Comparison of pharmacokinetics and milk elimination of flunixin in healthy cows and cows with mastitis Lindsey W. Kissell, PhD; Teresa L. Leavens, PhD; Ronald E. Baynes, DVM, PhD; Jim E. Riviere, DVM, PhD; Geof W. Smith, DVM, PhD

RUMINANTS

Objective—To determine whether pharmacokinetics and milk elimination of flunixin and 5-hydroxy flunixin differed between healthy and mastitic cows. Design—Prospective controlled clinical trial. Animals—20 lactating Holstein cows. Procedures—Cows with mastitis and matched control cows received flunixin IV, ceftiofur IM, and cephapirin or ceftiofur, intramammary. Blood samples were collected before (time 0) and 0.25, 0.5, 1, 2, 4, 8, 12, 24, and 36 hours after flunixin administration. Composite milk samples were collected at 0, 2, 12, 24, 36, 48, 60, 72, 84, and 96 hours. Plasma and milk samples were analyzed by use of ultra–high-performance liquid chromatography with mass spectrometric detection. Results—For flunixin in plasma samples, differences in area under the concentration-time curve and clearance were detected between groups. Differences in flunixin and 5-hydroxy flunixin concentrations in milk were detected at various time points. At 36 hours after flunixin administration (milk withdrawal time), 8 cows with mastitis had 5-hydroxy flunixin concentrations higher than the tolerance limit (ie, residues). Flunixin residues persisted in milk up to 60 hours after administration in 3 of 10 mastitic cows. Conclusions and Clinical Relevance—Pharmacokinetics and elimination of flunixin and 5-hydroxy flunixin in milk differed between mastitic and healthy cows, resulting in violative residues. This may partially explain the high number of flunixin residues reported in beef and dairy cattle. This study also raised questions as to whether healthy animals should be used when determining withdrawal times for meat and milk. (J Am Vet Med Assoc 2015;246:118–125)

F

lunixin is an NSAID approved for use in beef and dairy cattle for the modulation of inflammation in endotoxemia and the control of pyrexia associated with bovine respiratory disease and acute mastitis. When administered in accordance with label instructions, withdrawal time for violative residues in milk is 36 hours. In milk, the marker residue is a metabolite of flunixin, 5-hydroxy flunixin, the tolerance limit of which is 2 ng/mL. Nonsteroidal antiinflammatory drugs such as flunixin are reportedly the second most prescribed class of drugs by dairy veterinarians1 and the most frequently administered analgesics in cattle.2 Because flunixin is the only NSAID approved for use in dairy cattle in the United States and is routinely From the Department of Population Health and Pathobiology and the Food Animal Residue Avoidance Databank, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606 (Kissell, Baynes, Smith); Pharmacokinetics Consulting, Cary, NC 27513 (Leavens); and Department of Anatomy and Physiology and the Food Animal Residue Avoidance Databank, College of Veterinary Medicine, Kansas State University, Manhattan, KS 66505 (Riviere). Supported by the Food Animal Residue and Avoidance and Depletion Program. Presented in abstract form at the American Academy of Veterinary Pharmacology and Therapeutics Biennial Symposium, Potomac, Md, May 2013. Address correspondence to Dr. Kissell ([email protected]). 118

Scientific Reports

ABBREVIATIONS AUC0–∞ AUMC LOD LOQ λz MRT Vdarea Vdss

Area under the plasma concentration-time curve from time zero to infinity Area under the first moment curve Limit of detection Limit of quantification Rate constant associated with the terminal elimination phase Mean residence time Volume of distribution for the terminal elimination phase Volume of distribution at steady state

used on dairy farms, there is a potential for violative residues. Although flunixin is regularly used on dairy farms, milk samples in the United States are not routinely tested for 5-hydroxy flunixin residues.3 This may be partially a result of information in early reports,4,5 which suggested that flunixin residues were not a concern because residues were not detected in milk after IV or IM administration. However, more recent research with more sensitive analytic equipment clearly has revealed that 5-hydroxy flunixin and flunixin accumulate in milk at concentrations well above the tolerance limit.6,7 A recent flunixin surveillance study8 in which samples were obtained from 500 tanker truck loads of milk JAVMA, Vol 246, No. 1, January 1, 2015

Materials and Methods Animals—Twenty lactating Holstein cows from a dairy farm in North Carolina were used in a prospective controlled clinical trial. Cows weighed between 545 and 676 kg (1,199 and 1,487 lb). Ten cows were identified as having naturally occurring mastitis as defined by abnormal appearance of milk with a red or swollen mammary gland. Some mastitic cows had systemic signs of other diseases, whereas others did not. Ten healthy cows were then matched to the cows with mastitis on the basis of parity (ie, current lactation number), number of days in lactation, and milk production to serve as control cows. This study was approved by the North Carolina State University Institutional Animal Care and Use Committee. Mastitis was diagnosed during the morning milking for all cows. Within 2 hours after the diagnosis of clinical mastitis, a catheter was aseptically placed in a jugular vein. Heart rate and rectal temperature were recorded for each cow, and 5 mL of milk was aseptically collected from each mastitic gland for aerobic bacterial culture prior to treatment. Experimental design—One dose of flunixin megluminea (2.2 mg/kg [1 mg/lb], IV) was administered to all cows in the jugular vein contralateral to the vein in which the catheter had been placed. All cows also received ceftiofurb (2.2 mg/kg, IM, q 24 h for 3 days). For intramammary antimicrobial treatment, cows were allocated into 2 groups (groups 1 and 2), each of which consisted of 10 cows (5 JAVMA, Vol 246, No. 1, January 1, 2015

