unloading phase of vertical GRF, which was not seen in AIS. Even though, the amplitude of displacement of the individual markers during side bending was almost similar to the left and to the right in both groups, the spinal projection angles differed. AIS moved the segments with similar amplitude in the same direction resulting in a small projection angle, while controls moved the spinal segments individually. The relationship between total COP and COM showed that controls had a consistent strategy while AIS lacked in consistency. During trunk rotation, the segment angle of the pelvis was kept stable when the shoulder segment rotated in controls; the AIS subjects rotated both segments in the same direction.

The results of these studies indicate that quantitative analysis of the voluntary movement strategies by means of GRF patterns and body segment changes in space highlights and increases the understanding of the relation of adaptive changes caused by spinal deformity, such as AIS. Moreover, these results suggest that the CNS governs the interaction between posture and voluntary movement in different way in AIS than in control subjects.

MELATONIN: A POSSIBLE ROLE IN THE PATHOGENESIS OF HUMAN IDIOPATHIC SCOLIOSIS

J. Dubousset, M. Machida

Our previous experimental studies using chickens and rats suggest that a defect of melatonin secreted from the pineal body contributes to the etiology of the experimental scoliosis. The present study was undertaken to examine the possibilities that a defect of melatonin may also exist in human idiopathic scoliosis. We measured the melatonin levels during 24-hour period in 45 adolescents, 30 patients with idiopathic scoliosis and IS age matched normal controls. In all subjects the melatonin levels showed diurnal variations with low in the day and high at night. We found a highly significant decrease in the melatonin levels in patients with progressive scoliosis throughout the night when compared to those with stable scoliosis or control subjects. Melatonin levels were 1/3 or 35% less in the progressive scoliosis when compared to the stable scoliosis patients; the levels in the stable group and the control were similar. Melatonin levels also appeared to be a possible predictor of spine curve progression, i.e., the lower melatonin levels associated with the greater prognosis of curvature. There was no correlation between degree of curvature of the spine and the melatonin level.

SCOLIOSIS: IN DEFENCE OF BIOLOGY

C.J. Goldberg, D.P. Moore, E,E. Fogarty, FE. Dowling Children's Research Centre & Our Lady's Hospital for Sick Children, Crumlin, Dublin 12, Ireland

Since the time of Hippocrates. the multiple causes of spinal deformity have been recognised but simple mechanical means have been applied to its control. Investigators, while acknowledging general ignorance of the pathogenesis, have concentrated on a disease concept operating through the mechanical effect of faulty posture, unspecified gene action and sub-clinical neuromuscular defect. A re-appraisal from first principles led to a radically different approach which can explain many long-accepted features of spinal deformity as well as some of the irritating minor observations that defy mechanical explanation.

This study took as starting point the observation that most adolescent idiopathic curves have a right convexity in the lower thoracic region. Hand-preference and other modalities of cerebral lateralisation were investigated in adolescent idiopathic scoliosis. Dermatoglyphics (palm prints) in all categories of spinal deformity were studied, following previous reports from other centres of increased anatomical asymmetries in idiopathic scoliosis. Left-right differences in the organisation of both the motor system and a cerebral cognitive function suggested effects beyond the nervous system in idiopathic scoliosis. Analysis of dermatoglyphics (palm prints) in several different classes of spinal deformity (congenital, all idiopathic, miscellaneous and minor asymmetry) showed a pattern of asymmetries that are compatible with theories of developmental instability.

Biomechanical analyses of forces acting in scoliosis describe well the established deformity, but cannot be their own causation. Biological theories of growth and development can better account for the manifold observations in this condition.

GROWTH PATTERNS IN CHILDREN WITH CONGENITAL VERTEBRAL ANOMALY

C.J. Goldberg, D.P Moore, E.E. Fogarty, F.E. Dowling Children's Research Centre and Orthopaedic Department, Our Lady's Hospital for Sick Children, Dublin 12, Ireland There is a general impression that children with congenital spinal anomalies are of shorter stature than their peers, although it is unclear whether this is due to the bony anomalies themselves, to spinal surgery at an early age, to concomitant cardiac or renal disease or to a primary failure of growth.

