| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology and Intensive Care, Faculty of Medicine, University of Miyazaki, Kiyotake, Miyazaki, Japan; and Division of Clinical Chemistry and Informatics, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto University, Kumamoto, Japan.
Address correspondence and reprint requests to Masahiko Taniguchi, MD, PhD, Department of Anesthesiology and Intensive Care, Faculty of Medicine, University of Miyazaki, 5200 Kihara, Kiyotake, Miyazaki 889-1692, Japan. Address e-mail to mtanigu{at}med.miyazaki-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract IntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Rats were assigned to four groups (n = 6 in each) and received intrathecally 30 µL of 1% levobupivacaine complexed with 0 (control), 50, 100, or 200 mM of G2-ß-CD. Muscle tone and tail flick latency were studied after intrathecal administration. Four more groups (n = 6) of rats received a sciatic nerve block with 0.5% or 1% levobupivacaine complexed with either 0 or 100 mM of G2-ß-CD. The time course of changes in proprioception, motor function, and nociception after circumferential subcutaneous administration to the sciatic nerve was examined.
RESULTS: With the intrathecal block, all rats stopped tail movement immediately after injection of the local anesthetic. The mean time to recovery of muscle tone with 1% levobupivacaine complexed with 100 mM (80.0 ± 8.9 min) and 200 mM (91.7 ± 7.0 min) of G2-ß-CD was significantly longer than that of the control group (38.3 ± 3.1 min), but tail flick latency was not prolonged. With the sciatic nerve block, all rats were temporarily immobilized after levobupivacaine injection. The anesthetic effects of 0.5% levobupivacaine with 100 mM of G2-ß-CD were twice as long as those for 0.5% levobupivacaine alone, and those of 1% levobupivacaine with 100 mM of G2-ß-CD were 1.4 times longer than those for 1% levobupivacaine alone.
CONCLUSIONS: The complex of levobupivacaine with G2-ß-CD prolonged the anesthetic effect of levobupivacaine in both intrathecal and sciatic nerve blocks in rats.
|
Introduction |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prolonged delivery systems for local anesthetics using biocompatible drug carriers, such as cyclodextrins and liposomes, have been studied (4). The cyclodextrin-based drug carrier system is a well established technology. More than 15 drug delivery systems, which contain cyclodextrins or their derivatives and are intended for oral and injectable use, have been approved by regulatory agencies in various countries (5). Cyclodextrins are cyclic oligosaccharides consisting of 6, 7, and 8 
-1,4 linked glucopyranose units, respectively, and are called -, ß-, and 
-cyclodextrins. They have a slightly apolar and hydrophobic internal cavity into which various guest molecules can be accommodated; thereby, they can act as suitable carrier materials to eliminate undesirable properties of drug molecules and can be used to prepare new and differentiated dosage forms of the drugs. Cyclodextrins have been reported to prolong the anesthetic effects of short-acting opioids after intrathecal administration through inclusion complex formation and to be compatible with spinal tissue (6,7).
In the present study we used the cyclodextrin derivatives, maltosyl-ß-cyclodextrin (G2-ß-CD), in which one of the primary hydroxyl groups of ß- cyclodextrin is substituted by maltose through the -1,6 glycosidic linkage (Fig. 1). Our preliminary study showed that G2-ß-CD forms an inclusion complex with levobupivacaine in a 1:1 molar ratio in an aqueous solution, as assessed by nuclear magnetic resonance spectroscopy and solubility analysis. In addition, G2-ß-CD has been shown to increase the solubility of levobupivacaine hydrochloride and levobupivacaine free base in aqueous media at pH ranging from neutral to alkaline and to decrease the surface tension-decreasing ability of the drug.
|
The present study was designed to determine whether the complex of levobupivacaine with G2-ß-CD is effective to use in intrathecal block and to determine whether this complex extends the duration of local anesthetic effects in a sciatic nerve block.
|
METHODS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study was approved by both the Animal Research and the Use Committees of University of Miyazaki. Male Sprague-Dawley rats weighing 260– 330 g were used. All rats were maintained at the Experimental Animal Center of the University of Miyazaki under a controlled ambient temperature of 23 ± 1°C and a humidity of 50% ± 10% with a 12-h light–dark cycle. Rats were allowed free access to food and water and were handled for 7–10 days before the commencement of this study, so that they would be familiar with the experimental environment and procedures.
