Smith-Magenis Syndrome

The abstracts were prepared by Mr C. R. Weatherley. Correspondence should be addressed to him at 1 The Quadrant, Wonford Road, Exeter EX2 4LE.

NEW CANDIDATE GENES FOR ADOLESCENT IDIOPATHIC SCOLIOSIS A Child, D Milewicz, G Coulton

Cariological Sciences, St George s Hospital medical School, London: Department of Medical Genetics, University of Texas at Houston: Department of Biochemistry, Charing Corss and Westminster Medical School, London.

Collaborative studies of 8 UK two and three generation families with adolescent idiopathic scoliosis (AIS) have ruled out some connective tissue genes as being causative (eg elastin) and raised the possibility of kinkage to fibrillin2, the gene which causes Beals Contractural Arachnodactyly. Mutations in the neonatal Beals syndrome region of gene FBN-2, on chromosome 5. remember mutations inthe neonatal region of the Marfan syndrome gene, for fibrillin- I on chromosome 15.

The only identified animal model scoliosis gene, mouse kyphosis gene Ky is homologous to a region of a human chromosome. Linkage studies to this region in 5 UK AIS families have been negative. New connective tissue genes, such as latent transforming growth factor-B binding proteins (LTBPs) which are homologus to fibrillins, may play a role in causing human scoliosis phenotypes.

We are grateful to the British Scoliosis Research Foundation for support through the Robert Luff Fellowship.

A GENE FOR AUTOSOMAL RECESSIVELY INHERITED SPONDYLOCOSTAL DYSOSTOSIS MAPS TO CHROMOSOME 19Q13.1-Q13.3 P D Turnpenny, M P Bulman, T M Frayling, K Abu-Nasra, A T Hattersley, S Ellard Institute of Clinical Science, University of Exeter and Department of Clinical Genetics, Royal Devon and Exeter Hospital, Exeter

Spondylocostal dysostosis (SD) is a form of multiple vertebral segmentation defect which is heterogendous in causation. It generally gives rise to a nonprogressive scoliosis, sometimes in association with other congenital anomalies. Both autosomal dominant and autosomal recessive (AR) monogenic forms are reported. Recent access to the largest reported SD family demonstrating AR inheritance has made it possible to undertake gene mapping in this condition. This is a large inbred Israeli-Arab kindred where SD has emerged in three disparate branches, resulting in six living affected individuals. These were sampled together with nine obligate carriers and 13 individuals at high carrier risk. Genome-wide screening using Perkin Elmer optimised fluorescently labelled PCR primers and a semi-automated ABI 377 DNA fragment analyser has been undertaken. Amplification of microsatellite markers throughout the genome has enabled genetic linkage to be established by the technique of autozygosity (or homozygosity) of descent. Significant linkage has been demonstrated to chromosome region 19ql3.1-113.3, a genetic interval spanning approximately ill centimorgans between markers D19S425 and D19S412 (Genethon map of CEPH families). This is the first mapping of a human gene for a monogenic form of SD. Narrowing the region may be possible by analysing further families who are allelic for this monogenic form of congenital spinal deformity.

THE SPINAL MANIFESTATIONS OF SMITH MAGENIS SYNDROME

J Spilsbury, S Acharya, S Upadhyay Smith Magenis syndrome is a rare chromosomal disorder with a very variable clinical presentation. Children affected are developmentally slow and suffer from severe mental retardation. They have a high pain threshold and selfmutilate. It has been reported that 50% of them suffer from "mild to moderate scoliosis". The worldwide literature however consists of only one paper covering this issue, and none of the 20 patients were examined. All diagnoses were made radiologically.

So far 17 SMS children have been examined and had thorough histories investigated. Hospital notes and radiological investigations have been examined, six of 17 patients clinically have a thoraco-lumbar kypho-scoliosis valrying from 5 - 113 degrees Cobb angle. One patient has required surgery, though others should probably have undergone surgery. Eight of 17 patients clinically have symptoms of spinal stenosis affecting their walking distance, but no radiological sings to support this have been found, and this is being investigated electro-physiologically at present.

The syndrome will be discussed in greater detail and recommendations for radiological screening will be made.

THE EFFECTS OF MELATONIN THERAPY ON SCOLIOSIS DEVELOPMENT IN PINEALECTOMISED CHICKENS

K M Bagnall, M Moreau, J Raso, J Mahood and X Wang Department of Cell Biology and Anatomy, Universty' of Alberta, Edmonton, Alberta, Canada T6G 2H7

Introduction: Removal of the pineal gland in young chickens consistently produces scoliosis which has characteristics very similar to those seen in adolescent idiopathic scoliosis. Melatonin has been implicated as the main cause of the scoliosis bsut our finding that not all pinealectomised chickens develop scoliosis despite zero levels of serum melatonin casts doubt on this. Accordingly, this study was designed to assess the effects of melatonin therapy on scoliosis development following pinealectomy in young chickens.

Methods: A pilot study involved sacrificing five. 5 day old chickens at hourly intervals and measuring serum melatonin levels using radioimmunassay techniques to reveal the circadian rhythm of melatonin in these birds. A second pilot study involved injecting fifty, 5-day old chickens intraperitoneally with melatonin at a dose of 2.5 mg/kg. Ten of these birds were sacrificed at hourly intervals commencing 30 minutes after injection and serum melatonin levels were measured as before to confirm this doese as suitable for reproducing the serum melatonin cycle determined in the first pilot study. This does of melatonin wa sused in the subsequent experiments. In the first main experiment, seventy five, 3-day old chickens underwnet pinealectomy. Twenty five of these chickens were then given daily injections of melatonin as described above. Another twenty five chickens were given daily injections of similar amounts of saline to act as a control. The final twenty five birds were not given any injections and acted as the baseline control.

In the second main experiment, fifty 3 day old chickens underwent pinealectomy. They were all housed in the same room and radiographs were teakn at weekly intervals as described above. Two weeks after pinealectomy, when scoliosis presence can first be determined, all the birds were given daily injections of melatonin as described above. All birds in each experiment were housed together in one room and were radiographed at weekly intervals while anaesthetised and in a supine position. Scoliosis development was assessed from the radiographs. All birds were euthanised 5 weeks after pinealectomy.

Results: Approximately 50% of the chickens in each group developed scoliosis and the types of curve that formed, their severity, and their progression was not affected by any melatonin treatment. Specifically, mellatonin treatment commencing immediately after pinealectomy did not prevent the formation of scoliosis and the introduction of melatonin therapy once scoliosis development was confirmed, failed to prevent its progression in comparison with controls.

Discussion: The results of this study cast doubt on the involvement of melatonin in the production of scoliosis following pinealectomy in young chickens remains elusive. Further study in this area is warranted because the model is so attractive and consistent in its similarities to adolescent idiopathic

DOES THE WEDGING OF THE VERTEBRAE OF THE SCOLIOTIC CURVE DEVELOP FIRST FOLLOWING PINEALECTOMY IN YOUNG CHICKENS?

