During the last decade or more, the anchors used for instrumentation in scoliosis surgery are predominantly transpedicular screws, according to Suk. The long term radiographical feature of screw fixation after scoliosis surgery is not previously studied.

A consecutive series of 81 cases with AIS operated on with an all screw construct has been studied by means of low dose CT postoperatively and at 2 years postoperatively. There were 67 females and 14 males, with a mean age of 18.3 ± 3 years.

In 26 / 81 (32 %) there were signs of loosing of one or more screws, at a maximum 3 screws. We observed loosened screws in the upper thoracic region in 16 cases, in the thoracolumbar 6 and in lumbar area in 4. Mean pre-op Cobb angle was 56 in cases of loosening and 53 of intact screw fixation (n.s.), the correction rate was 69% in loosened vs 70% among intact screws (n.s.). In males there were signs of loosening in 8/14 (57%) and in females 18/67 (27%). Among cases with loosening, 14% had suboptimal screw positioning postoperatively, in intact cases it was observed in 11% (n.s.). In the whole group there were signs of suboptimal screw positioning 12%. Clinically, 1 case had a loosened L4 screw replaced; and at all 21/26 had no complaints and 5/26 reported minor pain or discomfort. 1/26 had a minor proximal junctional kyphosis about 10°, in 3/26 there was a pull-out of some few mms. With plain radiography loosening could be observed in 11/26 cases; 5 were in the lumbar region.

In a consecutive series of 81 adolescents with idiopathic scoliosis who had underwent scoliosis surgery according to Suk, one third showed, 2 years after the intervention, some minor screw loosening, assessed by low dose CT. One patient had one lumbar screw replaced and only 5 patients reported minor discomfort. Males were more prone to develop screw loosening.

Aims. Loosening of pedicle screws is a major complication of posterior spinal stabilisation, especially in the osteoporotic spine. Our aim was to evaluate the effect of cement augmentation compared with extended dorsal instrumentation on the stability of posterior spinal fixation. Materials and Methods. A total of 12 osteoporotic human cadaveric spines (T11-L3) were randomised by bone mineral density into two groups and instrumented with pedicle screws: group I (SHORT) separated T12 or L2 and group II (EXTENDED) specimen consisting of T11/12 to L2/3. Screws were augmented with cement unilaterally in each vertebra. Fatigue testing was performed using a cranial-caudal sinusoidal, cyclic (1.0 Hz) load with stepwise increasing peak force. Results. Augmentation showed no significant increase in the mean cycles to failure and fatigue force (SHORT p = 0.067; EXTENDED p = 0.239). Extending the instrumentation resulted in a significantly increased number of cycles to failure and a significantly higher fatigue force compared with the SHORT instrumentation (EXTENDED non-augmented + 76%, p < 0.001; EXTENDED augmented + 87%, p < 0.001). Conclusion. The stabilising effect of cement augmentation of pedicle screws might not be as beneficial as expected from biomechanical pull-out tests. Lengthening the dorsal instrumentation results in a much higher increase of stability during fatigue testing in the osteoporotic spine compared with cement augmentation. Cite this article: Bone Joint J 2016;98-B:1099–1105

Fracture of a pedicle is a rare complication of spinal instrumentation using pedicular screws, but it can lead to instability and pain and may necessitate extension of the fusion. Osteosynthesis of the fractured pedicle by cerclage-wire fixation and augmentation of the screw fixation by vertebroplasty or temporary elongation of the fixation, allows stabilisation without sacrifice of the adjacent healthy segment. We describe three patients who developed a fracture of the pedicle in the most caudal instrumented vertebra early after lumbar spinal fusion.

During revision surgery the pedicles were reduced and secured by a soft cerclage wire bilaterally. Fusion was obtained at the site of the primary instrumentation and healing of the pedicles was achieved. Cerclage wiring of the fractured pedicle seems to be safe and avoids permanent extension of the fusion without the sacrifice of an otherwise healthy segment.