Intrathecal Block
After obtaining baseline values for muscle tone and tail-flick latency, the rats were anesthetized by inhalation of a low concentration of sevoflurane (Sevofrane, Maruishi, Osaka), which allowed recovery within 2–3 min. The rats were held in a prone, elevated lumbar position, and their backs shaved and sterilized with 10% povidone iodine. The study anesthetic solutions were injected intrathecally with a 30-gauge needle on a 50-µL syringe (Hamilton, NV) at the L4–5 interspace with a positive indication of tail movement (8). This movement of the tail is "electrical shock-like" and is observed only when the needle is inserted into the intrathecal space. From immediately after injection until awakening from sevoflurane anesthesia, the rats were placed in a prone, horizontal position.
The rats were randomly divided into four groups (n = 6) to receive an intrathecal injection of one of the following doses of levobupivacaine complexed with G2-ß-CD. The solutions were prepared aseptically and administered intrathecally in a blinded, randomized fashion at a dose of 1% levobupivacaine complexed with 0 mM (control group), 50, 100, and 200 mM of G2-ß-CD. The pH and osmotic pressure of the solutions were adjusted to be pH 5.05–5.32 and 300 mOsm, respectively, by titration of hydrochloric acid and sodium chloride. All solutions were injected in a volume of 30 µL over 5 s. Animals were observed to identify the time to the onset of the nerve block. Muscle tone and tail-flick latency were assessed and were measured at 5-min intervals after injection in a blinded fashion.
General skeletal muscle tone was graded by manual manipulation and visual observation according to the scale proposed by Bromage et al. (9) for humans, which we modified for the rat model as follows: 3 = normal tone, free movement of the hindlimb; 2 = weak hypotonia of the hindlimbs/body posture; 1 = moderate hypotonia of the hindlimbs/ body posture; and 0 = inability to support the body on the hindlimbs and flat body posture.
To determine the tail-flick latency, a heat stimulus (xenon bulb, 24 V, 150 W) was directed at a portion of the tail approximately 5 cm from the tip. The time between stimulus onset and withdrawal of the tail was defined as the response latency. A tail-flick testing device (model 33B, IIITC, Inc., USA) was calibrated to give an average baseline latency of approximately 3–5 s. To prevent tissue damage, the heat stimulus was stopped if no response occurred within 10 s (cut-off time). Tail-flick testing began after confirmation of the animal’s ability to move its tail after an injection of a local anesthetic.
After the experimental series, rats had a laminectomy under deep pentobarbital anesthesia on the first day after injection. We confirmed the 30-gauge needle intrathecal injection hole in the dura. Rats were then perfused with formalin, and the spinal cords were removed from the 10th thoracic to the first sacral vertebra. Spinal cords were fixed in formalin and embedded in acrylic polymer (JB-4 TM plus embedding kit, Polysciences, Inc., USA). The spinal cords were then sliced in 1-µm transverse sections at the level of the cauda equina, stained with toluidine blue, and histologically evaluated under a light microscope. Examinations were performed on four slices near the level of the needle hole in a blind, randomized fashion. Sections were rated as severe (necrosis, axonal degeneration, large hemorrhage, and loss of myelin), mild (inflammatory cells, edema, and limited lesion of hemorrhage), or normal. Sections of a rat were evaluated if the hemorrhage was limited to the arachnoidea spinalis (subdural space) or to only one slice; in this case, a rat section was rated as mild. When there were many findings in a rat slice, the slice was classified as having the highest degree of damage.
Sciatic Nerve Block
Once baseline values were obtained, a unilateral sciatic nerve was blocked using brief sevoflurane anesthesia. A 30-gauge needle was inserted between the greater trochanter and ischial tuberosity with the bevel turned to the femoral head. When the needle encountered the body of the ischium, the local anesthetic solution was injected over a 5-s period (10).