A preliminary search was made through the scoliosis database to determine the extent of the problem. Records were reviewed and sequential heights and weights were plotted against clinical growth standards for Irish children. A total of 133 children (83 girls and 50 boys) born since 1969 have been identified with height and weight documentation sufficient to draw conclusions about their growth trajectories. 70% were consistently below the 50th. centile for age on both parameters, which is statistically significant. This height discrepancy could not be attributed to surgery or to the existence of other major anomalies in all cases.

This finding confirms the general impression of diminished stature in congenital vertebral anomaly. However, most children are above the 3rd centile, even at maturity, although there are a few marked exceptions, and 12%c have been consistently below the third centile. Those who have been referred for endocrine assessment have not shown any hormonal abnormality and the reason for the short stature is not clear, since it is evident even in children with lesser vertebral anomalies who have not required corrective or prophylactic surgery.

There are several possible mechanisms which might singly or in combination contribute to this general diminution of stature: intercurrent ill-health, paucity of growth centres,' endocrine disturbance and surgery. In addition, we cannot at present rule out the possibility that some of these children are showing familial short-stature, raising the possibility that this might itself predispose to congenital vertebral anomaly. To explain the observation totally in terms of vertebral malformation may be an oversimplification.

MELATONIN RECEPTORS IN CHICKEN PARASPINOUS TISSUES

G. G. Gupta, J. Kelly, L. Lattanza, R. Rattay, D. Pendergrass, B. Millington

Section of Orthopaedic Surgery, Children's Mercy Hospital, Kansas City, Missouri, USA

Machida reported a pinealectomized chicken model in which pineal gland transplantation markedly reduced scoliosis development. Other centers have reproduced the pinealectomized chicken model. The chicken pineal gland excretes the majority of serum concentrations of melatonin. Melatonin effects end organs via melatonin receptors. The high affinity, low affinity, and intra nuclear receptors are well characterized in several mammilian and non-mammalian species in various portions of the central nervous system and retina. The high affinity of 125-lodomelatonin and the melatonin receptor allows non-neural tissues to be evaluated. In previous studies, our laboratory confirmed the presence of melatonin receptors in chicken paraspinous muscles. This study evaluated paraspinous muscle melatonin receptor binding and affinity in pinealectomized and non-pinealectomised chickens. Six mountain hubbards from the same hatching were divided into pinealectomised and control groups. Pinealectomy was performed with in one week of hatching. The average melatonin receptor total binding for control was 1.98 +/- 0.15. The average melatonin receptor total binding for pinealectomised chickens was 2.02 +/- 0.56. Values are standardised per mg of protein. The following table gives values per specimen for binding. We did not find a significant difference between the melatonin receptor total binding in pinealectomised versus control birds. We did see an increase in standard deviation in the pinealectomised birds. Limitations of this study include the small study size. Only one pinealectomised bird had scoliosis, The binding assays were reproducible and reflected published values for the melatonin receptor binding.

MOTOR UNIT ANALYSIS OF PARASPINALIS MUSCLES IN IDIOPATHIC SCOLIOSIS

B. Koo, J. C. Y Cheng, X, Guo

Departments of Pediatrics and Orthopedics, Princes of Wales Hospital, Chinese University of Hong Kong, Hong Kong

The objective of this study was a quantitative Electrophysiological Analysis of Motor Units of Paraspinalis Muscles of Patients with Idiopathic Scoliosis for Evidence of Asymmetry. 24 female patients with idiopathic scoliosis were investigated. Their ages ranged from 11-15 years (mean 12.58 + 1.32 years). Cobb's angles were measured for each patient from x-ray spines. Electromyography (EMG) was performed on paraspinalis muscles of both sides at level of maximum curvature. Studies were conducted with concentric EMG needles using a Nicolet Viking Ile EMG equipment with a filter setting of 2 Hz - 10 kHz.

