Study Design: Prospective cohort study.

Objective: To assess the safety and efficacy of an intra-operative gamma probe in the surgical treatment of osteoid osteomas and osteoblastomas arising from the spine.

Summary of background data: Spinal osteoid osteomas and osteoblastomas are difficult to localise and may present adjacent to neural structures. Complete surgical excision of the nidus is a pre-requisite for curative resection.

Methods: All patients with a presumptive diagnosis of osteoid osteoma or osteoblastoma were investigated with plain radiography, computed tomography, magnetic resonance imaging and a technitium bone scan. Nine patients underwent surgical excision. 600 MBq of 99m technitium HMDP was administered intravenously three hours prior to surgery. A sterile cadmium telluride detector connected to a digital counter/ratemeter was used to detect gamma radiation emitted by the tumour intra-operatively to assist with localisation and confirmation of complete excision.

Results: Between October 1995 and September 2006, nine patients required surgical excision for seven osteoid osteomas and two osteoblastomas arising from the spine. All patients were between the ages of 9–31 years and presented with back or neck pain. All tumours involved the posterior elements of the spine. Three patients had previous failed treatment including CT-guided radiofrequency ablation and surgical excision. In all cases the counts per second (cps) dropped significantly following excision. For the osteoid osteoma cases, the mean cps dropped from 203.8 (range 60–515) to 72.5 cps (range 10–220) post-excision. For the osteoblastoma cases the mean cps dropped from 373.5 (range 67–680) to 40.5 cps (range 16–65) post-excision. Histological examination confirmed complete excision in all cases. The mean follow-up was 4.5 years (range 0.5 – 11 years). All patients reported disappearance of the characteristic pre-operative pain.

Conclusions: The use of an intra-operative gamma probe helps to localise and confirm complete excision of osteoid osteoma and osteoblastoma arising from the spine. Accurate localisation results in safe excision with maximal conservation of surrounding normal bone, whilst minimising operative time, blood loss, hospital stay and risk of recurrence.

Introduction We report the two-year clinical outcome of a prospective randomised trial comparing Femoral Ring Allograft (FRA) to a Titanium Cage (TC) in circumferential lumbar spinal fusion.

Methods 83 patients recruited to the study fulfilled strict entry requirements (> 6 months chronic discogenic low back pain, failure of conservative treatment, one or two-level discographically proven discogenic pain source). 38 patients were randomised to receive FRA, 45 patients were randomised to receive TC. Posterior stabilisation was achieved with translaminar or pedicle screws. Patients completed questionnaires including Oswestry Disability Index (ODI), Visual Analogue Scale (VAS) and the Short-Form 36 (SF-36) pre-operatively and 6, 12 and 24 months post-operatively.

Results Results were available for all 83 patients with a mean follow-up of 26.5 months (range 24–75 months). Baseline demographic data (age, sex, smoking history, number of operated levels, pre-operative outcome measures) showed no statistical difference between groups (p< 0.05). For patients receiving the FRA, mean VAS (back pain) improved 2.0 points (p=0.01), mean ODI improved 14 points (p=< 0.01), and mean SF-36 scores improved by > 11 points in all domains (p< 0.03) except general health and emotional role. For patients receiving the TC, mean VAS improved 1.2 points (p=0.002), mean ODI improved 5 points (p=0.02); SF-36 improved significantly in only one of eight domains (bodily pain).

Discussion Clinical outcome following circumferential lumbar fusion with FRA readily achieves the accepted mean clinically important differences (MCID). Fusion with TC does not achieve the MCID. The use of TC for circumferential lumbar fusion appears not to be justified.

Objective: To report on the long-term follow-up (mean 14.2 years) of patients with early onset idiopathic scoliosis treated with convex epiphysiodesis (CE) and Luque trolley instrumentation (LTI) without fusion. To evaluate factors influencing curve progression during the adolescent growth spurt in such patients.

Methods: Thirty-one cases of early onset idiopathic scoliosis with documented progression were surgically treated with CE and LTI without fusion between 1984 and 1992. Twenty-three (14 male, 9 female) of 31 cases had reached a minimum of 16 years of age at follow-up. These 23 cases were reviewed. The overlapped ‘L’ configuration of Luque trolley was used for the first 14 cases (prior to 1988), the overlapped ‘U’ configuration was used subsequently in the following 9 cases.

