Materials and Methods

Dogs—Eight adult sexually intact male dogs of various breeds (2 Pointers, 2 Cocker Spaniels, 2 Gordon Setters, and 2 Golden Retrievers), aged between 6 and 10 years and weighing 15 to 35 kg, were used. The dogs were determined to be healthy on the basis of results of a physical examination and full biochemical and hematologic analyses. Animal care and handling protocols complied with the EC council Directive 86/609 EEC, recognized and adopted by the Italian Government (DL 27/1/1992, No. 16). The study protocol was approved by the ethics committee of the University of Pisa and transmitted to the Italian Ministry of Health.

Treatments and sample collection—In each dog, a catheter was placed into the right jugular vein to facilitate blood sample collections. Dogs were randomly assigned to 2 treatment groups (4 dogs/group) in an open crossover study. Following a 12-hour period of food withholding overnight, each dog received a single dose of zolpidem^a (0.50 or 0.15 mg/kg) orally. The 0.15 mg/kg dose was chosen because it is approximately equivalent to the clinical dose administered to 60- to 70-kg humans. By use of the jugular catheter, a blood sample (2 to 3 mL) was collected before (0 minutes) administration of zolpidem and at 5, 15, 30, and 45 minutes and 1, 1.5, 2, 4, 6, 8, 10, and 24 hours after administration of zolpidem. Each sample was placed in a collection tube containing lithium heparin. A 1-week interval was allowed to elapse between the first administration of zolpidem and the second administration of zolpidem to ensure complete metabolism and excretion. Dogs initially given the lower dose were given the higher dose in the second experiment, and vice versa. During the second experiment, blood samples were collected at the same time points used in the first experiment. Each blood sample was centrifuged at 1,500 × g within 30 minutes after collection. The harvested plasma was then stored at −20°C until use. All samples were analyzed within 30 days after collection.

Clinical evaluations to assess pharmacodynamics—Clinical variables in the dogs were recorded in both experiments immediately before drug administration (0 minutes) and at 0.25, 0.5, 0.75, 1, 1.5, 2, and 4 hours following drug administration. Dogs receiving the higher dose of zolpidem were also evaluated at 6 hours after administration. Animals were videotaped from 15 minutes before and up to 4 hours after drug administration. Heart rate and respiratory rate were monitored via direct auscultation of the heart and lungs, respectively. Systolic and diastolic arterial pressures were monitored noninvasively with a veterinary oscillometric monitor.^b The sedation status was evaluated through the use of a numeric scoring system (Appendix 1).¹⁰ Level of agitation was evaluated on a 5-point scale as follows: 0 = calm, 1 = slightly agitated, 2 = agitated, 3 = very agitated, and 4 = dysphoric and vocalizing (Appendix 2). This scale was created for the present study. Any delayed adverse effects were also recorded. Three veterinarians (who were blinded to the dose the dogs received [including DAP]) independently assessed the sedation and agitation status on the basis of recorded video footage; the median values of the 3 observer scores for sedation and for agitation were used for the statistics. Rectal temperature measured with a digital thermometer was recorded at 0, 1, 2, and 4 hours during both experiments.

Chromatographic instrumentation and conditions—The HPLC system^c consisted of a high-pressure mixer pump, column oven, spectrofluorometric detector, and 20-μL loop. Data were processed with appropriate software.^c The chromatographic separation assay was performed with a reverse-phase C₁₈ analytic column^d (inner diameter, 250 × 4.6 mm; particle size, 5 μm) maintained at room temperature (approx 23°C). The analytic method was derived from a previously described protocol¹¹ with slight modifications. The mobile phase consisted of a 40:60 (vol/vol) mixture of acetonitrile^e and buffer (15mM KH₂PO₄,^f adjusted to a pH of 6.0 with NaOH^f) at a flow rate of 1.0 mL/min. Excitation and emission wavelengths were set at 254 and 400 nm, respectively.

