In human plasma, the total concentration of non-protein binding (NPB) drugs is equal to the free drug concentration because NPB drugs do not or hardly bind to plasma proteins. Thus, centrifuge ultrafiltration (CF-UF) has been used in the determination of the concentration of NPB drugs in human plasma. However, with only a common centrifugation, the recovery and the reproducibility were not as excellent as expected. In addition, we discovered that the values of the volume ratio of ultrafiltrate to sample solution (Vu/Vs) were different and could not be well controlled, which may affect the determination of the drug concentration. The problem also affected the determination of other NBP drugs. In the present work, we used biapenem as a representative drug to study the effect of Vu/Vs on the analysis of NPB drugs concentration in human plasma. The results showed that a Vu/Vs value of less than 0.4 had no effect on the analysis of free drug concentration, while a Vu/Vs value of more than 0.4 was associated with increased recovery rate and overestimation of drug concentration. Therefore, to maintain a Vu/Vs value of less than 0.4 and even at a constant value is the key to accurately determine the concentration of NPB drugs in plasma. Fortunately, with an HFCF-UF device, the Vu/Vs could be well controlled and kept at 0.08 in this study. The recovery rates were almost 100% and the analysis precision was greatly improved. In pharmacokinetics studies, this method was successfully employed to determine the concentration of biapenem with excellent accuracy and reproducibility. HFCF-UF may become a feasible platform for the determination of NPB drugs.

A simple sample preparation method has been developed for the determination of amoxicillin in human plasma by hollow fiber centrifugal ultrafiltration (HF-CF-UF). A 400-μL plasma sample was placed directly into the HF-CF-UF device, which consisited of a slim glass tube and a U-shaped hollow fiber. After centrifugation at 1.25 × 10(3) g for 10 min, the filtrate was withdrawn from the hollow fiber and 20 µL was directly injected into the high-performance liquid chromatography (HPLC) for analysis. The calibration curve was linear over the range of 0.1-20 µg/mL (r = 0.9996) and the limit of detection was as low as 0.025 µg/mL. The average recovery and absolute recovery were 99.9% and 84.5%, respectively. Both the intra-day and inter-day precisions (relative standard deviation) were less than 3.1% for three concentrations (0.25, 2.5 and 10 µg/mL). The sample preparation process was simplified. Only after a single centrifugal ultrafiltration can the filtrate be injected directly into HPLC. The present method is simple, sensitive and accurate. It could be effective for the analysis of biological samples with high protein contents, especially for the biopharmaceutical analysis of drugs that use traditional isolation techniques for sample preparation such as the protein precipitation method.

Aim: A urine drug concentration method was developed for the evaluation of oral ibandronte absorption based on the fact that ibandronate is excreted unchanged by the kidneys. Methodology: Ibandronate was isolated from the urine matrix by coprecipitated with 2.5 M CaCl2 and 1 M K2HPO4 in basic conditions. After a liquid–liquid extraction, the analytes were derivatized with trimethylsilydiazomethane prior to detection. Results: The calibration curves exhibited excellent linearity (r > 0.99) between 1 and 250 ng/ml in human urine. Ibandronate was recovered >86.5% with the inter- and intraday relative standard deviations less than 15%. Conclusion: The method is selective, accurate, practical and was successfully applied to study the influence of vitamin D3 on the absorption of ibandronate.