Article pubs.acs.org/JAFC
Excretory, Secretory, and Tissue Residues after Label and Extra-label Administration of Flunixin Meglumine to Saline- or Lipopolysaccharide-Exposed Dairy Cows David J. Smith,*,† Weilin L. Shelver,† Ronald E. Baynes,# Lisa Tell,§ Ronette Gehring,⊥ Mengjie Li,⊥ Terry Dutko,Δ J. W. Schroeder,‡ Grant Herges,† and Jim E. Riviere⊥ †
Biosciences Research Laboratory, USDA-ARS, 1605 Albrecht Boulevard, Fargo, North Dakota 58102, United States College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina 27607, United States § School of Veterinary Medicine, University of California, Davis, California 95616, United States ⊥ Institute of Computational Comparative Medicine, College of Veterinary Medicine, Kansas State University, Manhattan, Kansas 66506, United States Δ Office of Public Health Science, Midwestern Laboratory, USDA-FSIS, 4300 Goodfellow Boulevard, St. Louis, Missouri 63120, United States ‡ Department of Animal Sciences, North Dakota State University, P.O. Box 6050, Fargo, North Dakota 58108, United States #
S Supporting Information *
ABSTRACT: Twenty lactating dairy cattle were intravenously infused with either lipopolysaccharide (LPS) (n = 10) or sterile saline (n = 10). Five cattle in each group received three doses of flunixin meglumine administered by either intravenous infusion or intramuscular injection at 24 h intervals. Milk, urine, and tissues were collected. Thirty-six hours after the last flunixin administration, milk from six cows contained 5-hydroxyflunixin (5OHF) levels greater than the milk tolerance of 2 ng/mL; by 48 h, milk from two cows, a saline and a LPS-treated animal, had violative milk concentrations of 5OHF. A single animal treated with LPS and intramuscular flunixin contained violative flunixin residues in liver. The ratio of urinary flunixin/5OHF was correlated (P < 0.01; R2 = 0.946) with liver flunixin residues in LPS-treated animals, but not (P = 0.96; R2 = 0.003) in cows treated with saline in lieu of LPS. Violative residues of flunixin in dairy cattle may be related to LPS inhibition of flunixin metabolism. KEYWORDS: dairy, extra-label, flunixin, holstein, liver, metabolite ratio, milk, residue
■
INTRODUCTION
bases for these violations remains to be of interest to producers, regulatory agencies, and consumers. Although the United States has an excellent regulatory system for minimizing drug residues in meat and milk products, instances of violative residues do occur. For example, in 2011 the USDA FSIS confirmed 974 instances of violative chemical residues in inspector-generated samples screened using the kidney inhibition swab (KIS).4 Of these violations, 15% (142) were violations with flunixin. All of the flunixin violations were in cattle, with no violations in hogs. The major contributor to total flunixin violations was in dairy cows, comprising 63% of the total violations, with veal calves (23%) and beef cows and steers (14%) accounting for the remaining violations.4 Data from 20125 indicated a similar trend with 928 instances of violative chemical residues in inspector-generated samples screened using the KIS assay and confirmed with selective assays. Of these violations, 11% (101) were due to excessive flunixin residue. Similar to 2011, all of the flunixin residue violations occurred in cattle. Dairy cows comprised 58% of the
Flunixin meglumine is a nonsteroidal anti-inflammatory analgesic with antipyretic activities. Approval for flunixin use in cattle to treat pyrexia associated with bovine respiratory disease and to treat pyrexia and inflammation in endotoxemia has been granted1 by the U.S. Food and Drug Administration (FDA) Center for Veterinary Medicine (CVM). The label was extended in 20042 to include lactating dairy cattle for the control of pyrexia associated with mastitis. Under label specifications, flunixin meglumine may be administered to cattle as an intravenous infusion (iv) of 1.1−2.2 mg/kg body weight. Infusion should occur either once