K M Bagnall, M Moreau, J Raso, J Mahood, X Wang and J Zhao

Department of Anatomy and Cell Biology, University of Alberta, Edmonton, Alberta, Canada T6G 2H7 Introduction: Pinealectomy in young chickens consistently produces scoliosis which has many characteristics similar to those found in patients with adolescent idiopathic scoliosis (AIS). The mechanism behind this phenomenon remains elusive but there has been speculation that a major difference in the characteristics of the curves seen in patients with AIS is that there is vertebral deformity prior to curve formation in the chickens. This is converse to the situation that occurs in the human. Accordingly, this study was designed to examine this question.

Methods: Serial radiographs were collectled from 30 Mountain Hubbard chickens that had developed scoliosis during the five weeks following pinealectomy 3 days after hatching. Similar radiographs were collected from 20 chickens that had not developed scoliosis following pinealectomy as well as 10 control chickens that had not undergone pinealectomy. The radiographs had been taken at weekly intervals over a 5 weeks period while the chickens were anaesthetised and had been placed in a supine position. Assessment of the scoliosis was made from the radiographs. The wedge angle was measured for the apical and two adjacent vertebrae in those chickens that had developed scoliosis and for the equivalent T5-T7 vertebrae in those chickens that had not developed scoliosis and the normal controls.

Results: There was significant wedging of the future apical and adjacent vertebrae prior to scoliosis curve development in those chickens that ultimately developed scoliosis. However, there were also instances of vertebral wedging in some of the chickens that did not develop scoliosis although the wedging was not as severe or as frequent as in those chickens that did develop scoliosis. There were only a few instances of vertebral wedging in the normal chickens that had not undergone pinealectomy.

Discussion: These results confirm a major, significant difference in the development of the scoliosis curve following pinealectomy in young chickens when compared with patients with AIS. In the chickens there is vertebral wedging prior to the scoliosis curve development. It remains to be seen whether this difference can be attributed to the obvious differences in biomechanics between the spine of the chicken and that of the human. These results might also provide insight into the mechanism behind this phenomenon in young chickens.

Acknowledgements: The authors would like to thank the University of Alberta Hospitals Board, and the Edmonton Civic Employees for providing funds for this work.

CERVICAL OSTEOTOMY FOR ANKYLOSING SPONDYLITIS - A SAFER TECHNIQUE H Mehdian, P Licina, Splawinski, J K Webb The Centre for Spinal Studies and Surgery, University Hospital, Queen's Medical Centre, Nottingham UK

A 39 year old gentleman with ankylosing spondylitis presented with a severe fixed flexion deformity of the spine. He was initially treated with a multi-level extension osteotomy of the lumbar spine. This improved his posture but marked restriction of forward gaze persisted. A cervical osteotomy was planned and preoperative assessment determined a chin-brow to vertical angle of 60. Under general anaesthesia a halo ring was applied to the patient's skull. The patient was positioned prone, with his head resting on a Mayfield frame. Spinal cord monitoring with somatosensory evoked potentials was used throughout. The Cervifix device was used for reduction and fixation. This is a modular, posterior cervical system consisting of titanium lateral mass screws, thoracic pedicle screws and rods connected by clamps. This technique substitutes a temporary malleable rod for the titanium rod on one side. This allows controlled reduction of the osteotomy as this rod can bend and also slide through the thoracic clamps. This avoids inadvertent translation of the spine both in the sagittal and coronal planes. Once reduction is completed definitive contoured rods were inserted to maintain the correction. There were no neurological complications and the patient was discharged with a moulded cervical orthosis ten days after surgery. Following surgery his posture has markedly improved and horizontal gase was restored. Radiographs taken six months after surgery showed maintenance of correction and evidence of a solid fusion. The purpose of this report is to emphasise that use of the Cervifix device with the technique described above, provides a controlled method of reduction of cervico-thoracic osteotomy. It also reduces the risk of neurological catastrophe, achieves rigid internal fixation and eliminates the need for a halo vest post-operatively.

SPINAL OSTEOTOMY FOR THE MANAGEMENT OF SEVERE AND RIGID DEFORMITIES Y El Hawary, Y El Miligui Cairo, Egypt

This study will emphasise and demonstrate the correction of severe and rigid deformities of the spine by osteotomy of the spine. These cases did not include the ankylosing spondylitis cases.

There were 12 cases which were divided into two groups:

The first group (7 cases) were mainly congenital deformities ie congenital kyphosis or congenital kyphoscoliosis. Anterior and posterior osteotomy were performed in these cases to prevent stretching of the cord during correction.

The second group (5 cases) mainly consisted of idiopathic scolisosi cases but with rigid and severe curves. In addition, all these cases had previous trials for correction of their deformities. We only performed anterior release for these cases (not an osteotomy) because there was no fursion anteriorly

This was followed by a posterior proceudre which consisted of the following: Removal of the previous implants. Multiple (5 to 6) osteotomies performed in the fusion mass to allow correction.

Finally correction using the Isola instrumentation.

CORRECTIVE OSTEOTOMY IN ANKYLOSING SPONDYLITIS - OUTCOME AND COMPLICATIONS

S Kumar, D Sengupta, M Grevitt, 1 Webb Centre for Spinal Studies and Surgery, Queen's Medical Centre, Nottingham

Between 1992 and 1996,12 patients with ankylosing spondylitis had corrected osteotomy of the spine and instrumentation with the Universal Spine System (Stratec, Welwyn Garden City, UK)

Pain and loss of forward gaze were the commonest indication altough 3 patients had stress fractures. Nine patients had lumbar decancellisation closing wedge osteotomy, 3 had thoracic osteotomies and 3 had additional cervical osteotomies; There were 8 males and 4 females; the average age was 52 years (range 40-72). The average duration of the disease at the time of operation was 26 years.

The preoperative mean sagittal balance (the distance from the midsacral point ot a plumbline from C7) was 12 cms (range 4 cms - 29 csm). The mean preoperative thoracic kyphosis was 68 deg (range 40 - 90o) and lumbar lordosis was 31 deg (range 12 - 80 deg).

Postoperatively all patients had an improved sagittal balance, the mean value being 6.5 cm (range -3.5 - +15 cms). The mean postoperative lumbar lordosis was 45 deg (range 32 - 72 deg ) and the mean thoracic kyphosis was 57 deg (range 32 - 71 deg ). The lumbar lordosis improved by a mean of 14 deg after operation; the thoracic kyphosis improved by 11 deg; in those patients who had thoracic osteotomies, the average improvement was 26 deg.

One patient in the series died intraoperatively; there were 2 dural tears, and 3 superficial wound infections. Late complications included 3 implant failures and 2 pseudarthrosis. One patient developed an upward gaze after lumbar osteotomy and required a flexion closing wedge cervical osteotomy.

