| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Departments of *Anesthesiology and Hygiene and Space Medicine, Nihon University School of Medicine, Tokyo, Japan.
Address correspondence and reprint requests to Ryoji Iida, MD, PhD, Department of Anesthesiology, Nihon University School of Medicine, 30-1, Oyaguchi-Kamicho, Itabashi-Ku, Tokyo 173-8610, Japan. Address e-mail to ryoiida{at}med.nihon-u.ac.jp.
Abstract
BACKGROUND: Although intrathecal midazolam has been reported to produce antinociceptive effects mediated by 
-aminobutyric acid type A-benzodiazepine receptor complexes in the spinal cord, the effects of systemic midazolam on nociception remain unclear. We performed this study to examine the effects of IV-administered midazolam on somatosympathetic A
and C reflex discharges in brain-intact cats and decerebrate cats (with transection at midbrain level).
METHODS: Somatosympathetic Aand C reflexes were elicited in the inferior cardiac sympathetic nerve by electrical stimulation of myelinated (A) and unmyelinated (C) afferent fibers of the superficial peroneal nerve in 28 mature cats. After control somatosympathetic reflex responses were obtained, midazolam was administered IV to four groups of randomly allocated cats as follows: brain-intact cats at a dose of 0.03 mg/kg, brain-intact cats at a dose of 0.1 mg/kg, brain-intact cats at a dose of 0.5 mg/kg, and decerebrate cats at a dose of 0.1 mg/kg.
RESULTS: C reflex discharges were significantly augmented at the dose of 0.03 mg/kg and significantly depressed at the dose of 0.1 and 0.5 mg/kg in brain-intact cats. C reflex discharges were also significantly depressed at the dose of 0.1 mg/kg in decerebrate cats.
CONCLUSIONS: We have demonstrated that IV midazolam produces dose-related effects on somatosympathetic reflex discharges. The clinical implication of these findings is that the effect of midazolam on nociception depends on its dosage. It also appears that the infra-midbrain region plays a major role in mediating the depressive effects of midazolam on somatosympathetic C reflex discharges.
Systemically administered midazolam has been widely used for anesthetic premedication and as an adjuvant to anesthesia. Intrathecal midazolam has been reported to produce antinociceptive effects mediated by -aminobutyric acid type A (GABAA)-benzodiazepine receptor complexes in the spinal cord (1–6). However, the effects of systemic midazolam on nociception remain unclear. Although there have been reports indicating its antinociceptive effects (3–5,7,8), other findings have indicated hyperalgesic effects (9,10) or no effects on nociception (11–13).
In anesthetized animals, electrical stimulation of afferent somatic nerves has been demonstrated to elicit reflex responses in both cardiac and renal sympathetic nerves, i.e., a short-latency (approximately 80 ms) A reflex and a long-latency (approximately 400 ms) C reflex (14). These somatosympathetic reflexes (SSR) are induced by stimulation of afferent myelinated (A) and unmyelinated (C) fibers, respectively, ascending through the spinoreticular tract, and are mediated by neurons of the rostral ventrolateral medulla (RVLM) in the reticular formation (15,16). The C reflex discharge has been reported to be selectively depressed by IV administration of opiates in anesthetized animals, and can be considered a useful indicator for evaluating the antinociceptive properties of drugs (17,18), because C afferent fibers constitute a major proportion of the nociceptive afferent fibers. We have described the effects of analgesic drugs on SSR discharges (19,20).
The major goals of anesthesia include blunting or attenuation of adrenergic responses to surgical stimuli and providing lack of consciousness and movement in patients during surgery. In the last decade, it has been demonstrated that perioperative adrenergic ß-blockade reduces cardiac morbidity and mortality (21–23). The American College of Cardiology/American Heart Association has recommended that heart rate (HR) should be maintained at <80 bpm with the use of ß-blockers during the intra- and postoperative periods in surgical patients at increased risk for adverse cardiac events (24). Anesthesiologists are required to attenuate exaggerated sympathetic responses, which are associated with substantial increases in HR during surgery. It has thus become increasingly important to determine how anesthetic drugs affect SSR discharges.
