Low dose caffeine as a salivary tracer for the determination of gastric water emptying in fed and fasted state: A MRI validation study.

Improving our knowledge about human gastrointestinal physiology and its impact on oral drug delivery is crucial for the development of new therapies and effective drug delivery systems. The aim of this study was to develop an in vivo tool to determine gastric emptying of water by administration of a caffeine as a tracer substance followed by subsequent saliva caffeine analysis. For this purpose, 35 mg of caffeine were given to six healthy volunteers after a 10 h overnight together with 240 mL of tap water either on a fasted stomach or 30 min after the high-caloric, high-fat breakfast recommended for bioavailability/bioequivalence (BA/BE) studies. Caffeine was administered in form of an ice capsule in order to omit the contamination of the oral cavity with caffeine. Parallel to saliva sampling, magnetic resonance imaging (MRI) was applied in order to validate this novel approach. After administration of the ice capsule, MRI measurements were performed every 2 min for the first 20 min followed by further measurements after 25, 30, 35, 40, 50 and 60 min. Saliva samples were collected always 1 min after the MRI measurement in supine position in the MRI scanner and continued for further 240 min. The caffeine concentration in saliva was quantified after liquid-liquid extraction by a validated HPLC/MS-MS method. The obtained MRI data revealed a fast emptying of the co-administered water within 10 to 50 min in the fasted state and likewise in the fed state. Salivary caffeine kinetics showed a C from 150 to 400 ng/mL with a t from 20 to 90 min. MRI data were normalized by setting the maximum emptied volume to 100% and the salivary caffeine kinetics were normalized by setting C to 100%. In order to compare the results obtained by the MRI and the saliva method, the normalized data for each volunteer was correlated based on a linear regression. In the fasted state the mean slope for six comparisons was 0.9114 ± 0.1500 and the mean correlation coefficient was 0.912 ± 0.055. In the fed state, a mean slope of 0.8326 ± 0.1630 and a mean correlation coefficient of 0.887 ± 0.047 were obtained. Based on these results, we could show that salivary caffeine concentrations are suitable to describe the emptying of water as a non-caloric liquid from the fasted and the fed stomach. The presented technique provides a straight-forward, inexpensive and noninvasive method to assess gastric emptying of hydrophilic liquids, which can be broadly used in oral biopharmaceutics. Possible applications are the characterization of real-life conditions, specific populations (e.g. elderly people) and the better understanding of the contribution of gastric emptying to pharmacokinetic profiles of orally administered drugs.