Sample extraction—Each extraction procedure was performed in a 15-mL polypropylene vial. A 500-μL aliquot of a plasma sample was added to 100 μL of internal standard (5 μg of trazodone^g/mL) and vortexed for 60 seconds. Four milliliters of a 7:3 (vol/vol%) HPLC mixture of ethyl-ether^e and CH₂Cl₂^e was added, then the sample was vortexed (30 seconds), shaken (100 oscillations/min for 10 minutes), and centrifuged at 3,000 × g for 10 minutes at 10°C. Three milliliters of the supernatant was collected in a separate glass vial. The organic phase was evaporated under a gentle stream of nitrogen at room temperature and reconstituted with 500 μL of the mobile phase. Twenty microliters of this solution was injected onto the HPLC–fluorescence detector system. The mean recoveries of zolpidem^a and internal standard were 89 ± 6 and 93 ± 6, respectively. Deionized water was purified by a water purification system.^h

Quantification—The calibration curve of peak area versus concentration (ng/mL) of zolpidem was plotted. A least squares regression parameter for the calibration curve was calculated, and the concentrations of the test samples were interpolated from the regression parameter. Sample concentrations were determined by linear regression analysis with the following formula:

View Expanded

where Y is peak area, X is concentration of the standard (ng/mL), m is the slope of the curve, and b is the intercept with the y-axis. Correlation coefficient of the calibration curve was > 0.99.

When unknown samples were assayed, a control and a fortified blank sample were processed along with each set for quality control. Limit of detection and LOQ were determined as analyte concentrations giving signal-to-noise ratios of 1 and 5, respectively.

Pharmacokinetic evaluations—Compartmental pharmacokinetic evaluation for each individual dog's time versus plasma concentration profiles for zolpidem was performed with a pharmacokinetic software package.ⁱ The compartmental approach was preferred to the noncompartmental approach because its good predictive properties are generally independent of the size of the dose and dose interval.¹² The following variables were calculated: Cmax, Tmax, AUC_0–∞, half-life rate constant, and clearance. Area under the plasma concentration curve was calculated by the linear trapezoidal rule.¹³

Pharmacokinetic-pharmacodynamic evaluations—The t_se and Δt_se were determined by linear interpolation between the plasma concentration-time curve and a relevant plasma concentration. This concentration value, defined as the minimum concentration at which adverse effects occurred, was derived from the predicted plasma concentration value at the time at which the first adverse effect was detected in each individual dog. The t_se is equivalent to the time taken to reach this concentration, and Δt_se is equivalent to the period during which this plasma concentration is exceeded.

Statistical analysis—The pharmacokinetic statistical analyses were performed with an ANOVA. The parametrical pharmacodynamic parameters were evaluated with an ANOVA for repeated measures with a Dunnett post hoc test to compare the values determined at 0 minutes with values determined at subsequent time points. For the nonparametric data, a Friedman test with a Dunn post hoc test was used. The results are presented as mean ± SD for parametric analyses and as median and range for nonparametric analyses. All the analyses were conducted with a statistics program package.^j In all the analyses, differences were considered significant at associated values of P < 0.05.

Results

The analytic method was partially revalidated in dog plasma in terms of recovery, linearity, limit of detection, LOQ, recovery, precision, and accuracy. It was selective and specific for the zolpidem, with a limit of detection and LOQ of 0.3 and 1 ng/mL, respectively. The values of precision for zolpidem were always ≤ 6.1% (coefficient of variation), and accuracy was < 5.7%. In some samples, an unknown time-dependent peak was found. It was distinct from the peaks attributed to matrix impurities and continued to increase in concentration even after concentrations of zolpidem had begun to decrease (Figure 1).

View Full Size

Chromatographic curves generated in a study to evaluate the pharmacokinetics and pharmacodynamics of zolpidem after oral administration of a single dose (0.15 or 0.50 mg/kg) in healthy adult dogs. A—Chromatographic curve from blanks of canine plasma. B—Chromatographic curve of a canine plasma sample spiked with zolpidem (100 ng/mL [ZP]) and internal standard (1,000 ng of trazodone/mL [IS]). C—Chromatographic curve from plasma sample collected from a dog 1.5 hours after administration of zolpidem at a dose of 0.50 mg/kg. In all panels, y- and x-axis values represent μV (× 10⁴) and time (minutes), respectively.

Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1650

• Download Figure
• Download figure as PowerPoint slide

Pharmacokinetics—The pharmacokinetics of zolpidem were described by a 1-compartment model. Plasma zolpidem concentrations were still detectable in dogs following treatment with the lower or higher dose of zolpidem up to 4 and 6 hours after administration, respectively. The drug was both rapidly absorbed and eliminated from the plasma compartment. The mean plasma concentration versus time curves for both groups had a similar profile, with the same observed Tmax (1 hour; Figure 2). Time of Cmax differed widely among dogs; absorption was rapid in some dogs (Tmax, 30 minutes), whereas absorption was much more delayed in others (Tmax, 1.5 hours). The variance in Tmax did not appear to be breed associated; however, because each breed was only represented by small numbers of dogs, it was not possible to perform statistical analysis to definitely exclude a correlation. It is noteworthy that, in each dog, Tmax was constant regardless of the dose, demonstrating that this parameter was independent of dose.

View Full Size

Mean concentrations of zolpidem in plasma samples obtained from 8 healthy adult dogs before (0 minutes) and following oral administration of a single dose of zolpidem (0.15 mg/kg [white circles] or 0.50 mg/kg [black circles] in a crossover study. Bars represent SD. The horizontal dotted line represents the minimum concentration at which adverse effects occurred (MC_se); this concentration was derived from the predicted plasma concentration value at the time at which the first adverse effect was detected in each individual dog.

Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1650

• Download Figure
• Download figure as PowerPoint slide

The pharmacokinetic curves (Figure 2) were proportional to the administered dose. The ratio of doses administered in the present study was 3.3; this value was similar to the between-dose ratios for AUC_0–∞ (2.7) and Cmax (3.4; Table 1).

Clinical findings and pharmacodynamics—When dogs were given the lower dose of zolpidem, the sedation scores (data not shown) or agitation scores (Figure 3) did not vary significantly from the baseline scores. Also, cardiorespiratory variables (Figure 4) and temperature values (data not shown) did not differ significantly from baseline values following administration of the lower dose of zolpidem.

View Full Size

Median agitation scores recorded for 8 healthy adult dogs before (0 minutes) and following oral administration of a single dose of zolpidem (0.15 mg/kg [white circles] or 0.50 mg/kg [black circles]) in a crossover study. When dogs received 0.50 mg of zolpidem/kg, an agitation score was also assigned at 6 hours after drug administration. Level of agitation was evaluated on a 5-point scale (created for the present study specifically) as follows: 0 = calm, 1 = slightly agitated, 2 = agitated, 3 = very agitated, and 4 = dysphoric and vocalizing. The bolded and normal values reported in parentheses at each datum point represent the range of scores recorded for dogs treated with the higher and lower dose, respectively. *Value is significantly (P < 0.05) different, compared with the value at 0 minutes (baseline).

Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1650

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean ± SD values of heart rate (A), respiratory rate (B), systolic arterial blood pressure (SAP [C]), and diastolic arterial blood pressure (DAP [D]) recorded for 8 healthy adult dogs before (0 minutes) and following oral administration of a single dose of zolpidem (0.15 mg/kg [white circles] or 0.50 mg/kg [black circles]) in a crossover study. Data were not available for all 8 dogs at each time point because of severe paradoxical reaction. ∧Values derived from 4 dogs. §Values derived from 6 dogs.

