Abstract The potential of buccal mucosa as a site for systemic absorption has attracted increased attention in recent years creating a need for new predictive in-vivo models. The aim of this study was to evaluate anaesthetised and conscious Göttingen mini-pigs as a model for buccal drug absorption by testing pH-dependent absorption of metoprolol from a solid dosage form. Buccal tablets buffered to pH 6.2 and pH 8.9, oral liquid and intravenous injection were tested in four conscious and anaesthetised Göttingen mini-pigs in a non-randomised cross-over study. Blood samples were collected and processed before analysis by ultra-performance liquid chromatography with tandem mass spectrometry detection. An ex-vivo flow retention model was applied to study release and retention of the bioadhesive buccal tablets. The Tmax obtained from the two buccal conscious groups (55 ± 5 and 35 ± 5 min) were significantly different to the buccal anaesthetised groups (120 ± 0 and 165 ± 15 min) for buccal tablet pH 6.2 and pH 8.9, respectively. Also, the absolute bioavailability from the anaesthetised buccal tablet pH 8.9 (20.7 ± 4.0%) had a significant increase compared to all other buccal tablet groups. In conclusion, this study showed a pH-dependent absolute bioavailability of metoprolol when administrated as bioadhesive buccal tablets to anaesthetised mini-pigs. The anaesthesia was found to delay the time to reach maximal plasma concentration of metoprolol as compared to the conscious pig model when administrated as buccal tablets.