The present experiments were performed to examine the effects of IV-administered midazolam on SSR discharges in brain-intact cats and decerebrate cats (with transection at midbrain level). Midbrain transection enabled evaluation of the effects of midazolam on SSR discharges in the absence of supra-midbrain effects (20).
METHODS
Animal Preparation
This study was approved by the Animal Care and Use Committee of Nihon University School of Medicine. All the procedures in this study were performed in adherence to the guidelines supplied by the committee. Experiments were performed on 28 mature cats of either sex ranging in weight from 2.1 to 4.3 kg (Table 1), which were initially premedicated with IM ketamine hydrochloride (35 mg/kg) 30 min before the experimental procedure. The right radial cutaneous vein was cannulated for administration of fluid (lactated Ringer’s solution) and drugs. Subsequent surgical procedures were performed under infiltrative anesthesia with 0.5% lidocaine. Deep anesthesia was maintained with IV administration of 
-chloralose and urethane (total doses of 50 mg/kg and 100 mg/kg, respectively) throughout the experiments. The cats were paralyzed with IV administration of pancuronium bromide (total dose 3 mg/kg) and artificially ventilated with 50% oxygen-enriched air via a tracheal cannula. ETco2 was maintained at approximately 3% using a CO2 gas analyzer (1H21, NEC, Japan) through adjustment of the ventilation rate and/or volume control. HR was monitored with the electrocardiogram (2F52, NEC, Japan). Mean arterial blood pressure (MAP) was recorded via a pressure transducer (DPT-6003, Kawasumi, Japan) through a catheter inserted into the right common carotid artery. These traces were recorded with a thermal pen writing recorder (8K21-1-L, NEC, Japan). Esophageal temperature was maintained at 37.0°C–38.0°C with a heating pad and lamp. The animals were vagotomized bilaterally at the cervical level to prevent vagal input from affecting cardiac sympathetic nerve activity. The left inferior cardiac sympathetic nerve was exposed retropleurally after removal of the head and dorsal portions of the second, third, and fourth ribs, and severed as close to the heart as possible. The left superficial peroneal nerve was isolated and severed at the distal end. All the nerves prepared were placed on bipolar platinum electrodes, and immersed in warm paraffin oil throughout the experiments to preserve them and insulate them from surrounding tissue. SSR discharges were elicited by electrical stimulation of the left superficial peroneal nerve and recorded from the inferior cardiac sympathetic nerve. To elicit SSR discharges, electrical train pulse stimulations (four pulses of 0.5 ms duration at 20 Hz) were applied to the superficial peroneal nerve 16 times every 3.33 s at supramaximal intensities (20–50 V) by an electric stimulator (MS-25 System, Medelec, UK). SSR discharges were recorded from the inferior cardiac sympathetic nerve through a preamplifier (MS-25 System) with time constant set at 0.15 s. Sixteen consecutive reflex responses were averaged using an averaging instrument (MS-25 System). The averaged responses were displayed on a screen and recorded on a thermal recorder (MS-25 System). The magnitudes of SSR discharges were estimated by measuring the peak amplitudes and expressed as % of control responses.
|
For decerebration, the head of each animal was mounted on a stereotaxic frame in the prone position and fixed with ear bars after the intact carotid artery had been ligated to reduce blood loss during decerebration. The midbrain was transected at the midcollicular level with a blunt spatula through a hole bored in the occipital bone. The accuracy of sectioning was confirmed after completion of the experiment.
Experimental Protocol
After a stabilization period of at least 30 min, control SSR responses were obtained (Table 1). After control recording, midazolam was administered IV to four groups of randomly allocated cats as follows:
In all groups, A reflex and C reflex responses were elicited and measured at 1, 5, 10, 20, 30, 40, 50, and 60 min after IV administration of midazolam. At 65 min after IV administration of midazolam, flumazenil was administered IV at a fixed dose of 0.2 mg. SSR responses were again elicited and measured at 1, 5, 10, 20, and 30 min after IV administration of flumazenil. Representative reflex responses are shown in Figure 1. Resting HR and MAP were measured immediately before each electrical stimulation.
|
Statistical Analysis
One-way analysis of variance was used for statistical analysis of the time course of effects of midazolam (until IV administration of flumazenil), followed subsequently by the Dunnett test for comparison of the values at each time point with the control. Comparisons of amplitudes of SSR discharges at nadir depression between brain-intact cats and decerebrate cats were performed with the unpaired t-test. P values <0.05 were considered significant. All values are presented as the mean ± sem.
