Materials and Methods

Animals—Five sexually intact male llamas that were 1 to 2 years old (median weight, 124 kg; range, 92 to 181 kg) were used in the study. The llamas were determined to be in good health on the basis of history and results of physical examination and serum biochemical analyses. The university-owned llamas were dewormed and vaccinated and acclimated to their premises for 10 days prior to the study. Llamas were evaluated on 5 occasions, receiving each treatment at weekly intervals, according to a Latin square design. The study was approved by The University of Tennessee Institutional Animal Care and Use Committee.

Sedation—Food was withheld for 24 hours before anesthesia was induced. Approximately 1 hour prior to the start of the experiment, each llama was moved to a quiet room. Baseline values for rectal temperature, pulse, and respiratory rate were recorded once the llama became accustomed to its new environment. The llama then received 1 of the following 5 treatments: A (acepromazine maleate,^c 0.05 mg/kg), B1 (butorphanol,^d 0.1 mg/kg), AB (acepromazine maleate, 0.05 mg/kg, and butorphanol, 0.1 mg/kg), B2 (butorphanol, 0.2 mg/kg), or C (physiologic saline [0.9% NaCl] solution, 5 mL). The dose of each drug was prepared separately and then combined in a syringe to 5-mL volume with physiologic saline solution; the preparation was immediately injected into the semitendinosus or semimembranosus muscle by use of an 18-gauge, 1.5-inch needle. At 5minute intervals for a period of 30 minutes, an observer (EBB) evaluated the degree of sedation by use of a 6-point scale (Appendix) in which 0 indicated no sedation and 5 indicated maximum sedation.

Anesthesia—Thirty minutes after administration of the premedication treatment, and immediately prior to the administration of the TZ combination (2 mg/kg), respiratory rate and heart rate were recorded. The dose of TZ (prepared to a 5-mL volume with physiologic saline solution) was administered into the semitendinosus or semimembranosus muscle of the contralateral hind limb. The TZ combination doses are reported as the sum of tiletamine and zolazepam doses. After the llama became sternally recumbent and its neck could be extended without resistance, it was rolled into left lateral recumbency.

Recording and monitoring—Characteristics of anesthesia and recovery and selected cardiorespiratory variables were recorded for each llama during each experiment. These included times from TZ injection to sternal recumbency and to lateral recumbency, duration of antinociception, and times to subsequent sternal recumbency during recovery from anesthesia and to standing. In addition to the quality of recovery, heart rate, SpO₂, and blood pressures were monitored and recorded every 5 minutes while the llama was recumbent. Heart rate was monitored continuously by use of baseapex ECG electrodes.^e Hemoglobin oxygen saturation was estimated by use of a pulse oximetry probe^f placed on the tongue. Respiratory rate was determined by observation of thoracic excursions during a 1-minute period.

Systolic arterial blood pressure, DAP, and MAP were measured by use of an oscillometric device^g and recorded at 5-minute intervals. An appropriately sized cuff (neonatal size No. 4 or No. 5) was placed over the metacarpal artery of the nondependent forelimb and the limb positioned so that the cuff was approximately level with the base of the heart.

For blood gas analysis, arterial blood samples (1 mL) were collected percutaneously from the left femoral artery at 5 and 15 minutes after the llama was positioned in left lateral recumbency. Blood was collected anaerobically with a 1-mL syringe containing heparin and samples were analyzed^h with a cartridgeⁱ within 5 minutes of collection.

During anesthesia, antinociception was assessed every 5 minutes by clamping a digit with 10-inch Vulsellum forceps.^j The forceps was closed tightly to the first or second ratchet, depending on the digit size, for a maximum period of 1 minute or released sooner if the llama made a purposeful movement during application of this stimulus. Purposeful movement was defined as gross movement of the limbs or head and was considered to represent an absence of antinociception. Stimuli were applied after cardiorespiratory measurements were recorded. The order in which digits were clamped was randomized, and no digit was clamped on more than 2 occasions.