At an average follow up of 19 months (range 2 - 56 months) all showed appreciable improvement in forward gaze and reduction of flexion deformity.

COMPLICATIONS IN SURGERY FOR ANKYLOSING SPONDYLITIS H Nwaboku, J Lehovsky, T Morley The Royal National Orthopaedic Hospital

A report of 2 cases

Introduction: Ankylosing spondylitis is one of the common spondyloarthropathies characterised by skeletal and extra-skeletal manifestations. The skeletal complications are of particular interest to the orthopaedic surgeon. Spinal deformity producing severe kyphosis can seriously atect the activities of daily living and horizontal gaze. Today the surgical management of deformities at the cervicothoracic and lumbar regions involves an extension osteotomy with or without internal fixation and a protective cast. The operation may be performed under local or general anaesthesia requiring anaesthetists skilled in fibreoptic endoscopy. The procedure can be hazardous, the complications commonly reported being neurological. Rupture of the aorta has been recorded. We report two complications not reported in the literature.

The first case is in a 47 year old labourer who underwent a single level lumbar osteotomy and developed a false aneurysm. The operation was uncomplicated. However, the patient had a persistently low haemoglobin post operatively. He later complained of vague lower abdominal discomfort and subsequent investigation revealed a false aneurysm. This required surgery. The second was a 55 year old man who underwent a cervical osteotomy and developed an air embolism. During bone resection the epidural vein was injured. The anaesthetist observed a sudden derangement in the blood gases. This was quickly remedied by the insertion of a saline soaked swab inthe wound. Both patients made a good recovery with restoration of the horizontal visual field.

In conclusion surgery of the spine in ankylosing spondylitis can be rewarding but has inherent dangers. Vascular complications can be catastrophic. Constant awareness and a high index of suspicion may help reduce or reverse these complications.

SURFACE TOPOGRAPHY AND COBB ANGLES IN IDIOPATHIC SCOLIOSIS

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.

Introduction: While radiological assessment of the Cobb angle remains the gold standard for measurement of spinal deformity, not all aspects are adequately described by this linear device. Other aspects such as trunk balance and rib hump are difficult or impossible to appreciate in the abstracted format of a radiograph and changes in these remain subjective and unrecorded.

Background: A surface topography system (Quantec Imaging Systems) has been in use at this centre for three years. Initial studies were towards determining and minimising the margins of error due primarily to biological variation as previously described. This study examined whether the system thus operated can detect meaningful clinical change.

Materials and methods: Data for this study was taken from the normal clinical investigations in the spinal deformity clinics. The radiographs and contemporary Quantec scans of 57 girls with adolescent idiopathic scoliosis were examined for correlation between Cobb and Quantec spinal angle. 22 had contemporary serial radiographs and scans making it possible to compare progress as expressed by both methods.

Results: Correlation between Cobb angle as reported previously, by Sakka et al was good (Pearsonis coefficient = 0.8; p

Discussion: Correlation between Cobb angle and Quantec spinal angle was described by Sakka et al who also found surface topography useful in clinical follow-up (Sakka et al 1996). This study accords with theirs, but follow-up data, albeit on a small sample suggests unresolved problems. Wide error margins invalidate single scans for meaningful decisions in the clinical situation. While averaging a series of repositioned scans is more accurate and reproducible the technique prohibits rapid results in the clinical situation. Conclusion: Surface topography has potential to be a useful tool in the understanding and management of scoliotic deformity, but considerable work still needs to be done on the elaboration of criteria and interpretation.

A NEW SURFACE TOPOGRAPHICAL MEASURE OF SPINAL SHAPE IN SCOLIOSIS N J Oxborrow, S Gopal, A Walder, D Sharples, K Mardia, P A Millner, R A Dickson University Department of Orthopaedics, St James's Hospital, Leeds, Department of Statistics, University of Leeds.

Many patients with scoliosis followed in out patients undergo repeated radiographic examinations involving not insubstantial doses of radiation. What is clearly needed is a reliable method for the detection and monitoring of scoliosis without the use of ionizing radiation. A novel statistical analysis has been developed able to separate children with scoliosis from their straight backed counter parts from a 3-dimensional spine line created from radioagraphs. We investigate if this model can be applied to a spine line generated from surface topography.

14 subjects between the ages of II and 15 with scoliosis of varying aetiologies, curve patterns and magnitude (Cobb angle 13 - lOlo) underwent examination with both surface topography using the QUANTEC system and radiology. 14 straight backed, age and sex matched controls also underwent surface topography. Using the Quantec software a 3-D spine line and a Suzuki Hump Sum was obtained for each subject and the former analysed with principle component analysis to produce values for Log9L3) and Log (Tilt). Suzuki Hump Sum was a poor indicator of the presence of scoliosis with scoliotics having a value less than 15. However, plotting Log (L3) against Log (Tilt) a separation of the two groups was achieved. 4 patients were not detected but 3 of these had relatively minor curves (Cobb angle

We conclude that this analysis is useful in the study of spinal shape and is able to detect early scoliosis. The addition of the Suzuki Hump Sum improves the accuracy of this measure. Previous work has suggested a possibility of predicting progression of minor curves and a further prospective study is indicated.

EVALUATION OF VERTEBRAL ROTATION BY ULTRASOUND FOR THE EARLY DETECTION OF ADOLESCENT IDIOPATHIC SCOLIOSIS (AIS). R G Burwell, A S Kirby, E L Kirk, R K Aujla, R K Pratt, M A Bailey and J K Webb School of Biomedical Sciences and The Centre for Spinal Studies and Surgery, Queen's Medical Centre, Nottingham.

The early diagnosis of AIS is based on the detection of a rib or lumbar hump with the unaided eye and/or a mechanical device such as a commercial inclinometer (Scoliometer). In a previous paper to the Society we reported a new real-time ultrasound method for measuring vertebral and rib rotation in 20 school screening referrals with AIS (Kirby et al 1977). Rib rotation was less than vertebral rotation suggesting that the latter would be the better predictor of scoliosis curve severity. This paper evaluates vertebral rotation for the detection of lateral spinal curves of AIS in 50 adolescents referred consecutively to hospital in the Nottingham scoliosis school screening service (girls 39, boy 11, median age 15 years). Radiographically the curve types were thoracic 21, thoracolumbar 14 and lumbar IS with a mean Cobb angle of 17 deg (range 2 - 40 deg). Ultrasound laminal rotation was measured in the prone position from TI-SI by one of three observers (ASK, ELK, RKA). The capacity of ultrasound laminal rotation to predict the scoliosis curve (Cobb) angle was compared with the use of surface back shape angles calculated for 18 levels from angles of trunk inclinations (ATIs) measured by one observer (ROB) using a Scoliometer at each of 10 levels in the standing forward bending position. The findings show that ultrasound laminal rotation (and Scoliometer ATI) each with a threshold of 8 deg predicted Cobb angles of 20 deg or more with a sensitivity of 95% (79%), specificity of 68% (35%) and positive predictive value of 64% (43%). Overall, the ultrasound method is better than the Scoliometer method (w0.003, McNemar test). The findings suggest that in school screening for scoliosis ultrasound laminal rotation measured in the community may reduce the number of false positives who are referred to hospital and x-rayed.

