Introduction The latest Rome III revision of the Milwaukee classification describes 3 types of Sphincter of Oddi dysfunction (SOD). Type 1 is biliary-type pain, abnormal LFTs (>2 times upper limit of normal on ≥2 occasions) and a dilated bile duct (>8 mm). Type 2 is pain and either laboratory or imaging abnormalities and type 3 is pain alone. Type 1 is usually successfully treated with endoscopic sphincterotomy (ES), while Types 2 and 3 respond less consistently and Sphincter of Oddi manometry (SOM) is used to preselect those likely to benefit. There are however concerns with SOM. It is difficult to interpret, time consuming, expensive and associated with a high risk of pancreatitis - this applies to both the solid state and water perfused system, which we used. From 2007 to 2013, our unit performed SOM on type 2 and 3 patients. In 2014, our practice changed to quadrantic ampullary injections of botulinum toxin (100 U) to paralyse the SOD and allow assessment of the response. If clinical improvement persisted to 3 months, a further injection was given. Following two successful responses, ES was performed. Our aim was to assess the safety and efficacy of the two approaches.

Methods 50 patients were studied retrospectively - 34 were evaluated with SOM and 16 with botulinum.

Results 94.2% of patients were female and 76.9% had a prior cholecystectomy. 25 patients were Type 2 and 27 Type 3. In the SOM group (n = 34), 20 patients had high baseline or high frequency contractions and underwent ES. In this group 40% (n = 8) developed pancreatitis post procedure. In 5 patients SOM was normal and no ES was performed, with the pancreatitis rate being 40% (n = 2). In 9 patients technical problems prevented SOM being performed. Instead, these patients had ES, with the pancreatitis rate being 22.2%. Overall the intention to treat pancreatitis rate for SOM was 35.3%. Overall, this group required 149 bed days, with a median stay of 1 day (range 0–20). Three months following SOM; 77.3% had an improvement in symptoms compared to 45.5% at 12 months.

In the botulinum group (n = 16), there were 25 injections and no cases of pancreatitis were reported, a significantly lower rate than SOM (p = 0.005). One patient developed abdominal pain requiring a four day admission. Four patients had no response and had medical manangement instigated for functional disease. The remaining patients are proceding through the management algorithm.

Conclusion SOM, has a high risk of pancreatitis and following ES a modest response rate. The risks and cost do not support this strategy and so new management algorithms are required. Our pilot data suggests twice repeated botulinum injection is a safe investigative procedure and worthy of examination in larger trials.

Disclosure of Interest None Declared