per day as a single dose or twice per day as two doses separated by approximately 12 h for a maximum of three consecutive days. In practice it is often difficult to administer flunixin meglumine or other drugs by iv infusion to cattle, especially on a repeated basis. As a consequence, flunixin meglumine is commonly administered via intramuscular (im) or subcutaneous (sc) injection. In 2007, the CVM reminded veterinarians that im flunixin meglumine has the potential to cause violative residues in meat or milk products of cattle and that convenience is not an acceptable basis for extra-label flunixin use.3 Due to the persistent problem of violative flunixin residues, especially in dairy cattle,4,5 the © 2015 American Chemical Society
Received: Revised: Accepted: Published: 4893
March 25, 2015 April 24, 2015 April 27, 2015 May 7, 2015 DOI: 10.1021/acs.jafc.5b01509 J. Agric. Food Chem. 2015, 63, 4893−4901
Article
Journal of Agricultural and Food Chemistry 2011 flunixin violations; veal calves (19%) and beef cows and steers (14%) were the bulk of the remaining violations. The magnitude of flunixin residues has been reported sufficient to cause the contamination of bulk milk. For example, Kissell et al.6 measured a violation rate of 0.2% for residues of 5hydroxyflunixin (5OHF, the marker residue for flunixin in milk) in bulk milk from tanker trucks. Deyrup et al.7 have suggested that the probability of a violative flunixin residue increases in cull dairy cows that appear as “suspect”, with suspect including animals that appear to be lame, that appear to have mastitis, metritis, or that have lesions associated with injection. Kinetic studies in beef cattle8 have shown that rates of plasma flunixin depletion did not differ (P > 0.05) after subcutaneous or intravenous administration, but a study with larger numbers of dairy cattle indicated that the dosing route did influence terminal half-lives of flunixin.9 Another contributing factor to flunixin violations could be drug use with an insufficient preslaughter withdrawal period. Alternatively, violations could occur after flunixin is dosed to sick animals even after observation of the required 4 day withdrawal period. The latter instance could occur if illness affects the kinetics of flunixin elimination, as suggested by the data of Kissell et al.10 in mastitic cows treated with flunixin. For example, because flunixin tends to accumulate in inflamed tissues,11,12 its release from those tissues in animals with systemic infection could be altered significantly. In addition, flunixin efficacy is maintained even when plasma or serum concentrations are very low.13 Collectively, such observations suggest that nonplasma tissue pools of flunixin exist, but the locations of these “deep compartments” are unknown. Pharmacokinetic parameters of serum flunixin, however, did not differ after experimental induction of mastitis by mammary inoculation with Escherichia coli.14 In cows having naturally occurring mastitis, however, flunixin clearance (mL/h/kg) decreased and area under the curve (μg·h/mL) increased, relative to healthy cows.10 We hypothesized that route of administration or systemic inflammation could significantly alter flunixin kinetics and tissue residues in lactating dairy cows. If true, flunixin violations could occur in healthy or sick animals dosed intramuscularly with the maximum label dose (2.2 mg/kg bw) for the maximum dosing period of three consecutive days, but slaughtered with the label withdrawal period of 4 days. The systemic inflammatory responses observed during mastitis in cows can be mimicked by systemic lipopolysaccharide (LPS) administration.15 Therefore, the objective of this study was to determine the tissue residues of flunixin after intravenous (label) and intramuscular (offlabel) administration of flunixin meglumine to healthy and LPSchallenged dairy cows in regulated (milk, muscle, liver, kidney, adipose) and nonregulated (offal) tissues and urine.