ANALYSIS OF INTRAVERTEBRAL AXIAL ROTATION IN ADOLESCENT IDIOPATHIC SCOLIOSIS USING THREE-DIMENSIONAL MAGNETIC RESONANCE IMAGING.

D BirchalL D G Hughes, I Robinson, J B Williamson Hope Hospital, Salford M6 8HD

The objective of this study was to define the proportion of segmental axial rotation that occurs due to intravertebral deformity in patients with adolescent idiopathic scoliosis (AIS) using three dimensional magnetic resonance imaging.

Method: Patients with AIS were imaged with a Siemens 1 tesla impact scanner using dual echo steady state gradient echo TI-weighting (TR30 msec/TE9/45 msec/40 deg). Three dimensional volume images of the apical ten vertebrae were obtained in the axial plane and were post processed through multiplanar reconstruction. Using a novel and reproducible technique (1), sections through the superior and inferior end plates were selected, allowing axial reconstructions to be selected in the plane of each endplate. Axial rotation was measured by comparing angles subtended by datum points along the inner surfaces of the laminae with a vertical line drawn by the computer (global rotation) and also with a measurement determined by reference to a neutrally rotated vertebra (LS or TI, previously selected from plain radiographs) which had been included in the image qcquisition. Measurement of rotation at sequential endplates thereby allowed the proportion of intravertebral and intervertebral deformith within each scoliosis to be determined.

Results: 10 patients (8 girls and 2 boys, aged 12-19 years) with King type 3 or type 2 curves suitable for selective instrumentation were included in the study. Each patient had a right thoracic curve with a mean Cobb angle of 400 (range 30 deg to 61 deg). The mean change in axial rotation observed progressing from end vertebra to apical vertebra was 28 deg, ranging from 20 deg to 34 deg. The mean proportion of axial rotation within the overall scoliotic curve occurring on an intravertebral basis was 34%, with a range of 9% to 76%.

Conclusions: A variable but significant amount of the overall scoliotic deformity in patients with AIS occurs as a result of intravertebral rotation, an observation that has not been previously establihsed. Indeed, in half of the patients imaged in this study, intravertebral deformity contributed over 45% of the total axial rotation. These findings explain the observation that, even with the most sophisticated instrumentation strategies rotational ocrrection is often poor. This study implies that the assessment of endplate rotation with three dimensional MRI is a useful means of identifying a subgroup of patients in whom derotational surgery is likely to fail.

Reference: Measurement of Vertebral Rotation in Adolescent Idiopathic Scoliosis Using Three Dimensional Magnetic Resonance Imaging: Dr Birchall, D G Hughes, J Hindle, L Robinson, J B Williamson, Spine, 22, 2403-2407. MRI ABNORMALITIES IN "SYMPTOMATIC" ADOLESCENT IDIOPATHIC SCOLIOSIS S Kumar, H Surawera, A Saifuddin, M H Noordeen The Royal National Orthopaedic Hospital, Stanmore, Middlesex UK

The value of MRI scans in the investigation of neuroaxis abnormalities accompanying deformity in adolescent idiopathic scoliosis has previously been reported. The incidence of MRI abnormalities is reported as being between 067 and 4.4%.

Commonly used criteria for MRI scanning in adolescent idiopathic scoliosis include neurological signs and symptoms, pain, left sided thoracic or thoracolumbar curve, a Kyphotic component, rapid curve progression and associated syndromes.

We investigated the incidence of MRI neuroaxis abnormalities in adolescent idiopathic scoliosis patients referred with incriminating signs and symptoms. The whole spine was imaged using sagittal, coronal and axial TI-weighted images and spin echo T2-weighted images.

120 patients with adolescent idiopathic scoliosis presented with symptoms and signs warranting further investigation. 21 patients (17.5%) were found to have neuroaxis abnormalities. These included Chiari malformation (n = 7), Syrinx (n = 12), patulous intradural space (dural ectasia) (n = 4) diastematomyelia (n = 3), cervical disc protrusion (n = 1), lipomeningocele (n = 1) and, thickened flum (n = 1).

We found the incidence of neuroaxis abnormalities in "symptomatic" patients with adolescent idiopathic scoliosis to be greater than the reported range for adolescent idiopathic scoliosis in general.

The presence of unusual symptoms and signs are significant discriminating factors in identifying patients wtih adolescent idiopathic scoliosis requiring MRI scans.

THE NATURAL HISTORY OF ADOLESCENT IDIOPATHIC SCOLIOSIS 40" - 50" F Dowling, C Goldberg, D Moore, E Fogarty Children's Research Centre anad Orthopaedic Department, Our Lady's Hospital for Sick Children. Dublin 12, Ireland

Introduction: Adolescent idiopathic scoliosis may present early or late in its evolution, and prognostic indicators are not totally reliable in all cases. This is important at all degrees of curvature, in order to avoid both over treatment of minor cases and the neglect of progressive curves. Experience showed that, while a Cobb angle over 40 deg was a significant diagnosis, treatment decisions were still not automatic. In particular, curves of 40 deg - 50 deg warranted careful study and deliberation, as not all were progressive and not all were cosmetically unacceptable.

Methods: Records were retrieved from the scoliosis clinic data base of girls who either had Cobb angles between 40 deg and 50 at diagnosis, or who presented with lower curves but progressed into or beyond that range during follow-up. They were compared on standard demographic measures: age and maturity at diagnosis and at last review, initial and ending Cobb angles, and outcome, whether surgery or not.

Results: 112 girls were identified, of whom 48 had Cobb angles of 40 deg - 50 deg at diagnosis, and 64 had lower but progressive curves. It was found that those with higher curves at diagnosis were older and more likely to have reached menarche before diagnosis and to have done so at an earlier age. Children diagnosed with lower curves that progressed to beyond 40 deg had the same incidence of further progression and of surgery which was performed at the same age. Surgery was recommended for 62 (55.4%) overall on the basis of established and unacceptable deformity or of continued progression.

Conclusions: Scoliosis between 40 deg and 50 deg has a considerable risk (55%) of unacceptable deformity and surgery. This is not inevitable, however. Those for whom surgery has not been recommended are symptom free with good cosmesis that belies the radiographic evidence. The discrepancy between Cobb angle and outward appearance can be marked in this group, supporting the observation of Thulboume and Gillespie (J Bone and Joint Surg. 58-B(1):64-71) of low correlation between rib hump and Cobb angle. Better understanding of aetiology and natural history would improve decision making for this group of patients.