A total of 10 -15 motor units were analyzed from each side of each patient using the quantitative motor unit analysis software (Nicolet). Mean motor unit duration and mean motor unit amplitude from each side were calculated and correlated with the Cobb angle. Motor units were also analysed for waveform morphology and extent of polyphasia. There were no significant difference in motor units of concave side versus convex side when the Cobb angle is 30. However, in the group with Cobb angle >30o there is electrophysiological evidence of neurogenic changes in paraspinalis muscles of convex side with increase in duration and amplitude of motor unit potentials as well as increased polyphasia.

The neurogenic changes from the paraspinalis muscles of the convex side is likely secondary to the spinal deformity rather than primary cause of the scoliosis as they are seen only in the presence of more severe scoliosis. These emg neurogenic changes in patients with scoliosis also suggests possible intrinsic susceptibility of paraspinalis muscles to neurogenic injury resulting in muscle power imbalance between the two sides predisposing them to develop scoliosis.

EXPERIMENTAL SCOLIOSIS IN CHICKENS AND RATS

M. Machida, J. Dubousset Paris, France

There are a number of theories on the etiology and pathogenesis of idiopathic scoliosis. These include genetic, musculoskeletal, metabolic and chemical factors, and abnormalities of central nervous system. Experiments on various animal models have suggested possible anatomical or functional influence for each of those elements for the etiology of idiopathic scoliosis.

We have found that pinealectomy in chickens consistently produced scoliosis. The intramuscular implantation of the pineal gland into the pinealectomised chickens prevented the development the experimentally induced scoliosis. Also, intraperitoneal injections of melatonin reversed the effects of pinealectomy in chickens. Furthermore, we used rats instead of chicken because rats are mammals and phylogenetically close to human than chicken. The scoliosis developed only in pinealectomised bipedal but not quadrupedal rats. The melatonin was effective in preventing the development of scoliosis. Our studies suggested that melatonin synthesis or metabolism may be important for the symmetric development of paraspinal muscles and straight spine growth. However, melatonin deficiency secondary to pinealectomy alone does not produce scoliosis if quadrupedal condition is maintained. Bipedal condition, like chicken or human, plays an important role for scoliosis to develop. The findings suggest a critical influence of postural mechanism in the development of scoliosis.

A FEEDBACK LOOP CONTROLS SPINAL STABILITY

V J. Raso, M. J. Moreau, K. Bagnall, J. K. Mahood, D. L. Hill, H. Jiang

CHA - Glenrose Rehabilitation Hospital and University of Alberta, 10203 - III Avenue, Edmonton, Alberta CANADA TSG OB7

A theory for the development of idiopathic scoliosis is the presence of a proprioceptive defect that causes a miscalibration of the neural/musculoskeletal mechanisms that maintain the spine in a balanced upright position. The control of spinal alignment requires neural elements able to detect mechanical action and pass on this information for interpretation by higher structures.

We conducted three studies: (i) to identify ligaments that were sensitive to lateral bending, (ii) to determine if there were neural elements in these ligaments able to detect mechanical actions and (iii) to establish the pathways taken by the neural response to mechanical stress. The morphology, sites of attachment and dimensions of the superior costotransverse and the intertransverse ligaments were determined from thoracic level 7 to level 10 on 32 human cadavers. The contralateral location of the intertransverse ligament (ITI,) at the ends of the transverse process made them 4-6x more sensitive than midline ligaments to lateral bending. The neural structures in midline and lateral ligaments of the spine were identified in 36 White Leghorn chickens aged 0 to 13 weeks. Single nerve fibres were present from birth and by 7 weeks large nerve bundles and Ruffini corpuscles were visible. Ruffini corpuscles are able to detect stretching of a ligament. In the final part of this study, the ITL on the right side at T3-T4 in 4 week old chickens was mechanically stimulated repeatedly for 60 minutes. Various areas of the nervous system were sectioned and processed immunohistochemically to identify sites of Fos production.

