SciELO - Scientific Electronic Library Online

vol.58 issue3Breakage of epidural catheters: etiology, prevention, and managementAnesthesia for blalock-taussig shunt and pacemaker placement in an adult patient with univentricular heart: case report author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Anestesiologia

Print version ISSN 0034-7094On-line version ISSN 1806-907X

Rev. Bras. Anestesiol. vol.58 no.3 Campinas May/June 2008 



Comparison of histologic spinal cord and neurologic changes in guinea pigs after subarachnoid block with large volumes of racemic bupivacaine, 50% enantiomeric excess bupivacaine (S75-R25), and levobupivacaine


Comparación de las alteraciones histológicas de la medula espinal y neurológicas de hámsteres después de la anestesia subaracnoidea con grandes volúmenes de bupivacaina racémica, de mezcla con exceso enantiomérico de 50% de bupivacaína (S75-R25) y de levobupivacaína



Paulo de Oliveira Vasconcelos Filho, TSAI; Irimar de Paula Posso, TSAII; Mariza CapelozziI; Vera Luiza CapelozziIII

IDoutor pela FM/USP
IIProfessor Doutor da Disciplina de Anestesiologia da FM/USP
IIIProfessora Doutora da Disciplina de Imunopatologia da FM/USP

Correspondence to




BACKGROUND AND OBJECTIVES: Levobupivacaine has less central nervous system side effects than racemic bupivacaine, but its anesthetic effect is not as intense. The objective of this experimental study was to compare the adverse effects of large volumes of bupivacaine, S75-R25 bupivacaine, and levobupivacaine injected in the subarachnoid space of guinea pigs.
METHODS: Forty guinea pigs were divided in four groups. They were anesthetized with 100% O2 and 2% isoflurane, followed by puncture of the L6-S1 intervertebral space. In Group I, 2 mL of normal saline were injected; in Group II, 2 mL of 0.5% bupivacaine; in Group III, 2 mL of 0.5% S75-R25 bupivacaine, and in Group IV, 2 mL of 0.5% levobupivacaine. After the animal awakened, neurological exam was done at 0, 60, 120, and 180 minutes, and daily for one week. Animals were killed and underwent perfusion with 4% paraformaldehyde. After fixation, the spinal cord was isolated by dissection and analyzed histologically to evaluate the degree of spinal cord lesions.
RESULTS: Guinea pigs in the control group did not present nervous block. Those in Group II presented sensitive and motor block for more than 180 minutes. Animals in Groups III (S75-R25) and IV (levobupivacaine) developed sensitive and motor blockade at moment 0, but at 60 minutes the motor blockade was minimal. Histologic exam in Group I showed no changes. In Group II, severe spinal cord changes were observed. In Groups III and IV, spinal cord changes were mild.
CONCLUSIONS: Large volumes of levobupivacaine caused little damage in the central nervous system when compared with bupivacaine. Statistically significant changes were not observed between levobupivacaine and S75-R25 bupivacaine.

Key Words: ANESTHETICS, Local: bupivacaine, S75-R25 bupivacaine, levobupivacaine; ANESTHETIC TECHNIQUE, Regional: subarachnoid; ANIMAL: guinea pigs; COMPLICATIONS: neurologic lesion, neurotoxicity.


JUSTIFICATIVA Y OBJETIVOS: La levobupivacaína presenta menores efectos colaterales sobre el sistema nervioso central, si los comparamos con los inducidos por la bupivacaína racémica, sin embargo el efecto anestésico es menos intenso. Fue realizado un estudio experimental para comparar efectos adversos de grandes volúmenes de bupivacaína, de bupivacaína S75-R25 y de levobupivacaína cuando se inyectaron en el espacio subaracnoideo de los hámsteres.
MÉTODO: Cuarenta hámsteres se dividieron en cuatro grupos. Anestesiados con O2 a 100% e isoflurano a 2%, con posterior punción en el espacio intervertebral L6-S1. En los animales del Grupo I se administraron 2 mL de solución fisiológica a 0,9%; en el Grupo II, 2 mL de bupivacaína 0,5%; en el Grupo III, 2 mL de bupivacaína S75-R25 0,5% y en el Grupo IV, 2 mL de levobupivacaína 0,5%. Después del despertar, en los momentos 0, 60, 120 y 180 minutos, fue realizado examen neurológico diariamente por una semana. Los animales fueron sacrificados y sometidos a la perfusión con paraformaldeido a 4%. Después de la fijación, la médula espinal fue aislada por disección y analizada histológicamente para evaluar el grado de lesión medular.
RESULTADOS: Los hámsteres del grupo control no presentaron bloqueo nervioso. Los del Grupo II presentaron bloqueo sensitivo y motor por más de 180 minutos. En los Grupos III (S75-R25) y IV (levobupivacaína) hubo un bloqueo motor y sensitivo al momento 0 minuto, sin embargo al momento 60 minutos el bloqueo motor era mínimo. En el examen histológico, el Grupo I no presentó alteraciones. En el Grupo II fueron encontradas alteraciones medulares intensas. En el Grupo III y IV las alteraciones medulares fueron poco intensas.
CONCLUSIONES: La levobupivacaína en grandes volúmenes causó poco daño al sistema nervioso comparada a la bupivacaína. Entre la levobupivacaína y la bupivacaína S75-R25, no hubo diferencia estadística significativa.




