Unexplained abdominal pain after cholecystectomy has been attributed to sphincter of Oddi dysfunction, but no objective diagnostic criteria exist. Biliary excretion was quantitated by computer-assisted cholescintigraphy in 35 postcholecystectomy controls without symptoms, 9 patients with suspected sphincter of Oddi dysfunction (studied before and after sphincterotomy), and 18 patients with overt cholestasis from other causes (6 with extrahepatic obstruction and 12 with parenchymal liver disease). In patients with sphincter of Oddi dysfunction or with cholestasis, the time to attain maximal activity in the biliary system was significantly (p less than 0.05) longer, the percent of radiotracer excreted at 45, 60, and 90 min was less, and the emptying rate was slower compared with the controls. Cholecystokinin (0.02 U/kg X min) did not abolish biliary output, excluding a paradoxical response of the sphincter. After sphincterotomy, biliary activity peaked earlier and the percent excreted at 45 min increased but did not revert to normal. Relief of symptoms occurred in 8 of 9 patients. The one failure had normal emptying characteristics before sphincterotomy, and did not change after surgery. Another developed recurrent pain and a corresponding deterioration in biliary emptying on serial scans. Thus, functional obstruction at the sphincter of Oddi exists, is not due to any paradoxical response to cholecystokinin, and in the absence of overt cholestasis, can be detected by quantitative cholescintigraphy. Successful sphincterotomy may not completely restore biliary emptying to normal.