Ebook: Research into Spinal Deformities 1

Research on spine deformities does not go back more than 30 years; it begins in the mid- 1960's and has since then developed along two lines.

Ever since their start in 1965, the P.A. Zorab symposia have spurred orthopaedic surgeons and other scientists to look into different pathophysiological aspects of scoliosis and other spine deformities including epidemiologic, diagnostic, prognostic and therapeutic issues.

However, it is the spinal instrumentation of P.R. Harrington presented in 1965 that has had the greatest implications for the surgical treatment of spinal deformities as well as of injuries and pathologic conditions of the spine. The spectacular results of spinal instrumentation, particularly with regard to severe scoliotic deformity, surpass the highest expectations we had only thirty years ago.

Several scoliosis Societies have been established over the years : The Scoliosis Research Society in the US in 1966, The European Spinal Deformities Society in 1986, The European Spine Society in 1989 and the Société International de Recherche et d'Etude sur le Rachis, ín1995. Numerous other national scoliosis societies have also been established.

Three well reputed Journals, Spine (since 1975), Journal of Spine Disorders (since 1988) and the European Spine Journal, (since 1992) cover the field of spinal afflictions. However these journals, as well as the societies, deal almost exclusively with issues related to surgery and, to a lesser extend, bracing, screening and other clinical problems. Papers concerning pathophysiological aspects of scoliosis are rare.

Despite this, however, - as surgery ,a priori, is an unbiological necessity- research on the etiology, the pathogenesis and the pathomechanisms of idiopathic scoliosis is today booming. Also the technological achievements in recent decades have also stimulated research into non-invassive methods for registration of the back shape and the 3-D deviation and the kinematics of the scoliotic spine. The Moire fringe topography and its modifications, stereophotogrametry, and optoelectronic analysis of the movements of the spine and related new methods of registration of different variables of the deformity have spurred an increasing interest for the etiology, pathogenesis and mechanisms involved in the evolution of the thoracospinal deformity in idiopathic scoliosis and other spine deformities.

Biannual Symposia on Surface Topography of the Scoliotic Spine have been held since 1980, beginning in Vermont, USA, and on 3-D Scoliotic Deformities when the first meeting was held in 1992 in Montreal, Canada. After much discussion and deliberation it was found appropriate to form a common forum for presentation of research into spinal deformities with a theoretical orientation.

In the combined biannual meeting of Surface Topography and 3-D Scoliotic Deformities, in Pescara 1994, it was decided to form a new society: The International Research Society of Spinal Deformities, with the objective “to provide a forum for presention and encouragement of research relating to spinal deformities and to disseminate the results of such research”. It is expected that better understanding of the cause or causes of these pathological entities and their pathogenesis and pathomechanism with time will provide us with new additional knowledge, a prerequisite for the development of effective methods for no surgical treatment of the early scoliosis and in the long term for prophylaxis.

In the Pescara Meeting it was also decided that the first meeting of the new society was to be held in Stockholm, jointly organized by the Departments of Orthopaedics of the Karolinska Institute, at Huddinge and at Karolinska Hospitals. The “Stockholm-96” Meeting of the Society took place at Huddinge University Hospital, on June 16 to 19 ,1996, under the patronage of Her Majesty Queen Silvia.

In accordance with the spirit of the Society to merge different trends of research of spinal deformities into a common direction, the programme of the meeting was organized into separate sections comprising not only papers on surface back topography and 3-D spinal deformities but also on the etiology, pathogenesis and pathomechanism of idiopathic scoliosis, on the biomechanics and kinematics of the spine and on conservative and surgical tratment of spine deformities. Moreover, central issues were debated in five 1-hour-long Round Table Debates followed by discussions with participation of the audience.

Almost one hundred scientists from nineteen countries attended the meeting and reported the results of their research in altogether more than one hundred oral presentations and poster exhibitions.

The objectives of the Society to fuse together different trends of research on spine deformities was thus fulfilled.

Communication between the scientists is often hampered by the fact that presentations of distinctly different topics are held in biological, orthopaedic, mathematical, bioengenering and other “languages” not always comprehensible to all participants. Therefore, the meeting was preceeded by a half-day-long tutorial course, where terms and methods of research were defined and presented..