Bupivacaine hydrochloride is a local anesthetic, which has been available for more than 30 years, commonly used in regional blocks. Racemic bupivacaine, a mixture of isomeric forms (SR50), is the presentation available for clinical use. It has a long duration of action (4 to 12 hours), slow onset of action (10 to 30 minutes), and it has a pH between 4.6 and 6.0. Both forms of bupivacaine, R(+) and S(-), have anesthetic activity, but with different penetration on lipophilic layers and specific sodium channel receptors. Thus, there is a difference between the effects of the R(+) isomer, more toxic, and the S(-) isomer. The pharmacodynamics of racemic bupivacaine is marked by its cardiotoxicity when injected accidentally in the intravenous space 1-5.

Levobupivacaine is the S-enatiomer of bupivacaine. Pre-clinical protocols suggest that the potency of levobupivacaine is similar to bupivacaine, while exhibiting less cardio- and neurotoxicity 6-10.

The 50% enantiomeric excess mixture bupivacaine (S75-R25) resulted from the search for a local anesthetic with characteristics of long lasting neural blockade but with less systemic toxicity. It was demonstrated, in the experimental model of rat sciatic nerve, that S75-R25 maintains the local anesthetic properties of SR50. Afterwards, it was also shown in rats that the intravenous administration of a bolus dose of S75-R25 is less cardiotoxic than SR50, and this characteristic was attributed to the reduction of the R(+) isomer in the new drug 11,13.

In the decade of 1990, cases of paresis, paresthesias, and cauda equina syndrome, which were associated with the administration of a larger volume of local anesthetics in the subarachnoid space 14,15, were reported. Anesthetics were administered through microcatheters inserted through a small caliber, pencil-point needle 16,17. Patients who presented sequelae had received repeated injections due to inadequate block or injection of large volumes that should have been administered in the epidural space but were accidentally injected in the subarachnoid space. The cauda equina syndrome consists of urinary and bowel dysfunction, loss of sensitivity in the perineal area, and variable degrees of muscle weakness in the lower limbs 14,17,18.

The objective of this experimental study was to compare the effects of large volumes of bupivacaine and levobupivacaine administered in the subarachnoid space of guinea pigs. Spinal cord changes associated with changes in neurological exams, which allowed the confirmation of anesthetic block after the puncture, were compared.



After approval by the Research Project Analysis Commission (CAPPesq) of the Hospital das Clínicas of FMUSP, 40 female guinea pigs, weighing between 400 and 600 g, were used in this study. The animals were randomly divided in four groups, according to the solution that was administered. The model of a single puncture in guinea pigs was established in the institution after being published in 1998 19.

After fasting for 2 hours, the guinea pig was placed in ventral decubitus and anesthetized with 100% O2 and 2% isoflurane. A pulse oximeter was placed on the right anterior paw to determine the peripheral oxygen saturation. The two iliac tuberosities and the spinous process of the last lumbar vertebra were palpated. The lumbosacral space was immediately below the spinous process of the last lumbar vertebra. The area was shaved and cleaned, and anesthetized with 2% lidocaine. The animal was, then, positioned for subarachnoid puncture with a 23-gauge scalpel in the L6-S1 space. Animals received the different solutions randomly, being divided in four groups: in Group I, 2 mL of normal saline was administered (control group); in Group II, 2 mL of 0.5% bupivacaine; in Group III, 2 mL of 0.5% bupivacaine (S75-R25); and in Group IV, 2 mL of 0.5% levobupivacaine.