Stretching a ligament stimulates the sensory nerve-endings and activates the transcription of c-Fos in the neuron with subsequent production of Fos protein in the nerve cell body. Continued activation of stimuli by way of synapses to interneurons and neural nuclei describes a pathway of Fos production. By tracing the expression of c-Fos, we found that mechanical stimulation of the ITL generates fos protein along the nerves to the spinal cord and from the cord to the thalamus and vestibular nuclei. We concluded that there is a feedback control loop present and suggest that a defect in this control loop may cause idiopathic scoliosis.

WHAT IS ADOLESCENT IDIOPATHIC SCOLIOSIS, ANATOMICAL OR NEUROMUSCULAR ASYMMETRY? ELECTROPHYSIOLOGICAL INVESTIGATION AND CLINICAL IMPLICATIONS

P. Rogala, J. Huber, A. Nowakowski

Department of Orthopaedics and Department of Pathophysiology of Locomotor Organs, Karol. Marcinkowski University of Medical Sciences, Poznan, Poland

Stable development refers to the ability of an individual to maintain a consistent phenotype during growth, despite the destabilising forces of physiological stress. One characteristic of an ideal morphology is bilateral symmetry. We used Somatosensory Evoked Potentials (SEP's) during spinal cord monitoring in two generations of patients with A.I.S. We reviewed 29 girls aged 11 to 18 years with A.I.S. Global EMG amplitudes after effort tests were recorded by bipolar transcutaneous electrodes from m. rectus femoris, m. tibialis anterior, m. gastrocnemius, and m. extensor digitorum brevis of both lower extremities.

We found significant asymmetry in functional potentials of amplitudes of motor units in EMG leads from different group of muscles on the concave and convex side in idiopathic scoliosis. There was no significant difference between the left/right sizes of the bones. The secondary trunk configuration deformity in adolescent idiopathic scoliosis depends on the magnitude and direction of the scoliotic curves and rib cage deformity. The recorded global EMG from the other tested group of paravertebral and lower extremities muscles and electrophysiological tests of peroneal nerves showed small differences in spite of both sides of scoliosis. There is a neurological asymmetry in nonspinal sensory system with the defect of the motor brain cortex.

GH - IGF-I AXIS IN YOUNG PATIENTS WITH ADOLESCENT IDIOPATHIC SCOLIOSIS (AIS) AND SCHEUERMANN'S DISEASE (SD) COUPLED WITH AIS

S. Viola, F. Peter, 1. S. Nagy, I. Gyorgy Buda Children's Hospital, 1277 Budapest 23, POB 14, Hungary

Willner found higher values of GH in patients suffering from AIS . Our study confirms this publication to some extent especially in the cases with high progression potential. We found high resting GH levels only in AIS patients with Cd progression of more than 1 degree per month. Our results showed a great parallelism between the measured basal GH and IGF-I levels confirming the presentation of Scogland on serum Somatomedin C values. In summary we raise again the importance of GH-IGF axis in AIS and SD with idiopathic scoliosis and complete the previous observations with the suggestion that these parameters may indicate the trend of progression of AIS without or with SD. One hundred and six patients with AIS and 118 with SD were studied. 24 patients with adolescent idiopathic scoliosis (AIS) (21 female, 3 male) having an average age of 13,5 (minimum age 5, the maximum 17 years) as well as 20 adolescents with SD (14 male and 6 female) having average age of 13,5 years, (min. 6. max. 19 years) who were identified clinically and radiologically, were recruited for the hormone study. The bone age, height, weight, Tanner staging for puberty development. and Cobb degrees(Cd) for progression of scoliosis were measured in all patients, and the severity of SD was outlined by the Fisk graduation, the basal GH, IGF-I, (44 children) were also measured. The height percentile proved to be the highest in Tanner St. 1-3 (on the 80th percentile on average), but normal in Tanner stage 5.c. In the SD group the peak value is in Tanner st.3, but the height seemed to be above the average in all Tanner stages. The bone age in relation to the chronological age increased by I year on average in the AIS group, and decreased in the SD group by I year on average. The tested GH levels in AIS: five adolescents from this 24 AIS patients have elevated basal GH levels (21%). It seems to be the highest Cd per month values parallel with the elevated GH levels. Twelve have an IGF-I level score above 1.0 standard deviation (50%) in relation to the referencial average values according to the Tanner stages. There is a significant correlation between the elevation of scores and Cd per month (p

