Anesth Analg 2003;96:159-162
© 2003 International Anesthesia Research Society
ANESTHETIC PHARMACOLOGY
Serum and Cerebrospinal Fluid Concentrations of Midazolam After Epidural Administration in Dogs
Tomoki Nishiyama, MD, PhD,
Hisayoshi Tamai, MD, and
Kazuo Hanaoka, MD, PhD
Department of Anesthesiology, The University of Tokyo, Japan
Address correspondence and reprint requests to Tomoki Nishiyama, MD, PhD, 3-2-6-203, Kawaguchi, Kawaguchi-shi, Saitama 332-0015, Japan. Address e-mail to nishiyamat-ane{at}h.u-tokyo.ac.jp
 |
Abstract
|
|---|
The epidural administration of midazolam has analgesic effects that might be mediated by 
-aminobutyric acid type A receptors in the spinal cord. In this study, we examined both serum and cerebrospinal fluid (CSF) concentrations of midazolam after epidural administration to investigate the possibility of midazolam entering CSF directly from the epidural space. Five male mongrel dogs had catheters inserted in a femoral artery, the epidural space at L3-4, and the intrathecal space at the atlanto-occipital region under general anesthesia. Midazolam 1 mg/kg was epidurally administered, and arterial blood and CSF samples were collected until 240 min after the midazolam administration to measure midazolam concentration. Serum midazolam concentration increased and reached a peak at 30 min after the administration (224.8 ± 30.5 ng/mL) and then decreased to 25.8 ± 6.0 ng/mL at 240 min. Midazolam concentration in the CSF was less than the detection limit at 5 min, reached a peak at 30 min after the administration (7.2 ± 4.7 ng/mL), and decreased to 3.6 ± 3.3 ng/mL at 240 min. In conclusion, epidurally administered midazolam enters CSF, but CSF concentrations are only 3% of those in the systemic circulation.
IMPLICATIONS: Midazolam, which has spinally mediated analgesic potency, was epidurally administered in dogs, and serum and cerebrospinal fluid concentrations were measured. Epidurally administered midazolam enters the cerebrospinal fluid, but concentrations are only 3% of those in the systemic circulation.
|
Introduction
|
|---|
Midazolam, a water-soluble benzodiazepine, has analgesic properties mediated via the -aminobutyric acid (GABA)A receptor in the spinal cord (1). Intrathecally administered midazolam is effective on thermal or inflammatory-induced nociception in animals (2). Clinically, epidural midazolam produces analgesia in patients with postoperative wound pain (3–6). Serum concentrations of midazolam after an epidural administration were smaller than those producing sedative effects in humans (3). We hypothesized that epidurally administered midazolam enters cerebrospinal fluid (CSF) to exert spinally mediated analgesic effects and sedation in the brain. In this study, we examined both serum and CSF concentrations of midazolam after an epidural administration to investigate the possibility of midazolam entering the CSF directly from the epidural space.
|
Materials and Methods
|
|---|
After obtaining the approval of the Research Committee of the University of Tokyo, five male mongrel dogs weighing 8–12 kg (Nippon Bio-Supply, Tokyo, Japan) were studied. Anesthesia was induced with IM ketamine 20 mg/kg. An 18-gauge catheter was inserted into an antecubital vein, and lactated Ringer’s solution was administered at a rate of 10 mL · kg-1 · h-1 throughout the study. Thiopental 5 mg/kg and vecuronium 2 mg/kg were IV administered, and the trachea was intubated. Anesthesia was maintained with isoflurane 2% with 40% oxygen in air. Mechanical ventilation was adjusted to maintain end-tidal carbon dioxide tension between 30 and 35 mm Hg. Rectal temperature was monitored and kept at 37.0°C ± 0.6°C with a warming blanket and a heating lamp. A femoral artery was cannulated to measure blood pressure and heart rate and to collect blood samples. A Tuohy needle was introduced into the epidural space at L3-4, and a 20-gauge epidural catheter was inserted 5 cm cephalad. A 19-gauge catheter was inserted into the intrathecal space in the atlanto-occipital region to collect CSF.
Midazolam 1 mg/kg, diluted by saline to a total volume of 3 mL, was epidurally administered followed by 1 mL of saline to flush the catheter. Arterial blood and CSF samples (each 2 mL) were drawn before and at 5, 10, 30, 60, 90, 120, 180, and 240 min after the midazolam administration. The samples were centrifuged for 10 min at 3000g, and serum and supernatant of the CSF were stored at -20°C until assayed. Midazolam concentration was measured by gas chromatography (GC-7A, Shimazu Co Ltd, Tokyo, Japan). The detection limit was 5 ng/mL and the coefficient of variations 3.1%. After the study, dogs were killed with an overdose of IV thiopental (40 mg/kg), and the location of the epidural catheters was checked.