The rats were randomly divided into four groups (n = 6) to receive the sciatic nerve block of one of the following doses of 0.5% or 1% levobupivacaine complexed with either 0 (control group) or 100 mM of G2-ß-CD. The pH and osmotic pressure of the solutions were adjusted to be pH 5.05–5.32 and 300 mOsm, respectively, by titration of hydrochloric acid and sodium chloride. The solution was administered in a blinded, randomized fashion in a constant volume of 100 µL. The animals were observed over the period when the nerve block was detectable.
Proprioception, motor function, and nociception were measured before and at 5, 10, and 15 min postinjection and then at 15-min intervals until neurological deficit was no longer observed. The measurements were taken in the following order: (1) proprioception (tactile placing followed by hopping response), (2) motor function, and (3) nociception, within a 100-s timeframe. All tests were performed by an investigator blinded to the drugs administered.
Proprioception was based on the resting posture and the postural reaction, i.e., the tactile placing and the hopping response, respectively. For tactile placing, the rat was kept in a normal resting posture, and the hindpaw toes were flexed with their dorsi placed on the table; then, the rat’s ability to reposition the toes was evaluated. To measure the hopping response, the rat was placed with the hindlimbs on a supporting surface, and the front half of the animal was held up. One hindlimb at a time was lifted off the ground, and the animal’s body was moved laterally. The hopping response time was measured as the time at which the animal hopped normally with the weight-bearing limb in the direction of movement of the body to avoid falling over. With impaired proprioception, the animal would drag its leg on the table. This functional deficit was measured by the following scoring system: 3 = normal posture and gait with symmetrical postural reactions; 2 = toes are flat, a slight lameness is observed, and the postural reactions are slower on the affected side; 1 = toes are flat and ventroflexed and are not involved in walking, and the postural reactions are severely damaged; the rat does not correct the position of toes in tactile placing and cannot bear weight and follow in the hopping reaction; 0 = complete absence of postural reaction and dragging.
Motor function was evaluated by measuring the extensor postural thrust: the rat was held upright with the affected hindlimb extended so that the body’s weight was supported by the distal metatarsus and toes, and the extensor thrust was measured as the force applied to a digital platform scale (model 1458N, TANITA, Japan). If complete blockade of sciatic nerve function was achieved, only passive force, or the lateral hindlimb’s weight, was measured.
The withdrawal reflex to the mechanical stimulation of deep tissue was used to evaluate nociception. Forceps were used to apply pressure across the distal phalanx of the fifth toe. The force exerted by the forceps was decided with a protractor as a minimal force which all rats showed a withdrawal response to in our preliminary study. The withdrawal reflex was measured by the following scoring system: 4 = normal: general motor reaction, vocalization, and withdrawal of the hindlimb; 3 = slower withdrawal than that on the control side; 2 = not a complete response; no vocalization or general motor reaction, only a turn of the head; 1 = only a weak attempt to withdraw; 0 = no response.
Neurotoxicity of Cyclodextrin
We finally examined the neurotoxicity of intrathecal G2-ß-CD alone. The rats were randomly divided into four groups (n = 4) to receive intrathecal injections of one of the following solutions that had been prepared aseptically and was administered intrathecally in a blinded and randomized fashion. The solutions used were saline (0.9% NaCl) and G2-ß-CD at concentrations of 50, 125, and 250 mM. At all concentrations, the solution was injected in volumes of 30 µL over 5 s.
We watched for general neurological symptoms, such as somnolence, irritation, convulsion, paralysis, autotomy, or any other abnormal phenomenon. The tail-flick latency was measured before, and at 5, 10, 30, 60, 120, 240, 360, 480 min and 24 h after injection in a blinded fashion.
After the experimental series, rats were perfused with formalin, and the spinal cord was removed from the 12th thoracic to the first sacral vertebra under deep pentobarbital anesthesia. The spinal cords were embedded in acrylic polymer and then sliced in 1-µm transverse sections at the level of cauda equina. The sections were stained with toluidine blue and histologically evaluated under a light microscope. Sections were rated as severe, mild, or normal, as in the intrathecal protocol.
