- Email: [email protected]
Intrathecal mepivacaine and prilocaine are less neurotoxic than lidocaine in a rat intrathecal model
Intrathecal Mepivacaine and Prilocaine Are Less Neurotoxic Than Lidocaine in a Rat Intrathecal Model Tamie Takenami, M.D., Saburo Yagishita, M.D., Yoshihiro Nara, and Sumio Hoka, M.D. Background and Objectives: Histologic evidence of the comparative neurotoxicity of lidocaine, mepivacaine, and prilocaine is incomplete. We compared the intrathecal neurotoxicity in rats among these 3 drugs based on morphologic and neurofunctional findings. Methods: Rats (n ⫽ 169) randomly received 0.12 L/g of 0%, 2%, 5%, 7.5%, 10%, or 20% lidocaine, mepivacaine, or prilocaine or 25% glucose dissolved in distilled water via a chronically implanted intrathecal catheter. The effect of the agents on neurofunction was evaluated by movement of the hind limb (behavior test) and by sensory threshold (paw-stimulation test). The L1 spinal cord, the posterior and anterior roots, and the cauda equina were removed en bloc 5 days later and examined by light and electron microscopy. Results: A significant decrease in sensory threshold or irreversible hind-limb limitation was observed only in rats receiving 20% lidocaine. Morphologic abnormalities characterized by axonal degeneration were observed in rats receiving ⱖ7.5% lidocaine, 20% mepivacaine, and 20% prilocaine, at the posterior white matter and the proximal portion of the posterior root just at the entrance into the spinal cord. The incidence of lesions was significantly higher in rats receiving lidocaine than mepivacaine and prilocaine. Conclusion: It is suggested that intrathecal mepivacaine and prilocaine are less neurotoxic than highly concentrated lidocaine in a rat intrathecal model. Reg Anesth Pain Med 2004;29:446-453. Key Words:
Local anesthetics, Neurotoxicity, Pathology, Sensory impairment.
S
ince lidocaine first became available for spinal anesthesia in 1945, it has had a long history of safety. Short-acting intrathecal local anesthetics such as lidocaine are widely used.1,2 However, recent clinical reports3-5 and in vivo6,7 and in vitro experiments8,9 have questioned its safety and suggest that the neurotoxicity of lidocaine is greater than that of other commonly used local anesthetics. Histologic evidence of neurotoxicity among other From the Department of Anesthesiology, Kitasato University School of Medicine (T.T., Y.N., S.H.); and Department of Pathology, Kanagawa Rehabilitation Center (S.Y.), Kanagawa, Japan. Accepted for publication June 17, 2004. Supported by Grant-in-Aid for Scientific Research (C)(2) 13671611 and (B)(2) 14370494, Ministry of Education, Science, Sports and Culture, Japan. Presented in part at the Annual Meeting of the American Society of Anesthesiologists, Dallas, TX, October 17-21, 1999, and the American Society of Regional Anesthesia Pain and Medicine, Vancouver, Canada, May 10-13, 2000. Reprint requests: Tamie Takenami, M.D., Department of Anesthesiology, Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, Kanagawa 228-8555, Japan. E-mail: [email protected] © 2004 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/04/2905-0010$30.00/0 doi:10.1016/j.rapm.2004.06.013
446
short-acting local anesthetics is incomplete.10-12 The present study of intrathecal injection in rats was designed to investigate the comparative neurotoxicity of lidocaine, mepivacaine, and prilocaine, based on histologic characteristics of the primary lesion and changes in neurologic function.
Material and Methods The Animal Experimentation and Ethics Committee of the Kitasato University School of Medicine approved this study. Studies were performed on 169 male Wistar rats (12 weeks old, weighing 266-304 g). The animals were housed 3 per cage in the experimental facility for 1 week before the experiments and maintained on a 12-hour light-dark cycle at 22°C (room temperature). They were allowed free access to food and water. Surgical Procedure for Intrathecal Catheterization The rats were anesthetized by sodium pentobarbital (intraperitoneal 50 mg/kg), and the subarachnoid space was cannulated with a polyethylene
Regional Anesthesia and Pain Medicine, Vol 29, No 5 (September–October), 2004: pp 446 –453
Comparative Neurotoxicity of Local Anesthetics
•
Takenami et al.
447
tube (0.6 ⫻ 700 mm) through the atlanto-occipital membrane using the modified method of Yaksh and Rudy.13 The tip of the catheter was advanced 7 cm caudal, to the level of Th 13. The catheter was then fixed in the subcutaneous tissue. The rats were allowed to recover for 1 week before drug injection. Rats showing symptoms of traumatic nerve damage were excluded from the experiment. Intrathecal Drugs and Doses Seven days after intrathecal catheterization, the rats were divided at random into the following groups based on each concentration of lidocaine, mepivacaine, or prilocaine solution dissolved in distilled water. The 3 drugs (AstraZeneca, Tokyo, Japan) were prepared at concentrations of 2%, 5%, 7.5%, 10% or 20%, on the day of the experiment under aseptic methods by a pharmacist (S.M.). In addition, a vehicle of distilled water solution was used as the control and 25% glucose was used as a solution with high osmolarity close to 20% lidocaine. The total volume of the injected solution in each animal was 0.12 L/g body weight, in addition to 6 L/g for the dead space of the catheter, injected over 15 seconds. The subcutaneously embedded catheter was exposed under inhalational anesthesia with ether delivered through a face-snout mask, and each solution was administered intrathecally. Immediately after drug injection, ether inhalation was stopped (the total inhalation time was ⬍5 minutes), and the wound was closed after leaving the catheter under the subcutaneous tissue. Over several minutes, the rats were allowed to breathe room air until recovery from ether anesthesia. The osmolarity of each solution was measured by freezingpoint depression (Fiske One-10, Norwood, MA). Behavior Evaluation The rats’ behavior was evaluated by ability to walk with or without limitation. Evaluation was performed at 15 minutes, 30 minutes, and every hour for 4 hours immediately after spinal injection on the day of experiment (postinjection day [PID] 0) and every subsequent morning from the day after injection (PID 1) to the fourth day (PID 4). Paw-Stimulation Test A technician (Y.N.) blinded to the animal groups performed the paw-stimulation test. The latency of the hind-limb withdrawal response to radiant heat on the plantar surface was measured just before injection of each concentration of lidocaine, mepivacaine, or prilocaine (prelatency) and just before perfusion fixation (postlatency). Measurements
Fig 1. Sample for light and electron microscopic examination. The lumbar (L1) spinal cord with the anterior and posterior roots and the cauda equina was dissected and removed en bloc to make 4 samples. (A) Lumbar (L1) spinal cord with central portion of both roots. (B) Peripheral portion of posterior root and (C) anterior root just above dorsal ganglion. (D) Cauda equina.