cows with mastitis and 5 paired control cows). Mastitic cows of group 1 received cephapirinc (1 syringe [10 mL], intramammary, q 12 h for 3 days) in the affected gland, whereas mastitic cows of group 2 received ceftiofurd (1 syringe [10 mL], intramammary, q 24 h for 5 days) in the affected gland. The healthy cows of group 1 received intramammary treatments of cephapirin administered in the same gland as their matched cows with mastitis, and healthy cows of group 2 received intramammary treatments of ceftiofur administered in the corresponding gland. For example, if the mastitic cow was affected in the right rear mammary gland, the matched control cow also was given the same intramammary antimicrobial in the right rear mammary gland. Collection of blood and milk samples—Blood samples were collected from the jugular catheter into heparinized tubes before flunixin administration (time 0) and 0.25, 0.5, 1, 2, 4, 8, 12, 24, and 36 hours after administration. Blood samples were centrifuged at 1,690 X g for 10 minutes at 4°C; plasma was harvested and frozen at –20°C until analysis of flunixin and metabolite concentrations. Prior to flunixin administration, 5 mL of foremilk was manually collected from each gland of every cow; the 4 gland samples were pooled to provide a baseline composite sample for each cow. Additional composite milk samples were collected with a sampling devicee during milking. These composite milk samples were collected 2, 12, 24, 36, 48, 72, 84, and 96 hours after flunixin administration. All sample collection times, except for the sample collected at 2 hours, were times that corresponded with the farm’s regular milking schedule. Milk samples were immediately frozen at –20°C until analysis. Analysis of milk and plasma samples—Flunixin and 5-hydroxy flunixin concentrations were quantified in both plasma and milk by use of ultra–high-performance liquid chromatography with mass spectrometric detection. Plasma samples were extracted. Samples were thawed, and 0.3 mL of plasma was combined with 0.9 mL of 0.5% citric acid in acetonitrile. Samples were mixed in a sonicator for 5 minutes and then centrifuged for 10 minutes at 3,500 X g. The supernatant was loaded on a solid-phase extraction cartridge.f Eluate from the cartridge was collected, placed in a 55°C evaporator,g dried under a stream of nitrogen, reconstituted in 300 µL of mobile phase, and filtered through a 0.22-µm nylon syringe filter. Injection volume was 5 µL. Concentrations were derived by comparing peak areas for the samples with those for an external standard curve made from spiked plasma samples analyzed via the same procedures. Milk samples were also extracted. Samples (0.5 mL of milk) and 1.5 mL of 0.5% citric acid in acetonitrile were combined in a centrifuge tube, and the same procedures were used as those described for extraction and quantification of plasma samples. The ultra–high-performance liquid chromatography with a mass spectrometric detection systemh consisted of a high strength silica T3 column (1.8 µm inside diameter, 2.1 X 100 mm) and filter disk. The mobile phase was a mixture of acetonitrile and 0.1% acetic acid in water (68:32 [vol/vol]). A single quadrupole mass spectrometer was used in positive electrospray ionization mode. Ions with m/z of 297.0 and 313.0 were used for quantification of flunixin and 5-hydroxy flunixin, respectively. Column Scientific Reports

119

RUMINANTS

found that 1 of the 500 milk samples had 5-hydroxy flunixin concentrations above the tolerance limit. That study8 revealed that violative flunixin residues in milk occur in the dairy industry, although the cause of these violative residues remains unknown. Possible explanations for violative flunixin residues in milk include not observing the withdrawal time for milk, extralabel use of flunixin (eg, altering the dose, route of administration, frequency, or duration of treatment), and altered milk elimination as a result of a disease process. Investigators of several studies9–18 have reported alterations in plasma pharmacokinetics of drugs in diseased animals, compared with results for their healthy counterparts. Investigators of a recent study7 found a correlation between low milk production and prolonged flunixin elimination in milk. Considering that mastitis often results in a substantial decrease in milk production,19 violative flunixin residues may be prevalent in milk from mastitic cows. Since 2005, the USDA Food Safety Inspection Service has reported an increasing number of flunixin residue violations in meat from dairy cattle.20 This increase in the number of violations attributable to flunixin residues has led to flunixin becoming the second most common residue violation (behind only penicillin) in cull dairy cattle.20 These residue violations in meat have primarily been attributed to extralabel use of flunxin.21 However, delayed plasma clearance caused by a disease process may result in a prolongation of residues in meat and could contribute to the high number of flunixin violations in edible tissues. Therefore, the objective of the study reported here was to determine whether plasma pharmacokinetics and elimination of flunixin and 5-hydroxy flunixin in milk would differ between healthy cows and cows with clinical mastitis.