Treatment of Data
To assess the effect of local anesthetics on muscle tone, tail-flick latency, proprioception, motor function, and nociception, the measured values were converted to a percent maximum possible effect (%MPE). The increase in %MPE means an increase of neural blocking effect with local anesthetics. Muscle tone %MPE was estimated by the following formula: [(control – score)/control] x 100, where the control was the muscle tone before injection and the score was the muscle tone after injection. Tail-flick latency %MPE was calculated by the following equation: %MPE = [(tail-flick latency – control)/(cut-off time – control)] x 100, where the control latency was the time of withdrawal before injection and tail-flick latency was the time of withdrawal after injection. Motor function %MPE was calculated by the equation: %MPE = [(control – experiment)/ (control – residual)] x 100, where the control was the weight of the hindlimb before injection, the experiment was the weight of the hindlimb after injection, and the residual was the residual weight of the hindlimb in the absence of thrust. Proprioception and nociception %MPE were estimated by the formula: [(control – experiment)/control] x 100, where the control was the score before injection and the experiment was the score after injection.
The area under the time-effect curve (AUC) was calculated by integrating the %MPE x minutes using the trapezoidal integration method. To evaluate the duration of the effect of the local anesthetics, the time from the end of the intrathecal injection to the time of %MPE 50% (%MPE50T) was calculated from the time-effect curve.
To assess the neurotoxicity of G2-ß-CD, the measured values of tail-flick latency were converted to %MPE, calculated as [(tail-flick latency – control)/ (cut-off time – control)] x 100.
The values of %MPE, AUC, and %MPE50T with each local anesthetic solution are expressed as the mean ± sem.
Statistics
The differences among four groups of the intrathecal or sciatic nerve block in %MPE, AUC, and %MPE50T were compared using the one-way analysis of variance followed by the Bonferroni (Dunn) test. The data from the neurotoxicity study were analyzed using Mann–Whitney U-test with Bonferroni correction. P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mean times to the recovery of muscle tone after an intrathecal injection of 1% levobupivacaine and 1% levobupivacaine with 50, 100, and 200 mM of G2-ß-CD were 38.3 ± 3.1, 38.1 ± 4.8, 80.0 ± 8.9, and 91.7 ± 7.0 min, respectively (Fig. 2). The mean time to the recovery of muscle tone and the mean AUC of 1% levobupivacaine complexed with 100 and 200 mM of G2-ß-CD were significantly longer than those of 1% levobupivacaine alone.
|
The %MPE50Ts calculated from tail-flick latency after 1% levobupivacaine and 1% levobupivacaine complexed with 50, 100, and 200 mM of G2-ß-CD were 102.0 ± 10.7, 82.8 ± 9.5, 125.5 ± 12.2, and 132.8 ± 4.9 min, respectively (Fig. 3). The mean AUC of tail-flick latency of 1% levobupivacaine with 100 and 200 mM of G2-ß-CD was significantly longer than that of 1% levobupivacaine with 50 mM of G2-ß-CD but not significantly different from that of 1% levobupivacaine alone. The time to the recovery of tail movement was shorter with the test solutions than the time to the recovery from the sensory block. No histological abnormalities were found in the spinal cord.
|
Sciatic Nerve Block
The hindlimb of the injected side in all rats was temporarily immobilized during this study and recovered with full neurological function at its completion. The time course effects for proprioception, motor function, and nociception after sciatic nerve block were similar (Fig. 4). However, the antinociceptive effect was slightly shorter than either the proprioception or motor function in individual rats.
|
The %MPE50T of motor block after 0.5% levobupivacaine, 0.5% levobupivacaine with 100 mM of G2-ß-CD, 1% levobupivacaine, and 1% levobupivacaine with 100 mM of G2-ß-CD was 113 ± 4, 221 ± 12, 182 ± 7, and 252 ± 15 min, respectively (Table 1). The anesthetic effect of 0.5% levobupivacaine with 100 mM of G2-ß-CD was twice as long as that of 0.5% levobupivacaine, and that of 1% levobupivacaine with 100 mM of G2-ß-CD was 1.4 times longer than that of 1% levobupivacaine alone.
|
Neurotoxicity of Cyclodextrin
We observed no abnormal neurological symptoms in the saline and the three concentrations of G2-ß-CD groups. The changes for %MPE of tail-flick latency were not different among the groups. However, two rats that received 250 mM of G2-ß-CD showed a markedly increased tail-flick latency value (8.5–10 s) at 4 and 8 h after injection. The mean baseline value of tail-flick latency in all rats was 3.7 ± 0.08 s.
