Plasma disposition of gabapentin after the intragastric administration of escalating doses to adult horses

Abstract Background In humans, gabapentin an analgesic, undergoes non‐proportional pharmacokinetics which can alter efficacy. No information exists on the pharmacokinetics of dosages >20 mg/kg, escalating dosages or dose proportionality of gabapentin in horses. Hypothesis and Objectives Gabapentin exposure in plasma would not increase proportionally relative to the dose in horses receiving dosages ≥20 mg/kg. To assess the plasma pharmacokinetics of gabapentin after nasogastric administration of gabapentin at dosages of 10 to 160 mg/kg in adult horses. Animals Nine clinically healthy adult Arabian and Quarter Horses. Methods In a randomized blinded trial, gabapentin was administered by nasogastric intubation to horses at 10, 20 mg/kg (n = 3) and 60, 80, 120, 160 mg/kg (n = 6). Plasma was collected before and at regular times over 64 hours after administration of gabapentin. Gabapentin was quantified using a validated chromatographic method. Dose proportionality was estimated using a power model. Pharmacokinetic parameters were estimated using compartmental pharmacokinetic analysis. Results Plasma pharmacokinetics parameters of gabapentin were estimated after nasogastric administration at dosages of 10 to 160 mg/kg. Gabapentin plasma concentration increased with dose increments. However, the area under the concentration curve from zero to infinity and maximal plasma concentration did not increase proportionally relative to the dose in horses. Conclusions and Clinical Importance Gabapentin exposure in plasma is not proportional relative to the dose in horses receiving nasogastric dosages of 10 to 160 mg/kg.

anticonvulsant in humans and is currently used to control seizures in both humans and animals. [2][3][4] Gabapentin also plays an important role in pain management and was FDA-approved for treatment of humans with post-herpetic neuralgic pain, a type of neuropathic pain, in 2002. 5,6 Laminitis is an extremely painful, debilitating disease of horses, which is devastating to the animals and the animal owners because of the inability to control the pain in horse and the high financial cost/ loss. Neuropathic changes typical of those described in other neuropathic pain syndromes are present in horses with laminitis. 7 Neuropathic pain is often unresponsive to traditional analgesic treatment because of damage to and changes in a variety of the components in the pain pathway resulting in amplification and expansion of pain signals. Gabapentin is 1 of the few drugs that controls some forms of neuropathic pain in humans. 8 Although there is a paucity of information on the use of gabapentin in species other than humans, it has been administered to control neuropathic pain in small animals; however, dosing, and likely the type of disease, also complicates gabapentin efficacy in these dogs and cats. [9][10][11][12][13][14][15][16] Failure to prove efficacy is potentially the result of inadequate dosing. 17,18 Gabapentin is most commonly used to control the pain of laminitis and other conditions that cause neuropathic pain at 5-20 mg/kg twice daily 19,20 in combination with nonsteroidal anti-inflammatory agents. 18,[21][22][23] Treatment has not always been successful but the dosage regimens have been extrapolated from humans and efficacy of gabapentin with appropriate doses remains unknown. In order to determine efficacy of the drug, a dosing regimen that could potentially provide analgesia must be established.
After oral administration to horses, gabapentin is rapidly absorbed but the extent of absorption is relatively poor (mean oral bioavailability of gabapentin [±SD] was 16.2% ± 2.8%), 19 meaning that higher dosages might be required to achieve therapeutic plasma concentrations. In addition, in humans, gabapentin undergoes saturation pharmacokinetics so that the serum concentration of the drug is nonlinear to the dose administered (dose disproportionality). 24 12,14 Our main hypothesis is that doses of gabapentin between 10 and 160 mg/kg administered via nasogastric tube to horses will not increase the plasma concentration in a proportional manner. We addressed this hypothesis by evaluating the plasma pharmacokinetics of gabapentin in horses after a single nasogastric administration at 10, 20, 40, 60, 80, 120, and 160 mg/kg.