Table 1—Mean ± SD age, lactation number, number of days in lactation, milk production history, rectal temperature, and heart rate for 10 cows with mastitis and 10 healthy control cows enrolled in a study of pharmacokinetics and milk elimination of flunixin. Variable Age of cow (y) Current lactation No. No. of days in lactation Milk production prior to study (kg/d [lb/d]) 305-day mature equivalent milk production (kg [lb]) Rectal temperature (°C [ºF]) Heart rate (beats/min)

Control cows 4.8 ± 1.5 2.8 ± 0.9 153 ± 73 33.4 ± 3.8 (73.5 ± 8.4) 10,152 ± 2,030 (22,335 ± 4,466) 38.33 ± 0.61 (101.0 ± 1.1) 74.2 ± 10.4*

Mastitic cows 4.9 ± 1.8 3.0 ± 1.5 142 ± 74 29.9 ± 4.3 (65.8 ± 9.5) 8,155 ± 1,860 (17,940 ± 4,092) 38.89 ± 0.89 (102.0 ± 1.6) 93.6 ± 10.7

RUMINANTS

*Value differs significantly (P = 0.01) from the value for the mastitic cows.

temperature was 30°C, and sample temperature was 4°C. Run times were 2.2 minutes. The LOQ was determined as 10 times the SD of 6 blank samples. The LOD was determined as 3 times the SD of 6 blank samples. The LOD and LOQ for flunixin and 5-hydroxy flunixin in plasma were 10 and 20 ng/mL, respectively, and the linear range was 20 to 30,000 ng/mL. The LOD and LOQ for flunixin and 5-hydroxy flunixin in milk were 1 and 2 ng/mL, respectively, and the linear range for milk was 2 to 1,000 ng/mL. Relative SD for both interday and intraday accuracy was < 15% at all concentrations (r2 > 0.99). Pharmacokinetic analysis—A noncompartmental analysis of flunixin plas- Figure 1—Mean flunixin plasma concentration-time profile after administration of fluma concentration versus time profiles was nixin (time 0) to 10 cows with mastitis (squares) and 10 healthy control cows (triangles). performed with pharmacokinetic modelTable 2—Mean ± SD values for flunixin pharmacokinetic paraming software.i The AUC0–∞ and AUMC were calculated eters in plasma obtained after administration of flunixin to 10 by use of the linear trapezoidal rule; AUC0–∞ and AUMC healthy control cows and 10 cows with mastitis. subsequently were used to calculate clearance and MRT. The value for λz was estimated by means of linear regresParameter Control cows Mastitic cows P value* sion of the terminal phase of the logarithmic concentrat1/2λz (h) 3.68 ± 1.97 4.35 ± 1.82 0.9 tion-time profile and used to calculate the corresponding λz (h–1) 0.24 ± 0.14 0.18 ± 0.06 0.3 AUC0–∞ (µg•h/mL) 19.82 ± 6.68 42.98 ± 18.56 0.004 terminal elimination half-life for control and mastitic AUC0–tlast (µg•h/mL) 19.09 ± 6.43 41.49 ± 18.19 0.004 cows. The λz also was used to extrapolate AUC0–∞ and Clearance (mL/h/kg) 120.15 ± 31.70 67.02 ± 47.24 0.01 AUMC from the time of the last detectable concentraVdss (L/kg) 0.273 ± 0.13 0.165 ± 0.56 0.1 Vdarea (L/kg) 0.644 ± 0.42 0.383 ± 0.21 0.09 tion to infinity. Finally, Vdss and Vdarea were calculated MRT (h) 2.40 ± 1.92 3.14 ± 1.45 0.3 for flunixin. Statistical analysis—Statistical analyses were performed with the aid of commercially available software.j All values were expressed as mean ± SD, except for the value for terminal elimination half-life, which was expressed as the harmonic mean ± SD. Flunixin plasma pharmacokinetic parameters and flunixin and 5-hydroxy flunixin concentrations in milk were compared by means of Satterthwaite approximation for degrees of freedom. Values of P < 0.05 were considered significant. Results Cows—Ten cows with clinical mastitis were identified and enrolled in the study between May and August 2012; the 10 healthy control cows were enrolled during the same time period (Table 1). Mean heart rate of the mastitic cows was significantly (P = 0.01) different from that of the control cows; however, no difference 120

Scientific Reports

*Values were considered significantly different at P < 0.05. AUC0–tlast = Area under the curve from time 0 to the last detectable concentration. t1/2λz = Harmonic mean elimination half-life.

Table 3—Mean ± SD 5-hydroxy flunixin concentrations in milk (µg/L) obtained after administration of flunixin to 10 control cows and 10 cows with mastitis. Time after flunixin administration (h) 2 12 24 36 48

Control cows

Mastitic cows

P value*

12.2 ± 7.04 (n = 7) 22.2 ± 6.23 (n = 10) 5.76 ± 2.90 (n = 6) < LOQ (n = 10) < LOQ (n = 10)

5.04 ± 4.19 (n = 10) 10.5 ± 5.79 (n = 10) 5.83 ± 2.56 (n = 10) 3.75 ± 1.78 (n = 8) 0.96 (n = 3)