PYOGENIC VERTEBRAL OSTEOMYELITIS OF THE SPINE

A Stirling, M Rafiq, S Acharya, D Marks, A Thompson The Department of Spinal Szirgen, The Royal Orthopaedic Hospital Birmingham

Purpose: To analyse the initial presentation and treatment of pyogenic spondylo discitis and the consequences on eventual surgical requirement; to compare the outcome of conservative and surgical treatment and to analyse the outcome with different infecting organisms.

Methods and Results: A retrospective review of 53 cases of spontaneous pyogenic spondylodiscitis presenting over the six year period from 1991 to 1997 was conducted. Follow up ranged from three months, to six years with a mean of 22 months. In 12 patients the diagnosis was made within one month, in 19 one to three months, in 16 three to six months, and in six patients more than six months. At the time of referral 50% of patients had had one or more courses of antibiotics prior to obtaining a firm bacteriological diagnosis.

In 25 cases bacteriology following biopsy or surgery was positive; in the remainder the diagnosis was based on clinical findings, laboratory results and the MRI appearances.

Thirty five patients were treated conservatively with three or more months antibiotics. Resolution of symptoms occurred in 25 patients at three to six months of which 19 spontaneously fused and six had a nonunion. Nine patients with a persisting nonunion had mild symptoms at six to 12 months. Three patients had persisting pain sufficient to require time of work, of whom two had nonunions and one a spontaneous fusion. In this group 70% of those with S Aureus infection proceeded to spontaneous fusion in contrast to 30% with all other organisms,

Surgery was undertaken for neurological compromise or deterioration or for progressive vertebral collapse. Twelve patients had decompression and instrumented fusion and four patients decompression and fusion without instrumentation. Antibiotics were given for a similar period. Resolution of symptoms occurred in three to six months in 14 patients of whom 13 fused and one had a nonunion. Two patients had mild symptoms at a year one of whom had fused and the other had a nonunion.

Surgical intervention was required in 50% of those in whom antibiotics had been started prior to a firm bacteriological diagnosis in contrast to 20% when a bacteriological diagnosis was made prior to starting antibiotics. Of the nine patients with neurological compromise treated conservatively (either not fit for surgery or only minor neurological compromise) one one regained normal neurological function. In contrast all but one of the surgically treated group with neurological compromise regained normal function.

There were three deaths one from infection without surgical intervention and in the operative group one from postinduction myocardial infarction and one from pulmonary embolus.

Conclusions: Appropriate bacteriological diagnosis before commencing antibiotics greatly reduced eventual surgical treatment.

Infection with S aureus usually spontaneously fuses but with other organisms frequently does not. Surgical intervention usually resulted in neurological recovery.

Instrumentation in this group resulted in no related complications.

CLINICAL OUTCOME OF DECOMPRESSION IN POST-TUBERCULOUS KYPHOSIS WITH HEALED DISEASE

S Dalvie, V Laheri

King Edward VII Memorial Hospital, Bombay Purpose: To study the pattern of neurological recovery in patients with paraplegia of late onset associated with healed disease who were subjected to anterior decompression and arthrodesis.

Patients with severe post tuberculous angular kyphosis develop a spastic paraparesis many years after the initial disease. If it occurs in association with healed disease, the cause of the neurological dysfunction is the sharp internal gibbus and prolonged stretching of the cord, and the prognosis is not considered favourable.

The study included 15 children (age six to 17 years) presenting with paraplegia of grade II or III (Kumar, Goel) following two to 12 years after the initial disease. Duration of neurological symptoms was from eight to two months preceding surgery. Preoperative kyphosis ranged from 80 130. MRI was obtained in 11 patients.

Anterior decompression was carried out by thoracotomy, high thoracotomy or manubriostenotomy approaches. The cord was decompressed anteriorly and a strut-graft inserted to prevent further kyphosis. Intra-operative findings revealed healed disease with cartilaginous and bony internal gibbus causing compression and thinning of the cord. After decompression the cord expanded anteriorly and fell to a more relaxed position. Factors affecting neurological recovery were studied and included duration and severity of kyphosis and deficit. MRI findings and other epidemiologic data.

Results: Significant and functional recovery was achieved in 13 patients. Final status was normal in two patients, grade I in nine and grade II in two patients. Nemological deterioration from grade II to IV was seen in two patients. No pre-operative factor was found to have correlation with the neurological outcome. Neurological deterioration was attributed to handling of the already compromised cord. Conclusion: Favourable results can be expected provided a good decompression can be achieved without cord-handling. It is thus prudent to intervene at any stage of presentation, contrary to the previously held concept of withholding surgery till loss of ambulation.

DYNAMIC VERSUS RIGID SPINAL IMPLANTS A von Strempel, P de Muelenaere, G du Toit, A Neckritz Klinic III im Annastift, Heimchenstrasse I - 7, 30625 Hannover, Germany

Study design: Patients with lumbosacral disorders were stabilised with the Segmental Spinal Correction System (SSCS). The surgeon has the choice between a rigid pedicle screw and a screw with an articulated head (dynamic instrumentation).

Objectives: Biomechanical investigations that verified the advantages of the dynamic instrumentation were meant to be examined by means of the clinical study.

Summary of background data: By using rigid implants higher fusion rates are achieved than without implants. But implant breakages were reported. This lead to the development of a dynamic pedicle screw-rod system which should reduce the stress-shielding effect.

Methods: In a prospective multicentre study 219 patients were followed up for two years after having undergone fusion for low back disorders. The state of fusion was assessed by examining the trabecular pattern of the bone formation. Further implant failure, peri and post operative complications were recorded,

Results: One hundred and one patients received a monosegmental instrumentation (57 with rigid and 44 with dynamic screws) and 117 a multisegmental fusion exclusively with dynamic screws. The fusion rate at 24 months amounted to 91.2% for rigid screws and 93.2% for hinged screws. Implant related complications occurred more often in rigid than in dynamic instrumentation. Five of 228 rigid (2.2%) and only five of the 984 dynamic screws broke (0.5%).

Conclusion: The results show that the necessary stability for a fusion is also achieved with a dynamic instrumentation. The rate of implant related complications were reduced when compared to rigid instrumentation. The stress-shielding effect is reduced through the articulated connection between the rod and the screw.

CORRELATION BETWEEN MRI CHANGES OF LUMBAR MULTIFIDUS MUSCLE AND LEG PAIN. D Kader, F Smith and D Wardlaw Woodend Hospital, Aberdeen

Background and Purpose: In assessment of lumbar spine MR images the changes in paraspinal muscles are frequently ignored. There are indications that the Multifidus (MF) muscle is sensitive to different pathological changes in the lumbar spine eg radiculopathy, discs and facet degeneration. Our main aim was to investigate the significance of lumbar MF muscle atrophy and to evaluate the effect of low back pain (LBP), leg pain, disc degeneration and nerve root entrapment on lumbar MF muscle.