| MATERIALS AND METHODS
The study was approved by the Washington State University's Institutional Animal Care and Use Committee. In this blinded randomized experimental study, 9 clinically healthy adult horses were used. The horses were 2 Quarter Horse cross mares and 7 Arabian mares. The mean weight of the horses was 444 ± 15.68 kg. The equine age ranged from 8-23 years with a mean age of 13.2 years.
The horses were housed together on a dry lot with free choice water, selenium/mineral salt and were fed grass hay twice daily when they were not involved with the research project. While the study was being performed, the horses were brought into the WSU teaching facility stalls 24-48 hours before the first treatment and were fed a diet of grass hay daily and free choice water. The horses were held off feed the morning of the research project and were fed their standard portion of hay, 3 hours after gabapentin administration. Water was available at all times. The afternoon before the gabapentin administration, each horse had a jugular vein catheter placed aseptically.
The catheter was then heparin-locked (2.5 mL normal saline, At each sampling time, 2 blinded evaluators (the same 2 throughout the study) completed a full physical examination which included heart rate, respiratory rate, body temperature, mucous membrane color, capillary refill time, gastrointestinal sounds, and observation of or the presence of feces and urination of each horse. In addition, a sedation score was assigned using a sedation scoring system. 20 To ensure that sedation was identified if it occurred, the horses were continuously observed for the first 4 hours after gabapentin administration and a sedation score was assigned at every data collection time point throughout the entire study. The equine response to a loud clapping noise was done to help determine sedation level. The horses were also walked in a tight circle and straight line to look for any signs of ataxia. The Mayhew ataxia scale was utilized; grade 0, normal; grade 1, subtle gait abnormality that may get worse with head elevation. grade 2, moderate gait abnormalities noted at a walk; grade 3, easily recognizable gait abnormalities that are much worse when animal is going around obstacles or head is elevated; grade 4, easily seen gait abnormalities with the potential that the horse will fall easily or nearly fall when asked to walk or perform normal activities; grade 5, recumbent horse. 28

| Quantification of gabapentin in plasma from horses-High-performance liquid chromatography
Gabapentin was extracted from plasma samples using the pre-column derivatization, solid-phase extraction method of Mercolini et al. 29 Previously frozen plasma samples were thawed and vortex-mixed, and 100 μL were transferred to a clean tube, then 30 μL of internal standard (vigabatrin 10 μg/mL) added followed by 1 mL of 0.1 N HCL.
This mixture was loaded onto a preconditioned MCX cartridge (Waters MCX cartridge, Waters Corporation, Milford, Massachusetts).
Analysis of gabapentin in plasma samples was conducted using reversed-phase high-performance liquid chromatography. The system consisted of a 2695 separations module and a 2475 fluorescence detector (Waters). Separation was attained on a Waters Atlantis T3 4.6 × 250 mm (5 μm) preceded by a 5-μm Atlantis T3 guard column.
The mobile phase was a mixture of (A) 50 mM potassium phosphate dibasic buffer (pH 5.0) and (B) acetonitrile. Gradient elution was used to separate the analytes starting with 53% of solution A and 47% of solution B and was adjusted to 49% of solution A and 51% of solution B over 15 minutes, and back to initial conditions over 5 minutes. The flow rate was 1.1 mL/min. The fluorescence detector was set at an excitation of 300 and an emission of 500 with the gain at 10×. The column was at ambient temperature.
Standard curves for plasma analysis were prepared by spiking untreated plasma with gabapentin, which produced a linear concentration range of 25-10 000 ng/mL. Average recovery was 87% for gabapentin. Intra-assay variability ranged from 1.0% to 5.4%, whereas inter-assay variability ranged from 0.4% to 11% for gabapentin, respectively. The lower limit of quantification was 25 ng/mL.

| Estimation of pharmacokinetic parameters
Compartmental analysis was used to calculate primary and secondary (1-compartment model, first-order absorption, no lag time) (2-compartments model, first-order absorption, no lag time).
where K01 is the absorption rate post-nasogastric administration, assuming first-order absorption; D is the oral dose; V is the apparent volume of distribution; and, K10 is the elimination rate constant.

| Statistical analysis
The median AUC 0-∞ for each dose level was compared statistically using Kruskall-Wallis test using Prism 3.03 GraphPad software (La Jolla, California). The level of significance was set at P < .05.
No statistically significant findings were noted in any of the physical examination variables, ataxia, or sedation score of the horses.