■
Table 1. Study Design and Replication treatment intravenous +LPS
−LPS
+LPS
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
total n
5
5
5
5
trial trial trial trial trial
−LPS
intramuscular
A B C D E
replicate
a
b
Each treatment was replicated five times using a single cow per treatment per trial. bLipopolysaccharide (0.2 μg/kg BW). a
urine, and saliva were serially collected at predetermined intervals. Saliva data, however, are not reported in this treatise. A detailed animal use protocol was reviewed and approved by the North Dakota State University Institutional Animal Care and Use Committee (IACUC) prior to the initiation of the live phase of the study. Animals and Animal Husbandry. Lactating dairy cattle were purchased from either the North Dakota State University dairy herd (8 animals) or from a commercial dairy (12 animals) located in Jamestown, ND, USA. Upon delivery to the NDSU Animal Nutrition and Physiology Center (ANPC), each cow was weighed and placed in a tie stall with full access to water and a corn silage-based ration. Cows were milked twice daily at approximate 12 h intervals using a singlecow portable milking machine (InterPuls, Albinea, Italy). Animals were adapted to the ANPC barn facility for a minimum of 7 days prior to the start of the study. Animal weights, used as the basis for dosing LPS and flunixin, were collected 2−3 days prior to the actual dose date. Cow weights did not differ (P = 0.83) between the iv-saline, ivLPS, im-saline, and im-LPS treatment groups (645 ± 75, 607 ± 92, 606 ± 62, and 636 ± 93 kg, respectively; mean ± SD). LPS or Saline Administration. LPS purified from E. coli 0111:B4 (Sigma Chemical Co., St. Louis, MO, USA) was dissolved (10 μg/mL) in sterile saline. Immediately prior to administration of LPS or saline, cows were placed into a squeeze chute, and an 18 gauge, 6.25 cm, catheter (Milcath, Elanger, KY, USA) was placed into the jugular vein. Animals selected for LPS treatment were administered 0.2 μg/kg BW LPS (0.02 mL/kg BW) via iv infusion over a 1−2 min period; cows randomly selected for saline treatment were infused with 0.02 mL/kg saline in the same manner as LPS-dosed cows. Catheters were removed immediately after LPS or saline infusion. Flunixin Administration. Intramuscular flunixin meglumine (Banamine; Merck Animal Health; Millsboro, DE, USA) was delivered via 3.8 cm, 18 gauge, needles into neck muscles in locations caudal to the head and 7.6−10.2 cm abaxial to the spine (Supporting Information Figure S1). Sites for flunixin administration, and subsequent “injection site” recovery at slaughter, were marked by shaving a series of three sequential, vertically oriented patches (∼16 cm2 for each patch) for each injection day. Patches were also shaved on the necks of iv-treated animals to facilitate collection of equivalent muscles at slaughter. Between-day injections alternated between right and left aspects with separation between day 1 and 3 injections of approximately 10 cm (Figure S1). The maximum volume for a single injection site was 10 mL; because all animals required total daily doses of >10 mL, separate injections were delivered to the required volume in a vertical alignment with respect to the initial, within-day injection. The time of each flunixin treatment was recorded. Milk Collection. Cows were milked twice daily before and after the initiation of the study protocol with milk yield being measured gravimetrically. Milk was thoroughly mixed, and multiple aliquots (∼10 mL) of milk from each time point were frozen (≤ −20 °C) until analysis. Urine Collection. Urine samples were collected prior to dosing and at 2, 4, 8, 12, and 24 h on the initial dosing day and on dosing day 2; subsequent to dosing on day 3, urine was collected at 2, 4, 8, 12, 24,
MATERIALS AND METHODS
Study Overview. The study was designed and conducted as a 2 × 2 factorial experiment (Table 1) with main factors being route of drug administration (iv or im) and lipopolysaccharide challenge (LPS or saline). The study was conducted as five replicate trials for a total “n” of five cows per treatment, with each treatment being replicated once in each trial. Lactating Holstein cows were randomly assigned to receive either an iv infusion of 0.2 μg/kg BW of LPS (endotoxin) or saline. Two hours after LPS or saline infusion, each cow received 2.2 mg/kg BW of flunixin meglumine either by iv infusion or by im injection. Flunixin administration was repeated 24 and 48 h after the first flunixin dose; cows were slaughtered 96 h after the final flunixin administration. During the dosing and withdrawal periods, blood, milk, 4894
