Constant-Rate Infusion of Dexmedetomidine to Manage Thiopental Anesthesia during Intracranial Surgery in Cynomolgus Macaques (Macaca fascicularis)

Abstract

Macaques (Macaca spp.) are often used as animal models in biomedical research involving a neurosurgical approach. The development of new anesthetic techniques is pivotal for these studies. Studies in human anesthesia for intracranial surgery have shown that dexmedetomidine infusion reduces the incidence of cardiocirculatory complications in the perioperative period, reduces the need for supplemental analgesia, and provides an analgesic effect analogous to that of remifentanil. Data regarding the anesthetic effects of dexmedetomidine infusion in NHP including Macacaspp. are currently unavailable. The study population comprised 5 healthy cynomolgus macaques (Macaca fascicularis) that underwent intracranial surgery. On the day of surgery, the subjects were sedated with intramuscular ketamine (8 mg/kg) and dexmedetomidine (0.02 mg/kg). Anesthesia was induced with thiopental (3 mg/kg IV) and maintained by using constant-rate infusion of thiopental (3 mg/kg/h); analgesia was provided by constant-rate infusion of dexmedetomidine (0.012 mg/kg/h). Atipamezole (0.1 mg/kg IM) was administered at the end of the surgical procedure. The median heart rate increased after sedation, reaching its highest level at 60 min (91.0 ± 6.9 bpm); the highest systolic blood pressure (119.6 ± 10.5 mm Hg) occurred at 75 min. No animal experienced respiratory arrest, and all recovered within 6 min after atipamezole administration. In cynomolgus macaques, dexmedetomidine constant-rate infusion provided adequate analgesia and stable hemodynamic control. Using dexmedetomidine as an adjunct to thiopental-maintained anesthesia may be advantageous in healthy NHP undergoing intracranial surgery.

Cynomolgus monkeys (Macaca fascicularis) are often used as animal models in behavioral and biomedical research due to their phylogenetic affinity to humans. Chemical restraint and anesthesia are often required for the care of these species or for the purposes of the research in which they are involved.

The α2-adrenoreceptor agonists are a class of drugs widely used to produce sedation, anxiolysis, analgesia, muscle relaxation, and a sparing effect of injectable and inhalant anesthetics in different species.¹⁶ Moreover, they play a key role in NHP pharmacologic immobilization.^4,17,20,21 Various studies have demonstrated that medetomidine, a highly selective α2 agonist, produced a reliable loss of consciousness in rhesus monkeys when used alone or in association with ketamine, midazolam, or fentanyl.²⁰

Dexmedetomidine is the dextroisomer of medetomidine formulation and, at half the dose of medetomidine, it produces a similar effect.^2,8 Studies in human anesthesia for intracranial surgery have shown that dexmedetomidine infusion decreases the incidence of cardiocirculatory complications in the perioperative period, reduces the requirements of supplemental analgesia, and has an analogous analgesic effect to remifentanil.¹² In dogs, dexmedetomidine infusion has produced a marked reduction in cerebral blood flow in healthy animals.²² In rabbits, in the presence or absence of cryogenic lesions, dexmedetomidine infusion had minimal effects on intracranial pressure.²³

Various studies involving dogs have investigated the effect of dexmedetomidine at constant-rate infusion during isoflurane anesthesia.^6,10,11 At low doses, dexmedetomidine infusion (from 0.1 to 0.5 μg/kg/h) yielded acceptable cardiopulmonary changes, whereas a rate infusion of 3 μg/kg/h produced significant cardiorespiratory depression.^10,11 Only 2 studies have investigated the effects of a single dexmedetomidine bolus in NHP for chemical restraint, and no data have been reported regarding constant-rate infusion in this species.^15,19

We hypothesized that dexmedetomidine infusion in Macaca fascicularis undergoing intracranial surgery would provide analgesia, cardiocirculatory stability and a sparing effect of the thiopental dose necessary for maintaining a satisfactory plane of anesthesia. The aim of this study was to evaluate cardiocirculatory, respiratory, and anesthetic variables during dexmedetomidine infusion and thiopental general anesthesia in Macaca undergoing intracranial surgery.

Materials and Methods

Five male cynomolgus macaques (Macaca fascicularis; age, 5 to 12 y; weight, 3 to 11 kg) were involved in the study. Each primate underwent intracranial surgery for the purpose of anatomic and physiologic research that was approved by the Ethical-Scientific Committee of the University of Bologna and authorized by National Competent Authorities in accordance with Legislative Decree number 116/92 (enacting Council Directive 86/609/ECC). At their arrival, the macaques were certified free from herpes B virus, SIV, simian retrovirus, and simian T lymphotrophic virus. The animals had never been used previously in an experiment.

