Dexmedetomidine is a highly selective α2-adrenergic receptor agonist used in the ICU as a sedative and analgesic agent.1,2 It exerts its sedative and analgesic effects by stimulating α2-adrenergic receptors in the locus ceruleus and the spinal cord.2,3 The α2 : α1-receptor binding ratio for dexmedetomidine is approximately 1600 : 1, and the affinity for the α2-adrenergic receptor is eight times greater than that of clonidine, another common α2-adrenergic receptor agonist.1,2 An important clinical characteristic of dexmedetomidine is its weak depressant effect on respiratory drive,1,4,5 preventing the problems of decreased spontaneous ventilation in patients, whereas other agents used in the ICU, such as propofol and midazolam, may compromise safety during spontaneous ventilation.
Experimentally, there are conflicting results of studies on the effect of dexmedetomidine on respiration in vivo. Some authors have reported that it inhibits respiratory function,6,7 but it has been shown that high doses of dexmedetomidine increase resting ventilation.8 It is not surprising that there are conflicting results from studies in vivo because the respiratory system is complicated, consisting of several organs including the medulla oblongata and the carotid body.
The carotid body is a peripheral respiratory chemosensory organ which stimulates the respiratory drive in response to hypoxic stimulation.9,10 It has an important role in the respiratory system. Expression of α2-adrenergic receptors has been reported in the cat carotid body in vivo,11–13 and the α2-adrenergic receptor agonist guanabenz has been shown to depress the response of the carotid body to hypoxia in vivo.11 In addition, treatment with an α2-adrenergic receptor antagonist alone potentiated the chemoreceptor response to hypoxia in vivo.11,12 Clonidine showed inconsistent effects on carotid body activity, inhibitory or facilitative.13 To the best of our knowledge, the effects of dexmedetomidine on carotid body activity, especially the response to hypoxic stimulation, have not been reported in vitro.
We hypothesised that dexmedetomidine does not depress the response of the carotid body to low oxygen tension. The aim of the present study was to investigate the effect of dexmedetomidine on the activity of the carotid body and to confirm an α2-adrenergic receptor-specific action in vitro.
Animals and carotid body preparations
Ethical approval for this study (Ethical Committee Number 21–095) was provided by the Animal Ethics Committee of Nippon Medical School, Tokyo, Japan (Chairperson Professor Y. Sakuma) on 31 March 2009.
Experiments were performed on 10 carotid bodies removed surgically from anaesthetised male New Zealand white rabbits (body weight 2.19 ± 0.15 kg; n = 10; Saitama Experimental Animals Supply, Saitama, Japan). Six of the 10 preparations were perfused with only dexmedetomidine, and the remaining four preparations were used for dexmedetomidine in combination with yohimbine, an α2-adrenergic receptor antagonist. Anaesthesia was induced with thiamylal 25 to 50 mg administered via a 24-gauge catheter in the right marginal ear vein. A continuous infusion of thiamylal at a rate of 75 to 125 mg h−1 was adjusted thereafter to maintain adequate anaesthesia. No muscle relaxants were administered. A tracheostomy was performed after an anterior midline skin incision under local anaesthesia to the skin with lidocaine 1%. The lungs of the animals were ventilated mechanically with oxygen using an animal ventilator (KN-55; Natsume Co. Ltd, Tokyo, Japan).
The carotid sinus nerve was identified and carefully dissected free under a microscope. The carotid body was removed with the carotid arteries and the carotid sinus nerve en bloc after a bolus administration of heparin 1500 IU intravenously. Immediately after removal, the common carotid artery was flushed with 2 ml of modified Tyrode's solution (MTS) containing 120 mmol l−1 NaCl, 4.0 mmol l−1 KCl, 2.0 mmol l−1 CaCl2, 1.0 mmol l−1 MgCl2, 21.9 mmol l−1 NaHCO3, 1.9 mmol l−1 NaH2PO4 and 10.0 mmol l−1 D-glucose. The preparation was then placed into a perfusion chamber.