Data are reported as mean ± SD. Areas under the curve (AUC) of serum and CSF midazolam concentrations (0–240 min) were calculated by a computer. Statistical analysis was performed with one-way repeated-measures analysis of variance for the contrasts for blood pressure and heart rate, two-way repeated-measures analysis of variance for the contrasts for midazolam concentrations, and Student’s t-test for the AUC of midazolam concentrations by SPSSTM (SPSS Inc, Chicago, IL). A value of P < 0.05 was considered to be statistically significant.
|
Results
|
|---|
All catheters were located in the epidural space. Blood pressure decreased significantly at 5 and 10 min after the epidural midazolam administration and returned to baseline by 30 min (Fig. 1). Heart rate increased transiently at 5 and 10 min after the epidural midazolam administration (Fig. 1).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 1. Blood pressure (upper) and heart rate (lower). Closed circle: systolic blood pressure; Open circle: diastolic blood pressure; Closed triangle: heart rate. Bars indicate SD. *P < 0.05 versus the control value.
|
|
Serum midazolam concentration increased to 129.6 ± 28.8 ng/mL at 5 min after the midazolam administration, peaked at 30 min (224.8 ± 30.5 ng/mL), and then decreased to 25.8 ± 6.0 ng/mL at 240 min (Fig. 2). Midazolam concentration in the CSF was less than the detection limit (5 ng/mL) at 5 min, 6.2 ± 6.0 ng/mL at 10 min, peaked at 30 min after the midazolam administration (7.2 ± 4.7 ng/mL), and decreased to 3.6 ± 3.3 ng/mL at 240 min (Fig. 2). The AUC of midazolam concentration (0–240 min) was 20,873 ± 3026 ng · mL-1 · min-1 in serum and 769 ± 227 ng · mL-1 · min-1 in CSF.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 2. Midazolam concentration. Closed circle: serum midazolam concentration; Open circle: cerebrospinal fluid (CSF) midazolam concentration. Bars indicate SD. +P < 0.05 versus CSF concentration. The detection limit of the assay was 5 ng/mL.
|
|
|
Discussion
|
|---|
Epidurally administered midazolam entered the systemic circulation rapidly, but the CSF concentration was only approximately 3% of the concentration of midazolam in the serum. Epidural midazolam induced a transient decrease in blood pressure and an increase in heart rate.
We administered 1 mg/kg of midazolam epidurally in dogs, which is larger than an adequate analgesic dose (0.05 mg/kg) in human studies (3,4). However, we did not know the analgesic dose of epidural midazolam in dogs. In our preliminary study, we administered 0.05 mg/kg, 0.5 mg/kg, and 1 mg/kg. The former two doses gave the CSF midazolam levels less than detection limit; therefore, we chose 1 mg/kg.
Transient decreases in blood pressure and increases in heart rate were also reported in a previous human study but with a different time course (7). In that previous study, the decrease in blood pressure was attributed to the analgesic or sedative effects of midazolam. In the present study, we could not test analgesic or sedative effects because the study was performed under general anesthesia. Goodchild and Noble (8) reported that intrathecal midazolam did not change blood pressure and heart rate and blocked somato-sympathetic reflexes in humans. However, there are no reports of the direct effects of midazolam, especially epidurally or intrathecally administered, on thoracic or lumbar sympathetic activity. This should be investigated further.
The present study is the first to investigate serum and CSF concentrations simultaneously after an epidural administration of benzodiazepines. A serum concentration of 200 ng/mL is required for sedation by IV midazolam (9). However, in our previous study, the serum midazolam concentration was smaller than 200 ng/mL when patients slept after an epidural administration of midazolam (7). Therefore, the sedation induced by epidural midazolam might result from cranial spread within the CSF rather than systemic uptake. In the present study, the maximum CSF midazolam concentrations were 7.2 ng/mL, which are only approximately 3% of the serum concentration. In in vitro studies, 10-9 mol/L (0.33 ng/mL) of midazolam does not activate GABA neurons (10), whereas 10-7 mol/L (33 ng/mL) of midazolam augments the GABA response (11). The CSF concentration of midazolam in the present study was intermediate between these two concentrations. Whether the CSF concentration in our results is enough to activate GABA receptors in the spinal cord is not known. However, even smaller doses of midazolam activate Ca2+ mediated K+ conductance, which is another form of neuronal inhibition (10). Therefore, epidurally administered midazolam might act on the spinal cord directly through CSF.