The macaques were housed individually in appropriately sized cages, and environmental enrichment was provided in accordance with standard operating procedures. Twelve hours before the surgical procedure, food was withheld but water was available free choice. Two of the 5 animals were anesthetized to implant 2 metal plates on the skullcap for electrophysiologic studies; the remaining 3 macaques underwent craniotomy to inoculate anterograde tracers in the white matter of the brain for anatomic study.

On the day of the surgery, the macaques received ketamine (8 mg/kg IM; Ketavet, Intervet Productions, Latina, Italy) and dexmedetomidine (0.02 mg/kg IM; Dexdomitor, Elanco Animal Health, Hampshire, United Kingdom) in the gluteal muscles. The animals were immobilized for injection by using the retractable rear wall of the cage. Sedation time was defined as the time between the intramuscular injection and the moment when the animal was laterally recumbent on the floor of the cage (time 0). As soon as the macaques became recumbent, the clinical sedation score was determined in regard to response to firm pressure applied to the digits (grip reflex) and response to catheterization of the femoral vein (scored as ‘yes’ or ‘no’). The macaques were removed from the cages and carried to the operating theater. The left saphenous vein was catheterized by using a 22-gauge angiocatheter (Delta Med, Milan, Italy), and all monitoring devices were attached before the induction of anesthesia. A continuous lead II ECG was obtained by using a multiparameter monitor (Cardioline, Milan, Italy), and the pulse rate and arterial oxygen saturation were monitored by pulse oximetry (probe placed on the first anterior digit). The respiratory rate was determined by visually monitoring thoracic expansion over 1 min. Body temperature was measured by using a rectal temperature probe. Blood pressure was monitored by using a Doppler device (Minidrop ES-100 VX, Hadeco, Tokyo, Japan).³ In brief, the Doppler probe was placed over the common digital branch of the metatarsal artery on the dorsal aspect of the foot; the width of the cuff was approximately 40% of the limb circumference, as described previously.¹⁸ A no. 3 cuff was used for each animal. The hair was clipped before placing the probe. Ultrasound transmission gel (Aquasonic 100, Parker Laboratories, Fairfield, NJ) was placed between the probe and the skin to improve contact, and the volume of the Doppler machine was adjusted to obtain a clearly audible signal. Five measurements were performed; the highest and the lowest values were excluded, and the mean of the remaining values was calculated. All of these values and the presence of the grip reflex were recorded at 5-min intervals for the duration of the entire procedure. A forced-air warming blanket (Bair Hugger, 3M, Berkshire, United Kingdom) was used to maintain physiologic body temperature. After the first monitoring was completed, a bolus of thiopental (3 mg/kg IV; Pentothal, Intervet Productions) was administered, and infusion (3 mg/kg/h) was started by using an infusion pump (Angle syringe pump, CRIMO, Forlì-Cesena, Italy). Dexmedetomidine infusion (0.012 mg/kg/h) and lactated Ringer solution were administered (5 mL/kg/h) by using a peristaltic pump (B-Braun, Melsugen, Germany). Oxygen (2 L/min) was provided through a human pediatric facemask. The macaque's head was placed into the stereotaxis instrument, with the animal in the sphinx position. The animal was fixed to the stereotaxic frame by means of tapered ear bars which fit into the external auditory meatus. Adjustable infraorbital clamps, and a vertically adjustable palate bar prevented rotation of the skull. The stereotaxic apparatus limited the possibility of endotracheal intubation because of the encumbrance of the palate bar.

If a grip reflex or responses to stimuli were present, a bolus of thiopental (2 mg/kg IV) was administered. All these events were recorded. In the case of hypotension, defined as systolic blood pressure of less than 80 mm Hg, dopamine infusion from 5 to 10 μg/kg/min was administered IV. At the end of the procedures, atipamezole (0.1 mg/kg IM; Sedastop, Esteve, Oudewater, Netherlands) was administered. Time of recovery was defined as the time from the end of the procedure and cessation of the thiopental until the animal was able to sit up in the cage. To record any possible side effects, such as dysphoria and vomiting, the anesthetists observed the macaques until they were fully awake.

Statistical analysis.