The neurological exam was based on the neurological exam conducted by Ready et al. in rats in 1985 20. It was performed after awakening on moments 0, 60, 120, and 180 minutes, and repeated on the following days, for a total of seven days; the animal was, then, killed. The exam was restricted to three neurological functions: ability to walk, flexion of the hind paw, and cutaneous sensitivity, which received the following scores: 0 – normal; 1 – reduced; and 2 – absent. Ability to walk was evaluated by placing the animal on the ground and observing its attempts to walk. Flexion of the posterior paw is done, characteristically, when the animal is lifted from a surface. Cutaneous sensitivity was evaluated by the aversive response to the stimulus of a 100-millivolts stimulator in two skin areas. Evaluation of the amount of clinical lesions caused by the different solutions was determined by the sum of the scores of the three areas. Water and food were provided for the animals during the study.

After seven days of neurological exams, animals were killed with sodium pentobarbital (200 intraperitoneal). Afterwards, they underwent perfusion with 200 mL of 4% paraformaldehyde for exsanguination through a lesion in the atrium or right cardiac chamber and puncture of the left ventricle for the administration of the paraformaldehyde with an infusion pump. The spinal cord was dissected immediately, from the cervical to the sacral region, removed and fixed in buffered neutral formalin at 10% for 48 hours, followed by decalcification for another 48 hours. Macroscopic exam of the vertebral column and spinal cord was performed. Finally, the entire length of the vertebral column was cut in 5-mm thick slices, which were processed for the histological exam.

The histopathological specimens were examined under light microscopy with different magnifications. Analysis of the histological components of the blood vessels and neuronal structures of the spinal cord in different levels were done, as well as morphometric measurements, and compared among the four groups.

To compare the morphologic changes detected in each group, Analysis of Variance (ANOVA) was used, and values of p < 0.05 were considered significant.



Animals in the control group, which received normal saline, had no signs of loss of cutaneous sensitivity or compromise of motor function in any moment, indicating the absence of mechanical trauma due to the needle or volume injected. All animals that received local anesthetic developed nervous block, characterized by the absence of: cutaneous sensitivity in the lumbar and sacral regions; flexion of hind paws; and locomotion (ability to walk), when the animals awoke from general anesthesia. Group II (racemic bupivacaine) developed sensitive and motor blockade, which lasted more than 180 minutes, and residual block characterized by lack of sensitivity to electrical stimulation that persisted for 24 hours. Animals in Groups III and IV developed sensitive and motor blockade shortly after the puncture but, at 60 minutes, motor blockade was partially present, while the sensitive blockade lasted three hours (Figure 1).



Analysis of histological changes in the cervical and sacral spinal cord did not show significant differences among the four groups. Figures 2 and 3 illustrate the differences in cellular neuronal volume among the different groups. Both the thoracic (p = 0.002) and lumbar (p = 0.008) regions of the spinal cord of guinea pigs in Group II (bupivacaine) showed an important reduction in cellular neuronal volume when compared with the other groups.





In Figures 4 and 5 one can compare the volume of the edema in all groups. Increased edema volume in the thoracic (p = 0.015) and lumbar (p = 0.003) spinal cord were also observed. Although Group II (S75-R25) showed more morphological changes than Group IV (levobupivacaine), this difference was not statistically significant when compared with the control group. Figure 6 shows the histological view of the different groups.






In this study few histological changes in the spinal cords of guinea pigs exposed to large volumes of levobupivacaine and S75-R25 bupivacaine were observed, which were not statistically different than the control group. Administration of large volumes of bupivacaine was related with damages in the nervous system, which was statistically significant when compared with the other groups.

The results of the present study demonstrated that large volumes of levobupivacaine and S75-R25 bupivacaine produced reversible sensitive and motor blockades, even when administered accidentally. The behavior of levobupivacaine was similar to normal saline. Administration of large volumes of SR50 bupivacaine was responsible for significant changes in the spinal cord when compared with the other groups.

