Introduction: We report the result of cervical osteotomy in 11 patients using a controlled reduction technique and assess the safety and efficacy of this operation. Methods: Between 1993 and 2006, 11 patients with ankylosing spondylitis underwent correction of cervical kyphosis utilizing an extension osteotomy at the C7/T1 junction. The procedure was carried out under general anaesthesia with spinal cord monitoring. Lateral mass screws were placed from C3–C6 and thoracic pedicle screws placed from T2 to T5. After completion of the osteotomy, the reduction manoeuvre was carried out by the senior surgeon lifting the halo, while bilateral temporary malleable rods (fixed to cervical lateral mass screws) were allowed to pass through top loading thoracic pedicle screws, before tightening by the assistant when the desired position had been achieved. The temporary malleable rods were then replaced with definitive rods, thereby creating a solid internal fixation. A halo vest was maintained for 12 weeks to support the instrumentation and allow the fusion mass to develop.

Results: Surgery was performed on 10 males and one female. The mean age at surgery was 56 years (range 40–74). Duration of symptoms averaged 2.7 years (range 1–5 yrs). The average duration of surgery was 4.7 hours (range 3–6.5) with a mean blood loss of 1938cc (range 1000–3600). The mean follow up was 6.5 years (range 2–13). The mean pre-op chin brow vertical angle was 54º (range 20–70) reducing to 7º (range 2–20) at final follow-up. The mean pre-operative kyphotic angle was 19.2º reducing to minus 34º at final follow up. Restoration of normal forward gaze was achieved in all cases. No patient suffered spinal cord injury or permanent nerve root palsy.

Conclusion: Cervico-thoracic osteotomy is a potentially hazardous procedure. The technique described reduces the risk of translation during the reduction manoeuvre thereby reducing the risk of serious neurological injury.

Introduction: Conventional reduction techniques for high-grade isthmic spondylolisthesis do not address important anatomical constraints on the L5 and S1 nerve roots, thereby leading to a significant risk of neurological deficit. We describe a novel three-stage reduction technique carried out in one operative session that respects these anatomical constraints. We report the results in seven cases.

Methods: Between 2000 and 2006, four female and three male adolescents with high-grade spondylolisthesis (grade 3 or greater) underwent this 3 stage procedure which included: I) extensive posterior decompression of L5 and S1 nerve roots plus sacral dome osteotomy. II) anterior L5/S1 discectomy. III) reduction of spondylolisthesis with pedicle screw fixation and posterior lumbar interbody fusion using interbody cages. Somatosensory and motor evoked potentials were used during the procedure. Patients were followed up for a mean of 4 years (range1–6). Sagittal balance was restored and assessed by measuring sacral slope, lumbosacral angle, pelvic incidence and pelvic tilt.

Results: The mean age at surgery was 14.7 years (range 12–17) and average duration of symptoms was 13.7 months (range6–24). Mean operative time was 6.5 hours (range 5–8), with a mean blood loss of 2242cc (range1400–4200). The mean pre-op slip angle was 57°(range 45°–100°) and the mean post-op slip angle was 37.5°(range28°–57°). Anatomical reduction was achieved in six patients and one patient with spondyloptosis was reduced to grade 2. Sagittal balance was restored in all patients. There were no permanent neurological complications. One patient with grade 4 spondylolisthesis developed transient right L5 nerve root palsy which fully recovered within 3 months.

Conclusion: The safety and efficacy of this 3 stage reduction and stabilization procedure showed that immediate reduction of high grade spondylolisthesis with minimal risk of neurological deficit is possible. The procedure is technically demanding and should be performed by spinal surgeons familiar with the principles of anterior and posterior fusion.