Mostly retrospective series with limited use of sphincter of Oddi manometry have indicated that early complications are more common when endoscopic sphincterotomy is performed for sphincter of Oddi dysfunction than for common duct stones. The current study was undertaken to prospectively evaluate the frequency and type of complications of endoscopic sphincterotomy performed for sphincter of Oddi dysfunction compared with endoscopic sphincterotomy performed for other conditions. Four hundred twenty-three patients underwent sphincterotomy for sphincter of Oddi dysfunction (166), common duct stone(s) (163), tumor (60), and miscellaneous reasons (34). Patients were observed in the hospital for at least 24 hours after the procedure, and 30-day follow-up data were obtained. The overall complication rate was 6.9%, but complications were more frequent when sphincterotomy was performed for sphincter of Oddi dysfunction than for all other indications (10.8% vs. 4.3%; P = 0.009). Precut sphincterotomy was more frequently required in the sphincter of Oddi dysfunction group (21.1% vs. 11.7%, P = 0.009) but was no more likely to result in a complication (6.2%) than standard sphincterotomy. The risk of a complication was considerable for a small-diameter common bile duct (less than or equal to 5 mm), particularly when sphincterotomy was performed for sphincter of Oddi dysfunction (37.5%). The overall 30-day mortality rate was 1.7%, but the procedure-related mortality rate was believed to be 0.2%. It is concluded that endoscopic sphincterotomy for sphincter of Oddi dysfunction is more hazardous than for other conditions, particularly when a small common bile duct is present.