The chamber was perfused with MTS equilibrated with 95% O2 and 5% CO2 at a rate of 3.5 to 4.5 ml min−1 at a constant perfusion pressure of 45 cmH2O. The perfusate was applied to the bath and the carotid body preparation via a plastic tube attached to the common carotid artery (Fig. 1a). The temperature of the perfusate in the chamber was maintained at 37.5 ± 0.5°C with a regulated heating pump system (Heating Immersion Circulator MP; Julabo, Seelbach, Germany). The pH, pO2, pCO2 and K+ concentrations of the perfusate in the chamber were determined using a blood gas analyser (Stat Profile pHOx plus; Nova Biomedical, Waltham, Massachusetts, USA).
Carotid sinus nerve recording
In the perfusion chamber, the carotid sinus nerve was carefully cleared from the surrounding tissues and placed on to a silver electrode under the microscope. The nerve was covered with mineral oil for electrical isolation and to prevent it from drying out. Carotid sinus nerve activity was recorded from the whole carotid sinus nerve. A reference electrode was placed into the chamber. The nerve activity was amplified and filtered (0.1 to 10 kHz) with a differential alternating current amplifier (ER-1 differential extracellular amplifier; Cygnus Technology, Delaware Water Gap, Pennsylvania, USA). The nerve activity was recorded on a computer using a Power Lab 30 interface and Chart 5 software (AD Instruments, Colorado Springs, Colorado, USA). This method has been used in previous similar experiments to evaluate carotid body chemosensitivity and provides stable experimental conditions without damage to the preparation.14–16
Prior to starting recording, the preparation was perfused with MTS equilibrated with 95% O2 and 5% CO2 (high oxygen tension) for 20 to 30 min for stabilisation. During the following 3 min, the baseline action potentials of the whole sinus nerve were confirmed at high oxygen tension (first high oxygen tension). The preparation was then perfused with MTS equilibrated with 95% N2 and 5% CO2 (low oxygen tension) for 7 min to obtain the stimulated peak action potentials. Thereafter, the preparation was perfused with MTS equilibrated with 95% O2 and 5% CO2 for 20 min (second high oxygen tension). This 30 min procedure comprised one complete cycle, and the action potentials were obtained during four cycles for each preparation. The course of four cycles consisted of one control and three test sets.
Evaluation of effects of dexmedetomidine on the carotid body
In the three test sets, the preparation was perfused with MTS containing three different concentrations of dexmedetomidine (0.1, 1.0 or 10 nmol l−1; Hospira Japan, Osaka, Japan) in randomised order for each preparation at the first high oxygen tension for the first 3 min and at low oxygen tension for the following 7 min. The preparation was perfused with MTS without dexmedetomidine at the second high oxygen tension for 20 min to wash out the system and stabilise signals during each test set (Fig. 1b).
Receptor activity confirmation with yohimbine
We prepared four preparations for this protocol. The control preparations were perfused in the same manner as described in the protocol for dexmedetomidine administration. For the three test sets, the preparation was perfused with MTS containing 1.0 nmol l−1 yohimbine (Sigma-Aldrich, St Louis, Missouri, USA) for 30 min throughout the entire first test set. The preparation was then perfused with MTS containing a mixture of 0.1 nmol l−1 dexmedetomidine and 1.0 nmol l−1 yohimbine for the first 3 min and at low oxygen tension for the following 7 min. Next, the preparation was perfused with pure MTS at high oxygen tension for 20 min to wash out the system. Finally, the preparation was perfused with MTS containing 1.0 nmol l−1 yohimbine for 30 min in the same manner as the first test set (Fig. 1c).
After all tests, the viability of the preparation was confirmed. During the procedure, 200 to 250 μl of MTS in the chamber were sampled at the first high oxygen tension, low oxygen tension and at the second high oxygen tension to determine the pH, pO2, pCO2 and K+.