Citation: American Journal of Veterinary Research 73, 10; 10.2460/ajvr.73.10.1650

• Download Figure
• Download figure as PowerPoint slide

Following administration of the higher dose of zolpidem, several adverse effects were detected in each dog. Initially, the dogs vocalized, which was followed by restlessness, signs of anxiety, dysphoria, an acute rage reaction, excitement, hyperreflexia, and increased muscle spasticity. There was some variation in time until onset (range, 15 to 50 minutes) of the clinical signs; the duration of these adverse effects was approximately 1 hour. Because of the state of excitation, defined as a paradoxical reaction, some values of the physiologic variables could not be fully evaluated and were considered somewhat questionable. When administered the higher dose of zolpidem, some of the dogs (n = 4) became mildly sedated after resolution of the adverse effects. However, this apparent sedative effect was not considered significantly different from the baseline assessment; therefore, the sedation was classified as clinically irrelevant. Following resolution of the adverse effects, 6 dogs also developed hypersalivation and nausea. After administration of the higher dose of zolpidem, sedation score (data not shown), rectal temperature (data not shown), or cardiorespiratory variables did not differ significantly from baseline values at any time point. However, the agitation score at 0 minutes was significantly different from scores at 0.5, 0.75, 1, and 1.5 hours.

Pharmacokinetic-pharmacodynamic evaluations—The interval between drug administration and vocalization (the first noticeable adverse effect) was used to determine t_se. The mean plasma zolpidem concentration at which vocalization began was 60 ± 20 ng/mL. When dogs received the lower drug dose, this plasma concentration was never attained. When dogs received the higher drug dose, t_se and Δt_se were 25 ± 10 minutes and 65 ± 30 minutes, respectively; moreover, the maximum expression of adverse effects for each dog coincided with the Cmax of zolpidem. Following administration of the 0.50 mg/kg dose, the unknown plasma peak was identified in some dogs, whereas the adverse effects were detected in all dogs.

Discussion

In the present study, the pharmacokinetics and pharmacodynamics of zolpidem, given orally once at doses of 0.15 and 0.50 mg/kg, were assessed in 8 dogs. The active ingredient in zolpidem has been developed as a selective hypnotic drug with higher affinity for the α1-subtype GABA_A receptor, which is believed to be associated with sedation. The lower affinity of zolpidem for the other receptor subtypes is anticipated to result in fewer CNS effects than are typically associated with nonselective agents such as benzodiazepines. The pharmacokinetic profile of zolpidem allows people with insomnia to take the drug later at night without any residual cognitive impairment the next morning. The drug has rapid onset of action and increases total sleep duration and quality. Zolpidem's adverse effect profile in humans is satisfactory in that it appears to have low addiction potential.⁴ Given the findings in people, zolpidem is an attractive candidate for use in dogs, particularly in situations where mild sedation is required (eg, to facilitate handling of animals during diagnostic procedures).

The pharmacokinetic profile of zolpidem in dogs in the present study appeared to have similar features to that in humans: in both species, the pharmacokinetics are dose dependent and have a similar concentration pattern.^14,15 Data from dogs administered zolpidem also resembled data regarding use in humans, in that administration of the drug resulted in fast absorption with a wide variation in Tmax among subjects.¹⁴ However, differences in zolpidem behavior in humans and dogs were apparent and included a lower (approx 3-fold) plasma concentration in dogs, despite administration of equivalent doses.^14,15 This lower plasma concentration could potentially be accounted for by several factors (eg, the oral bioavailability may be lower in dogs than it is in people [70%]).⁴ It is also possible that metabolism of the drug in dogs is more rapid or widespread than in people (zolpidem metabolism principally relies on CYP 3A4 and, to a lesser extent, CYP 1A2 and CYP 2D6 in humans).¹⁶ The latter theory has accounted for similar differences encountered when other human drugs (particularly those metabolized by the same CYP enzymes) are administered to dogs.^17,18

In the present study, a high, time-dependent, unidentified plasma peak was evident in the chromatographic runs performed on plasma samples obtained from dogs that had received the higher dose of zolpidem. It was speculated that this represented a metabolite, although the metabolism of zolpidem does not generate active products in humans.¹⁶ Unfortunately, the lack of information about the structure of zolpidem prevents additional speculation. Further studies are necessary to investigate the nature of this chromatographic peak in canine plasma. Following administration of 0.50 mg of zolpidem/kg, this putative metabolite was found only in plasma from some of the study dogs but all dogs developed adverse effects, and it is unlikely that this metabolite is involved in paradoxical excitation.

