The major objectives of this project were to establish the pattern of basal hepatic bile flow and the effects of intravenous administration of cholecystokinin on the liver, sphincter of Oddi, and gallbladder, and to identify reliable parameters for the diagnosis of sphincter of Oddi spasm (SOS). Eight women with clinically suspected sphincter of Oddi spasm (SOS group), ten control subjects (control group), and ten patients who had recently received an opioid (opioid group) were selected for quantitative cholescintigraphy with cholecystokinin. Each patient was studied with 111-185 MBq (3-5 mCi) technetium-99m mebrofenin after 6-8 h of fasting. Hepatic phase images were obtained for 60 min, followed by gallbladder phase images for 30 min. During the gallbladder phase, 10 ng/kg octapeptide of cholecystokinin (CCK-8) was infused over 3 min through an infusion pump. Hepatic extraction fraction, excretion half-time, basal hepatic bile flow into the gallbladder, gallbladder ejection fraction, and post-CCK-8 paradoxical filling (>30% of basal counts) were identified. Seven of the patients with SOS were treated with antispasmodics (calcium channel blockers), and one underwent endoscopic sphincterotomy. Mean (+/-SD) hepatic bile entry into the gallbladder (versus GI tract) was widely variable: it was lower in SOS patients (32%+/-31%) than in controls (61%+/-36%) and the opioid group (61%+/-25%), but the difference was not statistically significant. Hepatic extraction fraction, excretion half-time, and pattern of bile flow through both intrahepatic and extrahepatic ducts were normal in all three groups. Gallbladder mean ejection fraction was 9%+/-4% in the opioid group; this was significantly lower (P<0.0001) than the values in the control group (54%+/-18%) and the SOS group (48%+/-29%). Almost all of the bile emptied from the gallbladder refluxed into intrahepatic ducts; it reentered the gallbladder after cessation of CCK-8 infusion (paradoxical gallbladder filling) in all eight patients with SOS, but in none of the patients in the other two groups. Mean paradoxical filling was 204% (+/-193%) in the SOS group and less than 5% (P<0.05) in both the control and the opioid group. After treatment, six of the SOS patients had complete pain relief and one, partial pain relief. The basal tonus of the sphincter is variable in patients with SOS, and allows relatively more of the hepatic bile to enter the GI tract than the gallbladder. Due to simultaneous contraction of the sphincter and gallbladder in response to CCK-8, most of the bile emptied from the gallbladder refluxes into intrahepatic ducts, and reenters the gallbladder immediately after cessation of hormone infusion. The characteristic features of gallbladder filling, emptying, and paradoxical refilling with cholecystokinin provide objective parameters for noninvasive diagnosis of SOS by quantitative cholescintigraphy.