Methods: A retrospective study on 78 patients (aged 17-72 mean = 46) with LBP with or without leg pain. Their MR images were visually analysed by three independent observers for signs of lumbar MF muscle atrophy, disc degeneration (DD) and nerve root entrapment. The observers were blinded to the clinical history and pain drawing charts.

Results: Overall interobserver agreement was good. MF muscle atrophy was present in 80% of the patients with LBP. The correlation between MF muscle atrophy and leg pain was significant (p

Conclusion: Atrophy of MF muscle should be considered in assessing MR images of lumbar spine. This may be helpful in diagnosing LDRS and Myofascial pain in absence of other abnormality in the scan. The clinical implication of that is to advocate early treatment with specific physiotherapy and perifacet injection.

THE ADVANTAGE OF SPINAL CORD EVOKED POTENTIAL MONITORING WHEN PERFORMING THE PLIF PROCEDURE S Tucker, F Taylor

The Royal National Orthopaedic Hospital Trust, Stanmore Posterior lumbar interbody fusion, using cages, is now a standard technique and beneficial in patients for whom nerve root decompression is indicated, in addition to interbody fusion.

The technique requires a wide posterior decompression and both dural and nerve root retraction to facilitate safe passage of the cage into the previously prepared disc space.

This paper presents the results of spinal cord monitoring of these patients during the PLIF procedure and its use for the detection of excessive S1 root retraction.

REPAIR AND REMODELLING OF A FRACTURE DISLOCATION AT THE THORACO-LUMBAR JUNCTION IN SKELETALLY IMATURE PATIENT

J Fabry, D Bracey, C Weatherley Spinal Unit, Princess Elizabeth Orthopaedic Centre, Exeter A 13 year old boy sustained a traumatic fracture dislocation of his spine at the T12/LI level without neurological deficit. The fracture was instrumented and fused using a Hartshill rectangle and sublaminar wires from TIO to L3 with only a limited correction being obtained at surgery. He was subsequently followed up for eight years and during this time there was a complete remodelling of the spine at the site of injury with a full restoration of the canal diameter.

This paper reviews the changes which occurred at the fracture site over the period of follow up. Measurements have been made of the kyphotic deformity of T12 on LI and the canal diameter. The significance of these changes will be discussed with regard not only to the management of such a fracture but with regard to the potential for correction of any spinal deformity in a skeletally immature individual.

ANALYSIS OF PRIMARY AND SECONDARY CURVES IN SCOLIOSIS FOLLOWING ANTERIOR INSTRUMENTATION

S Acharya, J Spilsbury, S Kohli, D Marks, A Thompson, P Pynsent

Spinal Unit, Royal Orthopaedic Hospital, Birmingham

This is a retrospective review of all scoliosis patients treated anteriorly at the Royal Orthopaedic Hospital, Birmingham from 1989 to 1996. The aim of this study is to assess the radiological outcome of the primary and secondary curve following surgery at their last follow up. There were 65 patients who had anterior surgery for scoliosis. Eighteen patients were excluded from the study (seven pre operative x-rays missing, six had both anterior and posterior surgery and five patients were sitters). Forty seven patients have been reviewed in this study. Thirty nine patients had idiopathic scoliosis. two patients had congenital scoliosis, and six patients had neuromuscular scoliosis. MRI scans were normal 21 patients, not available in 24 patients and abnormal in two patients. There were II males and 36 females. The mean age of diagnosis of scoliosis was 14 years while the mean age of operation was 17 years. The mean follow up was 32 months. Surgery had been performed with two types of instrumentation In the initial years a rigid anterior system (Webb Morley) was used in 21 patients and in the later years the Dwyer instrumentation was used in 26 patients. There were seven cases of metal failure with the Webb Morley system while there was only one case of screw pull out in the Dwyer's instrumentation. The pre operative primary curve had a mean of 58o which corrected to a mean of 25o six months after follow up. At final follow up the mean had deteriorated to 36. The mean secondary curve was 36 pre operatively which had corrected itself to a mean of 25 at six months of follow up but there was a marginal loss of correction at last follow up with the mean being 29o. The coronal balance improved by 36% while there was less improvement of sagittal balance ( 14%).

In conclusion the Dwyer system had better results than the Webb-Morley though there was loss of correction in both the systems. The compensatory curve seems to undergo a slow improvement with marginal deterioration at last follow up.

THE EARLY RESULTS OF TWO LOW PROFILE INSTRUMENTATION SYSTEMS G Robbins, B Okafor, M Edgar The Middlesex Hospital

The trend in scoliosis surgery is towards low profile systents, however, there are concerns with these implants with regard to:

- Increased blood loss - Increased operating time - Implant failure

Seventy idiopathic scoliosis cases, treated by posterior spinal fusion with two low profile systems (40 Modulock)Wrightlock and 30 Colorado) between 1994 and 1997, were reviewed with regard to operation time, intra-operative blood loss and correction obtained. There were no significant differences between these two implants with respect to any of these parameters. At six month review, there were two pedicle screw fractures in each group.

We compared our current series with out earlier study of 319 cases of Idiopathic Scoliosis treated with Harrington rods (+/- Luque wires) between 1974 and 1982 (Youngman and Edgar 1985).' Whilst the correction obtained with the low profile system was better (though not statistically significant) than the Harrington rods, the blood loss was much greater with over a 1400 ml loss in 80% of cases, compared to only 4% of the Harrington Rod cases.

Reference: 1. Youngman PME, Edgar MA "Spinal fusion and Idiopathic Scoliosis". Annal Royal Coll Surg 1985; Vol 67:313-7.

A TECHNIQUE TO MEASURE DEROTATION FORCES DURING SCOLIOSIS SURGERY V Raso, K Duke, D Hill, M Moreau, J Mahood, D Budney

Capital Health - GRH, 1023 lll Avenue, Edmonton, Alberta, Canada T5G OB7 Although safe, optimal, surgical correction of spinal deformities depends on the loads applied by the surgeon, little work has been done to measure these loads. The purpose of this study was to determine if the iderotationi loads can be measured during surgery reliably while not interfering or extending the procedure. A vise grip was modified to allow a strain gauged sleeve to slip over the handle. The hand grip was designed to be insensitive to the squeezing force and record only the net medial lateral force and axial torque transmitted to the hand grip by the surgeon; the loads applied to the spine can be determined directly from these loads. The load cell was calibrated with known weights between five and 65 Newtons. Data were collected during surgery while the spinal rod was rotated into a final aligned position. Forces and moments were recorded for about 30 seconds at 10 samples per second. The system was tested on three subjects - two AIS and one Retts Syndrome. Maximum forces applied by the surgeon ranged from 20 to 50 Newtons with the associated torsion forces ranging from three to II Newton-metres. There was little disruption in the procedure with this technique and the surgeon was able to review the loads applied during the surgery. We concluded that the hand grip provides a safe, convenient and reliable method to monitor translation forces and torsion forces applied during spine surgery. A study has begun to determine if there is a relation between applied loads and correction. SURGERY FOR SPINAL METASTASES H Taha, C Weatherley

Spinal Unit, Princess Elizabeth Orthopaedic Centre, Exeter

Surgery has become an accepted practice in the management of spinal metastases in selected cases. There remains, however. the belief that unless the metastasis is excised in toto the surgery is likely to be ineffective. In Exeter we have adopted the practice that most spinal metastases are best regarded as pathological fractures and have relied on adjuvant therapy to deal with tumour control. This paper reports the results of 43 cases operated upon by the senior author adopting this philosophy.