Zolpidem has a low incidence of adverse effects in humans. Over 30 placebo-controlled studies evaluating daytime cognitive functions in people taking zolpidem (5 to 10 mg) have revealed a satisfactory safety profile, compared with profiles of other hypnotics, such as flunitrazapam, nitrazepam, and triazolam.^7,19 In patients treated with zolpidem, common adverse effects of the drug included drowsiness (5%), dizziness (5%), headaches (3%), gastrointestinal symptoms (4%), memory problems (1% to 2%), nightmares (1% to 2%), and confusion (1% to 2%).²⁰ Less common adverse effects include ataxia or poor motor coordination, difficulty maintaining balance, euphoria or dysphoria, increased appetite and libido, impaired judgment and reasoning, uninhibited extroversion in social or interpersonal settings, and increased impulsiveness.⁷ Even though there are a few anecdotic reports of zolpidem administration in dogs, the rapid-onset paradoxical reactions associated with administration of the 0.50 mg/kg dose in this study were surprising. These findings support the observations of paradoxical reactions following high-dose zolpidem intoxication in dogs⁹ and with administration of some benzodiazepines in humans.^3,21 In contrast to the paradoxical excitation reaction that developed in 12 of 33 (36%) intoxicated dogs, as detailed by Richardson et al,⁹ results of the present study indicated that the paradoxical excitation reaction in fact affects a larger proportion of individuals and will occur at lower doses that could potentially be considered as therapeutic. Moreover, given that the omega-1 GABA_A receptor is found primarily in the cerebellum and is considered the target for zolpidem, the movement through the blood-brain barrier could affect the zolpidem activity. Some species-specific differences might be expected if a wider heterogeneous population would be considered. The sedation phase that became evident after resolution of the adverse signs in the dogs of this report was likely due to exhaustion following the period of excitation rather than effective hypnosis attributable to zolpidem's action on the receptors.

When dogs received the higher dose of zolpidem in the present study, t_se was < 1 hour, a finding that concurs with the reported interval from ingestion of higher doses (up to 21 mg/kg) of zolpidem to detectable adverse effects in dogs.⁹ The same study⁹ found that signs resolved within 12 hours after drug administration, whereas in the present study, they resolved within 2 hours after drug administration. This divergence may be due to the drastic dose differences between the 2 studies, and the finding may also support the hypotheses on species-specific, dose-related pharmacokinetics.

Current knowledge suggests that the pharmacokinetics of zolpidem are influenced by endocrine factors associated with CYP3A4 metabolism.²² It is thought that CYP3A activity is proportional to plasma concentrations of free testosterone because women routinely attain plasma zolpidem concentrations up to 50% higher than those attained by men and are comparatively more prone to adverse drug reactions, which is presumably the result of less effective biotransformation and excretion.^23,24 Compared with canine isoforms, CYP3A has several parallels.²⁵ Therefore, it is probable that female dogs and also neutered dogs are more likely to have adverse reactions following treatment with zolpidem.

Results of the present study were indicative of rapid-onset and dose-dependent pharmacokinetics for zolpidem in dogs. Unfortunately, the hypnotic effect in humans was replaced by signs of CNS stimulation at the higher dose (0.50 mg/kg), and no effect (therapeutic or adverse) was evident in dogs treated with the 0.15 mg/kg dose (a dose equivalent to the clinical dose in humans). Only at the higher dose did zolpidem administration result in adverse effects, and these effects spontaneously resolved within 2 hours. Although there is currently insufficient background information about the metabolism and receptor selectivity of zolpidem in dogs to fully understand its pharmacokinetics and pharmacodynamics, the present study's findings have suggested that zolpidem is not a recommended drug for inducing sedation in dogs.