The action potentials sampled from the whole carotid sinus nerve were discriminated from background noise by offline analysis with Chart 5 software based on amplitude threshold. The threshold was adjusted using the baseline action potentials during the control period and the same threshold was applied to all action potentials throughout the analysis with the same preparation. After noise reduction, the frequencies of action potentials were determined as activity of the carotid sinus nerve. The frequencies derived from the last 1 min in each first high oxygen tension period were regarded as the baseline activity. The peak activity at low oxygen tension was also determined, but was derived from 10 s of the highest frequencies during each low oxygen tension period. The changes in activity throughout the procedure (the difference in activity obtained by subtracting the activity at baseline from that at peak in each set) were evaluated statistically.
Two-way analysis of variance (ANOVA) for repeated measurements was used to compare the baseline and peak activities in each set throughout the experimental procedure. Post-hoc analysis was performed using the Bonferroni test. A Friedman test following Dunn's multiple comparison test was used to compare the difference between baseline and peak activity in each set throughout the experimental procedure. Data are presented as mean ± SEM. The gas and electrolyte data were analysed by two-way ANOVA for repeated measurements. A P value of less than 0.05 was considered statistically significant. Statistical analyses were performed with GraphPad Prism version 5.0b for Mac (GraphPad Software Inc., San Diego, California, USA).
The carotid body preparation showed stable action potentials at high oxygen tension (baseline activity) and increasing action potentials at low oxygen tension (peak activity). Typical action potentials are shown in Fig. 2a. This increase typically peaked 2 to 3 min after initiation of low oxygen tension stimulation (Fig. 2b). For the evaluation of the effects of dexmedetomidine, one preparation was excluded from analysis because it showed poor viability (n = 5). For the receptor activity confirmation with yohimbine (n = 4), all four preparations showed good viability. There was a significant change in activity between baseline and peak throughout the experimental procedure (Fig. 3a and b; P < 0.01). The increased action potentials during low oxygen tension returned to the baseline level immediately during the second high oxygen tension drug washout period. The pO2 was significantly reduced during the low oxygen tension period throughout this study, and then returned to the first high oxygen tension values during the second high oxygen tension period (Tables 1 and 2; P < 0.01).
Response of the carotid body to low oxygen tension during dexmedetomidine administration
At each set used for the evaluation of the effects of dexmedetomidine, carotid sinus nerve activity was significantly increased by stimulation by low oxygen tension from 203 ± 81 to 343 ± 89 Hz (control), from 244 ± 54 to 510 ± 128 Hz (0.1 nmol l−1), from 219 ± 58 to 435 ± 135 Hz (1.0 nmol l−1) and from 204 ± 57 to 427 ± 95 Hz (10 nmol l−1) (Fig. 3a; P < 0.01 for all). The difference in activity between baseline and peak at 0.1 nmol l−1 dexmedetomidine was significantly higher than that of control (Fig. 4a), with values of 140 ± 70 (control), 266 ± 116 (0.1 nmol l−1, P < 0.05), 215 ± 133 (1.0 nmol l−1, P > 0.05) and 223 ± 100 Hz (10 nmol l−1, P > 0.05).
Response of the carotid body to low oxygen tension during dexmedetomidine and yohimbine administration
At each set, carotid sinus nerve activity was significantly increased by stimulation by low oxygen tension (Fig. 3b; P < 0.01 for all comparisons). Yohimbine alone did not change the baseline and peak activity compared with control. There was no significant change in the difference in activity between baseline and peak (Fig. 4b), with values of 159 ± 50 (control), 166 ± 69 (yohimbine alone, P > 0.05), 136 ± 56 (dexmedetomidine with yohimbine, P > 0.05) and 143 ± 52 Hz (yohimbine alone, P > 0.05). This means that 1 nmol l−1 yohimbine blocked the facilitative effect of 0.1 nmol l−1 dexmedetomidine on the carotid body response to low oxygen tension.