DOI: 10.1021/acs.jafc.5b01509 J. Agric. Food Chem. 2015, 63, 4893−4901
Article
Journal of Agricultural and Food Chemistry 36, 48, 60, 72, and 96 h. Urine was collected by inducing micturition or during nonstimulated urination; samples were frozen (≤ −20 °C) until analysis. Blood Collection. Blood samples were collected by jugular venipuncture (21 ga needles) on treatment day 1 at 0, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 18, and 24 h. Subsequent to dosing on day 2, blood was drawn at 0.5, 1, 2, 4, 8, 12, and 24 h. After dosing on treatment day 3, blood was drawn at 0.5, 1, 2, 4, 8, 12, 24, 36, 48, 60, 72, 84, and 96 h. Blood samples were collected into BD Vacutainer tubes containing 158 units of sodium heparin, gently inverted eight times, and centrifuged at 1160g for 10 min. Plasma aliquots were distributed into replicate, labeled, 2 mL polypropylene vials and were placed in a freezer (≤ −20 °C) until analysis. Tissue Collection. Cows were slaughtered by captive bolt stunning followed by rapid exsanguination16 96 h after the last flunixin administration. Cattle were exsanguinated and eviscerated using standard industry practices. Day 1, 2, and 3 injection sites were removed so that approximately 500 g of tissue was gathered per site. Liver, kidney, skeletal muscle (longissimus dorsi), and adipose tissue were collected, diced, and placed into labeled containers. Livers were sampled, after removal of the gall bladder, by removing a cross section of 6 ± 2 cm width across all lobes. Kidneys were collected in their entirety, and the fibrous capsule was removed, diced, pooled, and frozen. All tissue samples were frozen at ≤ −20 °C until analysis. Specialty tissues including lung, tongue, heart, brain, tail base (oxtail), reticulum, abomasum, ileum, and mammary tissue were collected. Aliquots of the gastrointestinal tissues were removed, and contents were removed by rinsing with water, placed in labeled containers, and frozen until analyses. Sample Stability. On the basis of stability data published by Boner et al.17 flunixin in nonhepatic tissues is stable for a minimum of 2 months and in hepatic tissues is stable for a minimum of 1 year when stored at ≤ −20 °C. In addition, flunixin residues have been demonstrated to be stable after three freeze−thaw cycles.17 Jedziniak et al.18 established that 5OHF is stable when stored at −20 ± 2 °C for 9 months. Thus, additional stability studies were not conducted. Nevertheless, all tissue analyses were conducted within 1 year of tissue collection. Analyses. Milk flunixin and 5OHF concentrations were determined at the North Carolina State University Center for Chemical Toxicology Research and Pharmacokinetics as described by Kissell et al.10 The milk assay limit of quantitation was 0.1 ng/mL for each analyte. Plasma flunixin and 5OHF concentrations were quantified at the University of California Davis School of Veterinary Medicine, Veterinary Drug Residue Laboratory, as described by Buur et al.19 and modified with respect to the UPLC system by Shelver et al.8 In addition, the flunixin standard curve was expanded to include points at 0.5, 1, 2, 20, 200, 500, 1000, and 2000 ng/mL in the current study. Limits of detection and quantification were 0.1 and 0.5 ng/mL for flunixin and 0.3 and 0.9 ng/mL for 5OHF, respectively. Flunixin residues in liver and skeletal muscle were determined at both the USDA FSIS Midwestern Laboratory in St. Louis, MO, USA, and at the Biosciences Research Laboratory in Fargo, ND, USA. The FSIS Midwestern Laboratory used slight modifications of method CLG-FLX4.03.20 Briefly, the method employed a Waters Acquity UPLC triple-quadrupole instrument with a 1.7 μm, BEH C18, 2.1 × 50 mm column and gradient mobile phase consisting of 0.4% formic acid (solvent A) and 4 parts of CH3CN in 5 parts of MeOH with 0.18% formic acid (solvent B) at a flow rate of 0.650 mL/min. The solvent gradient (8 min total) was as follows: 0 min, 55:45 A/B; 0.42 min, 65:35 A/B; 5.53 min, 65:35 A/B; 5.64 min, 5:95 A/B; 6.36 min, 5:95 A/B; 6.42 min, 55:45 A/B; 8.00 min, 55:45 A/B. The FSIS method quantifies tissue residues to only half of the tolerance of 125 ng/g in liver and 25 ng/g in skeletal muscle. Flunixin free acid in edible tissues, injection sites, and specialty tissues was determined at the USDA ARS Fargo laboratory using the method of Boner et al.17 on which method CLG-FLX420 is based. However, matrix-matched standard curves were prepared at 0.25, 1.25, 6.25, 12.5, and 18.75 ng flunixin free acid/mL of extract corresponding