There were 24 males and 19 females with an age range of 38 - 83. The commonest metastases were from myeloma, the kidney, breast and prostate, with seven cases with an unknown primary. With the exception of three cases, all had pain on referral. Seventeen had a neurological deficit. Surgery involved an anterior procedure in three cases, all had pain on referral. Seventeen had a neurological deficit. Surgery involved an anterior procedure in three cases a posterior procedure in 33 and a combined procedure in seven.

Following surgery, 94% reported complete pain relief. No patient was downgraded neurologically and 53% of those with a neurologic deficit improved one Frankel grade or more. As a result of pain relief and neurological improvement, 63% regained their ability to walk. Only three cases had further spinal surgery and none at the same site. There were minimal postoperative complications. Survival ranged between three weeks and 66 months, with a mean of 17 months.

These results support the view that limited surgery with no attempt to excise totally the spinal metastasis is effective and avoids the morbidity associated with over zealous surgery.

THE RNOHT CLINICAL DATA FOR THE MOSS CASE VERTEBRAL BODY REPLACEMENT: FDA SUBMISSION

S Tucker, B Taylor

The Royal National Orthopaedic Hospital Trust, Stanmore

From April 1993 to January 1998, 83 Moss cage vertebral body replacements have been inserted for trauma, tumour and post traumatic reconstruction. This paper details 40 of these patients with a minimum follow up of twelve months (range 12 - 36 months for tumour patients; 12 - 48 months for trauma patients).

Ninety five per cent of patients underwent corpectomy; 5% total vertebrectomy. The patients have been assessed clinically and the residual symptoms of pain, neurological status and functional status recorded. Radiologically the patients have been assessed for migration of the cage, and cage integration by axial CT with 3D longitudinal reconstruction.

Our provisional results suggest that this implant facilitates good reconstruction of anterior and middle spinal columns (of Denis), with satisfactory osseous integration of the cages. Clinically, there is good symptomatic improvement and functional recovery, relative to the major conditions which necessitate this prosthesis.

TRANSPEDICULAR FIXATION (MINI OLERUD FIXATOR) IN THE TREATMENT OF VARIOUS SPINAL DISORDERS IN CHILDREN S Acharya, J Spilsbury. D Marks, A Stirling, A Thompson

Spinal Unit, Royal Orthopaedic Hospital, Birmingham Nine children underwent transpedicular stabilisation with the mini-olerud fixator. The indications were for deformity (four cases), tumour (four cases), and trauma (one case). The mean age of the children was 10.5 years (range six - 15 years). There were five females and four males. Three patients presented with scoliosis and one patient had low back pain with scoliosis. The other five patients presented with low back pain with two of them presenting with referred pain to their legs as well. Four patients had neurological deficit before surgery which improved after surgical stabilisation. Surgery consisted of neural decompression, internal fixation and bone grafting. The complications included one case of neurological deterioration, one superficial wound infection, one case of pneumothorax and in one case the superior pedicle screw cut out through the superior end plat. Instrumentation was removed in four patients due to prominence of metalwork. This was usually done at one year. The mean follow up was 16 months. The mean fusion rate was 91%. The clinical outcome was satisfactory in all. In children pedicle screw fixation is difficult and requires a high degree of accuracy to avoid major complications. To our knowledge the use of the miniolerud fixator has not been described before in the English literature. We describe our preliminary experience with this fixator

TREATMENT OF NEUROMUSCULAR SCOLIOSIS WITH LUQUE GALVESTON INSTRUMENTATION: RESULTS AND PITFALLS P Moens, L Vanden Berghe, G Fabry Department of Orthopaedic Surgery, UZ Pellenberg, Pellenberg, Belgium Treatment of musculoskeletal disorders in children with neuromuscular pathology is a challenge. They usually develop a deformity of the spine but also contractures muscle weakness, osteoporosis and general condition problems.

What are our goals of surgery? We must stop the evolution of the curve and restore the sitting balance. It means taking into account all the problems and sometimes solve the other problems before the spine surgery.

How can we reach our goals? We need an instrumentation which will allow a progressive correction of the spine on the pelvis, a stable fixation and early mobilisation without a cast. The ISOLA Luque Galveston technique is in our hands the best system to reach these goals.

For this paper we review 34 patients. Twenty were spastic paraplegia, two hypotonic CP, four SMA, two Rett syndromes, six other conditions.

We performed 30 posterior fusions, two anterior and posterior fusions. In one patient we had to stop surgery because of blood loss. In one patient we could not perform pelvic fixation.

For this paper we measure the pre and post operative Cobb angle and the sitting balance on x-rays. The mean pre operative Cobb angle was 65.3 deg (20 deg - 102 deg), the mean post operative Cobb angle was 33.lo. The mean pre operative tilt was 25.8 deg (0 deg - 55 deg) and post operative 7.5 deg. We have 6 over corrections (2 deg - 7 deg). The complications due to the technique were four pelvic perforations (one medial and three lateral without symptoms), one deep wound infection (debreidement without removal) and in the same patient a rod fracture 2.5 years after surgery. We noted also in SMA patients two temporary hypersensibilities and one temporary drop foot.

The complications related to the neuromuscular disorder are hip problems and are seen in five patients, in a sixth patient we operated on the hip before the spine surgery. We think that the pitfalls are more due to problems related to the neuromuscular disorder rather than to the technique itself. We may not forget we treat a patient and not only his scoliosis.

THREE-DEMENSIONAL CORRECTION OF SPINAL DEFORMITIES USING THE ISOLA SYSTEM

Y El Hawary Cairo, Egypt

One hundred patients with idiopathic scoliosis had surgical correction in the period between 10/96 and 10/97. The age varied from 12 to 36 years with a mean of 17.5 years. These 100 patients were divided into 82 females and 18 males.

Pre operative assessment included both clinical examination and an x-ray profile in the form of AP and lateral view (standing) and right and left bending (supine). All patients were corrected using the ISOLA spinal implant system and fusion was attempted by both allo and auto grafting.

These are our first 100 cases with a three dimensional system. Average correction was 75% and was maintained during follow up. One patient had post operative infection that subsided following surgical debridement. Two cases had dislodgement of the upper end of the construct which was reattached and did well afterwards.