All data are reported as mean ± 1 SD. The baseline heart rate, respiratory rate, and systolic blood pressure (measured after sedation with dexmedetomidine and ketamine) were compared between time points by using the Wilcoxon test (paired samples) (MedCalc 6.3, MedCalc Software, Ostend, Belgium) A P value of less than 0.05 was considered statistically significant.

Results

The sedation was smooth, and in 3 to 5 min, all of the cynomolgus macaques were lateral recumbent and unresponsive to manipulation. At the evaluation of the sedation score, the palpebral reflex was depressed, the menace and grip reflexes were lost, and muscle relaxation was good. No animal reacted during the positioning of the venous catheter.

The procedure time was 204 ± 121.2 min (mean ± 1 SD), depending on the type of surgery. The cardiovascular parameters are summarized in Figure 1. The heart rate increased after sedation increasing to 91 ± 7 bpm at 60 min. During the procedure, 2 macaques experienced sinus bradyarrhythmia consisting of a regular variation of rate associated with respiration. No other rhythm abnormalities were detected. The lowest systolic blood pressure measured in each animal was the baseline value (93.8 ± 5.7 mm Hg). During surgical manipulation, the highest blood pressure was measured at 75 min (119.6 ± 10.5 mm Hg). The mean arterial oxygen saturation according to pulse oximetry at the first measurement point was 97% and was maintained throughout the surgical period. No respiratory arrest occurred; the respiratory rate ranged from 18 to 24 breaths per minute. During the surgical period, the rectal temperature remained stable at approximately 36.5 °C with the aid of a forced-air warming blanket. Neither heart rate, blood pressure, mean arterial oxygen saturation, nor rectal temperature differed between any time points.

Figure 1.

Cardiovascular and respiratory parameters recorded from sedation (time 0) to the end of a neurosurgical procedure in cynomolgus macaques.

Thiopental was maintained through constant-rate infusion of 3 mg/kg/h in most of the animals. Only one macaque (at 120 min after induction) showed a marked grip reflex and response to stimulus (gross movement of the legs) in the absence of changes in cardiorespiratory parameters. In this case, a thiopental bolus (2 mg/kg IV) was administered, and thiopental infusion rate was increased to a maximum of 8 mg/kg/h until the disappearance of the grip reflex. In the remaining 4 monkeys, when there was an unexpected positive response to the grip reflex, a thiopental bolus (2 mg/kg IV; 2 times for 3 monkeys) was administered, however we did not observe hemodynamic or other gross movement associated with any surgical stimulus (incision of the skin, craniotomy, manipulation of the meninges). The average recovery time from anesthesia after atipamezole administration was 6.0 min (range, 4.5 to 7.0 min). All of the macaques awoke smoothly, without dysphoria or side effects.

Discussion

The present study confirmed that dexmedetomidine infusion in cynomolgus macaques undergoing intracranial surgery provided sufficient analgesia to maintain an adequate plane of anesthesia. The combination of ketamine and dexmedetomidine produced effective immobilization with an onset of less than 5 min in all 5 macaques treated.

In NHP, thiopental constant-rate infusion (15 to 17 mg/kg/h), used to maintain general anesthesia, has been reported to provide satisfactory chemical restraint.¹³ In the present study, the thiopental rate infusion was initially set at 3 mg/kg/h. Only 1 of the 5 macaques demonstrated a positive grip reflex and gross movement in response to surgical stimulus; administration of a thiopental bolus (2 mg/kg IV) was insufficient to achieve a stable anesthetic plane, so we increased the thiopental infusion rate to 8 mg/kg/h. This dosage is still lower than that previously reported in NHP.¹³ The choice of the initial thiopental infusion rate was justified because, in our experience, previously reported infusion rates¹³ associated with dexmedetomidine led to excessively deep sedation. In fact, the contemporary use of dexmedetomidine, as reported in small animal anesthesiology studies, decreased the intraoperative requirement of the drug used for the maintenance of general anesthesia.^8,10 Comparison with a control group of macaques anesthetized with thiopental infusion only might be useful in determining the dosages necessary to maintain anesthesia in the absence of dexmedetomidine infusion and in assessing the sparing effect of the sedative drug. However, the approved protocol did not include a control group and, therefore, the dosages were extrapolated from a previously published study.¹³