to tissue concentrations of 1, 5, 25, 50, and 75 ng/g, which allows quantitative analyses well below half the tolerance of skeletal muscle. With each sample set, blank tissues were fortified at 5 and 50 ng/g and extracted contemporaneously with incurred samples. Flunixin was measured using ultrahigh-performance liquid chromatography (Acquity, Waters, Milford, MA, USA) coupled to a tandem quadrupole mass spectrometer (Acquity; Waters) using the conditions described in FSIS method CLG-FLX4.03.20 Liver 5OHF concentrations were measured exactly as described for flunixin except that a water-based external standard curve equivalent to 1, 5, 25, 50, and 75 ng/g of tissue was used for quantitation based on the m/z 313 → 295 ion transition with qualifier m/z transitions of 313 → 280 and 313 → 109. Details of urinary flunixin and 5OHF analyses will be described elsewhere (Shelver et al., in preparation). Briefly, 1 part of urine and 4 parts of 6 N formic acid were mixed and heated at 120 °C for 2 h. Urine samples were then diluted to 1:100 in 50% acetonitrile in water containing 250 ng/mL of flunixin-d3 and 5-hydroxyflunixin-d3 as internal standards. If analyte concentration for a given urine sample was outside the range of the calibration curve, fresh urine aliquots were rediluted at either 1:10 in 50% aqueous acetonitrile or 1:250 using 1:100 control urine as diluent, depending on approximations from the original analysis, and reanalyzed. A matrix-matched calibration curve with appropriate dilution strength was run concurrently with all incurred samples. Within each set, blank urine was fortified at 1, 20, and 500 ng/mL each of flunixin and 5OHF and analyzed concurrently with unknowns. Quantitative analyses were performed on an Acquity UPLC (Waters Corp., Milford, MA, USA) coupled to a tandem quadrupole mass spectrometer with a heated electrospray ionization source operated in the positive ion mode (Acquity; Waters). The column was an Acquity UPLC HSS T3 (1.8 μm, 2.1 × 100 mm) maintained at 35 °C. The mobile phase was 0.1% acetic acid/ acetonitrile (32:68) at a flow rate of 0.4 mL/min. Ions were monitored in the multiple reaction monitoring mode with flunixin 297 → 279, flunixin-d3 300 → 282, 5OHF 313 → 295, and 5OHF-d3 316 → 298 serving as quantified transitions. Flunixin 297 → 264 and 297 → 259, flunixin-d3 300 → 264 and 300 → 112, 5OHF 313 → 280 and 313 → 109, and 5OHF-d3 316 → 280 and 316 → 112 transitions each served as qualifiers. Pharmacokinetic Parameter Estimates. Noncompartmental pharmacokinetic analyses of flunixin plasma concentration versus time profiles were performed on an individual cow basis using Phoenix NCA module (version 1.3; Certara, Cary, NC, USA). Estimates of Cmax and tmax after im injection, flunixin clearance at steady state (CLss/ F), volume of distribution at steady state (Vss/F), area under the plasma concentration−time curve from time zero to the time of the last observation (AUClast), mean residence time (MRT), and half-life (t1/2) were calculated using standard procedures. Statistical Analyses. Within collection time, the effects of LPS treatment and route of flunixin administration on milk yield were evaluated using a two-way analysis of variance (ANOVA). Across time points, there was no effect of route of flunixin administration (P > 0.05) on milk yield, but significant (P < 0.05) effects of LPS administration on milk yield were measured. Therefore, milk yield data are presented in terms of main effects only. The main effect of LPS on the milk flunixin/5OHF ratio was tested at each collection time point by a simple one-way ANOVA. For the 12, 24, 36, 48, and 72 h collection points, ratios were non-normally distributed (Shapiro−Wilk, P < 0.05) so the Kruskal−Wallis analysis of variance on ranks was employed. For the 84, 96, and 108 h collection points, ratios were normally distributed (P > 0.05) and ordinary oneway ANOVAs were conducted. Within tissue, flunixin concentrations were analyzed using a twoway ANOVA with main effects of endotoxin status (+LPS or saline) and route of flunixin administration (iv or im). Data were transformed if normality or equal variance assumptions were not met. The method of transformation for a particular set of means is included in tables summarizing means. For instances in which the ANOVA returned a significant main effect, the Holm−Sidak test was employed to infer differences in treatment means. 4895