A comparative study between the epidural and subarachnoid administration of bupivacaine, ropivacaine, and levobupivacaine in rats, by Kanai et al., demonstrated that levobupivacaine had greater anti-nociceptive effects than ropivacaine, and this was comparable to bupivacaine in concentrations of 0.25% to 0.75%. When ropivacaine and levobupivacaine were administered in the subarachnoid space, in equipotent concentrations, the motor blockade produce had similar duration. When compared with low concentrations of racemic bupivacaine, the duration of the blockade produced by both anesthetics was smaller. However, in high concentrations, the duration of the blockade was similar for all three drugs.

It would be interesting to know how useful experimental models are on identifying the neurotoxicity of local anesthetics. There are some experimental models used to study regional anesthesia in small animals 20-23. In 1998, a model of regional anesthesia in guinea pigs, with only a single paramedian puncture with a 23-gauge needle in the L6-S1 space, was described. The authors demonstrated, with the administration of methylene blue, that the puncture reached the subarachnoid space without damaging the spinal cord or nerve roots 19. Part of this model, without the methylene blue, was used in the present study; the presence of anesthesia was demonstrated by the clinical-neurological exam. The guinea pig was chosen because it is a docile animal and easy to evaluate, which was convenient for the present study. To decrease animal suffering, we decided to use general inhalational anesthesia, which also allowed total immobilization of the animal. Since the spinal cord of the guinea pig runs down to the lumbar region, the intervertebral space below the lumbar spine (L6-S1) was chosen 18,24.

Very little is known about the mechanisms by which local anesthetics produce nerve lesions. It is only known that, when clinical manifestations develop, the lesions are severe enough to produce loss of conduction in some populations of nerve fibers 15-17. It is known that, after the subarachnoid administration of local anesthetics, the metabolism of the spinal cord can decrease due to the intense sensitive and motor blockade. It is possible that the local anesthetic, in high concentrations, determines the irreversible blockade of intracellular transport, and this could contribute for the development of paresis and paralysis observed after regional blocks. Cell death is not necessary for the development of functional changes because toxic reactions in any population of cells or receptors can influence neurological function 16,20,25.

When large volumes of local anesthetics are inadvertently injected in the subarachnoid space, the tamponade and dilution that are normally seen with small volumes cease to exist. The possibility of greater exposure of the lumbar and sacral regions of the spinal cord and of the sacral nerve roots to the local anesthetic increases. The blood vessels of the spinal cord, that usually dilate, could undergo vasoconstriction, leading to spinal cord suffering 20,25.

The volume of anesthetic used in the present study was approximately four times greater than the standardized dose for anesthesia in animals. Prior pilot studies demonstrated that smaller volumes did not allow the identification of a number of neurologic lesions that differed significantly from the control group. For the development of cellular changes, such as of apoptosis and cell death, the concentration of the local anesthetic has to be higher than doses that produce irreversible blockade of sodium channels. Use of high concentrations can produce permanent neurologic changes 19,26.

The methodology of studies on the toxicity of local anesthetics includes intense manipulation of the study animals, including the insertion of catheters and the possibility of spinal cord damage. New studies try to find the minimal effective dose of local anesthetics (MLAC) and very few studies focus on the effects of large volumes of local anesthetics. The search for less traumatic methods leads to more confidence on the results.

Accidental injection of large volumes of local anesthetics in the subarachnoid space is a possible complication of epidural anesthesia 14,15. Over the last few years, the safety of this anesthetic technique has increased due to the possibility of administering local anesthetics with fewer side effects 9,13.

Large volumes of levobupivacaine caused little damage of the nervous system when compared with bupivacaine. Statistically significant differences were not observed between levobupivacaine and S75-R25.



01. Tetzlaff JE – The pharmacology of local anesthetics. Anesthesiol Clin North Am, 2000;18:217-233.         [ Links ]

02. Ritchie JM, Greene NM – Local Anesthetics, em: Gilman AG, Rall TW, Nies AS et al. – The Pharmacological Basis of Therapeutics. New York, Pergamon, 1990;311-329.         [ Links ]

03. Stoelting RK – Anestésicos Locais, em: Stoelting RK – Manual de Farmacologia e Fisiologia na Prática Anestésica. São Paulo, Artes Médicas, 1997;100-114.         [ Links ]

04. Stoelting RK – Anestésicos Locais, em: Barash PG, Cullen BF, Stoelting RK – Manual de Anestesiologia Clínica. São Paulo, Manole, 1991;223-242.         [ Links ]