0.030 0.008 0.900 — —

*Values were considered significantly different at P < 0.05. — = Not significant. JAVMA, Vol 246, No. 1, January 1, 2015

in rectal temperature was evident between the 2 groups (P = 0.1). Results of bacteriologic culture of milk obtained from 8 of the 10 mastitic cows indicated Klebsiella pneumoniae infection, and the remaining 2 cows had Escherichia coli infection. Plasma—Mean ± SD peak plasma concentration after flunixin administration was 18.35 ± 5.40 µg/mL and 24.45 ± 8.22 µg/mL for control cows and cows with mastitis, respectively; these values did not differ significantly (P = 0.06). Flunixin was detectable in the plasma of cows with mastitis for up to 24 hours after administration, which was 12 hours more than flunixin was detectable in the plasma of healthy cows (Figure 1). Clearance for the cows with mastitis was significantly (P = 0.01) reduced, compared with clearance for the healthy cows (Table 2). The AUC0–∞ for the mastitic cows was more than twice the AUC0–∞ for the control cows and was significantly (P = 0.004) different between the 2 groups. Plasma terminal elimination half-life (P = 0.9), Vdarea (P = 0.1), and Vdss (P = 0.09) did not differ significantly between the healthy and mastitic cows. Milk—Residues of 5-hydroxy flunixin above the tolerance limit were detected in milk samples from 8 of the mastitic cows at the labeled withdrawal time (36 hours after administration of flunixin) and were still detectable in milk samples from 3 mastitic cows up to 48 hours after administration (Table 3). However, no 5-hydroxy flunixin residues above the tolerance limit were detected in milk of the control cows > 24 hours after flunixin administration (Figure 2). Milk concentrations of 5-hydroxy flunixin were significantly different between the control cows and cows with mastitis for samples obtained at both 2 and 12 hours, with the mastitic cows having lower 5-hydroxy flunixin concentrations in milk at both time points. Flunixin residues in milk persisted for up to 60 hours after administration in some of the mastitic cows; in contrast, flunixin residues were detectable in milk of control cows for only up to 24 hours after administration (Figure 2). Flunixin concentrations in milk were significantly different between groups at all time points, JAVMA, Vol 246, No. 1, January 1, 2015

Table 4—Mean ± SD flunixin concentration in milk (µg/L) obtained after administration of flunixin to 10 control cows and 10 cows with mastitis. Time after flunixin administration (h) 2 12 24 36 48 60

Control cows

4.91 ± 3.52 (n = 6) 16.77 ±16.69 (n = 10) 3.18 ± 0.367 (n = 4) < LOQ (n = 10) < LOQ (n = 10) < LOQ (n = 10)

Mastitic cows

P value*

224.0 ± 232.2 (n = 10) 0.009 102.8 ± 88.8 (n = 10) 0.010 37.17 ± 36.02 (n = 10) 0.010 28.54 ± 32.01 (n = 7) — 11.37 ±15.69 (n = 7) — 13.02 ± 10.93 (n = 3) —

See Table 3 for key.

with the mastitic cows having substantially greater flunixin concentrations in milk, compared with flunixin concentrations in milk of control cows (Table 4). Discussion The objective of the present study was to determine whether plasma pharmacokinetics and elimination of flunixin and 5-hydroxy flunixin in milk of lactating dairy cattle would differ between cows with naturally occurring mastitis and paired control cows without mastitis. The significant difference in AUC0–∞ between the 2 groups was a result of a higher maximum plasma concentration and measureable concentrations of flunixin in the plasma for up to 24 hours after administration to mastitic cows. Clearance calculated for healthy cows in the present study was similar to findings reported in the literature.5,7,22–25 Reduced clearance in the mastitic cows may have been attributable to alterations in hepatic drug metabolism.9,26,27 This is similar to results for another NSAID, carprofen, for which clearance is reduced by 40% in mastitic cows, compared with that in healthy control cows.9 Inflammation and infection can greatly impact drug metabolism and protein binding,28 which may also contribute to alterations in clearance. In the present study, the mean plasma terminal elimination half-life for flunixin after IV administration corresponded with that found in other studies.4,5,7,23–25,29,30 Scientific Reports

121

RUMINANTS

Figure 2—Flunixin (A) and 5-hydroxy flunixin (B) concentrations in milk after administration of flunixin (time 0) to 10 mastitic cows (diamonds and dashed line) and 10 healthy control cows (squares and dotted line). In panel B, notice the tolerance limit for violative residues of 5-hydroxy flunixin in milk (solid horizontal line). Also notice that the scale on the y-axis differs between panels. a,bWithin a time point, values with different letters differ significantly (P < 0.05).