were repeated 6 times on both the left and right paws in each rat. The data were converted to percent maximum possible effect (%MPE), calculated as ([postlatency ⫺ prelatency]/[cutoff time ⫺ prelatency] ⫻ 100). Cutoff time was fixed as 20 seconds to prevent thermal injury. Tissue Preparation Five days after the intrathecal administration (PID 5), perfusion fixation was performed for tissue preparation subsequent to 2 test evaluations. The spinal cord (L1), including both the anterior and posterior roots, was prepared for light and electron microscopic studies, as in a previous study.14 After transcardiac perfusion fixation was performed, the lumbar spinal cord with the anterior and posterior roots and the cauda equina was dissected and removed en bloc, and 4 samples (from A to D) were used for morphologic examination (Fig 1). All specimens were embedded in epoxy resin. The semithin sections were stained with polychrome dyes. The ultrathin sections were double stained with uranyl acetate and lead citrate and examined under a JOEL FX2000 electron microscope (Nippon Denshi, Tokyo, Japan) at 100 kV. Skeletal muscle tissue of the hind limbs was examined in 3 rats with 20% lidocaine. The muscles
448
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004 Table 1. Osmolarity and pH of Each Solution Lidocaine
Concentration 2% 5% 7.5% 10% 20% Vehicle Distilled water 25% Glucose
Mepivacaine
Prilocaine
pH
mOsm/kgH2O
pH
mOsm/kgH2O
pH
mOsm/kgH2O
5.65 6.27 6.12 6.18 6.23
178 431 664 895 2,090
6.01 6.02 5.98 5.63 5.46
153 315 470 573 1,046
5.96 6.26 5.83 6.16 5.35
161 353 496 716 1,386
6.09 5.67
0 2,098
were fixed in 10% formaldehyde solution and embedded in paraffin wax. The thin sections were stained with hematoxylin-eosin and examined by light microscopy. A neuropathologist (S.Y.), who was blinded to intrathecal administration, examined the morphologic pathology. Statistical Analysis Values of %MPE were represented as mean ⫾ standard deviation and analyzed by Dunnett’s test for comparing a control (distilled water) mean to each group mean (1-way analysis of variance). Significant differences in the incidence of pathologic lesions after lidocaine injection were compared with the incidence after injection of the other 2 drugs by a chi-square t test. Recovery time of walking behavior was compared among groups for same concentration using a Kruskal-Wallis test. All statistical procedures were performed using Statview software version 4.5J (Abacus, London, England), and P ⬍ .05 was considered significant.
Results Thirteen rats were excluded from the study because of traumatic hind limb palsy caused by catheterization (n ⫽ 7), insufficient fixation (n ⫽ 4), or arachnoiditis (n ⫽ 2), which was confirmed by histologic investigation; a total of 156 rats were
analyzed. The number of rats in each group was as follows: control groups (distilled water, n ⫽ 5), 25% glucose (n ⫽ 6), lidocaine groups (2%, n ⫽ 9; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 10; 20%, n ⫽ 9), mepivacaine groups (2%, n ⫽ 9; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 9; 20%, n ⫽ 10), and prilocaine groups (2%, n ⫽ 10; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 10; 20%, n ⫽ 11). The pH and osmolarity of each drug solution are listed in Table 1. The pH ranged from 5.35 to 6.25. This variation was caused by difficulties in titrating, with an additional small amount of bicarbonate solutions that contained sulfuric acid for dissolution of the agents in distilled water. Neurofunctional Examination Behavior test. Rats receiving distilled water or 25% glucose could walk normally within 15 minutes after the intrathecal administration. The rats receiving ⱕ5% concentrations of the 3 drugs could walk normally within 1 hour. As the drug concentration increased, the time to recovery became prolonged (Table 2). All rats receiving 20% mepivacaine and 20% prilocaine could walk without limitation within 3 hours after the injections, whereas no rats receiving 20% lidocaine could walk even at 4 days after the injection. There was no significant difference in time to ambulation among
Table 2. Time to Normal Ambulation Concentration
Lidocaine
Prilocaine
Mepivacaine
2% 5% 7.5% 10% 20% DW 25%G
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (1 h-2 h) 3.5 h (3 h-4 h) Irreversible 0.25 h 0.25 h
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (0.5 h-2 h) 1.5 h (1 h-2 h)* 2 h (2 h-3 h)*
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (0.5 h-2 h) 1.5 h (1 h-2 h)* 2 h (2 h-3 h)*
NOTE. Hours in table are the time when rats could walk normally after drug administration. The data are expressed mean and range (minimum-maximum). Abbreviations: DW, distilled water; 25%G, 25% glucose solution. *P ⬍ .05 versus lidocaine.
Comparative Neurotoxicity of Local Anesthetics
•
Takenami et al.
449
rophages with destruction of myelin sheaths and axonal structures characterized the lesions. In cases with posterior root lesions with focal and sporadic disruption, posterior white matter damage was not observed, whereas in the cases with diffuse posterior root lesions, the marginal zone of posterior white matter tend to be injured (Fig 4). The character of the lesions: electron microscopic examination. In rats receiving 7.5% lidocaine, there was loss of neurofilaments and microtubules and degeneration of the mitochondria, which was compatible with a feature of axonal degeneration, whereas the myelin sheath was almost preserved (Fig 5A). However, in rats receiving 20% lidocaine, disruption of Fig 2. Sensory threshold estimated by paw-stimulation tests. A significant difference in sensory threshold was observed only in rats receiving 10% and 20% lidocaine versus vehicle. Rats receiving mepivacaine or prilocaine showed no significant difference in sensory threshold even at a drug concentration of 20%. %MPE ⫽ ([postlatency ⫺ prelatency]/[cutoff time ⫺ prelatency] ⫻ 100). Mean ⫾ standard deviation; vehicle, distilled water.
the 3 drugs at concentrations of ⬍10%. Conversely, rats receiving 10% and 20% of prilocaine and mepivacaine showed significantly faster recovery than rats receiving 10% and 20% lidocaine (P ⬍ .05), respectively. Paw-stimulation test. Figure 2 shows the effect of the 3 drugs on sensory threshold. A significant difference versus vehicle was observed only in rats receiving 10% and 20% lidocaine. The sensory threshold was decreased in those rats. The rats receiving mepivacaine and prilocaine did not show any significant difference in sensory threshold as compared with the vehicle (distilled water). Histopathologic Examination Lidocaine-, Mepivacaine-, and PrilocaineInduced Histopathologic Changes The distribution and incidence of the lesions: light microscopic examination. Figure 2 shows the extent of the lesions and location of the infusion catheter. Table 3 shows the incidence of the lesions with each drug. Significantly higher incidence was observed in rats receiving lidocaine than mepivacaine or prilocaine. The lesions were limited to the posterior white matter and the proximal portion of the posterior root just at the entrance into the spinal cord (within sample A) in rats receiving 7.5% to 20% lidocaine, 20% mepivacaine, and 20% prilocaine (Fig 3). Other areas, including the peripheral portion of both roots (samples B, C, and D), were intact. Lesions were not found in rats receiving vehicle or 25% glucose. Massive infiltration of mac-
Fig 3. Lesions and locations of the infusion catheter. A lesion in each group was found at the posterior root and posterior white matter. Locations of the catheter tip did not correlate with those of lesions. Black area, lesions showing diffuse disruption of both myelin sheath and axonal structure with macrophage infiltration; gray area, lesions showing focal or sporadic disruption of myelin and axonal structure with macrophage infiltration. ●, catheter tip with the lesions; Œ, catheter tip without the lesions.