RESULTS
The effects of IV-administered midazolam on SSR discharges are shown in Figure 2. Time courses of HR and MAP after the administration of midazolam are shown in Figure 3. Administration of 0.03 mg/kg of midazolam to brain-intact cats significantly augmented C reflex discharges from 10 to 40 min after administration, whereas A reflex discharges remained near the control level. Peak augmentation was 125.3% ± 13.0% of the control value at 20 min after administration. Administration of 0.1 mg/kg to brain-intact cats significantly depressed C reflex discharges from 5 to 10 min later, whereas effects on A reflex discharges were not significant. The nadir value of C reflex discharges was 81.7% ± 3.9% of the control at 10 min after administration. Administration of 0.1 mg/kg to decerebrate cats significantly depressed A reflex discharges from 1 to 60 min later, with recovery after IV administration of flumazenil. Administration of 0.1 mg/kg to decerebrate cats significantly depressed C reflex discharges from 1 to 40 min later. Nadir values were 43.8% ± 7.4% of the control at 10 min for A reflex discharges and 57.3% ± 8.7% of the control at 10 min for C reflex discharges. The magnitudes of depression of the C reflex discharge at 10 min were significantly greater in decerebrate cats than in the brain-intact cats (P = 0.034). Administration of 0.5 mg/kg to brain-intact cats significantly depressed A and C reflex discharges from 1 to 60 min later, with recovery after IV administration of flumazenil. Nadir values were 29.5% ± 5.1% of the control at 20 min for A reflex discharges and 22.6% ± 3.3% of the control at 30 min for C reflex discharges.
|
|
DISCUSSION
The precise effects of systemic administration of midazolam on nociception are unclear. Antinociceptive effects (3–5,7,8), hyperalgesic effects (9,10), and lack of effects on nociception (11–13) have been reported. In the present study, we found that IV administration of midazolam produced dose-related effects on SSR discharges in anesthetized cats.
Administration of 0.1 mg/kg and 0.5 mg/kg significantly depressed C reflex discharges in brain-intact cats. In particular, the dose of 0.1 mg/kg significantly depressed C reflex discharges but not A reflex discharges. These observations suggest that midazolam has antinociceptive effects at these doses. In the group treated with 0.5 mg/kg, C reflex discharges recovered rapidly and acutely after IV administration of flumazenil. Flumazenil competitively inhibits the binding of benzodiazepine to the GABAA receptor complexes and does not act on peripheral benzodiazepine receptors, which are not associated with the GABA system (25–27). In a preliminary study, flumazenil alone at the dosage we used was found not to affect SSR discharges (unpublished data). It thus appears that GABAA receptor complexes are involved in mediating the depressive effects of midazolam on C reflex discharges.
Depression of SSR discharges by midazolam was still apparent in decerebrate cats, suggesting that the infra-midbrain region, including the RVLM and the spinal cord, is a major site of action for the antinociceptive effects of midazolam. The magnitudes of depression of C reflex discharge were significantly greater in decerebrate cats than in brain-intact cats. It has been suggested that there might be tonic descending inhibitory modulation from the supra-midbrain region on the pathway of the SSR (20). A high density of specific benzodiazepine binding sites has been noted in the supra-midbrain region (28). It is possible that midazolam acts on the supra-midbrain region, inhibits descending inhibitory modulation, and thus attenuates the depressive effects on SSR discharges. Control amplitudes of SSR appeared to be greater in decerebrate cats than in intact cats (Table 1). This finding is consistent with the suggestion of Ogawa et al. (20) that there might be tonic descending inhibitory modulation from the supra-midbrain region on the pathway of the SSR.