The great majority , but not all, lecturers have subsequently submitted manuscripts of their presentations. They are included in this volume, which can now be distributed to the participants of the meeting and to other members of the Society.

As chairman of the Stockholm-96 meeting I wish to express sincere thanks to: the members of the international programme committee; M.Asher, P.Dangerfield, B.Drerup, E. Hierholzer, M.Moreland, A.Merolli, I.A.F. Stokes, D. Uyttendaele and the members of the local programme commitee; R. Hedlund, H. Hirschfeld, U. Nilsonne (co-chairman), S. Ollmar, H. Saraste (co-chairman), T. Widhe, and U. Willers, the contribution of whom to organizing and carrying through the meeting has been invaluable. The assistance of Ms MarjaLena Kiljander and Mr Pär Westblad in the preparation and the lay-out of this publication is also gratefully ackowledged by the authors. Moreover, the sponsoring of the Meeting, during the initial period of uncertainty, by the Department of Orthopadics at Huddinge University Hospital, the King Oscar II and Queen Sofia Golden Wedding Anniversary Foundation and the Swedish Medical Association is herewith ackowledged with gratitude. Last but not least the magnificent reception at the Town Hall of Stockholm hosted by the Town Council of Stockholm is acknowledged with high appreciation.

Stockholm, February 1997

John A. Sevastik, MD, PhD

Photogrammetry is an optical measurement technique which is becoming more accessible to non-photogrammetrists through easy-to-use digital software and hardware. A small number of fundamental considerations are outlined to assist users who may be unfamiliar with photogrammetry to plan photogrammetric measurement, and attention is drawn to a few basic aspects of photogrammetric system design which are crucial to obtaining the desired accuracy and output while ensuring that the technique meets cost and other constraints. Those contemplating the use of photogrammetry for medical purposes with easy-to-use products currently on offer should be aware of the breadth of options available for its implementation. A range of accuracies is available, usually with commensurate costs, but accuracy can in part be controlled by the user. The prospective user should also be aware of options for automated processing, which depend largely on the characteristics of the surfaces being measured. Photogrammetry almost always provides quantitative output, typically appearing as a simple list of numbers, and the conversion of this to useable information is not always simple.

Idiopathic scoliosis is generally considered as a uniform pathologic condition. However, its three groups, the infantile, the juvenile and the adolescent idiopathic scolioses and the seven patterns of the curve, except for the thoracospinal deformity, have nothing else in common but their unknown etiology, pathogenesis and pathomechanism. On the other hand, several known variations between the three groups of idiopathic scoliosis and between the patterns of the curve are challenging the undiscriminated use of the term today. Increased cognizance of the unknown cause, origin, and pathomechanism involved in the development of the thoracospinal deformity in the different forms of the condition would promote the ultimate task of relevant research, i.e. early treatment and in the long-term prophylaxis.

Asymmetry is a well known physical feature of scoliosis, expressed as the linear asymmetry of the spinal curvature found in the vertebral column and in other skeletal structures such as the skull and limbs. Fluctuating asymmetry (FA) is a deviation from bilateral symmetry in normally bilaterally symmetrical morphological traits. FA 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 in a sample of patients with idiopathic scoliosis. The results indicate that a positive relationship between the degree of spinal curvature and FA with severe curvature being associated with a high degree of FA. These findings indicate that a major phenodeviant (idiopathic scoliosis) is associated with general developmental instability.

The close association of spinal deformity with growth suggests that scoliosis may be a developmental response to perturbation of growth processes. Developmental criteria in 327 girls with late-onset idiopathic scoliosis were examined. Age at diagnosis correlated with apex (P=.002), Cobb angle (P=.006) and with age at menarche (P=.008). For left-right asymmetry, comparison of dermatoglyphics in 114 normal girls and 164 with AIS showed that in AIS there was increased directional asymmetry. When those with AIS were divided into school screening referrals (N=86) and others (N=78), the latter showed a significantly greater variance about the mean (P=0.0) suggesting that they form different subgroups of the population, with decreased developmental stability. These findings may be interpreted to indicate cephalo-caudal and left-right gradients and significant differences in the control of development in AIS.