450
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004
Table 3. Incidence of the Lesions at the Posterior Root and Posterior White Matter in Rats Receiving Intrathecal Injection of Lidocaine, Mepivacaine, and Prilocaine at Each Concentration Posterior Root
Posterior White Matter
Entry Distal Lateral Median 2% lidocaine (n ⫽ 9) 0% 2% mepivacaine (n ⫽ 9) 0% 2% prilocaine (n ⫽ 10) 0% 5% lidocaine (n ⫽ 10) 0% 5% mepivacaine (n ⫽ 10) 0% 5% prilocaine (n ⫽ 10) 0% 7.5% lidocaine (n ⫽ 10) 40% 7.5% mepivacaine (n ⫽ 10) 0%* 7.5% prilocaine (n ⫽ 10) 0%* 10% lidocaine (n ⫽ 10) 80% 10% mepivacaine (n ⫽ 9) 0%* 10% prilocaine (n ⫽ 10) 0%* 20% lidocaine (n ⫽ 9) 100% 20% mepivacaine (n ⫽ 10) 30%* 20% prilocaine (n ⫽ 11) 36%*
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
0% 0% 0% 0% 0% 0% 10% 0% 0% 60% 0%* 0%* 100% 20%* 27%*
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
the myelin sheath was noted in addition to axonal degeneration (Fig 5B). In rats receiving 20% mepivacaine or 20% prilocaine, axonal degeneration was more prominent than disruption of myelin structure in the posterior-root lesion in some cases similar to that in rats receiving 7.5% lidocaine, whereas degeneration of both axons and myelin sheaths became more frequent in rats with more severe posterior-root lesions similar to that in rats with ⱖ10% lidocaine. Thus, the characteristic morphologic findings in rats receiving mepivacaine or prilocaine were virtually identical in quality to those in rats receiving lidocaine. Skeletal Muscle Tissue Intrathecal 20% lidocaine did not produce any morphologic abnormalities in the skeletal muscle tissue, group atrophy, and angulated atrophic fibers, which are related with neurogenic changes,
*P ⬍ .05 versus lidocaine.
Fig 4. Light microscopic findings. The histologic abnormality was restricted to sample A (posterior white matter and central portion of the posterior root). The lesion after 7.5% lidocaine injection is restricted to the posterior root and the lesion is focal (A), whereas the lesion after 20% lidocaine injection extends to the posterior white matter (B) with diffuse disruption of the myelin sheath and axon. The lesions after 20% prilocaine (C) and 20% mepivacaine (D) are restricted to the posterior root and observed less at the posterior white matter compared with 20% lidocaine (black arrows indicate lesions). Semithin sections, polychrome stain, original magnification ⫻200. DH, dorsal horn; PR, posterior root; PW, posterior white matter.
Comparative Neurotoxicity of Local Anesthetics
Fig 5. Electron microscopic findings. (A) The 7.5% lidocaine group (original magnification 10,000⫻). Neurofilaments and microtubules disappear in the axoplasm in which disintegrated organelles are scattered, whereas the myelin structure is almost intact. Inset, original magnification 12,000⫻: control showing normal axoplasm containing normal-appearing neurofilaments and microtubules. (B) The 20% lidocaine group, original magnification 10,000⫻. Engulfed degenerating nerves in a cytoplasm (arrows) of a macrophage.
were not found. Disruption of myofibrils and increase of sarcolemmal nuclei, which are related with myopathic changes, were also not found.
Discussion This study showed that intrathecal administration of lidocaine, mepivacaine, and prilocaine at high concentrations caused virtually identical lesions characterized by axonal degeneration, which commonly started at the posterior root just at the entrance into the spinal cord (Obersteine-Radrich Zone) and extended to the posterior white matter by axonal degeneration. The results of this and our previous studies14,16 are not consistent with other
•
Takenami et al.
451
investigators who have shown various lesions after intrathecal administration of local anesthetics, such as vacuolation of the anterior and dorsal horn cells in the gray matter,17,18 central necrosis of the cord, and subpial vacuolation and chromatolysis.6,18-22 This inconsistency may be caused in part by different methodological techniques. For example, some of these investigations did not perform the application of a fixative solution. In that case, dissecting the spinal cord and processing the samples easily produced artifacts such as vacuolation and destruction of axon and myelin sheath. The location of the intrathecal catheter may also influence the histologic results. Although intrathecal administration of lidocaine, mepivacaine, and prilocaine caused identical lesions, significantly higher incidence was observed in rats receiving lidocaine than mepivacaine and prilocaine. Mepivacaine and prilocaine caused histologic abnormalities only at 20%, whereas lidocaine produced them at ⱖ7.5%. Thus, the neurotoxic lesions were only produced with clinically irrelevant concentrations of local anesthetics. However, 7.5% lidocaine is only 50% more concentrated than the 5% lidocaine that has been used clinically for spinal anesthesia. Therefore, our results suggest that lidocaine used in this rat model at concentration ⱖ150% of those used clinically cause neurotoxicity damage. Kishimoto et al.23 showed that intrathecal 2.5% prilocaine and 2.5% lidocaine produced virtually identical functional impairment and morphological damage in the cauda equina. This concentration for inducing tissue damage is lower than that in our study, but their total dose was relatively high and their exposure time was longer than ours (120 L for 2 hours). They also infused local anesthetics through a catheter placed near the cauda equina via a mechanical pump. In their model, maldistribution of anesthetics might have increased toxicity. Our results show that only rats receiving ⱖ10% lidocaine showed significant differences in sensory threshold versus the vehicle and also significantly longer recovery time in walking behavior. Moreover, only rats receiving 20% lidocaine had irreversible disturbance in walking. This increased sensitivity may be attributed to the damages in the posterior root or the posterior white matter, and the irreversible walking disturbance may also be caused by damage in the sensory, but not the motor, nervous system, or in skeletal muscle, as shown by their intact morphology. Conversely, rats receiving 7.5% lidocaine, 20% mepivacaine, and 20% prilocaine showed no significant alterations in sensory threshold and walking disturbance, although they