The present study did not determine the precise site of midazolam-mediated depression of SSR discharges. Numerous investigations have suggested that the antinociceptive effects of benzodiazepines are modulated in the dorsal horn of the spinal cord (3–5,7). However, endogenous GABA is involved in mediating the tonic inhibition of vasopressor neurons in the RVLM (29). It has been suggested that vasopressor responses of RVLM to nociception are inhibited by GABAA receptor binding (30,31). Specific benzodiazepine binding sites have also been identified in the intermediolateral nuclei of the spinal cord, which include sympathetic preganglionic neurons (32). These sites could not be excluded as possible sites of midazolam-mediated depression of SSR discharges.
In contrast, C reflex discharges were augmented by administration of 0.03 mg/kg in brain-intact cats. This finding suggests that, at this dose, midazolam reduces the nociceptive threshold. Augmentation of C reflex discharges at this dose of midazolam is consistent with the findings on the viscero-somatic reflex obtained by Crawford et al. (5). They demonstrated that, in rabbits, a small dose of midazolam (62.5 µg/kg) injected IM could produce hyperalgesic effects whereas a large dose (250 µg/kg) could produce antinociceptive effects (5). Studies based on hotplate and/or tail-flick tests have suggested that midazolam induces antinociceptive effects at the spinal level and hyperalgesic effects at the supraspinal level (10,33,34). Hence, the effects of midazolam on nociception appear to depend on its site of action and its dosage. In addition, it should be noted that our findings parallel those for barbiturates, in which spinothalamic tract neurons exhibited increased responses to peripheral C-fiber volleys and decreased responses to A-fiber volleys after small doses of pentobarbital (35).
There are a few limitations to this study. First, it is difficult with SSR experiments to compare absolute SSR discharges among groups, because the absolute amplitude of SSR discharges obtained exhibits a very large range of individual difference. SSR discharges were therefore computed as percentages of the control under conditions in which they were clearly elicited by supramaximal stimulation and stable control values had been confirmed. Second, the cats were premedicated with IM ketamine, which has antinociceptive properties. However, we have already examined the effects of ketamine on SSR discharges (36), and have demonstrated that the significant depressive effects of IV ketamine (10 mg/kg) on SSR discharges occur within 10 min in brain-intact cats and within 20 min in decerebrate cats, with spontaneous recovery of all averaged values to the control level within 60 min (36). Moreover, stable control values of SSR discharges were repeatedly confirmed before administration of midazolam in the present study. We therefore believe that ketamine had little effect on our results. Third, the animals had been vagotomized bilaterally at the cervical level to prevent vagal input from influencing cardiac sympathetic nerve activity in this study. The cardiac vagal nerves make a small contribution to the cardiac SSR in cats (37). It has been reported that stimulation of the superficial peroneal nerve at strengths sufficient to excite both A and C fibers gives rise to a brief period of bradycardia of small magnitude followed by more pronounced tachycardia in cats with intact vagal nerves, whereas cutting both vagal nerves results in the complete disappearance of bradycardia (37). This bradycardic response was thus because of the activation of vagal efferents to the heart. In clinical settings in which the vagal nerves are intact, SSR discharges might be somewhat suppressed.
In conclusion, our findings demonstrated dose-related effects of IV-administered midazolam on the SSR discharges in anesthetized cats. The clinical implication of these findings is that the effect of midazolam on nociception depends on its dosage. It also appears that the infra-midbrain region plays a major role in mediating the depressive effects of midazolam on the somatosympathetic C reflex discharges.
Footnotes
Accepted for publication May 14, 2007.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Int/0.03 mg: administration of 0.03 mg/kg to brain-intact cats. • Int/0.1 mg: administration of 0.1 mg/kg to brain-intact cats. 
Dec/0.1 mg: administration of 0.1 mg/kg to decerebrate cats. 
Int/0.5 mg: administration of 0.5 mg/kg to brain-intact cats. Flumazenil was administered IV 65 min after midazolam administration. Results are mean values ± sem for each data point (n = 7). Symbols (*) indicate significant difference (P < 0.05) from the control as determined by the Dunnett test.