This paper reports the use of real-time ultrasound methods to measure femoral anteversion (FAV), tibial torsion (TT), femoro-tibial rotation (FTR) at the knee and the summated rotational alignment (ΣRA = FAV+TT+FTR) in 537 healthy subjects age 5-18 years. The findings confirm knowledge that FAV decreases and lateral TT increases significantly with age in each of boys and girls. New findings relate to each of FTR (medial rotation of femur on tibia) and ΣRA. FTR, like TT, increases significantly with age in each of boys and girls. ERA changes significantly from being medially rotated in younger children to laterally rotated in older children with the cross-over from medial to lateral rotation occurring at 8-11 years of age; this explains the clinical observation that young children tend to in-toe as they walk and which resolves spontaneously with growth in most children. Overall, the data collected as means and standard deviations for each age and sex provide controls for comparison with data from subjects referred after school screening for scoliosis, having surgery for AIS and patients with patellofemoral pain syndrome which are considered elsewhere in this volume.

This paper reports a real-time ultrasound study of femoral anteversion (FAV) and tibial torsion (TT) in 78 subjects referred to hospital after school screening for scoliosis. Comparison with a control population of children matched for age and sex the shows no FAV asymmetry, but the left FAV of boys is significantly less than the left FAV of girls. In boys and girls combined, FAV asymmetry is weakly associated with Cobb angle (but not AVR) in each of thoracic curves (not sidewise) and thoracolumbar curves (sidewise). In girls alone, FAV asymmetry is associated with Cobb angle of thoracolumbar and lumbar curves combined (p=0.013, sidewise). In contrast, TT asymmetry is found in both boys and girls but is not significantly different from controls. The findings are discussed in relation to the neural control of proximal and distal limb musculature and spinal musculature in health and in AIS subjects.

The purpose of this study was to asses the relationship between iliac rotational asymmetry and the magnitude and pattern of the lumbar curve in patients with idiopathic scoliosis. We used the bisacro-iliac-ischial angle to differenciate between IRA and intra-pelvic torsional deformity. The results of this study supports the concept that the pelvis may be involved in the pathogenesis of idiopathic scoliosis.

The tails of growing rats were axially loaded in compression or distraction to investigate the hypothesis that vertebral and disc wedging during growth in spinal deformities is mechanically modulated. Twenty-eight 6-week-old Sprague-Dawley rats were assigned to one of three groups: compression loading, distraction loading, or sham (apparatus applied without loading). Two 0.7mm diameter stainless steel percutaneous pins were passed through two vertebrae and glued to 25mm diameter external ring fixators. Calibrated springs were installed on three threaded rods passing through the rings and compressed with nuts to apply compression or distraction forces between 25% and 75% of bodyweight. Radiographs made at weekly intervals for nine weeks were digitized to measure vertebral and disc growth rates. Loaded vertebrae grew at 68% of control rate for compressed vertebrae, and at 114% for distracted vertebrae (p<0.01 by ANOVA). Rate of change of disc thickness during the experiments was negative for all groups. However, comparison of radiographs before application of load with those made after sacrifice and removal of apparatus showed that compressed discs had reduced thickness averaging 0.49mm, distraction discs had average increased thickness of 0.13mm (p<0.05), and sham discs lost an average of 0.27mm of thickness. Results are compatible with the hypothesis that asymmetric loading of vertebrae and discs promotes scoliosis progression during growth.

Magnetic Resonance Imaging (MRI) has been applied to the examination of changes in the length of the spine which occurs between the recumbent and upright positions. Disc volume changes and their correlation with the variation in standing height were examined in seven normal female volunteers with the results indicating that dynamic time-dependent deformations are present within the intervertebral disks. These are attributed to fluid flowing to and from the disk and by visco-elastic deformation of the anulus fibres under the influence of osmotic and mechanical pressures. A significant difference between the sexes in the maximum volume of the disks was present, attributed to the increased lumbar lordosis in the female. Spinal length changes between recumbent and upright postures and the associated sexual dimorphism are biomechanical and physical effects which may have major clinical significance in the context of scoliosis aetiology.