Quality of recovery was subjectively evaluated by use of a 3-point scale as follows: score 1 = minimal struggling or paddling and standing on the first or second attempt; 2 = mild to moderate struggling and 3 attempts to stand; and 3 = moderate to excessive struggling or paddling during recovery and standing after ≥ 4 attempts.

Statistical analysis—A mixed-model ANOVA was used to examine the effect of treatment (ie, sedation) on physical and biochemical values among treatments. The independent variables treatment, time, and interaction between treatment and time were evaluated for their effect on the dependent variables heart and respiratory rate; PaO₂; PaCO₂; SpO₂; and DAP, systolic arterial blood pressure, and MAP. Week of treatment, llama, and the 3-way interaction between treatment, llama, and week of treatment were used as the error term in the model. A second mixed-model ANOVA was used to evaluate the effect of treatments on anesthetic and antinociceptive effects. The independent variables treatment, llama, week of treatment, and body mass were evaluated for their effect on the following dependent variables: time from induction of anesthesia until sternal recumbency, time from induction of anesthesia to lateral recumbency, time from induction of anesthesia to the end of antinociception, duration of lateral recumbency, time from induction of anesthesia to standing, and quality of recovery. By use of the W statistic of Shapiro-Wilk and observation of stem leaf diagrams, distributions of residuals from the models were used to ensure that model assumptions were met. Results are expressed as least squares mean ± SEM. For final analyses, a value of P ≤ 0.05 was considered significant.

Results

Acepromazine, butorphanol, or the combination of those 2 drugs did not have any effect on the heart rate or respiratory rate prior to administration of TZ. Sedation scores were zero for all treatments; therefore, those data could not be analyzed statistically. After TZ administration, times to sternal and lateral recumbency were not significantly different among treatment groups.

Duration of antinociception was significantly (P = 0.007) longer for treatments A (35 ± 7 minutes) and AB (41 ± 6 minutes) than it was for treatment C (7.5 ± 7 minutes; Table 1). Duration of antinociception for treatment AB was significantly longer than that associated with treatment B2 (14 ± 6 minutes). Time in lateral recumbency (ie, interval after injection of TZ to subsequent sternal recumbency) was significantly (P = 0.046) longer for treatment AB (97 ± 10 minutes) than for treatment C (50 ± 10 minutes).

Table 1—

Effect of IM treatment (least squares mean ± SEM) with acepromazine (0.05 mg/kg; A), a low dose of butorphanol (0.1 mg/kg; B1), a combination of acepromazine (0.05 mg/kg) and butorphanol (0.1 mg/kg; AB), a high dose of butorphanol (0.2 mg/kg; B2), or saline (0.9% NaCl) solution (5 mL; C) on times to sternal and lateral recumbency and duration of antinociception, time to standing, and quality of recovery following IM administration of a TZ combination (2 mg/kg) in 5 llamas (each of which received each treatment once).

View Table

Time to standing did not differ among treatments (Table 1). Recovery was considered smooth on most occasions, and there was minimal struggling or paddling; however, 3 llamas that received treatment A and 1 each that received treatments B1 and B2 made 4 attempts to stand. The quality of recovery did not differ among treatments.

Although cardiorespiratory function was monitored throughout the period of lateral recumbency, regardless of its duration, only the first 30 minutes of cardiovascular and respiratory data are reported (Table 2). During lateral recumbency, heart rate was increased significantly (P = 0.007) for treatment A (54 ± 3 beats per min), compared with treatments B1 (42 ± 3 beats per min) and C (40 ± 3 beats per min). The DAP and MAP values were significantly lower for treatment A than for treatment C.

Table 2—

Effect of treatments A, B1, AB, B2, or C on cardiovascular and respiratory values (least squares mean ± SEM) following IM administration of a TZ combination in 5 llamas (each of which received each treatment once).

View Table

The respiratory rate of llamas that received treatment B1 decreased significantly (P = 0.022) from baseline, compared with changes in respiratory rate of llamas that received treatment C (Table 2). The PaO₂, PaCO₂, and SpO₂ values did not differ significantly within treatments at 5 and 15 minutes, and values did not differ among treatments at either time point. One llama that received each of treatments A, B2, and AB had severe hypoxemia at 5 minutes (PaO₂ = 24, 28, and 26 mm Hg, respectively); however, there was no significant difference in these values among groups.