DOI: 10.1021/acs.jafc.5b01509 J. Agric. Food Chem. 2015, 63, 4893−4901
Article
Journal of Agricultural and Food Chemistry
■
RESULTS AND DISCUSSION Milk Yield. There was no (P > 0.05) effect of route of flunixin administration on milk yield (Figure 1; Supporting
were larger than parameter estimates of previous studies in lactating dairy cattle.9,10,14,24−26 This study differed from earlier studies in several respects: (a) cows were sequentially dosed with flunixin over a 3 day period; (b) the plasma flunixin assay was very sensitive, with an LOQ of 0.5 ng/mL, and blood samples were collected for 96 h after the final flunixin dose. In many instances, quantifiable flunixin residues were present 84− 96 h after dosing with the resulting pharmacokinetic parameters being influenced by the prolonged concentration−time curve relative to other studies. Within these data, however, LPS administration increased (P < 0.05) AUClast, clearance, and the volume of distribution in cows dosed with iv flunixin. For cows dosed im with flunixin, LPS administration increased (P < 0.05) clearance (CLss/F) and mean residence time. Milk Residues. Concentrations of 5OHF and flunixin in milk after withdrawal periods of 36, 48, and 96 h are shown in Table 3; there were no effects of either route (P > 0.50) or LPS status (P > 0.45) on 5OHF concentrations at either 36 or 48 h. Consistent with the results of Feely et al.,27 5OHF was typically measurable in milk for longer periods of time than was parent flunixin (Supporting Information Tables S2 and S3), corroborating the metabolite as an appropriate marker for milk. At the minimum milk discard time of 36 h, 6 of the 20 cattle (30%) had 5OHF concentrations that met or exceeded the 2.0 ng/mL tolerance for 5OHF in milk. Similarly, 5OHF exceeded 2 ng/mL 36 h after the last flunixin dose in 25% of the eight cattle studied by Feely et al.27 and in 8 of 20 cattle studied by Kissell et al.10 With a 48 h withdrawal time, 2 of the 20 animals (10%) in the present study had 5OHF levels greater than allowable under U.S. statutory limits: a single cow from the iv-flunixin (−)LPS group (label conditions; cow 1) and a single cow from the im-flunixin (+)LPS group (off-label conditions; cow 18). Again, this result does not seem remarkable given the fact that of the eight cows dosed by Feely et al.,27 a single cow secreted 17 ng/g of total radioactive residue (TRR) after a 48 h withdrawal period; assuming that 5OHF represented 20% of the total TRR in this milk, 5OHF would have exceeded tolerance levels at 3.4 ng/mL. In recent studies, Kissell et al.9 measured violative 5OHF levels in milk of two cows 36 h after the last im or subcutaneous flunixin administration, but residues were not present at a 48 h withdrawal time. In contrast, 5OHF residues were measurable in milk from 3 of 10 mastitic cows after a 48 h withholding time.10 Milk yield was considered a significant (P < 0.1) covariate with the rate of 5OHF elimination in cows by Kissell et al.9 Cows producing 30 kg of milk per day. In the current study, average daily milk production across all cows was 13.8 ± 7.5 kg/day (mean ± SD), indicative of the fact that each cow used for the study represented an animal selected for culling because of age, poor reproductive performance, or low milk production. None of the animals were selected for elimination from the herd because of mastitis. For these lowproducing cows, there was no relationship (R2 = 0.1509, P = 0.09) between milk yield and milk 5OHF concentration at 36 h of withdrawal. Kissell et al.10 noted that for mastitic animals, the relative quantity and duration of parent flunixin secreted into milk was surprisingly great, with three cows eliminating 13 ± 11 ng/mL after a 60 h withdrawal period. In the current study, parent flunixin was detectable in 12 of 20 milk samples collected 60 h after the last flunixin administration, but average concentrations
Figure 1. Effects of route of flunixin administration (top panel) or lipopolysaccharide (LPS) treatment (bottom panel) on mean milk yield (±SE) during the study period (10 cows per treatment group for each main effect). Route of drug administration had no effects (P > 0.05) on milk yield, but LPS treatment caused reductions (∗) in milk yields at 12 (P = 0.012), 24 (P = 0.004), and 36 h (P = 0.007). By 48 h, the effect of LPS on milk yield was marginal (P = 0.091) and, thereafter, negligible (P > 0.1).
