Septic arthritis of the hip remains a serious problem in our environment. The diagnosis is difficult and treatment is challenging. Particularly in patients presenting late, the outcome may be poor.

From January 1998 to December 2002, 27 children were admitted to the paediatric unit with the diagnosis of septic arthritis. Their ages ranged from 2 to 12 years and the duration of symptoms ranged from 3 days to 3 months. Pain, fever and a limp were the main clinical features. Only one child had bilateral involvement. Laboratory investigations revealed Staphylococcus in 20 children, Serratia in one, Pseudomonas in one and Haemophilus influenza in one. In four patients no growth was identified. Initial radiological findings ranged from no abnormalities in 18 children, to capsular distention in six and hip dislocation in three. Ultrasonography, CT scan and bone scan were also performed. All patients underwent surgical exploration. Twelve had septic arthritis, six pelvic osteomyelitis and five proximal femoral osteitis. There were two psoas abscesses, a buttock abscess and a case of inguinal lymphadenitis. Follow-up ranged from 3 months to 5 years.

Only eight children had a good outcome and full range of movement. There were six cases of ankylosis, 10 of avascular necrosis and three of dislocation.

Thorough clinical and radiographic examination is advocated in patients with apparent septic arthritis, as misdiagnosis is common. The infection may be in the pelvis, femoral neck or related soft tissues.

Tuberculous sacro-iliitis occurs in fewer than 10% of cases of skeletal tuberculosis. The diagnosis is usually delayed as more common causes of low back pain are sought. Treatment is mainly conservative, with very few patients requiring surgery.

In a retrospective analysis from 1994 to 2004, we reviewed 15 patients, ranging in age from 15 to 60 years, 13 of whom presented with lower back pain and difficulty with walking. Two patients had an abscess over the affected sacro-iliac joint. All patients had tenderness over the sacro-iliac joint. The Gaenslen and FABER stress tests were positive in all patients. Radiographs showed joint space widening, marginal sclerosis of the joints and peri-articular osteopoenia. Technetium 99 bone scan revealed increased uptake in the region of the sacro-iliac joint. CT scan revealed joint space widening, sclerosis and sequestra in the joint space. Only one patient had bilateral tuberculous infection. Two had had an associated lumbar spine lesion. All 15 patients underwent open biopsy. Histological and microbiological reports revealed chronic infection, with Mycobacterium tuberculosis the causative organism. An 18-month program of antituberculous medication was initiated. Ambulation followed wound healing. Follow-up ranged from 5 to 8 years. All 15 patients responded well to this conservative approach.

In patients with low back pain, sacro-iliac disease should always be included in the differential diagnosis. Thorough clinical and radiological examination and laboratory diagnosis is essential to exclude pyogenic infection and tumours.

Lisfranc injuries make up 0.2% of all fractures. With or without midfoot injuries, treatment requires early accurate diagnosis, anatomical reduction and stable internal fixation. Some surgeons prefer K-wire fixation, while others rely on rigid screw fixation, especially of the medial column. To assess the radiological and functional outcome of K-wire fixation of Lisfranc injuries, we carried out a prospective study between January 1999 and December 2000.

The ages of our 15 male and four female patients ranged from 15 to 47 years. Using the Quenu and Kuss system to classify injuries, we treated five isolated, nine homolateral and five divergent injuries. In eight patients there were associated midfoot injuries, and four had compound fractures. We treated 11 fractures with closed reduction and K-wires. Open reduction with K-wire fixation was carried out on eight fractures, including the four compound fractures, within 19 days of admission. All patients were kept non-weight-bearing in a short backslab, and the wires removed at six weeks. Follow-up times ranged from 4 to 19 months.

To assess functional outcome we used the American Orthopaedic Foot and Ankle Society’s midfoot scoring system, which has a maximum score of 100. The mean score of our patients was 70 (52 to 85). Mild or occasional foot pain and slight gait abnormality resulted in limitation of recreational activities. At three months, 15 patients were fully weight-bearing. A single case of superficial sepsis resolved, and there were no cases of implant failure or loss of reduction.

K-wire fixation following anatomical reduction is a satisfactory option for the treatment of tarsometatarsal injuries, especially when severe injuries involve the midfoot. The technique is minimally invasive and the K-wires are easily inserted and removed.