Pyogenic spondylitis is a neurological and life threatening condition. It encompasses a broad range of clinical entities, including pyogenic spondylodiscitis, septic discitis, vertebral osteomyelitis, and epidural abscess. The incidence though low appears to be on the rise. The diagnosis is based on clinical, radiological, blood and tissue cultures and histopathological findings. Most of the cases can be treated non-operatively. Surgical treatment is required in 10–20% of patients. Anterior decompression, debridement and fusion are generally recommended and instrumentation is acceptable after good surgical debridement with postoperative antibiotic cover.

Between September 2004 and December 2005 we carried out a prospective study of all cases of sepsis of the hip in childhood at a South African regional hospital with a large local population, and which also took referrals from nine rural hospitals. The clinical, radiological, ultrasound and bacteriological features were assessed. All the hips were drained by arthrotomy and the diagnosis was confirmed microbiologically and histologically. Hips with tuberculosis were excluded. The children were reviewed in a dedicated clinic at a mean follow-up of 8.1 months (3 to 18). There were 40 hips with sepsis in 38 patients. Two patients were lost to follow-up. Nine (24%) had multi-focal sepsis. Overall, 13 hips (34%) had a full and uncomplicated clinical and radiological recovery and 25 (66%) had complications. All patients treated by arthrotomy and appropriate antibiotics within five days of the onset of symptoms had an uncomplicated recovery. Initial misdiagnosis was associated with a delay to arthrotomy. However, 'deprivation', consultation with a traditional healer, maternal educational attainment and distance to a primary health-care facility were not associated with delay to arthrotomy. The early correct diagnosis of this condition, common in the developing world, remains a significant factor in improving the clinical outcome.

Background and objective: Endoscopic retrograde cholangiopancreatography (ERCP) is a technically demanding endoscopic procedure that varies from a simple diagnostic to a highly complex therapeutic procedure. Simple outcome measures such as success and complication rates do not reflect the competence of the operator or endoscopy unit, as case mix is not taken into account. A grading scale to assess the technical difficulty of ERCP can improve the objectivity of outcome data. Methods: A I to IV technical difficulty grading scale was constructed and applied prospectively to all ERCPs over a 12 month period at a single centre. The procedures were performed by two senior trainees and two experienced consultants (trainers). The grading scale was validated for construct validity and inter-rater reliability at the end of the study using the χ 2 test and κ statistics. Results: There were 305 ERCPs in 259 patients over the 12 months study period (males: 112, females: 147, age range 17-97, mean 70.3 years). There was overall success in 244 (80%) procedures with complications in 13 (4%): bleeding in five (1.6%), cholangitis in one (0.3%), pancreatitis in five (1.6%), and perforation in two (0.7%). Success rate was highest for grade I, 49/55 (89%), compared with grade IV procedures, 8/11 (73%). There was a significant linear trend towards a lower success rate from grade I to IV (p=0.021) for trainees, but not for trainers. Complications were low in grade I, II, and III procedures, 12/295(4%), compared with grade IV procedures, 1/11(9%). The inter-rater reliability for the grading scale was good with a substantial agreement between the raters (κ=0.68, p<0.001). Conclusion: Success and complications of ERCP by trainees are influenced by the technical difficulty of the procedure. Outcome data incorporating a grading scale can give accurate information when auditing the qualitative outcomes. This can provide a platform for structured objective evaluation. E ndoscopic retrograde cholangiopancreatography (ERCP) is an advanced endoscopic procedure that is not only technically challenging, but also associated with a risk of serious complications. The procedure itself varies from a simple, straightforward diagnostic ERCP to a highly specialised therapeutic ERCP. A recent survey in the UK showed wide differences between endoscopists in success rates (from 76% to 95%) and serious complications (0% to 16%).1 It is far from acceptable to have such a variation in practice and outcomes in any health care system. Studies have shown that skill and experience have a huge impact on the outcomes of ERCP. 3Success and complications depend not only on the endoscopist's experience, but also on the technical difficulty. Auditing ERCP procedures by simple outcome measures such as complication and success, without taking into account the experience and technical difficulty, does not reflect the actual competence of the unit or an individual endoscopist. Units dealing with advanced, therapeutic ERCP procedures may not be achieving the same success rat...