In this study twelve L5-S1 disc necropsy samples aging from 28 weeks (stillborn foetus) to 84 years were examined by histological and immunohistochemical methods. The histological investigation was performed by light microscope (magnification *100, *250 and *400) and hematoxylin-eosin. The immunohistochemical examination inquired type III and VI collagen arrangement in lumbar disc. Notably, type III and type VI collagen organization in seven specimens aging from 28 weeks to 1 month , previously described by no Author, differed from that observed in adulthood and elderly life.

This study was undertaken to evaluate the amount of disc and vertebral body wedging in scoliosis, as well as how the wedging changed over the progression of the scoliosis curve. The degree of vertebra and disc wedging was evaluated in fourteen patients with scoliosis by means of PA radiographs which were obtained early in the course of their scoliosis and late in the progression, prior to surgery. Each vertebral body endplate on the PA film was marked with a horizontal line corresponding to the inclination of the vertebra. Each of the lines was then digitized and the wedge angle of the vertebra and of the disc was calculated. Comparisons were made between the earlier and later curves for the apical five vertebra (Apex ± 2). Among the 70 vertebrae examined, later curves showed a statistically greater wedge angle of the vertebra than the disc. The findings of this study show that early in the scoliotic deformity, the wedging of the vertebra and the disc occur in equal proportions. However, as the curve progresses, the vertebra becomes more wedged with respect to the disc.

Thirty-one patients with idiopathic scoliosis were studied to identify changes in the nucleus pulposus(NP) using magnetic resonance imaging (MRI). The mean curvature and the mean age at initial assessment were 26and 12.6 years respectively. The mean follow-up time period after MRI scan was 3.6 years. Deviation of the NP near the apex of the curves was examined from two sets of MRI scans obtained before and after correction of the deformities. The subjects who showed good mobility or slight deviation of the NP(n=14) successfully completed brace treatment. Those subjects who showed an immobile NP(n=5) were shown to have progressive curves despite brace treatment. Unbraced subjects who showed mobility or slight deviation of the NP(n=10) demonstrated no progression at final follow-up. Two patients with Risser sign 4 and 2 years after menarche who showed an immobile NP were shown to have nonprogressive curves. The results indicate that deviation and mobility of the NP prior to application of a brace or prior to cessation of growth seem to be good prognostic indicators for curve progression.

This study evaluates anthropometric data from 110 pre-operative patients with AIS (90 girls, 20 boys). These patients are compared with data from 1362 healthy school children (693 girls, 669 boys). Holtain anthropometric equipment and standard techniques were used to make the anthropometric measurements. In pre-operative AIS, the standing height (corrected for the lateral curve), subischial height, leg lengths and upper limb lengths are increased with no significant increase in sitting height (corrected for the lateral curve) or weight. However, by evaluating ratios of one body segment length to another, there is length disproportion in AIS. The lower limbs are longer compared with the upper limbs which in turn are longer compared with the sitting height. In the upper limbs, all three length components are more asymmetrical in AIS than in healthy children. In right sided AIS curves, the upper arm is longer on the right by 7.7mm, whereas in left sided AIS curves, the upper arm is longer on the left by 2.1 mm. In the trunk, the biacromial and bi-iliac widths are both significantly increased in the AIS patients. The lateral chest diameter is not significantly different from that in healthy children and the AP chest diameter is significantly smaller in the AIS patients. The large extrathoracic skeleton (limbs) relative to the normal/decreased thoracic size and the upper limb asymmetry may all be risk factors in the development of AIS by altering spinal mechanics during movement. Other ligamentous or neuromuscular risk factors would also be required.

The aim of this study is to examine the coupling of lateral flexion and axial rotation in adolescent idiopathic scoliosis (AIS). Seventy six patients with AIS each had left and right supine lateral bending films before surgery. The vertebral rotation (Perdriolle) was measured from T5 to L4 on each radiograph. The change in rotation was calculated for each vertebra from left to right lateral bending = ‘lateral flexion induced rotational change (LFIRC)’

In AIS, the coupling of lateral flexion and axial rotation is similar to that seen in the normal spine. Curves which are stiff in frontal and transverse planes show little or no change in frontal plane or transverse plane rotation. Conversely, curves which are mobile in frontal and transverse planes show considerable change in frontal and transverse plane rotation. The presence of soft tissue tethers is the most likely explanation for the limited rotations of the stiff curves.

