The main health care gain in the correction of idiopathic scoliosis is cosmetic. Debate exists regarding the optimum implant method of fixation. The use of pedicle screws is the thoracic spine is common. Complications of implant placement are reported less frequently than they occur. The late development of neurological complications has not been reported before and the scoliosis society members need to be aware of the risk specifi-cally associated with increased kyphosis at the cranial end of the fusion. A 33 year old female underwent correction of a 72 degree right thoracic scoliosis. Pedicle screws were used and a costoplasty undertaken. Cord monitoring was satisfactory and there were no neurological symptoms or signs in the postoperative period. At six week review the patient was very pleased with the cosmetic improvement. At 8 weeks post operatively the patient became aware of a weakness in the right foot, at 10 weeks an early review was requested for what was thought to be a drop foot. In clinic at 11 weeks post op there was a sensory level at T5 with paretic gait and weakness grade 3 of the right leg. Imaging revealed an increase in the upper thoracic kyphosis and the upper right screw was confirmed as impinging on cord with MRI and CT. The screw was removed immediately and a rapid recovery occurred. Late complications of pedicle screws are not commonly reported. The upper thoracic spine may be a specific area of increased risk.
In the absence of randomised trials comparisons are often made between historical cohorts in an effort to compare new surgical techniques. This study simply compares two historical cohorts to assess the effect of time on outcome. Using the Oswestry disability index (ODI), low back outcome score (LBO) and visual analogue score (VAS) 305 elective spinal patients with 6 month surgical follow-up were reviewed. Cohort one was 1995–1999 and consisted of 152 cases. Cohort two was 153 cases operated upon during 2000–2005. The pre operative scores were remarkably similar, ODI of 57 in both groups. Suggesting the threshold for surgery remained unchanged with time. The mean improvement in outcome was greater in the later cohort. In the 1995–1999 cohort the improvements were ODI 23, LBO 16, and VAS 3.4, in the second cohort 2000–2005 the change was ODI 28, LBO 20 and VAS 3.9. Recorded complications reduced from 40 to 27. Incidental durotomy was similar at 15 and 16 in both. Anterior approaches resulted in a single venous tear in each group. Revision cases accounted for 16 early and 12 later cases. Probably the most significant difference was the reduction in the number utilising instrumentation. The early group had 53 instrumented cases out of 152, in the later group 35 out of 153 had instrumentation. The number of anterior fusion cases decreased by almost half from 15 to 8 in the later cohort.