Discussion

On the basis of results of the present study, butorphanol, acepromazine, or their combination did not induce sedation in llamas by 30 minutes after IM administration of the doses evaluated. To reduce variability in drug uptake, care was taken to ensure that drugs were injected deep into the muscle and that the same site was used for all administrations. A 30-minute period was considered adequate to allow IM premedications to take effect, because most drugs that are administered as aqueous solutions achieve peak plasma concentrations within that interval.¹⁶ For butorphanol, this supposition is supported by published data,⁶ which indicate that the mean plasma butorphanol concentration at 15 minutes was higher than the value at 30 minutes following IM administration of the drug to llamas.

The doses of butorphanol used in our study were based on published information that describes the pharmacokinetics and pharmacodynamics of single IV and IM doses (0.1 mg/kg) of butorphanol in llamas.⁶ Interestingly, it was reported⁶ that 2 llamas became sedated and 2 became excited following butorphanol administration; however, it was not mentioned whether these effects occurred following IV or IM administration. Behavioral effects associated with administration of butorphanol (0.1 or 0.2 mg/kg, IV) in awake sheep have also been reported; in those animals, butorphanol caused restlessness, chewing, and vocalization.⁹ No such behavioral effects were observed in the present study, perhaps because these effects are less likely to occur if the drug is given IM.

The reported sedative effects of butorphanol alone, or in combination with other agents, indicate that the drug's effects are variable. Butorphanol improved the quality of sedation in horses treated with the A₂-adrenoceptor agonist romifidine.¹⁷ The inability of butorphanol to potentiate the sedative effects of acepromazine in the present study may be a result of differences in drug interactions or species differences in response.

Acepromazine is widely used for premedication and sedation of domestic species; however, the degree of sedation achieved via acepromazine administration is variable, and the drug cannot be relied upon to induce sedation on every occasion.¹⁸ The dose of acepromazine used in the present study was within the accepted dose range for domestic species.¹⁹ It is possible that a higher dose of acepromazine may have resulted in sedation in the study llamas; however, it is generally accepted that there is little benefit in the use of high doses of acepromazine.¹⁸ Another possibility for the lack of a sedative effect following acepromazine administration was that insufficient time was allowed for signs of sedation to become obvious. This seems unlikely because the onset of sedation following IV administration of acepromazine to horses was approximately 20 minutes, although maximum sedation was not achieved until 33 minutes.²⁰ In another study,¹⁵ IM administration of acepromazine to horses 23 minutes prior to induction of anesthesia with ketamine improved the quality of induction, indicating that considerable absorption of acepromazine had occurred by that time. The inability to detect a sedating effect for acepromazine or butorphanol in the llamas in the present study could also be attributed to the lack of sensitivity of the sedation assessment scale. This llama sedation scale has been used for a decade in the authors' workplace and is based on clinical signs; however, it may not suitable for identifying subtle changes in behavior.

Administration of the TZ combination alone resulted in only a brief period of antinociception, making it unsuitable at this dose for most surgical procedures. The TZ combination dose was chosen on the basis of the findings of other experiments by one of our group, in which TZ administered IM at a dose of 2 mg/kg resulted in lateral recumbency and a short period of antinociception in llamas. The combination of TZ appears to have poor antinociceptive effects even at doses capable of causing immobilization. In llamas, higher doses of the TZ combination (4.4 mg/kg, IM) failed to block the response to noxious stimuli; however, the type of nociceptive stimulation was not described.^b A study²¹ in rats revealed that TZ, at doses that resulted in a general anesthesia-like state (ie, caused immobility, loss of consciousness, and loss of the righting reflex), had only minimal effects on the tail-flick test.

