Patients who undergo endoscopic sphincterotomy for common bile-duct stones, who have residual gallbladder stones, are referred for laparoscopic cholecystectomy. However, only 10% of patients who do not have this operation are reported to develop recurrent biliary symptoms. We aimed to assess whether a wait-and-see policy is justified. We did a prospective, randomised, multicentre trial in 120 patients (age 18–80 years) who underwent endoscopic sphincterotomy and stone extraction, with proven gallbladder stones. Patients were randomly allocated to wait and see (n=64) or laparoscopic cholecystectomy (56). Primary outcome was reoccurrence of at least one biliary event during 2-year follow-up, and secondary outcomes were complications of cholecystectomy and quality of life. Analysis was by intention to treat. 12 patients were lost to follow-up immediately. Of 59 patients allocated to wait and see, 27 (47%) had recurrent biliary symptoms compared with one (2%) of 49 patients after laparoscopic cholecystectomy (relative risk 22·42, 95% CI 3·16–159·14, p<0·0001). 22 (81%) of 27 patients underwent cholecystectomy, mainly for biliary pain (n=13) or acute cholecystitis (7). Conversion rate to open surgery was 55% in patients allocated to wait and see who underwent cholecystectomy compared with 23% in those who were allocated laparoscopic cholecystectomy (p=0·0104). Morbidity was 32% versus 14% (p=0·1048), and median hospital stay was 9 versus 7 days. Quality of life returned to normal within 3 months after either treatment policy. A wait-and-see policy after endoscopic sphincterotomy in combined cholecystodocholithiasis cannot be recommended as standard treatment, since 47% of expectantly managed patients developed at least one recurrent biliary event and 37% needed cholecystectomy. No major biliary complications arose, but conversion rate was high.