Purpose: Postoperative dislocation after revision total hip arthroplasty (rTHA) is a frequent complication. Certain risk factors have been well identified (greater trochanter non-union, history of repeated dislocation or infection, multiple operations), but the role of spinal morphology is not well known. The purpose of this prospective study was to determine the role of spinal morphology on postoperative dislocation.

Material and methods: Between September 2000 and March 2002, 49 patients underwent rTHA. The prospective analysis included a preoperative radiographic evaluation of the spinal morphology for lumbopelvic assessment using the Legave and Duval Beaupère criteria. A standard information card was used pre- intra- and postoperatively to record usual patient- and material-related risk factors of dislocation. Five patients experienced postoperative dislocation despite any apparent defect in implant position.

Results: Mean sacral slope was significantly different (p=0.006) between patients with and without dislocation. This difference remained significant (p=0.017) when limiting the study to the 33 patients who had no associated risk factor postoperatively (history of recurrent dislocation or infection, multiple operations, tight non-union of the greater trochanter).

Discussion: Our study demonstrated the role of lumbar morphology on the risk of postoperative dislocation. Spinal morphology modifies the pelvic orientation and thus landmarks habitually used for implantation. It also affects the amplitude of pelvic movement when moving from the sitting to standing position, requiring hip compensation, particularly extension.

Purpose: The Exeter technique opens new perspectives for the treatment of femoral bone loss observed at revision hip arthroplasty. Early migration of the implant, considered by the advocates of the technique to be beneficial when limited, can, in the absence of secondary instability, weaken the cement shield leading to early revision. Several publications on this topic have examined the improvement in primary stability achieved by modifying the impaction technique or by searching for the ideal size of the grafts. The purpose of the present study was to examine the reproducibility of this method and its effect on transformation of the allograft.

Material and methods: We performed a prospective analysis of outcome in 46 patients operated on since 1996. The Poste-Merle-d’Aubigné (PMA) clinical score and the Ling and Gie radiographic score as well as the SOFCOT score for substance loss were determined. We used frozen fragmented allografts without consideration of graft size. A standard sized femoral implant was used in all cases.

Results: Mean follow-up was 3 years (range 12 – 66 months). Four patients were not followed beyond 9 months because of major complications requiring revision surgery (infection, fracture of the femur, malposition) or patient death (stroke). For the remaining 42 patients, loss of femoral stock was scored I in 6, II in 23, III in 13. The functional score improved from 9.13±3.9 preoperatively to 16.07±2.5 postoperatively. Radiographically, bone lines were observed in the graft in 36 patients, associated with bone remodelling in ten. In six patients, the allograft exhibited a heterogeneous aspect. Three implants migrated 4 mm. Defective distal sealing was noted in all three. One prosthesis implanted in a varus position worsened before stabilising.

Discussion: This technique is a reliable method since primary stability of the implant was obtained in 90% of the cases and was maintained during long follow-up. This did not prevent graft remodelling.

Purpose: We analysed retrospectively 32 cases of posterior cervicothoracic fixation for spinal tumours. We evaluated spinal stability, spinal alignment, and associated complications.

Material and methods: Thirty-two patients underwent surgery: 27 men and five women, mean age 52 years, age range 17–72 years. We implanted 96 articular screws in C4 to C6, 54 screws in C7 and 180 pedicular screws in T1 to T8. Nineteen patients had primary lung cancer with spinal invasion, eleven had spinal metastases, one had a chondrosarcoma and one had a myeloma. For the first group of 19 patients, en bloc resection of the tumour with the vertebra was performed: four total vertebrectomies, 15 partial vertebrectomies. In a second group of 15 patients, palliative posterior fixation was performed with laminectomy decompression.

Results: Follow-up ranged from three to 54 months with a mean of 15 months. Mean survival after total or partial vertebrectomy was 16 months (range 3 – 54 months). Survival after palliative decompression was eleven months with a range from five to 19 months. There were no changes in the sagittal alignment in 30 patients: two patients developed mechanical complications late after surgery requiring revision. We did not have any case of screw, plate or rod fracture. There were no neurological complications related to screw insertion either at the thoracic level (180 screws) or the cervical level (96 screws in C4C5C6 and 54 screws in C7). A control scan was available for 21 patients and revealed a malposition of the implanted screws for 2.5% of the screws with no clinical impact.

Discussion: Posterior screw fixation is a good method to stabilise the cervicothoracic spine during tumour surgery. Articular cervical screws and transpedicular thoracic screws provide effective stability postoperatively. In addition, this type of instrumentation does not interfere should subsequent laminectomy or wider resection be necessary.