Butorphanol did not improve the antinociceptive effects of the TZ combination in our study. In contrast, butorphanol significantly enhanced the antinociceptive effect of TZ in the tail-flick test in rats.²¹ This discrepancy may be attributable to the difference in the methods used to measure antinociception and the type of nociceptive stimulus between studies. Butorphanol seems to be efficacious in thermal nociceptive assessments, such as the tail-flick test in rats; in sheep, the drug was efficacious in a thermal nociceptive experiment but not in a mechanical nociceptive experiment.⁹ Thus, on the basis of these findings, it is perhaps understandable why butorphanol had no major antinociception effects in our study, in which mechanical assessment of nociception was used.

Although there was no significant difference in the duration of antinociception between the butorphanol- and saline solution–treated llamas, the mean duration of antinociception after treatment with butorphanol was, nevertheless, approximately twice that associated with saline solution treatment. Failure to identify significant differences between the saline solution–treatment effect and effect of the high or low doses of butorphanol may have been a consequence of the effect of sample size. On the basis of the differences between findings after saline solution treatment and findings after treatment with butorphanol, 41 and 85 llamas would be needed in the low- and high-dose butorphanol treatment groups, respectively, to obtain statistical significance at the 0.05 level with a power of 80%.

Administration of acepromazine alone significantly increased the duration of TZ-induced antinociception, compared with the effect of administration of saline solution, as determined in the present study. These findings are consistent with the minimum alveolar concentration-reducing effect of acepromazine for inhalational anesthetic agents¹³; however, premedication with acepromazine did not improve the duration of antinociception in horses anesthetized with romifidine and ketamine.¹⁵ Prolongation of the antinociceptive effects of the TZ combination by another drug may be a result of sedating or antinociceptive actions of the drug in question. Although the sedative actions of acepromazine are well established, it is generally accepted that phenothiazines provide poor analgesia.¹⁹ Nevertheless, the phenothiazine prochlorperazine did increase the pain threshold in a rodent study involving hot-plate stimulation.²² The antinociceptive effect of prochlorperazine was attributed to a centrally mediated cholinergic mechanism, because the antinociception was abolished by the anticholinergic drug atropine.

In the llamas of the present study, combination of butorphanol and acepromazine did not significantly improve the duration of antinociception, compared with that associated with acepromazine alone. These findings are consistent with the interaction of acepromazine and butorphanol with volatile anesthetics in ponies, in which the drug combination did not reduce the minimum alveolar concentration by a greater margin than did acepromazine alone.¹² There is, however, reason to expect an improvement in antinociception with the combination of acepromazine and butorphanol, and there is evidence for an interaction between the dopaminergic and opioid receptor systems.²³ The neuroleptic drug droperidol significantly increased the analgesic effects of fentanyl, compared with effects achieved by fentanyl alone, in a mechanical nociceptive study of sheep.²³

Recovery from anesthesia was considered smooth and uncomplicated in all llamas after each experiment. In a previous study,^b the TZ combination (4.4 mg/kg IM) was administered as the sole anesthetic agent in llamas; muscle tremors, rigidity, and hyperthermia developed in 1 of 4 llamas during recovery from anesthesia. In another study⁴ in guanacos, TZ (5.0 ± 1. 1 mg/kg, IM) caused stumbling and disorientation during recovery from anesthesia. None of the adverse responses mentioned in those 2 reports were observed during the recovery phase in the llamas of the present study. On the basis of our assessment criteria, the recovery period was uneventful, which may be a result of administration of a comparatively lower dose of TZ.

Overall, heart rate among llamas was significantly higher for treatment A than it was for treatments B1 and C. The effects of acepromazine on heart rate are variable and perhaps dose dependent. In dogs, large doses of acepromazine cause bradycardia.²⁴ In horses, acepromazine (0.025 mg/kg, IM) prevents the bradycardia and hypertension induced by romifidine.¹⁵ Butorphanol had no significant effect on heart rate in our study, and this finding is in contrast to the effect of butorphanol in awake llamas, for which development of bradycardia was reported.⁶ The reported effects of butorphanol on heart rate are variable. In TZ-anesthetized rats, butorphanol caused transient bradycardia²¹ but did not significantly affect heart rate in TZ-anesthetized sheep.²⁵