Between T7 and T10, the magnitude of the vertebral rotation on the pre-operative AP radiograph affects the direction of the lateral flexion induced rotational change; this is probably due to a change in the line of action of the ligaments and muscles responsible for this rotational change.

Further work is needed to examine the relation of the variation in rotational change to the progression of the rib hump after surgical instrumentation and evaluate the predictive worth of these findings.

Lateral curvature is thought to cause asymmetrical loading of vertebral physes, resulting in asymmetrical vertebral growth according to the Hueter-Volkmann principle. A biomechanical model was used to calculate trunk muscle forces and intervertebral forces in a lumbar scoliosis and in a symmetrical spine, in order to quantify the degree of asymmetrical loading. The analysis included five lumbar vertebrae, the thorax and the sacrum/pelvis and 90 pairs of muscles. Five spinal geometries were analyzed: the mean spinal shape of 15 patients with left lumbar scoliosis having a Cobb angle of 38° and apex at L1-2 (reference or ‘100%’ geometry), and this geometry scaled such that it had 0%, 33%, 67%, and 132% of the lateral and axial asymmetry of the reference shape. The muscle and intervertebral forces for maximum efforts opposing moments applied to the T12 vertebra in each of the three principal directions were calculated. The loading state of each interspace was expressed as the resultant force, the lateral offset of the resultant force from the disc center, and the angle of the resultant force from the axial direction. These analyses indicate that lumbar scoliosis produces asymmetrical spinal loading characterized by shear forces tending to increase the scoliosis, but with little increase in the lateral offset of the resultant forces transmitted through motion segments. If scoliosis progression results from asymmetrical loading, it appears that it is the shear force component more than the lateral bending moment which is responsible for this.

In order to study the effect of angulation and asymmetric loading on the progression of spinal curvature a rat tail model was used to apply asymmetric loading across an isolated vertebra and to measure its growth response. Two 25 mm external ring fixators were glued to pins which had been inserted percutaneously into the 8th and 10th caudal vertebra of 6-week old Sprague-Dawley rats. Calibrated springs and 15° wedges mounted on stainless steel threaded rods passing through holes distributed around the rings imposed a 30° curvature and axially loaded the vertebrae in the instrumented tail region. Fluorochrome labels and radiographs were used to document the progression of vertebral wedging. Initially the wedging was entirely in the intervertebral discs, but by 6 weeks the discs and vertebrae were approximately equally wedged. Fluorochrome labeling confirmed that the vertebral wedging resulted from asymmetrical growth in the physes. The amount of longitudinal growth in these tails was similar to that occurring during human adolescent growth.

The concave and convex rib-vertebral angles (RVA) at levels T2 to T12 were measured on AP radiographs of 19 patients with right convex idiopathic thoracic scoliosis, and of ten patients with major thoracic right convex neuromuscular scoliosis. The difference between the angles of the concave and the convex sides, the RVAD, was calculated. The RVA:s were also measured on radiographs from three animal groups where spinal curves had been induced experimentally: in group I, by selective electrostimulation of a) the latissimus dorsi, b) the erector spinae and c) the intercostal muscles, in 16 rabbits; in group II, by manual bending of the spine of four dead rabbits; and in group III by mechanical elongation of one rib in eight rabbits. In both the idiopathic and the neuromuscular groups, the convex RVA was smaller than the concave RVA between levels T2 and T8, with a maximal difference between T4 to T5. From T9 to T12 the concave RVA was smaller than the convex. With increasing Cobb angle the RVAD:s increased linearly with the greatest difference at the second vertebra above the apex. In the experimental groups the pattern of the RVAD:s between T6 to T12 was similar to the findings of the clinical study. The typical pattern of the RVA:s of the concave and convex side seems to be independent of the underlying cause of the spinal curvature.