The route through which epidurally administered midazolam enters the CSF could not be determined in this study. Midazolam need not enter the systemic circulation to enter the central nervous system (12). After IM diazepam 10 mg, the CSF concentration was only 2%–3% of the plasma concentration (13). Free diazepam concentration in the plasma was in equilibrium with the CSF concentration (13). The entry of IM administered diazepam into the CSF seems to be by passive diffusion because the concentrations of diazepam in the CSF follow the plasma concentration (13). Because in the CSF there is no protein for binding, the rapid turnover rate of the CSF decreases diazepam concentrations (14). The protein binding of midazolam is 96%–97%. Therefore, in the present study, midazolam concentration in CSF, which is approximately 3% of serum concentration calculated by the AUC, is approximately in equilibrium with the free (not-protein bound) concentration in the blood. This suggests that midazolam may also enter the CSF by passive diffusion from the epidural space.
Neurotoxicity of midazolam to the spinal cord is still controversial (15–17). However, the small midazolam CSF concentration in the present study suggests that epidural administration of midazolam has a wide safety margin for neurotoxicity of the spinal cord. In conclusion, epidurally administered midazolam enters the CSF most likely by passive diffusion, but concentrations are only 3% of those found in the systemic circulation.
|
References
|
|---|
- Edwards M, Serrao JM, Gent JP, et al. On the mechanism by which midazolam causes spinally mediated analgesia. Anesthesiology 1990; 73: 273–7.[ISI][Medline]
- Nishiyama T, Hanaoka K. The synergistic interaction between midazolam and clonidine in spinally-mediated analgesia in two different pain models of rats. Anesth Analg 2001; 93: 1025–31.[Abstract/Free Full Text]
- Nishiyama T, Odaka Y, Hirasaki A, Seto K. Epidural midazolam for treatment of postoperative pain. Masui 1991; 40: 1353–8.[Medline]
- Nishiyama T. The post-operative analgesic action of midazolam following epidural administration. Eur J Anaesthesiol 1995; 12: 369–74.[ISI][Medline]
- Nishiyama T, Yokoyama T, Hanaoka K. Midazolam improves postoperative epidural analgesia with continuous infusion of local anaesthetics. Can J Anaesth 1998; 45: 551–5.[Abstract/Free Full Text]
- Nishiyama T, Hanaoka K. Effect of diluent volume on post-operative analgesia and sedation produced by epidurally administered midazolam. Eur J Anaesthesiol 1998; 15: 275–9.[ISI][Medline]
- Nishiyama T, Hirasaki A, Odaka Y, et al. Epidural midazolam with saline: optimal dose for postoperative pain. Masui 1992; 41: 49–54.[Medline]
- Goodchild CS, Noble J. The effects of intrathecal midazolam on sympathetic nervous system reflexes in man: a pilot study. Br J Clin Pharmacol 1987; 23: 279–85.[ISI][Medline]
- Crevat-Pisano PC, Dragna S, Granthil C, et al. Plasma concentrations and pharmacokinetics of midazolam during anaesthesia. J Pharm Pharmacol 1986; 38: 578–82.[ISI][Medline]
- Carlen PL, Gurevich N, Polc P. Low-dose benzodiazepine neuronal inhibition: enhanced Ca2+-mediated K+-conductance. Brain Res 1983; 271: 358–64.[ISI][Medline]
- Jansen H, Laursen AM. The effects of a benzodiazepine on the hyperpolarizing and the depolarizing responses of hippocampal cells to GABA. Brain Res 1981; 207: 214–7.[ISI][Medline]
- Henry RJ, Ruano N, Casto D, Wolf RH. A pharmacokinetic study of midazolam in dogs: nasal drop vs. atomizer administration. Pediatr Dent 1993; 15: 237–41.[Medline]
- Kanto J, Kangas L, Siirtola T. Cerebrospinal-fluid concentrations of diazepam and its metabolites in man. Acta Pharmacol Toxicol Copenh 1975; 36: 328–34.[Medline]
- Rall DP, Zubrod CG. Mechanism of drug absorption and excretion: passage of drugs in and out of the central nervous system. Am Rev Pharmacol 1962; 2: 109–28.
- Malinovsky JM, Cozian A, Lapage JY, et al. Ketamine and midazolam neurotoxicity in the rabbit. Anesthesiology 1991; 75: 91–7.[ISI][Medline]
- Schoeffler P, Auroy P, Bazin JE, et al. Subarachnoid midazolam: histologic study in rats and report of its effect on chronic pain in human. Reg Anesth 1991; 16: 329–32.[ISI][Medline]
- Svenson BA, Welin M, Gordh T Jr, Westman J. Chronic subarachnoid midazolam (Dorumicum) in the rat: morphological evidence of spinal cord neurotoxicity. Reg Anesth 1995; 20: 426–34.[ISI][Medline]
Accepted for publication September 24, 2002.
This article has been cited by other articles:
|
|
|
|
 
Y. K. Batra, R. Mahajan, S. Kumar, S. Rajeev, and M. Singh Dhillon
A Dose-Ranging Study of Intraarticular Midazolam for Pain Relief After Knee Arthroscopy
Anesth. Analg.,
August 1, 2008;
107(2):
669 - 672.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