Information Table S1); however, administration of LPS to lactating dairy cows reduced (P < 0.05) milk yield relative to the milk yield of saline-treated cows (Figure 1). The magnitudes and duration of milk yield reductions following iv LPS administration were comparable to those observed after iv infusion of E. coli LPS to lactating Holstein cows by Waldron et al.21 LPS-associated reduction in milk yields or milk fat yields are well characterized after dietary induction of ruminal LPS and the resulting inflammatory response in dairy cattle.22,23 Such results in cows used for the present study strongly suggest that LPS infusion was successful in mediating a systemic, but temporary, inflammatory response. Kinetic Parameter Estimates. Estimated pharmacokinetic parameters for parent flunixin are shown in Table 2. Across treatments, terminal plasma half-life, AUClast, Vdss/F, and MRT 4896
DOI: 10.1021/acs.jafc.5b01509 J. Agric. Food Chem. 2015, 63, 4893−4901
Article
Journal of Agricultural and Food Chemistry
Table 2. Flunixin Plasma Pharmacokinetic Parameter Estimates (Estimates Are Means ± Standard Deviations of Five Cows per Treatment) route of administration intravenous t1/2 tmax Cmax AUClast CLss Vdss CLss/F Vdss/F MRT
intramuscular
unit
−LPS
+LPS
h h μg/mL μg·h/mL mL/h/kg L/kg mL/h/kg L/kg h
11.6 ± 8.0
5.6 ± 1.7
parameter
11.9 40.2 151 5.82
± ± ± ±
4.7 16.7 62.8 3.11
9.7 29.5 205 8.19
36.7 ± 10.4
± ± ± ±
−LPS a
15.5 0.25 5.5 33.2
1.2 11.6a 42.4a 1.62a
± ± ± ±
+LPS
8.0 0.01 2.3 9.1
14.2 0.89 3.1 37.0
188 ± 48.7 4.19 ± 2.69 44.3 ± 8.3
41.1 ± 12.0
± ± ± ±
6.11 0.68 1.5 13.0
210 ± 69.7b 4.21 ± 1.83 62.1 ± 23.0b
Means differed (P < 0.05) between saline and LPS treatments within the iv flunixin group. bMeans differed (P < 0.05) between saline and LPS treatments within the im flunixin group. a
Table 3. Concentrations of 5-Hydroxyflunixin (5OHF) and Flunixin and in Milk of Cows at 36, 48, and 96 h after the Last Administration of Flunixin Megluminea intravenous saline (ng/mL) withdrawal period (h)
1
6
12
36 48 96
19.1 34.4 4.4
2.0 0.5
0.6 0.3 0.1
36 48 96
1.9 4.1 1.6
0.3
0.2 0.1
13 0.4 0.1
intramuscular +LPS (ng/mL)
19
3
8
0.7 0.1
0.4 0.4
1.0 0.2 0.1
0.3
0.6 1.0
1.1 0.1
10
saline (ng/mL)
15
17
2
5-Hydroxyflunixin 0.5 0.4 2.8 0.9 0.2 0.1 0.6 0.4
0.1
Flunixin 0.3 0.7 0.1 0.2
0.5 0.5 0.1
+LPS (ng/mL)
5
11
14
20
4
7
9
16
18
2.4 0.2
4.7 1.4 0.3
0.9 0.3
1.8 0.8 0.1
1.8 0.4
1.3 0.4 0.1
1.4 0.7 0.2
0.9 0.3
8.3 2.5 0.7
0.2
0.8 0.4
0.4 0.1
0.6 0.3
1.0 0.5 0.2
1.7 0.5 0.2
1.3 0.9 0.1
0.6 0.2
10.3 4.6 2.4
a
The tolerance for 5OHF in milk is 2.0 ng/mL; a minimum 36 h milk discard time is required. Milk values of 5OHF shown in bold exceed the tolerance at the indicated times. Concentrations of 5OHF and flunixin in milk collected from 0 to 96 h are shown in Tables S2 and S3 of the Supporting Information.
