Purposes of the study. The most common fracture of the cervical spine in the elderly population is a fracture of the odontoid peg. Such fractures are usually not displaced and these are commonly treated non-operatively. Rarely though, peg fractures are displaced and then their management is less straightforward. This is in part because the group of patients who sustain them frequently have complex and pre-existing medical co-morbidities and in part because a new neurological injury may have been sustained as a result of the peg fracture itself. Many options for the management of displaced peg fractures, both operative and non-operative have been described in the literature and discussion continues as to which technique is superior and in which patient population. The purpose of this study was to follow-up those patients who were managed operatively in our unit between 2007 and 2009. Methods and Results. We present our case series of 4 patients who sustained significantly displaced fractures of the odontoid peg with accompanying neurological injury, who were treated with posterior stabilisation using the Harms technique. Conclusions. We have found this method to be safe and reliable. It not only yields a good surgical outcome, but allows patients' rehabilitation to be optimised, maximising functional improvement. We believe the technique is superior to anterior stabilisation in this patient population because it utilises superior posterior bone quality and mechanical fixation. The approach in our unit is to treat elderly patients with displaced odontoid peg fractures according to the same principles as would be followed in managing those that have disabling injuries such as fractures of the femoral neck; early stabilisation and early mobilisation in those patients whose co-morbidities allow it

Traumatic atlanto-occipital dislocation in adults is usually fatal and survival without neurological deficit is rare. The surgical management of those who do survive is difficult and controversial. Most authorities recommend posterior occipitoaxial fusion, but this compromises cervical rotation. We describe a case in which a patient with a traumatic atlanto-occipital disruption but no neurological deficit was treated by atlanto-occipital fusion using a new technique consisting of cancellous bone autografting supported by an occipital plate linked by rods to lateral mass screws in the atlas. The technique is described in detail. At one year the neck was stable, radiological fusion had been achieved, and atlantoaxial rotation preserved.

The rationale behind this approach is discussed and the relevant literature reviewed. We recommend the technique for injuries of this type.