In a histological study, no abnormal lesion was seen in the saline group, and no significant findings were seen in rats that received G2-ß-CD, except for two rats that received 250 mM of G2-ß-CD, which had a small hemorrhage on the surface of the spinal cord.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Each cyclodextrin has the ability to form inclusion complexes with specific molecules; this depends on the correct fit of the molecule into the hydrophobic cyclodextrin cavity (5). However, natural cyclodextrins have relatively low solubility, both in water and organic solvents, which limits their use in pharmaceutical formulations. The introduction of a maltosyl group onto the hydroxyl group of cyclodextrins confers increased solubility with a slight temperature dependency, e.g., the solubility of G2-ß-CD in water is move than 50% (w/v) at 25°C, (5,11) similar to that of hydroxypropyl-ß-cyclodextrin. Such a property is particularly useful for designing aqueous injectable cyclodextrin solutions that are to be heat-sterilized. The metabolic fate of G2-ß-CD is suggestive of less renal toxicity compared with that of ß-cyclodextrin, as the nephrotoxicity of the parent ß-cyclodextrin at higher doses is ascribed to the crystallization of a less soluble ß-cyclodextrin or its cholesterol complex in renal tissue (12). In fact, G2-ß-CD is excreted as soluble 6-O-glucosyl-ß-CD, not as the less soluble ß-cyclodextrin (5). In addition, natural cyclodextrins possess hemolytic ability at a higher concentration (13,14), but G2-ß-CD shows no or only slight hemolytic action (15,16). Therefore, G2-ß-CD may be useful as a solubilizer for parenteral preparations such as injections because of its high solubilization ability, weak hemolytic ability, and high bioadaptability. For example, some biological surface membrane constituents such as cholesterol and phospholipids and some drugs with hydrophobic moiety seem to be good candidates as the guest molecules. The binding constant of the complex of 1% levobupivacaine with 200 mM G2-ß-CD is 60.2 M–1, as determined by nuclear magnetic resonance spectroscopy. Based on the binding constant of the complex, more than 90% of levobupivacaine is assumed to exist as a complexed form with G2-ß-CD in equilibrium immediately after injection. Subsequently, the complex is gradually dissociated into levobupivacaine and G2-ß-CD, due to dilution with the biological fluid at the injection site and the competitive binding with other guest molecules.
There are a number of reasons why cyclodextrin drug carrier systems can extend the local anesthetic effect. As local anesthetics are small-size molecules, their rapid redistribution from the injected site limits the duration of effective analgesia. The molecular weight of the complex of levobupivacaine with G2-ß-CD is estimated to be 1747 Da, which is approximately 6 times heavier than that of levobupivacaine alone. As large molecules diffuse more slowly from the injection site, the complex of local anesthetic and cyclodextrin is cleared more slowly from the effective site than the local anesthetic alone. The sustained release of local anesthetics from the carrier vesicle will prolong the anesthetic effect. In addition, excess cyclodextrin could potentially slow the redistribution of the drug further because the movement of the drug into the biophase is in competition with the free cyclodextrin molecules in the same space (7).
G2-ß-CD lacks detergency ability, which lowers the surface tension of the solution. The surface tension-decreasing ability of the compound would be proportional to the tendency of the compound to move to the biological interface between biological fluid and cell surfaces and to enter into the cells. Therefore, the surface tension-decreasing ability of the compound is used as a simple and reliable measure to predict the local tissue irritancy of the compound at nontargeted tissues in vivo. From the viewpoint of local tissue irritancy, the lack of surface tension-decreasing ability of G2-ß-CD is desirable as a biocompatible drug carrier. Furthermore, G2-ß-CD decreases the surface tension-decreasing ability of levobupivacaine; the surface tensions of 0.5% levobupivacaine solution were 59.1 and 62.9 dyn/cm2 in the absence and presence of 200 mM G2-ß-CD in water at 25°C. This indicates that G2-ß-CD has the potential to reduce the local tissue irritation associated with prolonged stay of levobupivacaine at relatively high concentrations at the site of administration.