RUMINANTS

The Vdarea and Vdss were also consistent with values reported in the literature5,7,23–25,29,30; however, they were greater than expected for a drug that is highly (99%) protein bound. The high number of violations for flunixin residues in tissues of dairy cattle may in part be attributable to disease-induced alterations in pharmacokinetics, particularly alterations in clearance of a drug. Investigators in numerous studies9–18 have reported changes in pharmacokinetics as a result of inflammation. In 1 study31 in horses, investigators found significantly greater concentrations of several NSAIDs in inflamed tissues, compared with the concentrations in healthy tissues. Similarly, there is an increased partition of both ceftiofur and erythromycin into infected bovine tissue chambers, compared with the partitions for healthy bovine tissue chambers.32,33 In rabbits, inflammation can increase capillary permeability, thus altering flunixin partitioning into tissues.15 In that same study,15 endotoxemic rabbits also had a significantly lower clearance and longer elimination half-life, compared with values for healthy control rabbits. A recent surveillance study34 found that cows culled because of clinical signs or evidence of disease had a significantly higher incidence of violative tissue flunixin concentrations at slaughter than did healthy-appearing dairy cows. The high number of flunixin tissue residues identified in culled dairy cattle may be related to mastitis-induced alterations in drug clearance. Milk concentrations of 5-hydroxy flunixin in the control cows at each time point in the present study were similar to milk concentrations in other studies.6,7,35 The persistence of 5-hydroxy flunixin residues in milk beyond the labeled withdrawal time may in part be related to a decrease in milk production as a result of mastitis.19 There are a paucity of reports describing the effect of milk production on drug elimination, especially for systemically administered drugs. However, several studies36–40 have found a correlation between low milk production and prolonged drug elimination for some intramammary drugs. A previous study7 conducted by our laboratory group found that milk production was a significant covariate for the depletion of 5-hydroxy flunixin concentrations in milk over time. In that study,7 cows producing < 20 kg (< 44 lb) of milk/d eliminated 5-hydroxy flunixin more slowly than did cows producing 30 kg (66 lb) of milk/d. In the present study, all of the mastitic cows produced < 6 kg (13.2 lb) of milk/d during the experimental period. Prior to clinical mastitis diagnosis, these cows were producing a mean ± SD of 29.9 ± 4.3 kg (65.8 ± 9.5 lb) of milk/d. The persistence of 5-hydroxy flunixin residues in milk beyond the labeled withdrawal time may also have been attributable to binding of 5-hydroxy flunixin to inflamed mammary gland tissues. In a study23 conducted to examine the pharmacokinetics and pharmacodynamics of flunixin in calves, investigators found that the drug binds to inflamed tissue and achieves high concentrations in inflammatory exudate. The accumulation of flunixin in inflamed tissue and slow clearance from exudate may partially explain the reason that 5-hydroxy flunixin residues were detectable in milk for up to 60 hours after administration. 122

Scientific Reports

Flunixin concentrations in the milk of control cows of the present study were comparable to concentrations reported on the US FDA website.35 The significantly greater concentration of flunixin in the milk from mastitic cows was unexpected. Several factors may have contributed to this finding. One explanation may be disruption of the blood-milk barrier. Both K pneumoniae and E coli are known to increase mammary gland vascular permeability and cause breakdown of the blood-milk barrier.41 This may have resulted in leakage of flunixin from the blood into the milk prior to hydroxylation by the liver. Considering that all cases of mastitis in the present study were caused by coliform bacteria, it is unclear whether the increase in flunixin concentrations in milk would be as pronounced with other causes of mastitis (ie, gram-positive organisms). Another potential explanation for the significantly greater flunixin concentrations in milk from mastitic cows may be related to impaired hepatic metabolism as a result of mastitis. Disease states can impair hepatic metabolism of drugs and alter clearance of the parent compound. Similarly, the prolonged persistence of flunixin residues in the milk of mastitic cows (up to 60 hours after administration) may also have been related to the decrease in milk production associated with mastitis as well as flunixin binding to inflamed mammary gland tissue. Mastitis induces physical and chemical changes in both milk and affected mammary glands; these changes have the potential to alter distribution and elimination of drugs through the mammary gland.38 Inflammation of a mammary gland leads to vascular permeability changes that often enhance systemic absorption and perhaps distribution of drugs into the udder. For example, gentamicin is not detected in plasma following intramammary administration in clinically normal mammary glands but is well absorbed in mammary glands of cows with mastitis.42 Similarly, in a study43 that involved the use of polymyxin B, the drug was not found in the blood or untreated mammary glands following intramammary administration in clinically normal cattle; however, substantial systemic absorption was evident in cows with experimentally induced coliform mastitis. Investigators of a study9 conducted to evaluate the influence of E coli endotoxin–induced mastitis on the pharmacokinetics of the NSAID carprofen found a significant reduction in systemic clearance, prolonged terminal elimination half-life, and increased milk carprofen concentrations in mastitic cows after IV administration, compared with results for healthy control cows. Finally, investigators of a study44 that involved the use of an intramammary preparation of cefoperazone sodium reported significantly greater systemic drug absorption, half-life in the milk, and MRT in cows with subclinical mastitis, compared with results for healthy control cows. The parenteral administration of ceftiofur to treat clinical mastitis in the present study represented extralabel use of this drug. It was a standard procedure on this farm to administer ceftiofur IM to all cattle that had mastitis and moderate to severe signs of systemic disease (particularly if milk production decreased substantially). Investigators of another study45 found that IM adminisJAVMA, Vol 246, No. 1, January 1, 2015