| Plasma concentration of gabapentin following simulated multiple dose administrations
For the simulation of plasma gabapentin concentrations after multidose drug administration, data from each horse and dose level (10, 20, 60, 80, 120, 160 mg/kg) was modeled using compartmental analysis and the modeling procedures described above. The mean pharmacokinetics data was used to simulate plasma concentrations of gabapentin at each dose level and 3 dosing intervals scenarios: every 8, 12, and 24 hours over a period of 120 hours.

| Determination of dose proportionality and statistical analysis
The dose proportionality was tested using a standard method known as power model approach 31 T A B L E 1 Plasma pharmacokinetic parameters (median [range]) derived from a 1-or 2-compartmental model with first-order absorption for gabapentin in horses after a single oral administration at 10, 20, 80, 120, and 160 mg/kg of body weight where Ln is the Napierian logarithm, AUC 0-last , and Cmax are the dependent variable jth observation in the ith subject and dose the independent variable, η 1 the random intercept (B 0 ) and η 2 the random slope (B 1 ) and error ε ij. Dose proportionality would be declared when the (1 − α) × 100 confidence interval (CI) of the slope lies entirely within the critical region: Where Ln is the Napierian logarithm, θ high ( 160 mg/kg), and θ low (10 mg/kg) are the preestablished limits of the CI at the lower (0.80) and high ends (1.25), respectively. The advantage of this approach is that it not only tests dose proportionality but also provides evidence of the degree disproportionality. 30,32,33 3 | RESULTS

| Physical examination variables and sedation scores
There were no differences among dosage levels in heart rates, respiratory rates, and body temperatures and all remained within normal limits. Subjectively, appetite, urine, and fecal production/ consistency did not change. Sedation was identified in 1 horse at exactly 2 hours after administration of 120 and 160 mg/kg of gabapentin. The highest sedation score was 2 but returned to 0 after 1 hour. All other sedation scores and all ataxia scores remained 0 throughout the study.

| Estimation of pharmacokinetic parameters and simulated plasma concentration of gabapentin following multiple dose administrations
Following the nasogastric administration of gabapentin at all doses, the drug was detected in all horses (Figure 1). Pharmacokinetic parameters were determined by compartmental analysis are presented in Table 1. For all dose levels and horses, the percent AUC extrapolated to infinity was <17%. Simulated plasma concentration of gabapentin following multiple dose administrations are displayed in Figure 2. The median AUC 0-∞ after the administration of gabapentin at 10 mg/kg was lower than the median AUC 0-∞ obtained after the administration of gabapentin at 120 (P = .02) and 160 mg/kg (P = .008).  (Table 1). This is in contrast to the results reported by 3.4 and 7.7 hours (6.7-11.9 hours) after the oral administration of 5 and 20 mg/kg, respectively. 19,20 The reason(s) for these discrepan-  The fact that horses were sedated with xylazine to facilitate passage of the nasogastric tube to deliver gabapentin is a potential limitation as xylazine slows intestinal motility, 41 which could alter the uptake of orally administered drugs. However, the effects of the xylazine were antagonized with atipamazole, which is shown to return intestinal motility to normal. 41 Clinically, all horses had normal gastrointestinal borborygmi and fecal production, subjectively indicating normal motility. All horses received xylazine/ atipamazole, thus any potential impact would be uniform across all dosages.

| Determination of dose proportionality
Another limitation is the fact that the pharmacokinetics of gabapentin following a single-dose administration was assessed in a relatively small number of adult horses and therefore these findings should be confirmed with a larger population of animals receiving multiple dosages, as would occur in clinical treatment of chronic pain.
Multiple dosage regimen studies would provide valuable information about potential adverse effects or toxicoses not detected in our study, such as previously mentioned hepatic and renal dysfunction [36][37][38] ( Figure 2). Results of this study suggest that gabapentin might accumulate in plasma if it is administered at a <30-hour dosing interval until steady state is reached.
This study did not include IV administration of gabapentin. A commercially available injectable solution was not available at the time and thus bioavailability and IV pharmacokinetics could not be performed. Finally, no pharmacodynamic assessment was performed at varying dosages; therefore, there is no ability to recommend a dose to treat neuropathic pain based on our results.
In conclusion, this study reports the disposition of gabapentin after the administration of multiple escalating doses. However, safety and efficacy of repeated dose administration higher than 20 mg/kg have yet to be confirmed.