Objectives: We aimed to compare the antibiotic concentrations obtained using the volumetric absorptive microsampling (VAMS) devices with those determined in plasma from conventional venous blood collected within the frame of a pharmacokinetic study of amoxicillin (AMO), metronidazole (MET), azithromycin (AZI), commonly used for periodontal treatment. The suitability and overall, acceptability of the VAMS approach was also ascertained by both participants of the pilot study and dentist practitioners.

Materials and methods: Twelve volunteers (6 subjects without periodontal problems (PH), and 6 individuals affected with periodontitis (PP)) were administered 500 mg each of amoxicillin, metronidazole, and azithromycin. Paired venous blood (VB) and capillary VAMS samples were collected at 2-, 6-,10-, 24-, 48- and 96-hours post-antibiotics administration. Antibiotic concentrations were determined using multiplex liquid chromatography coupled tandem mass spectrometry (LC-MS/MS). Statistical analyses included Mann-Whitney U tests and t-tests.

Results: Significant differences in antibiotic concentrations were observed between VAMS and venous blood (VB) collection methods, across different time points for the three antibiotics (p < 0.05). AMO concentrations in VB were 3.5-fold higher (p < 0.01) than in VAMS at early time points (2, 6, 10 h (h)). MET levels in VB were 1.5-fold higher than in VAMS at 2 h and 6 h, (p < 0.01), but this difference disappeared after 10 h. Alternately, while AZI levels were similar in VB and VAMs 2 h after administration, AZI concentrations in VB and VAMS declined non parallelly, with VB levels decreasing to about 60 to 25% of those measured in VAMS over the observed 96 h interval. Antibiotic exposures were not different in the PH and PP groups. Differences in antibiotics concentrations determined in VB and VAMS samples are a direct consequence of (i) the matrices used for analyses (plasma in VB, vs. whole blood with VAMS), (ii) the subjects' hematocrit, and (iii) the distinct cell distribution pattern of antibiotics with AMO characterized by a weak penetration in red blood cells (RBC) while AZI tends to progressively concentrate into RBC. MET was present at higher concentrations in plasma until 6 h which thereafter tended to re-equilibrate equally in plasma and RBC.

Conclusion: Though VAMS yielded significantly different results compared to plasma, it effectively reflects the concentration evolution of the antibiotics and could be an alternative in pharmacokinetic studies and therapeutic monitoring.