Dexmedetomidine produces bradycardia and initial hypertension, as do other α2-adrenoreceptor agonists.^14,15 In the present study, assessment of pretreatment baseline values for heart rate and blood pressure was infeasible due to risk to the handler and the need to minimize stress to the subjects. The normal resting heart rate of cynomolgus macaques is between 100 to 200 bpm.⁹ After sedation, the first recorded heart rate was 68.4 ± 7.4 bpm, whereas the mean heart rate ranged from 69.6 to 94.0 bpm during the procedure and therefore can be defined as bradycardia when compared with the normal values previously reported.⁹ The bradycardia was due to the action of the dexmedetomidine on central and peripheral α2 adrenoreceptors, as described previously for NHP and in small animals.^4,16 The bradycardia was accompanied by a respiratory sinus arrhythmia, which might have been related to the action of dexmedetomidine on central α2-adrenoreceptor agonists, with a reduction of sympathetic outflow.¹⁶ In the event of life-threating bradycardia or arrhythmias, atipamezole should be administered to reverse the cardiocirculatory side effects of dexmedetomidine.⁷ Another potential explanation for the bradycardia is an increase in intracranial pressure consequent to the surgical procedure, as has been reported in human patients.¹ However, the bradycardia in our macaques began before the intracranial surgical procedure was initiated.

Throughout the intracranial surgery, the systolic blood pressure remained stable, at 93.8 ± 5.7 to 119.6 ± 10.5 mm Hg, in all 5 macaques. The lowest value was measured at the first time point. This finding seems to be in contrast to the biphasic arterial pressure (initial hypertension followed by hypotension) due to dexmedetomidine injection in dogs.¹⁴ The effect in our study might be explained by a decrease in central sympathetic tone that overcame the peripheral cardiovascular effects.¹⁶ None of our macaques experienced hypotension, and none received dopamine infusion.

In the current study, the respiratory rate showed an initial decrease, but the arterial oxygen saturation according to pulse oximetry remained within the normal range throughout surgery. Pulse oximetry is widely considered the optimal method for noninvasive continuous monitoring of the oxygen saturation of arterial hemoglobin; however to overcome the limitations of this technology, a blood-gas evaluation of the oxygen partial pressure should be performed.⁵ In the present study, the pulse oximetry did not drop below 97%, and the wave tracing on the monitor remained stable throughout surgery; we therefore considered blood-gas analysis to be unnecessary. However the anesthetist should evaluate capnography and end-tidal CO₂ to confirm adequate ventilation. A criticism regarding the anesthesia management of these 5 macaques was the inability to intubate them due to the position of their heads in the stereotaxic apparatus and the lack of blood-gas analysis to control the pCO₂. During intracranial surgery, monitoring of end-tidal CO₂ or pCO₂ enables detection of hypercapnia. In fact, hypercapnia decreases the blood pH, which induces vasodilation in the brain and consequently increases cerebral blood flow and intracranial pressure.

In small animals, α2-adrenoreceptor agonists typically induce vomiting due to their direct action on the chemoreceptor trigger zone.¹⁶ Similarly, in NHP, vomiting occurs during the perioperative period; therefore, endotracheal intubation is recommended to avoid aspiration pneumonia, which is a risk even in food-fasted animals. However, endotracheal intubation was not possible in the current study, due to interference from the palate bars of the stereotaxic instrument. Regardless, none of the 5 macaques vomited during the perioperative period, nor were secretions or foreign materials detected in the mouth at the end of the procedure.

Increased intracranial pressure is often a complication of intracranial surgery, and anesthetic drugs can exacerbate the hemodynamic changes in brain perfusion. We did not evaluate intracranial pressure directly, but the surgeons did not report any increase in bleeding or tissue edema during the procedure, and the anesthetist did not note any dysphoria during recovery. Dexmedetomidine reduces cerebral blood flow through α2-adrenoreceptor–mediated vasoconstriction, and atipamezole administration allowed rapid recovery from anesthesia, as has previously been reported after the administration of medetomidine in NHP.^17,21

The current study confirmed that the use of dexmedetomidine infusion in healthy cynomolgus macaques undergoing intracranial surgery provided stable sedation and adequate analgesia for neurosurgical procedures. Furthermore, our findings demonstrated that dexmedetomidine administered through constant-rate infusion to cynomolgus macaques as an adjunct to thiopental-maintained anesthesia provided good cardiovascular stability and reduced the dose of thiopental necessary for maintaining adequate sedation for neurosurgical procedures, compared with dosages previously reported.¹³ During the perioperative period, the macaques’ hemodynamic parameters remained in acceptable clinical ranges in the absence of cerebral bleeding or edema. Additional studies to determine the effects of dexmedetomidine on the cerebral blood flow and associated changes in the intracranial blood pressure of NHP are warranted.