The values of pH and pCO2 were in the range of 7.45 to 7.48 (mean) and 4.00 to 4.50 kPa (mean), respectively, throughout the experiments evaluating the effects of dexmedetomidine (Table 1) and in the range of 7.44 to 7.48 (mean) and 4.48 to 4.95 kPa (mean), respectively, during the receptor activity confirmation with yohimbine (Table 2). There were no significant changes in either parameter, indicating that there was no effect on the response of the carotid body. K+ concentrations in the chamber were also stable throughout the experiments (Tables 1 and 2).
Dexmedetomidine is a highly selective α2-adrenergic receptor agonist providing sedative and analgesic effects without respiratory depression in the clinical setting.17,18 In contrast, other hypnotic agents such as propofol and midazolam generally depress the respiratory system. It is known that these agents inhibit the carotid body response to low oxygen tension stimulation in vitro.14–16 In the present study, we demonstrated that dexmedetomidine does not depress the activity of the carotid body under conditions of high oxygen tension nor depress the carotid body response to low oxygen tension stimulation, whereas 0.1 nmol l−1 dexmedetomidine facilitates the low oxygen tension response in vitro.
Controversial effects of α2-adrenergic receptor agonists
An α2-adrenergic receptor agonist has been reported to inhibit the carotid body chemoreceptor response to hypoxic stimulation, whereas α2-adrenergic receptor antagonists potentiate this response.11–13 The α2-adrenergic receptor agonist guanabenz has been shown to decrease the baseline activity of the carotid body during ventilation with room air in vivo in a dose-dependent manner in cats. This effect was blocked by the α2-adrenergic receptor antagonist SKF-86466.11,12
However, clonidine, another α2-adrenergic receptor agonist, elicits variable effects on the carotid body chemosensory discharge, with either inhibitory or slightly facilitatory responses in vivo.13 After blocking α2-adrenergic receptors with SKF-86466, clonidine stimulated the chemosensory discharge during hypoxia.13 It was suggested that facilitation of carotid body activity with α2-adrenergic receptor agonists might occur via I1-imidazoline receptor activation, rather than via α2-adrenergic receptors.13 It seems that activation of α2-adrenergic receptors results in inhibition of the carotid body response, and this is inconsistent with the findings of the present study.
The majority of these studies were performed in vivo. The respiratory system is complicated, consisting of several organs including the medulla oblongata and the carotid body. Respiratory cycles are controlled not only by the carotid body but also the medulla oblongata and central respiratory chemosensory organ in vivo. It has been shown that clonidine, an α2-adrenergic receptor agonist, affects the medulla oblongata.19 Thus, it is difficult to evaluate where the agent is acting to produce its effect. The agents affects many sites and what we can observe is the result of cross-interaction in in vivo studies, even if these experimental designs were well organised. However, experimental designs can be simplified to test one organ in vitro. Thus, it is convincing that different findings were obtained in vivo and in vitro. In addition, the nicotinic acetylcholine receptor (nACh) receptor should be considered as an important factor for the confounding effects. This will be discussed below.
Effector site of dexmedetomidine
We found that the facilitative effect of 0.1 nmol l−1 dexmedetomidine on the carotid body response to low oxygen tension stimulation was blocked after administration of 1 nmol l−1 yohimbine, an α2-adrenergic antagonist, indicating that it was mediated via α2-adrenergic receptor activation. This is rational, but another effector site, such as the nACh receptor, cannot not be excluded as a candidate for a mediating receptor of dexmedetomidine. It is known that the nACh receptor on the carotid body, not the γ-aminobutyric acid A (GABAA) receptor, mediates effects of midazolam and propofol which are GABAA receptor agonists.14,16 Like these agents, dexmedetomidine might be mediated via the nACh receptor which has an important role at the carotid body, but this was not evaluated in the present study.