452
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004
exhibited the same morphologic damages in the posterior root. One of the potential reasons why they did not exhibit any significant changes in sensory threshold is that localized and mild lesions could spare the remaining intact nerve fibers, which could compensate for the dysfunction of injured nerves. We are unable to offer an explanation for the lack of neurotoxicity changes in the ventral roots. Factors other than the neurotoxicity of local anesthetics may be important in the histologic and functional deterioration in rats receiving intrathecal injection. At first, hyperosmolarity of an intrathecal solution may increase the histologic damage. However, previous experiment studies6,16 have shown that the hyperosmorality of an intrathecal solution does not correlate with the extent of neurotoxicity. In our study, differences in the osmolarity of intrathecal solution ranged from 153 to 2090 mOsm · kgH2O. We observed discrepancies in measured osmolarities among 3 anesthetics at equivalent concentrations, such as 2,090, 1,046, and 1,386 mOsm · kgH2O for 20% of lidocaine, mepivacaine, and prilocaine, respectively. These discrepancies may be explained in part by intermolecular attachment or intermolecular constraint in the solutions, which is apt to occur at a high concentration. In our study, 10% lidocaine (895 mOsm · kgH2O) caused neurotoxic lesions in excess of those lesions caused by 20% mepivacaine and prilocaine (both ⬎1,000 mOsm · kgH2O). In addition, 25% glucose (osmolarity ⬎2,000 mOsm · kgH2O and similar to 20% lidocaine) did not induce any lesions. Therefore, it is unlikely that the observed histopathologic differences could be explained entirely by changes in osmolarity. Secondly, mechanical injuries caused by the implanted catheter may cause tissue damage. However, the spinal lesions that we observed did not correlate with catheter location as shown in Fig 3. In addition, we collected samples from below the catheter tip to avoid any mechanical injury induced by the catheters. Thus, it is unlikely that the implanted catheter caused mechanical injury. The fact that the higher concentrations of local anesthetics were delivered to the thoracic spinal cord may have played an undetermined role in our findings. In summary, we compared the neurotoxicity of intrathecal lidocaine, mepivacaine, and prilocaine in rats. The primary location and character of the neurotoxic lesions was virtually identical among the 3 drugs. However, the histologic damage and neurofunctional impairment was significantly more severe with highly concentrated lidocaine than with mepivacaine and prilocaine. All neurotoxicity observed in this study occurred in a rat model at clinically irrelevant concentrations of local anes-
thetic. The clinical relevance of those findings to humans cannot be extrapolated from this study.
Acknowledgment The authors would like to thank Shinichi Sakura, Assistant Professor of Shimane University for his kind advice, and Yumiko Sugiura, Toshihiko Satoh, Setsuko Murase, and Electron Microscope Laboratory Center of Kitasato University School of Medicine for their help in this experiment.
References 1. Liu SS, McDonald SB. Current issue in spinal anesthesia. Anesthesiology 2001;94:888-906. 2. Carpenter RL. Hyperbaric lidocaine spinal anesthesia: Do we need an alternative? Anesth Analg 1995;81: 1125-1128. 3. Hampl KF, Heinzmann-Weidmer S, Luginbuehl I, Harms C, Seeberger M, Schneider, MC, Drasner K. Transient neurologic symptoms after spinal anesthesia: A lower incidence with prilocaine and bupivacaine than with lidocaine. Anesthesiology 1998;88: 629-633. 4. Pollock JE, Liu SS, Neal JM, Stephenson CA. Dilution of spinal lidocaine does not alter the incidence of transient neurologic symptoms. Anesthesiology 1999; 90:445-450. 5. Freedman JM, Li DK, Drasner K, Jaskela MC, Larsen B, Wi S. Transient neurologic symptoms after spinal anesthesia: An epidemiologic study of 1,863 patients. Anesthesiology 1998;89:633-641. 6. Ready LB, Plumer MH, Haschke RH, Austin E, Sumi SM. Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology 1985;63:364-370. 7. Drasner K, Sakura S, Chan VWS, Bollen AW. Persistent sensory deficit induced by intrathecal local anesthetic infusion in the rat. Anesthesiology 1994;80: 847-852. 8. Lambert L, Lambert D, Strichartz G. Irreversible conduction block in isolated nerve by high concentration of local anesthetics. Anesthesiology 1994;80:10821093. 9. Radwan IAM, Saito S, Goto F. The neurotoxicity of local anesthetics on growing neurons; a comparative study of lidocaine, bupivacaine, mepivacaine and ropivacaine. Anesth Analg 2002;94:319-324. 10. Liguori GA, Zayas VM, Chisholm MF. Transient neurologic symptoms after spinal anesthesia with mepivacaine and lidocaine. Anesthesiology 1998;88: 619623. 11. Drasner K. Lidocaine spinal anesthesia: A vanishing therapeutic index? Anesthesiology 1997;87:469-472. 12. Rowlingson JC. To avoid “transient neurologic symptoms.” The search continues. Reg Anesth Pain Med 2000;25:215-217. 13. Yaksh TL, Rudy TA. Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976;17: 1031-1036.
Comparative Neurotoxicity of Local Anesthetics 14. Takenami T, Yagishita S, Arai M, Asato F, Hoka S. Intrathecal lidocaine causes posterior root axonal degeneration near entry into the spinal cord in rats. Reg Anesth Pain Med 2002;27:58-67. 15. Sakura S, Hashimoto K, Bollen AW, Ciriales R, Drasner K. Intrathecal catheterization in rats; Improved technique for morphologic analysis of drug-induced injury. Anesthesiology 1996;85:1184-1189. 16. Takenami T, Yagishita S, Arai M, Asato F, Hoka S. Neurotoxicity of intrathecally administered tetracaine commences at the posterior root near entry into the spinal cord. Reg Anesth Pain Med 2000;25: 372-379. 17. Bahar DFL, Cole G, Rosen M. Neurotoxicity of the subarachnoid infusion of bupivacaine, lignocaine or 2-chlorocaine in the rat. Br J Anaesth 1985;57:424-429. 18. Bahar DFL, Cole G, Rosen, Vickers MD. Histopathology of the spinal cord after intrathecal cocaine, bupivacaine, lignocaine and adrenalin in the rat. Eur J Anaesthsesiol 1984;1:293-297. 19. Adams HJ, Mastri AR, Eicholzer AW, Klipatrick G.
20.
21.
22.
23.
•
Takenami et al.
