Spinal infections can result in devastating consequences for the patient. Surgery is indicated in certain circumstances. Our institution’s surgical intervention for the treatment of spinal infection was studied. The aim was to identify characteristics helpful for future management. Areas of particular interest were the causative agent, organism identification methods, indications for surgery, materials used for anterior column reconstruction and the ultimate outcome.

A retrospective investigation of all patients receiving surgery for spinal infections between the years of 2004 to 2009 was conducted. All surgical postoperative infections were excluded. Twenty patients were identified in the nominated study period. Twelve males and eight females aged in between 15 to 83 with an average age of 59.4 years. The offending organism was a gram positive in half (50%) of the study group. Staphylococcus aureus was overwhelmingly the representative bacterium (90%). Five cases (25%) were attributed to gram negative organisms while tuberculosis was present in four (20%). No organism was identified in one. Blood cultures identified the causative agent in just over half (55%) of the cases while seven (35%) relied on surgically obtained tissue. Computer tomography guided biopsy was positive in only one case. The most common reason for surgical intervention was a combination of neurological deficit and failing medical management. The second most common reason was in cases with an unidentified organism along with neurological issues. Anterior column reconstruction using cortical strut allograft was performed in four patients while another four received a synthetic spacer device. Two patients had both allograft bone and a spacer device. Only single case of autograft anterior column reconstruction was identified. Two deaths occurred following surgery while all surviving patients with preoperative neurological deficits improved. All surviving patients are currently infection free.

The surgical management of spinal infections can result in advantageous outcomes for the patient, especially in the setting of neurological deterioration or failing medical management. The use of allograft bone and synthetic spacer devices in the midst of infection does not appear to prevent successful organism eradication. The possibility of tuberculosis being the offending bacterium cannot be overlooked.

A lumbar laminectomy is a commonly performed surgical procedure for the decompression of neural structures. The aim of this human cadaveric study is to establish the extent of pars interarticularis remaining at each lumbar level when a laminectomy is performed to the medial edge of the pedicle.

Seven human cadavers with intact lumbar spines were obtained for this study. The lumbar spine was dissected from the body and segmental disarticulation of each level was performed. The isolated lumbar levels had laminectomies performed exposing the neural canal. The vertical alignment of the laminectomy was orientated in line with the medial aspect of the ipsilateral pedicle. The remaining lateral pars interarticularis was measured with a calliper. The procedure was performed bilaterally at each isolated lumbar segment.

Five males and two female cadavers with an age range of sixty-eight years to ninety-five years at the time of death. Fourteen lumbar segments of each respective level were available for study except at L5, where only twelve was possible due the presence of a transitional vertebra in one of the specimens. Taken to the nearest mms, the average width of the remnant pars interarticularis at the L1 level was 4 mm, range 3–6 mm (SD 0.95); L2 6 mm, range 5–7 mm (SD 0.77); L3 8mm, range 4–9mm (SD 1.34); L4 11mm, range 9–14 mm (SD 1.31) and L5 16mm, range 13–17 mm (SD 1.15). One way analysis of variance for each of the groups were performed to establish that the difference recorded was greater than that expected by chance (p< 0.05). The results predictably established the gradual narrowing of the pars interarticularis as the levels ascend cranially from L5.

The medial wall of the pedicle could be used as an indirect means to establish a satisfactory remnant of the pars interarticularis following a laminectomy in the lower lumbar spine, at the levels of L3 to L5. However in the upper two levels direct visualisation of the pars is recommended.

The aim was to determine the longterm results of the Oxford unicompartmental knee replacement implanted by a single surgeon in a community hospital setting.

The results of the first 97 cases performed by a single surgeon between Feb 1991 Feb 1999 were retrospectively reviewed. Demographic and operative data were recorded and patients were assessed with Knee Society knee score and x-ray.

Ninety seven operations were performed on 83 patients. Antero-medial osteoarthritis was the predominant preoperative diagnosis present in 96, gout was present in 2 and avascular necrosis in one. The average age was 70, follow-up 104.3 months with maximum 170 months. Kaplan-Meier analysis was performed and survival was 88% at 124 months. 20 patients (21 knees) had died and 10 (9 patients) had required revision leaving 65 knees available for review. No cases were lost to follow-up. Of the 10 revisions 7 had been revised to a total knee, 4 for lateral wear or pain, one for femoral loosening, 1 for tibial subsidence and 1 for infection. Four of these patients had a previous high tibial osteotomy. The remaining 3 revisions included 2 1iner exchanges for fracture and wear in one patient with varus knees at 11 years and 1 liner exchange for dislocation.

The Oxford UKA gives satisfactory longterm results in antero-medial osteoarthritis. We would caution against using this prosthesis where a previous high tibial osteotomy has been performed.