Effects of drugs on nACh receptors might be responsible for the differences between previous studies and the present study. Activation of the carotid body is mediated via nACh receptors20,21 and is blocked by neuromuscular relaxants.22,23 In in vivo studies, neuromuscular relaxants are often used and it is possible that activity of the carotid body is modified. No neuromuscular relaxant was administered in the present study. Findings of in vivo and in vitro experiments should be evaluated carefully based on the use of neuromuscular relaxants.
Differences of species and agents
Most studies of the carotid body have been performed in cats,11–13 rats20 and rabbits.14–16 Various agents have been tested in various settings both in vivo and in vitro, but no differences between species have been identified.
Guanabenz and clonidine were used as α2-adrenergic receptor agonists in previous studies.11–13 One of the differences between these agents and dexmedetomidine is the α2 : α1-receptor binding ratio. The ratio for dexmedetomidine is approximately 1600 : 1, and the affinity for the α2-adrenergic receptor is eight times greater than that of clonidine. Expression of α2-adrenergic receptors in the cat carotid body has been demonstrated in vivo.11–13 It has also been suggested that facilitation of carotid body activity with clonidine might occur via I1-imidazoline receptor activation rather than via α2-adrenergic receptors.13 Therefore, dexmedetomidine might act on α2-adrenergic receptors more specifically than clonidine (or guanabenz) to yield the different effects on the carotid body response, rather than acting on I1-imidazoline receptors. The manner in which α2-adrenergic receptors and I1-imidazoline receptors modify the response of the carotid body should be investigated.
Potentiating or attenuating effects of dexmedetomidine were not observed at concentrations of 1.0 and 10 nmol l−1. It seemed that there was a dose-dependent effect (Fig. 3a), but no statistical significance was found. Dexmedetomidine 0.01 to 0.05 mmol l−1 has a dose-dependent inhibitory effect on action potential conduction of sciatic nerves in frogs in vitro.24 Although these concentrations are extremely high compared with those in the present study, it suggests that dexmedetomidine might attenuate the conduction of the carotid sinus nerve in this study. Therefore, the potential facilitating effects of 1.0 and 10 nmol l−1 dexmedetomidine on the carotid body response might be cancelled by decreased conduction of its afferent nerve from which we obtained the action potentials. This is the reason that only 0.1 nmol l−1 dexmedetomidine was tested with yohimbine. It is difficult to evaluate the effect of yohimbine if conduction of the afferent nerve decreases.
Concentration of dexmedetomidine
The three different concentrations of dexmedetomidine (0.1, 1.0 and 10 nmol l−1) used in the present study were low compared with previous studies. The doses were determined based on the effective plasma concentrations of dexmedetomidine for sedation in humans (0.7 to 1.9 ng ml−1)4 and the protein-binding ratio of dexmedetomidine in human plasma (94%).25 Thus, we speculated that in vitro concentrations of 0.18 to 0.48 nmol l−1 were equivalent to the effective concentrations in human plasma.
Effects of pH, carbon dioxide partial pressure, oxygen partial pressure and potassium ion
There was a significant difference in pO2 between baseline and peak, whereas pH, pCO2 and K+ were stable throughout the experimental procedure (Tables 1 and 2). Thus, apart from the test drugs, only pO2 may have influenced the results of our study. The levels of high and low oxygen tension were comparable to those in previous studies.14–16
We demonstrated that dexmedetomidine does not depress the activity of the carotid body under high oxygen tension conditions nor the response to low oxygen tension stimulation, whereas 0.1 nmol l−1 dexmedetomidine facilitates the response via α2-adrenergic receptor activation.
Assistance with the study: none declared.
Financial support and sponsorship: none declared.
Conflicts of interest: none declared.
