Individuals with diabetes mellitus are reported to have a twofold to threefold increase in the incidence of cholesterol gallstones. A frequently cited but unproven pathophysiologic mechanism for this phenomenon is reduced gallbladder muscle function, which results in stasis and allows for cholesterol gallstone crystal formation and gallstone growth. To date, gallbladder motor function has not been investigated in a well-characterized diabetic population. Therefore, using radionuclide cholescintigraphy, gallbladder filling and subsequent emptying produced in response to an infusion of the octapeptide of cholecystokinin in 30 diabetic patients and 20 control individuals were studied. No difference in any parameter used to assess gallbladder filling was demonstrated in the diabetics when compared with controls. In contrast, gallbladder emptying induced with cholecystokinin-octapeptide (20 ng/kg body wt . h) was reduced in diabetics compared with controls (55% +/- 5% vs. 74% +/- 4%, p less than 0.01). The peak emptying rate in the diabetics was also decreased (5.0% +/- 0.5% per minute) compared with the controls (7.0% +/- 0.6% per minute, p less than 0.02). The observed decreased gallbladder emptying found in diabetics was not related to obesity, type of diabetes, diabetic control, or presence or absence of peripheral neuropathy. The most severe impairment of gallbladder emptying occurred, however, in diabetics with an associated autonomic neuropathy. This subgroup demonstrated a significant reduction in the percentage of gallbladder emptying (40% +/- 8% vs. 62% +/- 5%, p less than 0.04) and the peak ejection rate (3.5% +/- 0.5% per minute vs. 5.6% +/- 0.6%, p less than 0.02) compared with the diabetics without autonomic neuropathy.