DISCUSSION GROUP 4:

ENDORGAN

Chairman - Dr Jill Urban, Dept Physiology, University of Oxford, UK In group 4 we examined the end-organs - ie the tissues ir which the scoliotic deformity is actually manifest - princi pally vertebral bodies and intervertebral discs. There were several main points which arose. 1. Are there subgroups of AIS which can now be distinguished and which could repay further study" This question arose from a study of bone density in AIS patients in Hong Kong - In a large study (250 controls, 75 patients) bone density of 20% of AIS patients was > 2 standard deviations below that of age matched controls, and in these patients, unlike controls, there was little increase in bone density with age. We concluded it would be useful to repeat the study among other scoliosis population groups. This study also demonstrates that there are distinct subgroups of AIS and others may come to light by careful examination of endorgan tissues. Another example of a distinctive group are those with a fibrillin defect.

2. There is a lack of knowledge of normal spinal tissue. This problem affected the conclusions of many of the studies discussed. Tissue can be obtained from scoliotic discs and other spinal structures from segments removed at surgery. This tissue can be studied for its composition, cellular activity, innervation etc. There is now an increasing body of knowledge about the composition of scoliotic discs. Sally Roberts has mapped differences in changes towards convexity relative to concavity in scoliosis. However, because there is no information on aged matched controls, we have no idea if the relative levels of these components are normal or higher or lower in either region of scoliotic discs and thus have no idea of how these changes may affect functional properties of the disc.

Our group felt that setting up a tissue bank for study of spinal structures would be invaluable for the study of changes in these endorgans in scoliosis - and also in other spinal disorders.. We did not discuss in detail how this could be organised but concluded that we are most likely to obtain suitable tissue from organ donors. This would aid in drawing up an atlas of composition of normal relative to scoliotic discs and may help isolate at present unsuspected subgroups of AIS.

3. There is a lack of a good animal model for scoliosis. Whether the defect is primary or secondary, we need some means to follow the consequences of change longitudinally. At present this is not really possible in humans, except by the use of imaging. This gives some indication of change at a structural but not a biochemical or cellular level. Another animal model of potential importance in study of scoliosis is transgenic mouse. Here a gene is deleted from /inserted in to the mouse and the effect of this gene deletion (or insertion) on the mouse's structure and function can be followed. This technique has great potential importance. If a gene concerned with scoliosis is identified, it will still be necessary to determine how and whether a this gene can produce a scoliotic phenotype; the transgenic mouse provides and opportunity for this. However present experience with these mice shows that the system has enormous redundancy and transgenic KO s of molecules thought to be essential - eg tenascin which is apparently switched on selectively during development, or type X collagen apparently essential during bone growth and calcification - were apparently normal or with only most subtle of defects.

4. The final question was the question of reversibility. At what stage of endorgan deformity is the scoliotic defect no longer reversible - for example, when are the changes in the disc or other soft tissues irreversible so that it is no longer sensible to brace. For example, scoliotic discs calcify extensively and occlude path for transport of nutrients from blood supply disc - once calcified, changes in cells and matrix of the disc are probably irreversible. However other changes may be irreversible at an earlier stage and these needs to be identified. Possibly again the use of an animal model, while realising there limitations, would aid progress in this respect.