Results: Mean age at operation was 4.3 years (range, 1.5 – 9 years). Mean pre-operative Cobb angle was 65° (range 30° – 95°), and immediate post-operative Cobb angle was 28° (range 10° – 60°). Mean follow-up was 14.2 years (range, 7–19 years). Four cases required insertion of longer Luque rods (mean age of 7.5 years).

Definitive spinal fusion was required in thirteen cases at a mean age of 14.5 years (range 12–23 years), due to progression of scoliosis in 9 cases (mean Cobb angle 55°), and the development of junctional kyphosis in 4 cases. In ten cases the correction obtained was maintained through skeletal maturity (mean Cobb angle at final follow-up 33°). These cases did not require definitive spinal fusion.

The mean growth within the instrumented segment was 3.2 cm (42% of the expected growth). Progression of scoliosis was predicted by pre-operative apical convex rib-vertebra angle (RVA) (p=0.002). Excessive growth within the instrumented segment was predictive of junctional kyphosis but not of scoliosis progression. Age at operation and initial curve magnitude were not found to be significant predictive factors. 72% of overlapped ‘L’ rod construct (10 cases), and 33% of overlapped ‘U’ rod construct (3 cases) had documented curve progression within the adolescent growth spurt and required definitive spinal fusion.

Conclusions: CE and LTI was effective in controlling early onset idiopathic scoliosis, whilst still allowing significant growth. Pre-operative convex RVA was predictive of curve progression. The overlapped ‘U’ rod construct was more effective than the overlapped ‘L’ rod construct in preventing curve progression.

Introduction: Surgical treatment is indicated in Scheuermann’s disease with severe kyphotic deformity, and/or unremitting pain. Proximal or distal junctional kyphosis and loss of correction have been reported in the literature, due to short fusion level, overcorrection, or posterior only surgery with failure to release anterior tethering. We reviewed surgically treated Scheuermann’s kyphosis cases, to evaluate the factors affecting the sagittal balance.

Methods and results: 35 cases (22 male, 13 female) of Scheuermann’s kyphosis were treated surgically in this centre during 1993–1999. Mean age at operation was 21.5 years (14–53 years). The kyphosis was high thoracic (Gennari Type I) in two cases, mid thoracic (Type II) in 11 cases, low thoracic or thoraco-lumbar (Type III) in eight cases, and whole thoracic (Type IV) in 14 cases. Mean pre-operative kyphosis (Cobb angle) was 81° (range 70° to 110°). Ten cases (mean kyphosis 77°) had one stage posterior operation only with segmental instrumentation. Twenty-five cases had combined anterior and posterior (A-P) surgery. Fifteen cases (mean kyphosis 81°) had one stage thoracoscopic release and posterior instrumentation, and 10 cases (mean kyphosis 89°) had open anterior release, followed by second stage posterior instrumentation. Minimum follow-up was 14 months (mean 45 months, range 14–140 months). The mean post-operative kyphosis was 47.2°. Kyphosis correction achieved ranged from 39% after posterior surgery only, to 42% after thoracoscopic A-P surgery, and 48% after open A-P surgery. Mean loss of correction was 12° after posterior only surgery, 9.5° after thoracoscopic A-P surgery, and 6° after open A-P surgery. Four cases of open A-P surgery had additional anterior cages to stabilise the kyphosis before posterior instrumentation; a mean 55% kyphosis correction was achieved in this group, and there was no loss of correction. Younger cases, under 25 years (n=16) had significantly better kyphosis correction (p< 0.05). Two cases (6%) developed distal junctional kyphosis due to fusion short of the first lordotic segment, requiring extension of fusion. Four cases (12%) developed proximal junctional kyphosis requiring extension of fusion; all of them had primary posterior surgery only. Location of the curve (Gennari Type) had no significant influence on the initial curve, degree of immediate correction, or loss of correction. Complications included infection (4 cases), pneumothorax (1 case), haemothorax (1 case), instrumentation failure (3 cases); 3 cases had persistent back pain.

Conclusion: Combined anterior release and posterior surgery achieves and maintains better correction of Scheuermann’s kyphosis. Loss of correction, and proximal junctional kyphosis are more frequent after posterior surgery only, and short fusion. Use of cages anteriorly prevents loss of correction. Correction is better achieved in younger patients, but is not influenced by the location of the curve.

