Jump to Main Content
Development of a physiologically based pharmacokinetic model for flunixin in cattle (Bos taurus)
- Teresa L. Leavens, Lisa A. Tell, Lindsey W. Kissell, Geoffrey W. Smith, David J. Smith, Sarah A. Wagner, Weilin L. Shelver, Huali Wu, Ronald E. Baynes, Jim E. Riviere
- Food additives & contaminants 2014 v.31 no.9 pp. 1506-1521
- blood, cattle, cross-over studies, drug residues, flunixin, food animals, hydrolysis, intramuscular injection, intravenous injection, kidneys, liver, metabolites, milk, milk production, partition coefficients, pharmacokinetics, prediction, simulation models, subcutaneous injection, uncertainty
- Frequent violation of flunixin residues in tissues from cattle has been attributed to non-compliance with the USFDA-approved route of administration and withdrawal time. However, the effect of administration route and physiological differences among animals on tissue depletion has not been determined. The objective of this work was to develop a physiologically based pharmacokinetic (PBPK) model to predict plasma, liver and milk concentrations of flunixin in cattle following intravenous (i.v.), intramuscular (i.m.) or subcutaneous (s.c.) administration for use as a tool to determine factors that may affect the withdrawal time. The PBPK model included blood flow-limited distribution in all tissues and elimination in the liver, kidney and milk. Regeneration of parent flunixin due to enterohepatic recirculation and hydrolysis of conjugated metabolites was incorporated in the liver compartment. Values for physiological parameters were obtained from the literature, and partition coefficients for all tissues but liver and kidney were derived empirically. Liver and kidney partition coefficients and elimination parameters were estimated for 14 pharmacokinetic studies (including five crossover studies) from the literature or government sources in which flunixin was administered i.v., i.m. or s.c. Model simulations compared well with data for the matrices following all routes of administration. Influential model parameters included those that may be age or disease-dependent, such as clearance and rate of milk production. Based on the model, route of administration would not affect the estimated days to reach the tolerance concentration (0.125 mg kg ⁻¹) in the liver of treated cattle. The majority of USDA-reported violative residues in liver were below the upper uncertainty predictions based on estimated parameters, which suggests the need to consider variability due to disease and age in establishing withdrawal intervals for drugs used in food animals. The model predicted that extravascular routes of administration prolonged flunixin concentrations in milk, which could result in violative milk residues in treated cattle.