OSTEOPENIA IN ADOLESCENT IDIOPATHIC SCOLIOSIS

J. C. Y. Cheng, X. Guo, H. L. Sher

Dept. Orthopaedics & Traumatology, The Chinese University of Hong Kong

The present study tested the hypothesis that adolescents with idiopathic scoliosis have an increased risk of osteoporosis compared with the general adolescent population. Baseline bilateral proximal femoral bone mineral density (BMD) was evaluated using dual-energy x-ray absorptiometry (DEXA) on 75 girls with adolescent idiopathic scoliosis (AIS) (aged 12 to 14 years old, Cobb's angle of the major curve between 12 and 46 degree). 220 age and maturity matched healthy girls served as normal control.

This study supported the hypothesis that adolescents with idiopathic scoliosis are at increased risk of osteoporosis compared to general paediatric population. Moreover the persistently lower BMD and the lower rate of increase of BMD in these patients could predict a significantly lower peak bone mass in to adulthood with all the long term problems associated with osteoporosis.

MATRIX TURNOVER IN THE SCOLIOTIC DISC

S. Roberts, J. Menage, E. H. Evans, V. C. Duance. J. Crean, S. M. Eisenstein

Centre for Spinal Studies, Robert Jones and Agnes Hunt Orthopaedic Hospital, Oswestry, Shropshire. SY10 7AG

The mechanical properties of the intervertebral disc are a reflection of the composition of the matrix, the major components being collagen, proteoglycan and water. It is known that the composition differs across the wedged, scoliotic disc, with the thinner disc at the concavity having less proteoglycan, water and collagen than the contralateral tissue. How this difference arises, however, is not known. We have investigated indices of catabolism and anabolism to determine if there is a difference in the balance of these processes at opposite sides of the disc. The enzymes, matrix metalloproteinases (MPs) are capable of degrading all the known macromolecular constituents of the disc matrix and are therefore taken as an indication of catabolism. Indicators of anabolism are reducible collagen crosslinks, found in newly synthesised collagen, and expression of the proteoglycan neo-epitopes, 3B3(-) and 7D4. These are found at elevated levels in young developing connective tissues and in some repair processes. All these parameters have been measured and compared in two localities in each scoliotic disc, one nearest the convex aspect of the curve and the other nearest the concavity, in patients with idiopathic scoliosis undergoing corrective anterior fusion. The levels of MMPs 2 and 9 at the concave side of the scoliotic disc were similar to those found in non-scoliotic discs (approximately 1.0 units/mg dry weight). However, towards the convexity higher levels were found (approximately 1.5-2.0 units/mg). The concentration of crosslinks was also lower at the concave side than at the convex side, with the values from non-scoliotic discs falling between the two. Similar trends were seen with the proteoglycan neoepitopes. Hence there are lower levels of all parameters measured in the thinnest part of the scoliotic disc nearest the concavity, than at the opposite side. These results suggest that there is the greatest matrix turnover, both synthesis and degradation, towards the convexity. However, relative to comparable, non-scoliotic discs the level of MMPs (taken here to indicate catabolism) are of a similar level at the concavity. One mechanism which could explain the relative differences with location would be if the loads differed across the scoliotic disc since applied load is known to influence cell metabolism.

THE ROLE OF THE NEUROCENTRAL CARTILAGE IN THE PATHOGENESIS OF "IDIOPATHIC" SCOLIOSIS - AN UPDATE

K. Tucker, B.A. Taylor

RNOH, Stanmore, Middlesex, UK

The purpose of our study is to demonstrate: The significance of the neurocentral cartilage in the aetiology of scoliosis and to experiment as to whether early, minimally invasive, surgery could be effective in preventing progression of the scoliosis and achieve correction of the associated rotational deformity. It has been postulated that the neurocentral cartilage has a fundamental importance in the pathogenesis of idiopathic scollosis, but this is not supported by adequate experimental evidence.