Mean arterial blood pressure and DAP were lower in acepromazine-treated llamas than in saline solution– treated llamas, and these findings are consistent with the vasodilating effects of acepromazine.¹⁸ Butorphanol had no significant effect on blood pressure, a finding consistent with the effects of butorphanol in TZ-anesthetized goats.¹⁰ In sheep anesthetized with TZ, butorphanol (0.5 mg/kg, IV) caused a decrease in MAP, an increase in systemic vascular resistance, and a decrease in cardiac output.²⁵

Arterial pressure monitoring by use of oscillometric devices is a noninvasive method in routine clinical use; however, the position of the cuff in relation to the patient's aortic root and the size of the cuff can cause erroneous readings.²⁶ In the present study, error was minimized by positioning the cuff at approximately the level of the heart base and by use of an appropriately sized cuff.²⁶

A significant decrease in respiratory rate was detected in the llamas following treatment with the low dose of butorphanol (B1) but not following treatment with the high dose of butorphanol (B2). Because there was no difference in PaCO₂ between the low-dose butorphanol treatment and the saline solution treatment, the tidal volume must have increased in llamas receiving the low dose of butorphanol. An increase in PaCO₂ was detected after all of the treatments, indicating that hypoventilation was occurring; however, these values were within the expected range for spontaneously breathing recumbent animals. Opioids are generally regarded as respiratory depressants; however, a ceiling for the respiratory depressant effects of butorphanol has been reported.¹⁸ In TZ-anesthetized goats, butorphanol had no significant effect on respiratory rate, PaCO₂, or PaO₂ when goats were placed in dorsal recumbency.¹⁰ The lack of significant differences in SpO₂, PaO₂, and PaCO₂ between the butorphanol treatments used in our study indicated that butorphanol-induced respiratory depression in the llamas was not dose dependent.

For all 5 treatments, PaO₂ was low 5 minutes after lateral recumbency and values ranged from 35 to 57 mm Hg; values for awake llamas at sea level are typically in the range 80 to 95 mm Hg.²⁷ Several mechanisms are thought to be responsible for the decreased PaO₂ during anesthesia and recumbency. Hypoventilation increases the alveolar partial pressure of CO₂ and contributes to hypoxemia; however, the degree of hypoventilation was not severe in the present study and would have had only a small effect on PaO₂. Major causes of hypoxemia during anesthesia appear to be associated with ventilation-perfusion mismatching and intrapulmonary shunting from perfusion of nonventilated areas of the lungs. Anesthesia induces changes in lung ventilation that are secondary to a decrease in functional residual capacity and regional differences in pleural pressure. Additionally, anesthesia causes changes in lung perfusion secondary to decreases in cardiac output and pulmonary vascular resistance and inhibition of hypoxic pulmonary vasoconstriction.

After all 5 treatments, the SpO₂ values remained relatively high despite the low values for PaO₂. The pulse oximeter used in our study was not validated for use in llamas, and we are unaware of any study to validate the use of any commercially available pulse oximeter in llamas. Nevertheless, the SpO₂ values are consistent with published values for llamas.²⁸ In a previous study,²⁸ llamas maintained arterial oxygen hemoglobin saturation values of 94%, 90.4%, 84.8%, and 71.5% at PaO₂ values of 71.7, 57.3, 43.1, and 31.9 mm Hg, respectively. As an adaptation to high-altitude living, hemoglobin of llamas has a high affinity for oxygen^27-29 (P₅₀ value [PaO₂ at 50% saturation of hemoglobin] of approx 21 mm Hg at sea level) and enhanced kinetics of oxygen use.²⁷

In adult male llamas, acepromazine, butorphanol, or the combination of those drugs did not induce sedation at the doses evaluated in the present study. Prior administration of acepromazine and acepromazine with butorphanol significantly prolonged the analgesic effect of TZ, compared with the effect of treatment with saline solution, but butorphanol alone had no significant effect on TZ-induced antinociception. Llamas treated with TZ alone or TZ with either acepromazine or acepromazine and butorphanol may develop transient hypoxemia; however, because of the unique physiology of llamas in relation to hemoglobin saturation and oxygen delivery, the authors do not considered this to be a serious problem.