JAVMA, Vol 246, No. 1, January 1, 2015

ever, disease states can profoundly alter the pharmacokinetic behavior of a drug.9–18 The most profound differences in pharmacokinetic responses are generally associated with hepatic, renal, and cardiovascular disease, but other processes such as inflammation, endotoxemia, and stress can also significantly alter a drug’s absorption, distribution, metabolism, and elimination.48 Although data are limited on the effects of disease on the pharmacokinetics of drugs in cattle, there is evidence to suggest that the use of healthy animals for establishment of drug withdrawal periods may not be appropriate. For example, a study49 in the late 1970s revealed that following intramammary infusion of several antimicrobials, only minimal concentrations were found in the kidneys and liver of healthy cows. Detectable concentrations of antimicrobials were found only in the liver and kidneys for 24 hours, and residues were not detected in the meat of these cattle. In contrast, antimicrobial concentrations in the tissues of mastitic cows were much higher and persisted longer. Differences in drug pharmacokinetics have been described for oxytetracycline in cows with theileriosis.17 Following IM administration, infected cattle had significantly prolonged absorption and terminal elimination half-life, MRT, AUC0–∞, and bioavailability, compared with results for healthy cows.17 Additionally, 5 of 20 calves with respiratory disease died after administration of theophylline during a field trial,50 whereas all 20 calves that received a placebo survived. A subsequent study18 revealed that calves with pneumonia had significantly higher plasma concentrations of theophylline, compared with concentrations in healthy calves. Similarly, greater secretion of ceftriaxone into milk was also detected in cows with endometritis after IV administration, compared with results for control cows.51 Finally, use of a population pharmacokinetic model revealed the need for a longer withdrawal interval than the FDAapproved withdrawal time52 for flunixin in meat, which suggested a need for pharmacokinetic studies in both healthy and diseased animals. The present study provided strong evidence that withdrawal times for milk determined in healthy cattle may not be appropriate for cows with clinical mastitis. Pharmaceutical companies must conduct studies to determine the efficacy of various drugs for treating a specific disease or condition during the approval process, so it would appear logical that pharmacokinetic and residue studies could be performed with the same animals or under similar conditions. a. b. c. d. e. f. g. h. i. j.

Banamine, Merck Animal Health, Summit, NJ. Naxcel sterile powder, Zoetis, Kalamazoo, Mich. ToDAY, Boehringer Ingelheim Vetmedica, St Joseph, Mo. Spectramast LC sterile suspension, Zoetis, Kalamazoo, Mich. Metatron sampler, Westfalia Surge, Naperville, Ill. Supelco hybrid SPE-phospholipid cartridge, Sigma-Aldrich, St Louis, Mo. TurboVap LV evaporator, Zymark Corp, Hopkinton, Mass. Acquity, Waters Corp, Milford, Mass. Phoenix WinNolin, version 1.3x, Pharsight Corp, St Louis, Mo. SAS, version 9.1, SAS Institute Inc, Cary, NC.

References 1.

Sundlof SF, Kaneene JB, Miller RA. National survey on veterinarian-initiated drug use in lactating dairy cows. J Am Vet Med Assoc 1995;207:347–352. Scientific Reports

123

RUMINANTS

tration of ceftiofur in cows with severe coliform mastitis reduced the proportion of cases that resulted in death in the form of culling. Although there has been a prohibition on the extralabel use of cephalosporins in the United States, these drugs can still be used for nonlabeled indications as long as the administration is in accordance with the labeled dose and duration.46 The dose of 2.2 mg/ kg, IM, every 24 hours for 3 days is in accordance with the label direction and would not represent illegal use of this drug. Control cows without mastitis also received 3 doses of ceftiofur so that they were treated in the same manner as the mastitic cows. Administration of antimicrobials to one group and not the other could potentially have affected the pharmacokinetics of flunixin. Cephapirinc is labeled for intramammary administration every 12 hours for a total of 2 doses; therefore, administration of this drug every 12 hours for 3 days represented extralabel use. The protocol for this farm at the beginning of the study was to administer cephapirin via the intramammary route every 12 hours until the milk became visually normal. Because of the need to have a consistent duration of treatment for all cows in the study, farm personnel elected to use intramammary administration of cephapirin for 6 total treatments (every 12 hours for 3 days) in all cows regardless of the duration of clinical mastitis. Toward the middle of the study, farm personnel elected to change intramammary treatment to ceftiofur because most cases had been caused by coliform bacteria. Ceftiofurd was administered via the intramammary route once a day for 5 days, which was in accordance with label directions. Treatment of only the mastitic cows with antimicrobials could have potentially resulted in a drugdrug interaction, which may have altered the pharmacokinetics and confounded comparisons between healthy and mastitic cows. Therefore, the same treatment regimen administered to the mastitic cows was also administered to the control cows. Violative drug residues in milk are a primary concern for the dairy industry. Residues in milk as a result of administration of NSAIDs can result in major economic losses to producers and pose a potential health hazard to consumers. Strict financial penalties and suspension of a producer’s permit to sell milk are possible outcomes of violative drug residues detected in milk. To prevent economic losses to producers, it is imperative that the labeled withdrawal time is accurate for the disease condition in which the drug is administered. Problems arise when there is discord between the labeled withdrawal time, which is calculated on the basis of data from healthy animals, and the time required for diseased animals to eliminate a drug. Withdrawal times are calculated by performing a statistical tolerance limit procedure on residue data from the depletion curve of the drug residue in milk or a target tissue. The US FDA procedure predicts with 95% confidence a withdrawal time when the residue in milk or a target tissue in 99% of the animal population receiving the drug is at or below tolerance limits.47 Part of the approval process for a veterinary drug requires that residue studies be conducted in healthy animals. One assumption of the approval process is that there will be no change in the drug’s pharmacokinetics when administered to diseased animals, compared with results for healthy animals. How-

2. 3. 4. 5.

RUMINANTS

6. 7.

8. 9.

10. 11. 12.

13. 14.

15. 16.

17.

18. 19. 20. 21. 22. 23.