Objective: This study investigates whether fixation down to lumbar spine only can prevent pelvic tilt compared to pelvic fixation, in the surgical treatment of Duchenne Muscular Dystrophy (DMD).

Design: Retrospective and prospective clinical outcome study, with long-term follow up.

Materials and Methods: Nineteen cases of DMD with scoliosis had early stabilisation (mean age 11.5 years, range 9–16) with sublaminar wires and rods, and pedicle screws up to the lumbar spine. This cohort was followed up for a mean 4.2 years (3–10 years). 31 cases in another centre had late stabilisation (mean age 14.5 years, range 10–17), with Luque rod and sublaminar wire fixation, and pelvic fixation using L-rod (22 cases) configuration or Galveston technique (9 cases) and were followed up for 4.6 years (0.5–11.5 years). Post-op morbidity, Cobb angle correction and pelvic obliquity data were collected retrospectively and prospectively for comparison.

Results: In the lumbar fixation group FVC was 58%, the mean Cobb angle and pelvic obliquity were 19.8° and 9° preoperative, 3.2° and 2.2° direct postoperative, and 5.2° and 2.9° at final follow up respectively. The mean estimated blood loss was 3.3 litres and average hospital stay 7.7 days. In the pelvic fixation group FVC was 44%, the mean Cobb angle and pelvic obliquity were 48° and 19.8° preoperative, 16.7° and 7.2° direct postoperative, and 22° and 11.6° at final follow up respectively. The mean blood loss (4.1 litres) and the average hospital stay (17 days) were significantly higher (p< 0.05) compared to the lumbar fixation group. The pelvic fixation group had higher complication rate at the lower end of fixation. No progression of the pelvic obliquity was noted in the lumbar fixation group during follow up

Conclusion: Lumbar fixation may be adequate for scoliosis in DMD, if the stabilisation is performed early, before the pelvis becomes tilted, and scoliosis becomes significant. The caudal pedicular fixation in the lumbar spine stops rotation of the spine around the rods, and prevent pelvic tilt to occur. Pelvic fixation may be necessary in presence of established pelvic obliquity and larger scoliosis, but is associated with higher morbidity and complications.

Objective: To demonstrate possible advantages of combined (motor and sensory) versus single modality (either motor or sensory) intraoperative spinal cord monitoring

Design: Retrospective and prospective clinical study

Materials and Methods: One hundred and twenty-six consecutive operations in 97 patients had peroperative monitoring the lower limb motor evoked potentials (MEPs) to multi- pulse transcranial electrical stimulation (TES), and tibial nerve somatosensory evoked potentials (SEPs). Seventy-nine patients had spinal deformity surgery, and eighteen had surgery for trauma, tumor or disc herniation

Results: Combined motor and sensory monitoring was successfully achieved in 104 of 126 (82%) operations. Monitoring was limited to MEPs alone in two, and SEPs alone in eighteen cases. Neither MEPs nor SEPs were obtainable in two cases with Friedreich’s ataxia. Significant evoked potentials (EP) changes occurred in one or both modalities in 16 patients, in association with instrumentation (10) or systemic changes (6). After appropriate remedial measures, SEPs recovered either fully or partially in all cases (8/8) and MEPs in 10/15. New neurodeficits developed post-operatively in six of the sixteen patients with abnormal EPs, including two in whom SEPs had either not changed or recovered fully after remedial measures. One patient developed S3–5 sensory loss despite full recovery of both SEPs and MEPs. Two patients without neurological consequences had persistent MEP changes. Normal MEPs (but not SEPs) at the end of the operation correctly predicted the absence of new motor deficits. There were no false negative MEP changes.

Conclusion: MEPs are more sensitive than SEPs, but may rarely raise false positive alarm. SEPs are unaffected by anaesthetics and can be monitored more frequently. Combined monitoring is safe, complimentary to each other, and increases sensitivity and predictivity of adverse neorological consequences. True incidence of false positive MEP or SEP changes are difficult to define. Remedial measures after monitoring changes may help cord ischaemia to recover and absence of neurological deficit, therefore, may not indicate a false positive monitoring change.