Celiac disease is associated with impaired cholecystokinin (CCK) release. The mechanism by which CCK release is impaired is poorly understood and seems to be related to the mucosal atrophy or to decreased stimulation due to reduced intraduodenal nutrient hydrolysis. The aims of our study were to evaluate basal and postprandial CCK in celiac patients presenting with distinctive types of mucosal lesions (normal, infiltrative and atrophic), and to study the role of protein hydrolysis on CCK release. Plasma CCK was measured in 20 celiac patients (normal mucosa: n=6; infiltrative type: n=6; atrophic type=8) and 9 controls, before and after ingestion of a polymeric or a semi-elemental meal. Significant decreases in basal CCK plasma (B 0.6 [95% CI, 0.3-1.3] pmol/l; p<0.003) and postprandial CCK area under curve (AUC 34 [19-61] pmol/l x 120 min, p<0.0001) were observed in patients with an atrophic mucosa compared with treated patients (B 1.6 [1.0-2.4] pmol/l, AUC 267 [172-414] pmol/l x 120 min) or healthy volunteers (B 1.0 [0.7-1.4] pmol/l, AUC 186 [131-264] pmol/l x 120 min). A significant defective CCK release was also observed in patients with an infiltrative type: B 0.4 [0.2-0.7] pmol/l and AUC 56 [31-101] pmol/l x 120 min; p<0.0001. Administration of a semi-elemental diet did not correct the defective CCK release. In conclusion, the decreased CCK levels observed in celiac patients are not strictly related to the mucosal atrophy but rather to the lymphocytic infiltrate. Administration of a predigested meal did not correct the impaired CCK release. Some inhibitory mechanism could be involved in the CCK cell dysfunction observed in celiac patients presenting with lesser degrees of disease activity.