24. 124

Fajt VR, Wagner SA, Norby B. Analgesic drug administration and attitudes about analgesia in cattle among bovine practitioners in the United States. J Am Vet Med Assoc 2011;238:755–767. National Milk Drug Residue Data Base. Fiscal year 2011 annual report. Available at: www.kandc-sbcc.com/nmdrd/fy-11.pdf. Accessed Jul 17, 2013. Benitz AM. Pharmacology and pharmacokinetics of flunixin meglumine in the bovine. In: Proceedings. 13th World Cong Dis Cattle 1984;928–930. Anderson KL, Neff-Davis CA, Bass VD. Pharmacokinetics of flunixin meglumine in lactating cattle after single and multiple intramuscular and intravenous administrations. Am J Vet Res 1990;51:1464–1467. Feely WF, Chester-Yansen C, Thompson K, et al. Flunixin residues in milk after intravenous treatment of dairy cattle with 14Cflunixin. J Agric Food Chem 2002;50:7308–7313. Kissell LW, Smith GW, Leavens TL, et al. Plasma pharmacokinetics and milk residues of flunixin and 5-hydroxy flunixin following different routes of administration in dairy cattle. J Dairy Sci 2012;95:7151–7157. Kissell LW, Baynes RE, Riviere JE, et al. Occurrence of flunixin residues in bovine milk samples from the USA. Food Addit Contam Part A Chem Anal Control Expo Risk Assess 2013;30:1513–1516. Lohuis JACM, Van Werven T, Brand A, et al. Pharmacodynamics and pharmacokinetics of carprofen, a non-steroidal anti-inflammatory drug, in healthy cows and cows with Escherichia coli endotoxin-induced mastitis. J Vet Pharmacol Ther 1991;14:219–229. Jha K, Roy BK, Singh RCP. The effect of induced fever on the biokinetics of norfloxacin and its interaction with probenecid in goats. Vet Res Commun 1996;20:473–479. Gips M, Soback S. Norfloxacin pharmacokinetics in lactating cows with sub-clinical and clinical mastitis. J Vet Pharmacol Ther 1999;22:202–208. Rao GS, Ramesh S, Ahmad AH, et al. Effects of endotoxininduced fever and probenecid on disposition of enrofloxacin and its metabolite ciprofloxacin after intravascular administration of enrofloxacin in goats. J Vet Pharmacol Ther 2000;23:365–372. Ismail M, El-Kattan YA. Comparative pharmacokinetics of marbofloxacin in healthy and Mannheimia haemolytica infected calves. Res Vet Sci 2007;82:398–404. Lucas MF, Errecalde JO, Mestorino N. Pharmacokinetics of azithromycin in lactating dairy cows with subclinical mastitis caused by Staphylococcus aureus. J Vet Pharmacol Ther 2010;33:132–140. Elmas M, Yazar E, Uney K, et al. Pharmacokinetics of flunixin after intravenous administration in healthy and endotoxaemic rabbits. Vet Res Commun 2006;30:73–81. Rantala M, Kaartinen E, Valimaki M, et al. Efficacy and pharmacokinetics of enrofloxacin and flunixin meglumine for treatment of cows with experimentally induced Escherichia coli mastitis. J Vet Pharmacol Ther 2002;25:251–258. Kumar R, Malik JK. Influence of experimentally induced theileriosis (Theileria annulata) on the pharmacokinetics of a longacting formulation of oxytetracycline (OTC-LA) in calves. J Vet Pharmacol Ther 1999;22:320–326. Langston VC, Koritz GD, Davis LE, et al. Pharmacokinetic properties of theophylline given intravenously and orally to ruminating calves. Am J Vet Res 1989;50:493–497. Ostergaard S, Grohn YT. Effects of diseases on test day milk yield and body weight of dairy cows from Danish research herds. J Dairy Sci 1999;82:1188–1201. USDA Food Safety Inspection Service. Red book 2005–2010. Available at: www.fsis.usda.gov/Science/2005_Red_Book/index. asp. Access Jul 3, 2013. Smith GW, Davis JL, Tell LA, et al. FARAD Digest: extralabel use of nonsteroidal anti-inflammatory drugs in cattle. J Am Vet Med Assoc 2008;232:697–701. Hardee GE, Smith JA. Pharmacokinetics of flunixin meglumine in the cow. Res Vet Sci 1985;39:110–112. Landoni MF, Cunningham FM, Lees P. Determination of pharmacokinetics and pharmacodynamics of flunixin in calves by use of pharmacokinetic/pharmacodynamic modeling. Am J Vet Res 1995;56:786–794. Odensvik K. Pharmacokinetics of flunixin and its effect on Scientific Reports

25.

26. 27.

28. 29. 30. 31. 32.

33.

34.

35.

36.

37. 38. 39. 40.