453
Morphologic effects of intrathecal etidocaine and tetracaine on the rabbit spinal cord. Anesth Analg 1974; 53:904-908. Rosen MA, Baysinger CL, Shnider SM, Dailey PA, Norton M, Curtis JD, Collons M, Davis RL. Evaluation of neurotoxicity after subarachnoid injection of large volume of local anesthetic solutions. Anesth Analg 1983;62:802-808. Oka S. Matsumoto M, Ohtaka K, Kiyoshima T, Nakakimura K, Sakabe T. The additional epinephrine to tetracaine injected intrathecally sustains an increase in glutamate concentrations in the cerebrospinal fluid and worsens neuronal injury. Anesth Analg 2001;93:1050-1075. Sakura S, Bollen AW, Ciriales R, Drasner K. Local anesthetic neurotoxicity does not result from blockade of voltage -gated sodium channels. Anesth Analg 1995;81:338-346. Kishimoto T, Bollen AW, Drasner K. Comparative spinal neurotoxicity of prilocaine and lidocaine. Anesthesiology 2002;97:1250-1253.
Local anesthetics, Neurotoxicity, Pathology, Sensory impairment.
S
ince lidocaine first became available for spinal anesthesia in 1945, it has had a long history of safety. Short-acting intrathecal local anesthetics such as lidocaine are widely used.1,2 However, recent clinical reports3-5 and in vivo6,7 and in vitro experiments8,9 have questioned its safety and suggest that the neurotoxicity of lidocaine is greater than that of other commonly used local anesthetics. Histologic evidence of neurotoxicity among other From the Department of Anesthesiology, Kitasato University School of Medicine (T.T., Y.N., S.H.); and Department of Pathology, Kanagawa Rehabilitation Center (S.Y.), Kanagawa, Japan. Accepted for publication June 17, 2004. Supported by Grant-in-Aid for Scientific Research (C)(2) 13671611 and (B)(2) 14370494, Ministry of Education, Science, Sports and Culture, Japan. Presented in part at the Annual Meeting of the American Society of Anesthesiologists, Dallas, TX, October 17-21, 1999, and the American Society of Regional Anesthesia Pain and Medicine, Vancouver, Canada, May 10-13, 2000. Reprint requests: Tamie Takenami, M.D., Department of Anesthesiology, Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, Kanagawa 228-8555, Japan. E-mail: [email protected] © 2004 by the American Society of Regional Anesthesia and Pain Medicine. 1098-7339/04/2905-0010$30.00/0 doi:10.1016/j.rapm.2004.06.013
446
short-acting local anesthetics is incomplete.10-12 The present study of intrathecal injection in rats was designed to investigate the comparative neurotoxicity of lidocaine, mepivacaine, and prilocaine, based on histologic characteristics of the primary lesion and changes in neurologic function.
Material and Methods The Animal Experimentation and Ethics Committee of the Kitasato University School of Medicine approved this study. Studies were performed on 169 male Wistar rats (12 weeks old, weighing 266-304 g). The animals were housed 3 per cage in the experimental facility for 1 week before the experiments and maintained on a 12-hour light-dark cycle at 22°C (room temperature). They were allowed free access to food and water. Surgical Procedure for Intrathecal Catheterization The rats were anesthetized by sodium pentobarbital (intraperitoneal 50 mg/kg), and the subarachnoid space was cannulated with a polyethylene
Regional Anesthesia and Pain Medicine, Vol 29, No 5 (September–October), 2004: pp 446 –453
Comparative Neurotoxicity of Local Anesthetics
•
Takenami et al.
447
tube (0.6 ⫻ 700 mm) through the atlanto-occipital membrane using the modified method of Yaksh and Rudy.13 The tip of the catheter was advanced 7 cm caudal, to the level of Th 13. The catheter was then fixed in the subcutaneous tissue. The rats were allowed to recover for 1 week before drug injection. Rats showing symptoms of traumatic nerve damage were excluded from the experiment. Intrathecal Drugs and Doses Seven days after intrathecal catheterization, the rats were divided at random into the following groups based on each concentration of lidocaine, mepivacaine, or prilocaine solution dissolved in distilled water. The 3 drugs (AstraZeneca, Tokyo, Japan) were prepared at concentrations of 2%, 5%, 7.5%, 10% or 20%, on the day of the experiment under aseptic methods by a pharmacist (S.M.). In addition, a vehicle of distilled water solution was used as the control and 25% glucose was used as a solution with high osmolarity close to 20% lidocaine. The total volume of the injected solution in each animal was 0.12 L/g body weight, in addition to 6 L/g for the dead space of the catheter, injected over 15 seconds. The subcutaneously embedded catheter was exposed under inhalational anesthesia with ether delivered through a face-snout mask, and each solution was administered intrathecally. Immediately after drug injection, ether inhalation was stopped (the total inhalation time was ⬍5 minutes), and the wound was closed after leaving the catheter under the subcutaneous tissue. Over several minutes, the rats were allowed to breathe room air until recovery from ether anesthesia. The osmolarity of each solution was measured by freezingpoint depression (Fiske One-10, Norwood, MA). Behavior Evaluation The rats’ behavior was evaluated by ability to walk with or without limitation. Evaluation was performed at 15 minutes, 30 minutes, and every hour for 4 hours immediately after spinal injection on the day of experiment (postinjection day [PID] 0) and every subsequent morning from the day after injection (PID 1) to the fourth day (PID 4). Paw-Stimulation Test A technician (Y.N.) blinded to the animal groups performed the paw-stimulation test. The latency of the hind-limb withdrawal response to radiant heat on the plantar surface was measured just before injection of each concentration of lidocaine, mepivacaine, or prilocaine (prelatency) and just before perfusion fixation (postlatency). Measurements
Fig 1. Sample for light and electron microscopic examination. The lumbar (L1) spinal cord with the anterior and posterior roots and the cauda equina was dissected and removed en bloc to make 4 samples. (A) Lumbar (L1) spinal cord with central portion of both roots. (B) Peripheral portion of posterior root and (C) anterior root just above dorsal ganglion. (D) Cauda equina.
