Management of femoral shaft fractures in the transitional age group (8 to 14 years) remains controversial. The aim of this retrospective review of 50 consecutive patients was to evaluate the outcomes of various treatment methods.

Fifty patients (39 boys and 11 girls) with femoral shaft fractures treated between January 1998 and March 2003 were followed-up for a mean period of 30 months (9 to 54), using clinical and radiological parameters as well as patient questionnaires. Motor vehicle accidents had caused the femoral fractures in 56%. The fractures were closed in 88% of the patients and there were associated injuries in 16%.

Depending on the degree of comminution and fracture pattern, patients were treated in one of five ways: initial traction, manipulation under anaesthesia and spica immobilisation (16 patients), non-locking nail (18 patients), locking nail (10 patients), plate and screws (four patients) and external fixation (two patients). The mean time to union was 11 weeks (9 to 28). There was an 8% reoperation rate. Complications included limb length discrepancy in eight patients, malrotation in eight, angulation in 12, entry site pain in five, ectopic calcification in six, delayed union in three, knee stiffness in six and scar problems in four. The mean time away from school was 6.5 weeks (2 to 20). At the last follow-up 86% of patients were satisfied with the outcome.

The use of non-locking nails is cost effective, and healing time with non-locking nails is no longer than with locking nails.

The results of fluoroscopically-guided closed needle (3.5-mm diameter) biopsies performed under general anaesthesia on 70 consecutive patients with lumbar spine pathology were reviewed. There were 36 men and 34 women with a mean age of 38 years. Sixty-four patients were Frankel-E, four were Frankel-D and two were Frankel-C.

The mean duration of the procedure (three tissue cores) was 17 minutes. Consultants performed 42% of the procedures and registrars the rest under supervision. All specimens were subjected to histological, cytological and microbiological evaluation. The results show a histological diagnostic yield of 88.57%. Of the 47 infective lesions, 34 were due to tuberculosis, three were pyogenic and 10 were reported as chronic non-specific inflammations.

In the 14 patients with neoplastic lesions, nine were metastatic and five were primary bone tumours. Histological diagnosis was inconclusive in six patients. Traction artefacts were noted in two patients with sclerotic lesions and one patient revealed normal bone. In tuberculosis, the culture was positive in only 29% of patients. The combined Lowenstein-Jensen and mycobacterial growth indicator tube media improved detection time and recovery rate of mycobacteria from smear negative specimens by 10%. The overall diagnostic yield for tuberculosis was 72.34%.

The diagnostic accuracy and safety of fluoroscopically-guided closed needle biopsy is comparable to that of CT scan and it is more cost effective. It should be an integral part of the management algorithm for spinal pathology.