41. 42. 43. 44. 45.

prostaglandin F2α metabolite concentrations after oral and intravenous administration in heifers. J Vet Pharmacol Ther 1995;18:254–259. Odensvik K, Johansson MI. High-performance liquid chromatography method for determination of flunixin in bovine plasma and pharmacokinetics after single and repeated doses of the drug. Am J Vet Res 1995;56:489–495. Salam Abdullah A, Baggott JD. Influence of induced disease states on the disposition kinetics of imidocarb in goats. J Vet Pharmacol Ther 1986;9:192–197. van Gogh H, Watson ADJ, Nouws JFM, et al. Effect of tickbourne fever (Ehrlichia phagocytophila) and trypanosomiasis (Trypanosoma brucei) on the pharmacokinetics of sulfadimidine and its metabolites in goats. Drug Metab Dispos 1989;17:1–6. Petrovic V, Teng S, Piquette-Miller M. Regulation of drug transporters during infection and inflammation. Mol Interv 2007;7:99–111. Abo-El-Sooud K, Al-Anati L. Pharmacokinetic study of flunixin and its interaction with enrofloxacin after intramuscular administration in calves. Vet World 2011;4:449–454. Jaroszewski J, Jedziniak P, Markiewicz W, et al. Pharmacokinetics of flunixin in mature heifers following multiple intravenous administration. Pol J Vet Sci 2008;11:199–203. Landoni MF, Lees P. Comparison of the anti-inflammatory actions of flunixin and ketoprofen in horses applying PK/PD modeling. Equine Vet J 1995;27:247–256. Clarke CR, Bourne DW, Lauer AK, et al. Distribution of intramuscularly administered erythromycin into subcutaneous tissue chambers before and after inoculation with Pasteurella haemolytica. Res Vet Sci 1993;54:366–371. Clarke CR, Brown SA, Streeter RN, et al. Penetration of parenterally administered ceftiofur into sterile vs. Pasteurella haemolytica–infected tissue chambers in cattle. J Vet Pharmacol Ther 1996;19:376–381. Deyrup CL, Southern KJ, Cornett JA, et al. Examining the occurrence of residues of flunixin meglumine in cull dairy cows by use of the flunixin cull cow survey. J Am Vet Med Assoc 2012;241:249–253. Freedom of information summary. Supplemental new animal drug application (NADA 101–479). Available at www.fda.gov/downloads/ AnimalVeterinary/Products/ApprovedAnimalDrugProducts/ FOIADrugSummaries/ucm064910.pdf. Accessed May 21, 2013. Mercer HD, Geleta JN, Schultz JE. Milk-out rates for antibiotics in intramammary infusion products used in the treatment of bovine mastitis: relationship of somatic cell counts, milk production level, and drug vehicle. Am J Vet Res 1970;31:1549–1560. Whittem T. Pharmacokinetics and milk discard times of pirlimycin after intramammary infusion: a population approach. J Vet Pharmacol Ther 1999;22:41–51. Gehring R, Smith GW. An overview of factors affecting the disposition of intramammary preparations used to treat bovine mastitis. J Vet Pharmacol Ther 2006;29:237–241. Smith GW, Gehring R, Riviere JE, et al. Elimination kinetics for ceftiofur hydrochloride after intramammary administration in lactating dairy cows. J Am Vet Med Assoc 2004;224:1827–1830. Stockler RM, Morin DE, Lantz RK, et al. Effect of milk fraction on concentrations of cephapirin and desacetylcephapirin in bovine milk after intramammary infusion of cephapirin sodium. J Vet Pharmacol Ther 2009;32:345–352. Bannerman DD, Paape MJ, Hare WR, et al. Characterization of the bovine innate immune response to intramammary infection with Klebsiella pneumoniae. J Dairy Sci 2004;87:2420–2432. Sweeney RW, Fennell MA, Smith CM. Systemic absorption of gentamicin following intramammary administration to cows with mastitis. J Vet Pharmacol Ther 1996;19:155–157. Ziv G, Schultze WD. Pharmacokinetics of polymyxin B administered via the bovine mammary gland. J Vet Pharmacol Ther 1982;5:123–129. Cagnardi P, Villa R, Gallo M, et al. Cefoperazone sodium preparation behavior after intramammary administration in healthy and infected cows. J Dairy Sci 2010;93:4105–4110. Erskine RJ, Bartlett PC, VanLente JL, et al. Efficacy of systemic ceftiofur as a therapy for severe clinical mastitis in cattle. J Dairy Sci 2002;85:2571–2575. JAVMA, Vol 246, No. 1, January 1, 2015

46. US FDA. Cephalosporin extralabel drug use; order of prohibition. Fed Regist 2012;77:735–744. 47. US FDA. Guidance for industry: FDA approval for new animal drugs for minor uses and for minor species. Available at: www.fda.gov/ downloads/AnimalVeterinary/GuidanceComplianceEnforcement/ GuidanceforIndustry/UCM052375.pdf. Accessed Jul 16, 2013. 48. Martinez M, Modric S. Patient variation in veterinary medicine: part I. Influence of altered physiological states. J Vet Pharmacol Ther 2010;33:213–226. 49. Nouws JFM, Ziv G. Tissue distribution and residues of antibiotics in normal and emergency slaughtered dairy cows after intramammary treatment. J Food Prot 1978;41:8–13.

50. McKenna DJ, Koritz GD, Neff-Davis CA, et al. Field trial of theophylline in cattle with respiratory tract disease. J Am Vet Med Assoc 1989;195:603–605. 51. Kumar S, Srivastava AK, Dumka VK, et al. Plasma pharmacokinetics and milk levels of ceftriaxone following single intravenous administration in healthy and endometritic cows. Vet Res Commun 2010;34:503–510. 52. Wu H, Baynes RE, Leavens TL, et al. Use of population pharmacokinetic modeling and Monte Carlo simulation to capture individual animal variability in the prediction of flunixin withdrawal times in cattle. J Vet Pharmacol Ther 2013;36: 248–257. RUMINANTS

JAVMA, Vol 246, No. 1, January 1, 2015

Scientific Reports

125