were only 0.9 ng/mL. For cows 1 and 18 (which had violative milk residues at 48 h), parent flunixin remained at relatively high levels (>1 ng/mL) until slaughter at 96 h. On the basis of the data of Kissell et al.10 one might predict that cow 1 was mastitic or sick, and the milk record of cow 1 does suggest that this cow was not well. On the first day of dosing, cow 1 produced a total of 16 kg of milk; thereafter, the daily milk production decreased progressively from 8.3, 4.0, 3.2, 1.1, 0.2, to 0.1 kg on days 2−7, respectively (Supporting Information Table S1). Although the cows were not mastitic when purchased, we cannot rule out the development of mastitis during the study period. Cow 18 was in the im flunixin (+)LPS treatment group and was a low-producing cow, secreting an average of 2.0 kg of milk per day. Figure 2 shows the ratio of flunixin to 5OHF secreted into milk across the study period for the control and LPS-treated cows. Cows pretreated with 0.2 μg/kg LPS consistently secreted greater (P < 0.05) ratios of flunixin/5OHF in milk than control cows, suggesting that either the transport or metabolism of flunixin in LPS-treated animals was altered relative to control animals. These data are consistent with the findings of Kissell et al.,10 who demonstrated that mastitic cows secreted flunixin and 5OHF in ratios ranging from 6 to 45 from 2 to 48 h after flunixin dosing, whereas control cows had milk secretions containing ratios ranging from 0.4 to 0.8 for 24 h after dosing (neither flunixin nor 5OHF was detected in milk of
Figure 2. Ratio of flunixin to 5OHF secreted into milk by cows treated with LPS [(+)LPS] or with saline [(−)LPS]. Data are expressed as the mean ratios ± SEM of the concentration of flunixin to 5OHF in milk (n = 10 per treatment) by collection time. Asterisks indicate probabilities of significant differences between treatment means (∗, P < 0.05; ∗∗, P < 0.01).
control animals after 24 h). Similarly, healthy cows used by Jedziniak et al.28 secreted flunixin and 5OHF into milk at ratios estimated to be on the order of 0.1. Metabolite ratios in the current study were not as pronounced as those measured by Kissell et al.10 but were highly consistent with the notion that inflammation may alter flunixin metabolism and disposition. Tissue Residues. Mean concentrations of flunixin in edible and specialty tissues are shown in Table 4. There was no effect 4897
DOI: 10.1021/acs.jafc.5b01509 J. Agric. Food Chem. 2015, 63, 4893−4901
Article
Journal of Agricultural and Food Chemistry
Table 4. Residues of Flunixin in Traditionally Edible and Specialty Tissues of Cows Treated with 2.2 mg/kg Body Weight of Flunixin Meglumine for Three Consecutive Days and Slaughtered 96 h after the Final Flunixin Dose route of administration intravenous
P value
intramuscular
item
−LPS
+LPS
−LPS
+LPS
pooled SEM
route
LPS
route × LPS
n livera (ng/g) kidneya (ng/g) skeletal musclea (ng/g) adipose tissueb (ng/g) injection sitec (ng/g) mammarya (ng/g) lunga (ng/g) hearta (ng/g) tongue (ng/g) reticulum (ng/g) abomasum (ng/g) ileumc (ng/g) brain (ng/g)
5 31.3 15.4 1.9 3.3 4.6x 9.2 17.5 7.1 6.2 5.2 7.1 9.6 mammary, ileum, abomasum, heart, reticulum, tongue > skeletal muscle, adipose tissue > brain. Residue analysis of specialty tissues indicated that flunixin residues would not represent a food safety risk for consumers of those tissues. This assertion is based on the fact that flunixin concentrations in specialty tissues were