In the spinal and sciatic nerve block studies, the upper limit of the G2-ß-CD concentration was restricted by the osmotic pressure setting of the solution and the safety margin of G2-ß-CD. In our preliminary studies, we used the concentrations of G2-ß-CD up to 200 mM both for spinal and sciatic blocks. In the case of the sciatic nerve block, G2-ß-CD, even at the lower concentration (100 mM), prolonged the anesthetic effect of levobupivacaine. Therefore, we chose the concentration of G2-ß-CD at 100 mM for the sciatic nerve block studies. In the neurotoxicity studies, we chose higher concentrations of the cyclodextrin up to 250 mM in order to confirm the safety margin of the cyclodextrin.
Our study showed that the complex of cyclodextrin and levobupivacaine prolonged the effects of the local anesthetic in the sciatic nerve block to a greater extent than in the spinal block. One reason for this difference would be the environmental differences in the injection sites. The release of levobupivacaine from its cyclodextrin complex depends on the concentration gradient of levobupivacaine; therefore, the release of levobupivacaine is expected to occur more rapidly in the fluid environment of the intrathecal space than in the solid environment of the circumference of the sciatic nerve.
Meert and Melis (6) showed that hydroxypropyl-ß-cyclodextrin significantly increased the duration of analgesia after both epidural and intrathecal administration of sufentanil and bupivacaine in rats. However, the complex of bupivacaine and cyclodextrin had no anesthetic effect because of the small dose (80 µg) of bupivacaine. Fréville et al. (17) showed that epidurally administered complexes of bupivacaine and hydroxypropyl-ß-cyclodextrin could prolong the duration of the action and the systemic uptake of bupivacaine in rabbits. However, sulfobutylether7-ß-cyclodextrin delayed the onset of the motor block and did not modify the duration after bupivacaine administration near the sciatic nerve in rabbits (18). Overall, the anesthetic effects of the complex of local anesthetics with cyclodextrin have been controversial. Regarding analgesic opioids, Yaksh et al. (7) found that cyclodextrin could enhance the action of intrathecally administered opioids.
This study has two limitations. First, we did not compare G2-ß-CD alone, as a control vehicle, with the G2-ß-CD–levobupivacaine complex in the same protocol. In general, cyclodextrins are thought to be pharmacologically inactive. The tissue toxicity induced by cyclodextrin is very low in humans and animals (5,16). We also showed that there is no difference in the tail flick response between the cyclodextrin alone and reference control. However, two rats received G2-ß-CD at a high concentration of 250 mM showed an increased tail-flick latency value and mild histological damage. Moreover, the number of models studied was small, and not all potential toxicities were evaluated. Our single study is not definitive for evaluating the safety of this complex. Second, sensory function was not easily distinguishable from motor function during the sciatic nerve block because the motor block affected sensory assessment. As shown in our study, the time course effects for proprioception, motor function, and nociception after the sciatic nerve block were similar. Although we adapted a well-established neurological evaluation for the rat (10), the interrelation of neurological function should not be completely ignored.
In conclusion, the complex of levobupivacaine with G2-ß-CD prolonged the anesthetic effect of levobupivacaine in both intrathecal and sciatic nerve blocks in rats. This complex may be used to prolong the effect of a long-acting local anesthetic such as levobupivacaine 1.4–2.0 times through a conduction block and to supply adequate postoperative analgesia. Further studies are necessary to fully establish the safety and efficacy of this interesting complex for prolonging local anesthetic block.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
Supported in part by Grant-in-Aids for Scientific Research (B) (12470323) and for the Promotion of Science (B) (13557131) from the Ministry of Education, Science, Sports and Culture of Japan, Tokyo, Japan, and by the Yokohama Clinical and Pharmacological Research Foundation, Nagoya, Japan and by Maruishi Pharmaceutical Co., Ltd., Osaka, Japan.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
-, ß-, and -cyclodextrins on human erythrocytes. Eur J Biochem 1989;186:17–22.[Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
50% After Sciatic Nerve Block