1. Carollo DS, Nossaman BD, Ramadhyani U. Dexmedetomidine
: a review of clinical applications. Curr Opin Anaesthesiol
2. Coursin DB, Coursin DB, Maccioli GA. Dexmedetomidine
. Curr Opin Crit Care
3. Kamibayashi T, Maze M. Clinical uses of α2
-adrenergic agonists. Anesthesiology
4. Ebert TJ, Hall JE, Barney JA, et al. The effects of increasing plasma concentrations of dexmedetomidine
in humans. Anesthesiology
5. Venn RM, Hell J, Grounds RM. Respiratory effects of dexmedetomidine
in the surgical patient requiring intensive care. Crit Care
6. Nishida T, Nishimura M, Kagawa K, et al. The effects of dexmedetomidine
on the ventilatory response to hypercapnia in rabbits. Intensive Care Med
7. Fernandes FC, Ferreira HC, Cagido VR, et al. Effects of dexmedetomidine
on respiratory mechanics and control of breathing in normal rats. Respir Physiol Neurobiol
8. Nguyen D, Abdul-Rasool I, Ward D, et al. Ventilatory effects of dexmedetomidine
, atipamezole, and isoflurane in dogs. Anesthesiology
9. Prabhakar NR. O2
sensing at the mammalian carotid body
: why multiple O2
sensors and multiple transmitters? Exp Physiol
10. Prabhakar NR, Overholt JL. Cellular mechanisms of oxygen sensing at the carotid body
: heme proteins and ion channels. Respir Physiol
11. Kou YR, Ernsberger P, Cragg PA, et al. Role of alpha 2-adrenergic receptors in the carotid body
response to isocapnic hypoxia. Respir Physiol
12. Prabhakar NR, Kou YR. Inhibitory sympathetic action on the carotid body
responses to sustained hypoxia. Respir Physiol
13. Ernsberger P, Kou YR, Prabhakar NR. Carotid body
I1-imidazoline receptors: binding, visualization and modulatory function. Respir Physiol
14. Jonsson MM, Lindahl SG, Eriksson LI. Effect of propofol on carotid body
chemosensitivity and cholinergic chemotransduction. Anesthesiology
15. Akada S, Fagerlund MJ, Lindahl SG, et al. Pronounced depression by propofol on carotid body
response to CO2 and K+-induced carotid body
activation. Respir Physiol Neurobiol
16. Kim C, Shvarev Y, Takeda S, et al. Midazolam depresses carotid body
chemoreceptor activity. Acta Anaesthesiol Scand
17. Belleville JP, Ward DS, Bloor BC, Maze M. Effects of intravenous dexmedetomidine
in humans. I: Sedation, ventilation, and metabolic rate. Anesthesiology
18. Hsu YW, Cortinez LI, Robertson KM, et al. Dexmedetomidine
pharmacodynamics. Part I: Crossover comparison of the respiratory effects of dexmedetomidine
and remifentanil in healthy volunteers. Anesthesiology
19. Rives JE, Bernard DG. α2
Adrenergic receptors and the central control of breathing in the cane toad, Bufo marinus. Respir Physiol
20. Nurse CA, Zhang M. Acetylcholine contributes to hypoxic chemotransmission in co-cultures of rat type 1 cells and petrosal neurons. Respir Physiol
21. Varas R, Alcayaga J, Iturriaga R. ACh and ATP mediate excitatory transmission in cat carotid identified chemoreceptor units in vitro. Brain Res
22. Jonsson M, Kim C, Yamamoto Y, et al. Atracurium and vecuronium block nicotine-induced carotid body
chemoreceptor responses. Acta Anaestheiol Scand
23. Igarashi A, Amagasa S, Horikawa H, Shirahata M. Vecuronium directly inhibits hypoxic neurotransmission of the rat carotid body
. Anesth Analg
24. Kosugi T, Mizuta K, Fujita T, et al. High concentrations of dexmedetomidine
inhibit compound action potentials in frog sciatic nerves without α2
adrenoceptor activation. Br J Pharmacol
25. Talke P, Richardson CA, Scheinin M, Fisher DM. Postoperative pharmacokinetics and sympatholytic effects of dexmedetomidine
. Anesth Analg