were repeated 6 times on both the left and right paws in each rat. The data were converted to percent maximum possible effect (%MPE), calculated as ([postlatency ⫺ prelatency]/[cutoff time ⫺ prelatency] ⫻ 100). Cutoff time was fixed as 20 seconds to prevent thermal injury. Tissue Preparation Five days after the intrathecal administration (PID 5), perfusion fixation was performed for tissue preparation subsequent to 2 test evaluations. The spinal cord (L1), including both the anterior and posterior roots, was prepared for light and electron microscopic studies, as in a previous study.14 After transcardiac perfusion fixation was performed, the lumbar spinal cord with the anterior and posterior roots and the cauda equina was dissected and removed en bloc, and 4 samples (from A to D) were used for morphologic examination (Fig 1). All specimens were embedded in epoxy resin. The semithin sections were stained with polychrome dyes. The ultrathin sections were double stained with uranyl acetate and lead citrate and examined under a JOEL FX2000 electron microscope (Nippon Denshi, Tokyo, Japan) at 100 kV. Skeletal muscle tissue of the hind limbs was examined in 3 rats with 20% lidocaine. The muscles
448
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004 Table 1. Osmolarity and pH of Each Solution Lidocaine
Concentration 2% 5% 7.5% 10% 20% Vehicle Distilled water 25% Glucose
Mepivacaine
Prilocaine
pH
mOsm/kgH2O
pH
mOsm/kgH2O
pH
mOsm/kgH2O
5.65 6.27 6.12 6.18 6.23
178 431 664 895 2,090
6.01 6.02 5.98 5.63 5.46
153 315 470 573 1,046
5.96 6.26 5.83 6.16 5.35
161 353 496 716 1,386
6.09 5.67
0 2,098
were fixed in 10% formaldehyde solution and embedded in paraffin wax. The thin sections were stained with hematoxylin-eosin and examined by light microscopy. A neuropathologist (S.Y.), who was blinded to intrathecal administration, examined the morphologic pathology. Statistical Analysis Values of %MPE were represented as mean ⫾ standard deviation and analyzed by Dunnett’s test for comparing a control (distilled water) mean to each group mean (1-way analysis of variance). Significant differences in the incidence of pathologic lesions after lidocaine injection were compared with the incidence after injection of the other 2 drugs by a chi-square t test. Recovery time of walking behavior was compared among groups for same concentration using a Kruskal-Wallis test. All statistical procedures were performed using Statview software version 4.5J (Abacus, London, England), and P ⬍ .05 was considered significant.
Results Thirteen rats were excluded from the study because of traumatic hind limb palsy caused by catheterization (n ⫽ 7), insufficient fixation (n ⫽ 4), or arachnoiditis (n ⫽ 2), which was confirmed by histologic investigation; a total of 156 rats were
analyzed. The number of rats in each group was as follows: control groups (distilled water, n ⫽ 5), 25% glucose (n ⫽ 6), lidocaine groups (2%, n ⫽ 9; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 10; 20%, n ⫽ 9), mepivacaine groups (2%, n ⫽ 9; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 9; 20%, n ⫽ 10), and prilocaine groups (2%, n ⫽ 10; 5%, n ⫽ 10; 7.5%, n ⫽ 10; 10%, n ⫽ 10; 20%, n ⫽ 11). The pH and osmolarity of each drug solution are listed in Table 1. The pH ranged from 5.35 to 6.25. This variation was caused by difficulties in titrating, with an additional small amount of bicarbonate solutions that contained sulfuric acid for dissolution of the agents in distilled water. Neurofunctional Examination Behavior test. Rats receiving distilled water or 25% glucose could walk normally within 15 minutes after the intrathecal administration. The rats receiving ⱕ5% concentrations of the 3 drugs could walk normally within 1 hour. As the drug concentration increased, the time to recovery became prolonged (Table 2). All rats receiving 20% mepivacaine and 20% prilocaine could walk without limitation within 3 hours after the injections, whereas no rats receiving 20% lidocaine could walk even at 4 days after the injection. There was no significant difference in time to ambulation among
Table 2. Time to Normal Ambulation Concentration
Lidocaine
Prilocaine
Mepivacaine
2% 5% 7.5% 10% 20% DW 25%G
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (1 h-2 h) 3.5 h (3 h-4 h) Irreversible 0.25 h 0.25 h
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (0.5 h-2 h) 1.5 h (1 h-2 h)* 2 h (2 h-3 h)*
0.5 h (0.5 h-1 h) 1 h (0.5 h-1 h) 1 h (0.5 h-2 h) 1.5 h (1 h-2 h)* 2 h (2 h-3 h)*
NOTE. Hours in table are the time when rats could walk normally after drug administration. The data are expressed mean and range (minimum-maximum). Abbreviations: DW, distilled water; 25%G, 25% glucose solution. *P ⬍ .05 versus lidocaine.
Comparative Neurotoxicity of Local Anesthetics
•
Takenami et al.
449
rophages with destruction of myelin sheaths and axonal structures characterized the lesions. In cases with posterior root lesions with focal and sporadic disruption, posterior white matter damage was not observed, whereas in the cases with diffuse posterior root lesions, the marginal zone of posterior white matter tend to be injured (Fig 4). The character of the lesions: electron microscopic examination. In rats receiving 7.5% lidocaine, there was loss of neurofilaments and microtubules and degeneration of the mitochondria, which was compatible with a feature of axonal degeneration, whereas the myelin sheath was almost preserved (Fig 5A). However, in rats receiving 20% lidocaine, disruption of Fig 2. Sensory threshold estimated by paw-stimulation tests. A significant difference in sensory threshold was observed only in rats receiving 10% and 20% lidocaine versus vehicle. Rats receiving mepivacaine or prilocaine showed no significant difference in sensory threshold even at a drug concentration of 20%. %MPE ⫽ ([postlatency ⫺ prelatency]/[cutoff time ⫺ prelatency] ⫻ 100). Mean ⫾ standard deviation; vehicle, distilled water.
the 3 drugs at concentrations of ⬍10%. Conversely, rats receiving 10% and 20% of prilocaine and mepivacaine showed significantly faster recovery than rats receiving 10% and 20% lidocaine (P ⬍ .05), respectively. Paw-stimulation test. Figure 2 shows the effect of the 3 drugs on sensory threshold. A significant difference versus vehicle was observed only in rats receiving 10% and 20% lidocaine. The sensory threshold was decreased in those rats. The rats receiving mepivacaine and prilocaine did not show any significant difference in sensory threshold as compared with the vehicle (distilled water). Histopathologic Examination Lidocaine-, Mepivacaine-, and PrilocaineInduced Histopathologic Changes The distribution and incidence of the lesions: light microscopic examination. Figure 2 shows the extent of the lesions and location of the infusion catheter. Table 3 shows the incidence of the lesions with each drug. Significantly higher incidence was observed in rats receiving lidocaine than mepivacaine or prilocaine. The lesions were limited to the posterior white matter and the proximal portion of the posterior root just at the entrance into the spinal cord (within sample A) in rats receiving 7.5% to 20% lidocaine, 20% mepivacaine, and 20% prilocaine (Fig 3). Other areas, including the peripheral portion of both roots (samples B, C, and D), were intact. Lesions were not found in rats receiving vehicle or 25% glucose. Massive infiltration of mac-
Fig 3. Lesions and locations of the infusion catheter. A lesion in each group was found at the posterior root and posterior white matter. Locations of the catheter tip did not correlate with those of lesions. Black area, lesions showing diffuse disruption of both myelin sheath and axonal structure with macrophage infiltration; gray area, lesions showing focal or sporadic disruption of myelin and axonal structure with macrophage infiltration. ●, catheter tip with the lesions; Œ, catheter tip without the lesions.
