obstruction. Laboratory and clinical factors that might influence prognosis were reviewed to attempt to externally validate the previously identified factors and model for risk stratification. RESULTSThe median (range) age of the 49 patients identified was 71 (36-91) years, and the median survival was 174 (14-602) days. Tumours were of urological origin in 66% of patients. Patients with prostate cancer had nephrostomy tubes indwelling for a mean of 279 days, vs 190 days ( P = 0.07) for patients with tumours not of prostatic origin. A serum albumin level of > 30 g/L ( P ≤ 0.001), serum sodium < 135 mmol/L ( P = 0.019) and three or more events related to dissemination of cancer ( P = 0.04) were factors associated with a significantly shorter mean survival. Complications related to the nephrostomy tube were experienced by 39% of patients. The model proved useful in stratifying these patients into different risk groups ( P = 0.002). CONCLUSIONConsistent with a previous report we showed that a low serum albumin level and events related to metastatic disease were indicative of a poor prognosis. We also found that a low serum sodium level might be associated with a worse prognosis. We externally validated a model for stratifying patients into different prognostic groups. Palliative decompression is associated with significant morbidity. KEYWORDS

Pathways incorporating routine imaging will improve early diagnosis, but has not been proven to reduce complication rates or hospital length of stay. On the basis of this systematic review conclusions can therefore not be drawn about the pathways described and their benefit to the diagnostic process for patients presenting with abdominal pain.

Large balloon sphincteroplasty of the sphincter of Oddi is a safe and effective adjunct to endoscopic biliary sphincterotomy in difficult to treat choledocholithiasis. The current study demonstrated large balloon sphincteroplasty did not result in more complications despite a tendency to being used for larger stones.

Surgery aiming at an R0 resection remains the mainstay of treatment. We propose the most effective way to grow the knowledge base in New Zealand is the establishment of a national register, thereby allowing better clinical decision-making by interpretation of a larger data set.

No abstract