Pedicle screw fixation in the spine.

In August 1993, in the USA, an FDA advisory committee of experts reviewed the scientific data on pedicle screws and decided that not enough was known about the shortterm problems or the long-term safety and effectiveness of these devices to allow their use to be approved. Manufacturers were therefore forbidden to promote the use of screws in the spine, and patients were warned that pedicle screws were experimental. Before discussing the significance of this decision it is appropriate to examine the history of the use of pedicle screws and their clinical indications. Although surgeons had previously inserted screws into the pedicles to immobilise the facet joints in the performance of spinal fusion, the era of the pedicle screw began when Roy-Camille et al (1970, 1976) reported the use of a screw-plate device for the treatment of spinal fractures. The screw of the plate went down the pedicle and into the body of the vertebra, traversing all three columns of the spine (Denis 1984). Their results were first published in French and during the next few years several surgeons in France used their technique or modifications of it (Louis 1986). There were obvious mechanical advantages of this form of fixation over sublaminar wires or hooks attached to the posterior elements, and the fact that the spinal canal was not violated soon attracted surgeons in other countries to its use. Manufacturers hastened to invent and market devices which used the pedicle screw principle, altering the design of the original Roy-Camille plate to deal with actual or perceived problems experienced during its use. The development of rod-screw systems made contouring easier, and encouraged the use of pedicle screws in the treatment of scoliosis. There are now many such systems on the market, but the rate at which they are modified and remodified suggests that the biomechanical input to their design may not always be adequate. In the treatment of fractures of the spine, pedicle screws allow easy manipulation and reduction of displaced vertebrae, even if the posterior elements are fractured. Their use facilitates decompression of the neural elements by distraction, avoiding the need for laminectomy, and