450
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004
Table 3. Incidence of the Lesions at the Posterior Root and Posterior White Matter in Rats Receiving Intrathecal Injection of Lidocaine, Mepivacaine, and Prilocaine at Each Concentration Posterior Root
Posterior White Matter
Entry Distal Lateral Median 2% lidocaine (n ⫽ 9) 0% 2% mepivacaine (n ⫽ 9) 0% 2% prilocaine (n ⫽ 10) 0% 5% lidocaine (n ⫽ 10) 0% 5% mepivacaine (n ⫽ 10) 0% 5% prilocaine (n ⫽ 10) 0% 7.5% lidocaine (n ⫽ 10) 40% 7.5% mepivacaine (n ⫽ 10) 0%* 7.5% prilocaine (n ⫽ 10) 0%* 10% lidocaine (n ⫽ 10) 80% 10% mepivacaine (n ⫽ 9) 0%* 10% prilocaine (n ⫽ 10) 0%* 20% lidocaine (n ⫽ 9) 100% 20% mepivacaine (n ⫽ 10) 30%* 20% prilocaine (n ⫽ 11) 36%*
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
0% 0% 0% 0% 0% 0% 10% 0% 0% 60% 0%* 0%* 100% 20%* 27%*
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
the myelin sheath was noted in addition to axonal degeneration (Fig 5B). In rats receiving 20% mepivacaine or 20% prilocaine, axonal degeneration was more prominent than disruption of myelin structure in the posterior-root lesion in some cases similar to that in rats receiving 7.5% lidocaine, whereas degeneration of both axons and myelin sheaths became more frequent in rats with more severe posterior-root lesions similar to that in rats with ⱖ10% lidocaine. Thus, the characteristic morphologic findings in rats receiving mepivacaine or prilocaine were virtually identical in quality to those in rats receiving lidocaine. Skeletal Muscle Tissue Intrathecal 20% lidocaine did not produce any morphologic abnormalities in the skeletal muscle tissue, group atrophy, and angulated atrophic fibers, which are related with neurogenic changes,
*P ⬍ .05 versus lidocaine.
Fig 4. Light microscopic findings. The histologic abnormality was restricted to sample A (posterior white matter and central portion of the posterior root). The lesion after 7.5% lidocaine injection is restricted to the posterior root and the lesion is focal (A), whereas the lesion after 20% lidocaine injection extends to the posterior white matter (B) with diffuse disruption of the myelin sheath and axon. The lesions after 20% prilocaine (C) and 20% mepivacaine (D) are restricted to the posterior root and observed less at the posterior white matter compared with 20% lidocaine (black arrows indicate lesions). Semithin sections, polychrome stain, original magnification ⫻200. DH, dorsal horn; PR, posterior root; PW, posterior white matter.
Comparative Neurotoxicity of Local Anesthetics
Fig 5. Electron microscopic findings. (A) The 7.5% lidocaine group (original magnification 10,000⫻). Neurofilaments and microtubules disappear in the axoplasm in which disintegrated organelles are scattered, whereas the myelin structure is almost intact. Inset, original magnification 12,000⫻: control showing normal axoplasm containing normal-appearing neurofilaments and microtubules. (B) The 20% lidocaine group, original magnification 10,000⫻. Engulfed degenerating nerves in a cytoplasm (arrows) of a macrophage.
were not found. Disruption of myofibrils and increase of sarcolemmal nuclei, which are related with myopathic changes, were also not found.
Discussion This study showed that intrathecal administration of lidocaine, mepivacaine, and prilocaine at high concentrations caused virtually identical lesions characterized by axonal degeneration, which commonly started at the posterior root just at the entrance into the spinal cord (Obersteine-Radrich Zone) and extended to the posterior white matter by axonal degeneration. The results of this and our previous studies14,16 are not consistent with other
•
Takenami et al.
451
investigators who have shown various lesions after intrathecal administration of local anesthetics, such as vacuolation of the anterior and dorsal horn cells in the gray matter,17,18 central necrosis of the cord, and subpial vacuolation and chromatolysis.6,18-22 This inconsistency may be caused in part by different methodological techniques. For example, some of these investigations did not perform the application of a fixative solution. In that case, dissecting the spinal cord and processing the samples easily produced artifacts such as vacuolation and destruction of axon and myelin sheath. The location of the intrathecal catheter may also influence the histologic results. Although intrathecal administration of lidocaine, mepivacaine, and prilocaine caused identical lesions, significantly higher incidence was observed in rats receiving lidocaine than mepivacaine and prilocaine. Mepivacaine and prilocaine caused histologic abnormalities only at 20%, whereas lidocaine produced them at ⱖ7.5%. Thus, the neurotoxic lesions were only produced with clinically irrelevant concentrations of local anesthetics. However, 7.5% lidocaine is only 50% more concentrated than the 5% lidocaine that has been used clinically for spinal anesthesia. Therefore, our results suggest that lidocaine used in this rat model at concentration ⱖ150% of those used clinically cause neurotoxicity damage. Kishimoto et al.23 showed that intrathecal 2.5% prilocaine and 2.5% lidocaine produced virtually identical functional impairment and morphological damage in the cauda equina. This concentration for inducing tissue damage is lower than that in our study, but their total dose was relatively high and their exposure time was longer than ours (120 L for 2 hours). They also infused local anesthetics through a catheter placed near the cauda equina via a mechanical pump. In their model, maldistribution of anesthetics might have increased toxicity. Our results show that only rats receiving ⱖ10% lidocaine showed significant differences in sensory threshold versus the vehicle and also significantly longer recovery time in walking behavior. Moreover, only rats receiving 20% lidocaine had irreversible disturbance in walking. This increased sensitivity may be attributed to the damages in the posterior root or the posterior white matter, and the irreversible walking disturbance may also be caused by damage in the sensory, but not the motor, nervous system, or in skeletal muscle, as shown by their intact morphology. Conversely, rats receiving 7.5% lidocaine, 20% mepivacaine, and 20% prilocaine showed no significant alterations in sensory threshold and walking disturbance, although they
452
Regional Anesthesia and Pain Medicine Vol. 29 No. 5 September–October 2004
exhibited the same morphologic damages in the posterior root. One of the potential reasons why they did not exhibit any significant changes in sensory threshold is that localized and mild lesions could spare the remaining intact nerve fibers, which could compensate for the dysfunction of injured nerves. We are unable to offer an explanation for the lack of neurotoxicity changes in the ventral roots. Factors other than the neurotoxicity of local anesthetics may be important in the histologic and functional deterioration in rats receiving intrathecal injection. At first, hyperosmolarity of an intrathecal solution may increase the histologic damage. However, previous experiment studies6,16 have shown that the hyperosmorality of an intrathecal solution does not correlate with the extent of neurotoxicity. In our study, differences in the osmolarity of intrathecal solution ranged from 153 to 2090 mOsm · kgH2O. We observed discrepancies in measured osmolarities among 3 anesthetics at equivalent concentrations, such as 2,090, 1,046, and 1,386 mOsm · kgH2O for 20% of lidocaine, mepivacaine, and prilocaine, respectively. These discrepancies may be explained in part by intermolecular attachment or intermolecular constraint in the solutions, which is apt to occur at a high concentration. In our study, 10% lidocaine (895 mOsm · kgH2O) caused neurotoxic lesions in excess of those lesions caused by 20% mepivacaine and prilocaine (both ⬎1,000 mOsm · kgH2O). In addition, 25% glucose (osmolarity ⬎2,000 mOsm · kgH2O and similar to 20% lidocaine) did not induce any lesions. Therefore, it is unlikely that the observed histopathologic differences could be explained entirely by changes in osmolarity. Secondly, mechanical injuries caused by the implanted catheter may cause tissue damage. However, the spinal lesions that we observed did not correlate with catheter location as shown in Fig 3. In addition, we collected samples from below the catheter tip to avoid any mechanical injury induced by the catheters. Thus, it is unlikely that the implanted catheter caused mechanical injury. The fact that the higher concentrations of local anesthetics were delivered to the thoracic spinal cord may have played an undetermined role in our findings. In summary, we compared the neurotoxicity of intrathecal lidocaine, mepivacaine, and prilocaine in rats. The primary location and character of the neurotoxic lesions was virtually identical among the 3 drugs. However, the histologic damage and neurofunctional impairment was significantly more severe with highly concentrated lidocaine than with mepivacaine and prilocaine. All neurotoxicity observed in this study occurred in a rat model at clinically irrelevant concentrations of local anes-
thetic. The clinical relevance of those findings to humans cannot be extrapolated from this study.