Scoliosis has been produced by epiphysiodesis of the neurocentral cartilage in pigs (Beguiristain, 1980) using a single screw. However the technique involved was an open one, and numerous attempts to produce animal models of scoliosis have been thus flawed, as the associated denervation of the spine is enough to produce a neurogenic scoliosis on its own. It is our aim to perform percutaneous epiphysiodesis of the neurocentral cartilage in the White Pig. Contralateral epiphysiodesis of the neurocentral cartilages, will then be performed to arrest progression of the scoliosis and achieve correction of the initial deformity. The results will be recorded and analysed by serial radiographs; Computerised Axial Tomography with multi-planar reconstruction; histology and electron microscopy: Tetracycline bone labelling; photography.

CHANGES TO ENDPLATE PERMEABILITY IN SCOLIOTIC DISCS

M. R. Urban, J. C. T. Fairbank, C. P. Winlove, J. P. G. Urban

Centre for Biological and Medical Systems, Imperial College, London

In scoliosis the intervertebral disc deforms, becoming irreversisbly wedge-shaped, losing height and swelling pressure on the concave side. This change arises in part from a change in the activity of disc cells. Here we discuss that the role of nutrient supply in inducing this change in cellular activity. The intervertebral disc is avascular. Nutrients, required by the disc cells for maintainance and repair of the tissue, are supplied by blood vessels at the margins of the disc. The nucleus and inner annulus cells rely on capillaries from the vertebral bodies which penetrate the subchondral bone and terminate above the cartilage endplate. Scoliosis appears to induce premature expression of type X collagen and calcification in the disc matrix itself and particularly in the cartilage endplate where in vitro studies suggest that it could decrease permeability and thus transport into the disc.

We have measured transport into between 3 and 5 discs/ patient of 5 scoliotic patients (19 discs in all) aged 5-19 yrs using N20 administed during anaesthesia as a tracer. The amount of N20 in the disc at a known time after initition of N20 administration was a measure of endplate permeability. N20 was measured electrochemically by inserting a custom-built needle micro-electrode into the disc during surgery. We found that endplate permeability varied from disc to disc of the same spine, decreasing towards the apex of the curve. The permeability at the apex was 5-15% compared to 25-40% 2 levels distant from it.

We have found that changes in disc biochemistry are most extreme in the apical disc; this disc has the lowest cell density independant of level, the hightest lactate concentration and the lowest proteoglycan content and swelling pressure. Since disc cells require an adequate supply of nutrients for maintaining tissue homeostasis, the biochemical results indicate that changes in nutrient supply, induced by loss of permeability in scoliotic discs, particularly at the apex of the curve, have a deleterious effect on disc cell activity. This ultimately results in loss of matrix components and possibly leads to the deformed disc so evident in scoliosis.

ZORAB 1998 ADOLESCENT IDIOPATHIC SCOLIOSIS PHENOTYPE CRITERIA

J Fairbank

Nuffield Orthopaedic Centre, Oxford

AIS is well defined for the demands of clinical practice. It is less easy to define in genetics research, where phenotypes need to be distinguished both in the younger child with a developing deformity and mildly affected adults. It is also important to identify those in a pedigree with abnormal manifestations of scoliosis.

Paricipants at the Zorab Symposium in Oxford struggled with a definition of an AIS. This, at first sight complex, form is an attempt to summarise what was said in a fashion which is useful to gene hunters. It is not designed for use for your clinical practice

The three vertical columns are "normal", "true AIS" (in 3 degrees of severity - just detectable/arguable, moderate and severe), and "atypical spinal deformity". One form descibes one individual in a pedigree. There are many methods of assessment. Some may well be better than others.

This form allows the examiner to indicate which method(s) have been used to define that particular individual. There is no limit to the number of methods used on one individual. (For example, I have indicated at least 4 different surface topography systems, and doubtless there are others. Most people would use only one of these, if at all). MRI is rarely going to used on this group of individuals, but will be in some cases. If it has, then there is space to use it.

This form can be reduced to contain only those methods used by that particular investigating group.

This is a draft document, and I am open to comments and suggestions for improvements

The form can be found on www.ndos.ox.ac.uk/pzs

Copyright British Editorial Society of Bone & Joint Surgery 1998
Provided by ProQuest Information and Learning Company. All rights Reserved