SURGICAL TREATMENT OF PAINFUL ADULT SCOLIOSIS: A COMPARISON OF TECHNIQUE P Parisini, T Greggi, M Di Silvestre, A Migletta, B Maiorana

Rizzoli Orthopaedic Institute, Bologna, Italy Pain and degenerative changes with progression of the deformity are generally the primary indications to the surgical treatment of adult scoliosis. The purpose of our study is to evaluate whether new techniques, based on segmental and tridimensional correction, can achieve better results. We examined 38 patients affected with scoliosis (Group A), aged 45 years on average (32 to 63), treated from 1987 to 1997 by posterior arthrodesis and instrumented using these new system. Indications to surgery included progressive deformity (43.8%), pain (30%), respiratory problems (25%) and cosmetic appearance (1.2%). We compared the results with those obtained in a second group (Group B) with similar clinical features, formerly treated at our department by Harrington/Luque technique, associated in five cases with an anterior arthrodesis. Group B wore a plaster cast post operatively for five months on average, while Group A a brace only for four months. At a follow up of three years on average (one to nine) we registered a minor loss of correction in group A (Table I). Pain improvement was 84% in Group A and 79% in Group B. The incidence of complications was 15.7% for Group A against 44% for Group B.

Since the aim of scoliosis management in adults is not cosmetic improvement but pain resolution, we can conclude that the new techniques (segmental correction and pedicle fixation) can be applied with a relatively low rate of severe complications and good results, in terms of pain relief and reasonable correction of the deformity. Moreover, in our opinion, these techniques allow to avoid anterior arthrodesis in many cases, if compared to the old systems.

SURGICAL TREATMENT OF ADULT SCOLIOSIS SELECTION OF LEVELS AND RESULTS L Vanden Berghe, P Moens, G Fabry University Hospital Pellenberg, Belgium Scoliosis in adults usually begins in childhood or adolescence and are often progressive idiopathic. congenital or muscular curves. Some lumbar curves can arise de novo in adults because of disc and facet joints degeneration, osteoporosis and fractures.

The symptoms caused by adult scoliosis are pain, progressive deformity and functional problems. Thoracic curves can also lead to cardiopulmonary problems. Pain is more frequent in lumbar or thoracolumbar curves and is usually in the curve on the concave or convex side. It may also be at the lumbosacral area. Sometimes radicular complaints may be presenting symptoms. Pain can also arise at the contact area between the ribcage and the pelvis. Progression of the deformity may also be a complaint. The patient may notice a waist-line difference, a decompensation increase and rotational prominence. Patients with adult scoliosis may also present with pulmonary problems such as progressive dyspnoea, mostly exertion. This often accompanies a painful curve that is decreasing the patients activities.

Painful and progressive curves are often an indication for surgery. If the curve is painful but not progressive, the patient has to decide if the pain is severe enough to consider surgical treatment. If the pain is at the level of the curve, surgical stabilisation of the curve will usually lower the pain. If the pain is below the curve or at the lumbosacral junction, surgical stabilisation of the curve will often not influence the pain, and may even increase the pain. Progressive functional problems. or respiratory problems may also be an indication for surgery.

In most cases a posterior approach with posterior instrumentation and fusion is preferred. With adequate release and removal of the facet joints, reasonable correction can be achieved. With the newer methods of instrumentation with screws, hooks and sublaminar wires a stable fixation can be achieved and pseudarthrosis is not a major problem. In very severe cases or if the lumbosacral junction is also fused a complimentary anterior release and fusion may be indicated. Traction and bending x-rays are helpful to determine the flexibility of the curves.

The levels of instrumentation depend on the type of the curves and the symptoms of the patient. In the frontal plain the whole curve is instrumented, but if there are degenerative and painful levels below the curve, they may have to be included. In some cases with severe lumbosacral pain, even this area may have to be included in the fusion. In the sagittal plain, kyphosis is often present and when instrumenting lumbar or dorsolumbar curves care must be taken not to stop at the apex of the thoracic kyphosis but to include the whole kyphosis in the fusion.

Between 1990 and 1995, 45 patients had surgery for painful or progressive adult scoliosis at the University Hospital of Louvain. Thirty eight patients were available for follow up. The levels of fusion, clinical and radiographic results are discussed.

The mean age at operation was thirty eight years (21-56) and there were 35 female patients and three male patients. Twenty four patients presented with pain,15 patients complaned about progression of the deformity, three patients had cardio pulmonary complaints, five patients had neurological complaints and 14 patients complained of functional problems. Regarding the pain distribution four patients had only low back pain, nine patients had mainly deformity pain and eight patients had low back and deformity pain. Four patients only had leg pain. There were six thoracic only curves, 23 double major curves, three thoraco-lumbar curves and seven lumbar curves. The mean thoracic Cobb angle pre op was 62 (34o - 1355) and the mean lumbar Cobb angle was 58 (22 - 90o). The lower level of instrumentation was D12 in two patients. LI in three patients, L2 in one patient, L2 in 14 patients, L4 in 14 patients, LS in one patient and St in four patients. In eight patients an anterior release was performed and in seven patients a thoracoplasty to improve the cosmesis. Types of instrumentation were CDI in 17 patients, GDLII in eight patients and ISOLA in 14 patients.

The objective results showed a frontal correction of 47% of the thoracic curves and 40% of the lumbar curve. The sagittal contour was not statistically changed. The subjective results were satisfactory with 65% improvement of the low back pain in 65% of the patients and 48%, improvement of the deformity pain in 48% of the patients. Daily activity was increased in 50% of the patients and not changed in 33%. Ninety-two per cent of the patients felt better about their outlook and seventy one percent of the patients were very satisfied with the surgical treatment and 25% were satisfied. Sixty nine percent of the patients would choose for the operation again.

Early and late completions were low with no neurological complications, and no deep infections. It can be concluded that the surgical treatment of adult scoliosis provides satisfactory results and has a reasonably low complication rate.

RISK FACTORS FOR THE EVOLUTION OF ADULT DEGENERATIVE SCOLIOSIS V Lykomitros, G Sapkas. Z Kostas, D Korres, J Spilsbury, T Pantazopoulos

Laboratory for the Research of Musculo skeletal system, Orthopaedic Department Medical School, Athens University, Athens, Greece The cases of 175 women (mean age 67 years) with degenerative scoliosis were examined in our department. Only patients with no previous history of scoliosis before the age of 20 and with generative lumbar curves greater than tOa were included.

Disc degeneration above and below L4 and the corresponding development of deformity in the sagittal and coronal plain were analysed. The potential for the deformity to increase due to osteoporotic microfractures and the subsequent destabilisation of the spine was also investigated.

We found that the important factors leading to curve progression and therefore requiring surgery include: (1) curves over 30o (2) curves with Nash and Moe grade II and III rotation (3) an intercrest line through the LS vertebra (4) vertebral lateral translation of 6 nun or more (5) osteoporosis with a Z score of -2 or more in a lateral DEXA measurement.

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