Acknowledgment The authors would like to thank Shinichi Sakura, Assistant Professor of Shimane University for his kind advice, and Yumiko Sugiura, Toshihiko Satoh, Setsuko Murase, and Electron Microscope Laboratory Center of Kitasato University School of Medicine for their help in this experiment.
References 1. Liu SS, McDonald SB. Current issue in spinal anesthesia. Anesthesiology 2001;94:888-906. 2. Carpenter RL. Hyperbaric lidocaine spinal anesthesia: Do we need an alternative? Anesth Analg 1995;81: 1125-1128. 3. Hampl KF, Heinzmann-Weidmer S, Luginbuehl I, Harms C, Seeberger M, Schneider, MC, Drasner K. Transient neurologic symptoms after spinal anesthesia: A lower incidence with prilocaine and bupivacaine than with lidocaine. Anesthesiology 1998;88: 629-633. 4. Pollock JE, Liu SS, Neal JM, Stephenson CA. Dilution of spinal lidocaine does not alter the incidence of transient neurologic symptoms. Anesthesiology 1999; 90:445-450. 5. Freedman JM, Li DK, Drasner K, Jaskela MC, Larsen B, Wi S. Transient neurologic symptoms after spinal anesthesia: An epidemiologic study of 1,863 patients. Anesthesiology 1998;89:633-641. 6. Ready LB, Plumer MH, Haschke RH, Austin E, Sumi SM. Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology 1985;63:364-370. 7. Drasner K, Sakura S, Chan VWS, Bollen AW. Persistent sensory deficit induced by intrathecal local anesthetic infusion in the rat. Anesthesiology 1994;80: 847-852. 8. Lambert L, Lambert D, Strichartz G. Irreversible conduction block in isolated nerve by high concentration of local anesthetics. Anesthesiology 1994;80:10821093. 9. Radwan IAM, Saito S, Goto F. The neurotoxicity of local anesthetics on growing neurons; a comparative study of lidocaine, bupivacaine, mepivacaine and ropivacaine. Anesth Analg 2002;94:319-324. 10. Liguori GA, Zayas VM, Chisholm MF. Transient neurologic symptoms after spinal anesthesia with mepivacaine and lidocaine. Anesthesiology 1998;88: 619623. 11. Drasner K. Lidocaine spinal anesthesia: A vanishing therapeutic index? Anesthesiology 1997;87:469-472. 12. Rowlingson JC. To avoid “transient neurologic symptoms.” The search continues. Reg Anesth Pain Med 2000;25:215-217. 13. Yaksh TL, Rudy TA. Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976;17: 1031-1036.
Comparative Neurotoxicity of Local Anesthetics 14. Takenami T, Yagishita S, Arai M, Asato F, Hoka S. Intrathecal lidocaine causes posterior root axonal degeneration near entry into the spinal cord in rats. Reg Anesth Pain Med 2002;27:58-67. 15. Sakura S, Hashimoto K, Bollen AW, Ciriales R, Drasner K. Intrathecal catheterization in rats; Improved technique for morphologic analysis of drug-induced injury. Anesthesiology 1996;85:1184-1189. 16. Takenami T, Yagishita S, Arai M, Asato F, Hoka S. Neurotoxicity of intrathecally administered tetracaine commences at the posterior root near entry into the spinal cord. Reg Anesth Pain Med 2000;25: 372-379. 17. Bahar DFL, Cole G, Rosen M. Neurotoxicity of the subarachnoid infusion of bupivacaine, lignocaine or 2-chlorocaine in the rat. Br J Anaesth 1985;57:424-429. 18. Bahar DFL, Cole G, Rosen, Vickers MD. Histopathology of the spinal cord after intrathecal cocaine, bupivacaine, lignocaine and adrenalin in the rat. Eur J Anaesthsesiol 1984;1:293-297. 19. Adams HJ, Mastri AR, Eicholzer AW, Klipatrick G.
20.
21.
22.
23.
•
Takenami et al.
453
Morphologic effects of intrathecal etidocaine and tetracaine on the rabbit spinal cord. Anesth Analg 1974; 53:904-908. Rosen MA, Baysinger CL, Shnider SM, Dailey PA, Norton M, Curtis JD, Collons M, Davis RL. Evaluation of neurotoxicity after subarachnoid injection of large volume of local anesthetic solutions. Anesth Analg 1983;62:802-808. Oka S. Matsumoto M, Ohtaka K, Kiyoshima T, Nakakimura K, Sakabe T. The additional epinephrine to tetracaine injected intrathecally sustains an increase in glutamate concentrations in the cerebrospinal fluid and worsens neuronal injury. Anesth Analg 2001;93:1050-1075. Sakura S, Bollen AW, Ciriales R, Drasner K. Local anesthetic neurotoxicity does not result from blockade of voltage -gated sodium channels. Anesth Analg 1995;81:338-346. Kishimoto T, Bollen AW, Drasner K. Comparative spinal neurotoxicity of prilocaine and lidocaine. Anesthesiology 2002;97:1250-1253.