05. Strichrtz GR, Cocino BG – Local Anesthetics, em: Miller RD – Anesthesia. 3rd Ed., New York, Churchill Livingstone, 1990;437-465.         [ Links ]

06. McClellan KJ, Spencer CM – Levobupivacaine. Drugs, 1998;56: 355-362.         [ Links ]

07. Foster RH, Markham A – Levobupivacaine: a review of its pharmacology and use as a local anaesthetic. Drugs, 2000;59: 551-579        [ Links ]

08. Simonetti MPB, Valinetti EA, Ferreira FMC – Avaliação da atividade anestésica local da S(-) bupivacaína: estudo experimental in vivo no nervo ciático de rato. Rev Bras Anestesiol,1997;47: 425-434.         [ Links ]

09. Kopacz DJ, Allen H, Thompson G – A comparison of epidural levobupivacaine 0.75% with racemic bupivacaine for lower abdominal surgery. Anesth Analg, 2000;90:642-648.         [ Links ]

10. Vladimirov M, Nau C, Mok WM et al. – Potency of bupivacaine stereoisomers tested in vitro and in vivo. Anesthesiology, 2000; 93:744-755.         [ Links ]

11. Simonetti MPB, Ferreira FMC – Does the D-isomer of bupivacaine contribute to the improvement of efficacy in neural block? Reg Anaesth Pain Med 1999;24(suppl):43.         [ Links ]

12. Simonetti MPB, Bird RA – Evaluation of new local anesthetics obtained through the manipulation of the enantiomeric ratio of bupivacaine on the central nervous system of the rat. Int Monitor, 2000;12:129.         [ Links ]

13. Delfino J, Vale NB – Bupivacaína levógira a 0,5% pura versus mistura enantiomérica de bupivacaína (S75-R25) a 0,5% em anestesia peridural para cirurgia de varizes. Rev Bras Anestesiol, 2001;51:474-482.         [ Links ]

14. Ben-David B, Rawa R – Complications of neuraxial blockade. Anesthesiol Clin North Am, 2002;20:669-693        [ Links ]

15. Munnur U, Suresh MS – Backache, headache, and neurologic deficit after regional anesthesia. Anesthesiol Clin North Am, 2003;21:71-86.         [ Links ]

16. Beardsley D, Holman S, Gantt R et al. – Transient neurological deficit after spinal anesthesia: local anesthetic maldistribution with pencil point needles? Anesth Analg, 1995;81:314-320.         [ Links ]

17. Lambert DH, Hurley RJ – Cauda equina syndrome and continuous spinal anesthesia. Anesth Analg, 1991;72:817-819.         [ Links ]

18. Drasner K, Rigler M, Sessler DI et al. – Cauda equina syndrome following intended epidural anesthesia. Anesthesiology, 1992; 77:582-585.         [ Links ]

19. Capelozzi M, Arantes FM, Paiva PSO et al. – Spinal anesthesia increases pulmonary responsiveness to methacholine in guinea pigs. Anesth Analg, 1998;87:874-878.         [ Links ]

20. Ready LB, Plumer MH, Haschke RH et al. – Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology, 1985; 63:364-70.         [ Links ]

21. Bahar M, Rosen M, Vickers MDV. Chronic cannulation of the intradural or extradural space in the rat. Br J Anaesth, 1984,56:405-10.         [ Links ]

22. Durant PAC, Yaksh TL – Epidural injections of bupivacaine, morphine, fentanil, lofentanil, and DADL in chronically implanted rats. A pharmacologic and pathologic study. Anesthesiology 1986, 64: 43-53.         [ Links ]

23. Kannai Y, Tateyama S, Nakamura T et al. – Effects of levobupivacaine, bupivacaine and ropivacaine on tail-flick and motor function in rats following epidural or intrathecal administration. Reg Anesth Pain Med, 1999,24:444-452.         [ Links ]

24. Cooper G, Schiller A – Guinea Pig Anatomy. 1st Ed., Harvard University, 1980;152-153.         [ Links ]

25. Ganem EM, Vianna PTG, Marques M et al. – Efeitos da administração subaracnóidea de grandes volumes lidocaína a 2% e ropivacaína a 1% sobre a medula espinal e as meninges. Estudo experimental em cães. Rev Bras Anestesiol, 2003;53:351-360.         [ Links ]

26. Myers RR, Sommer C– Methodology for spinal neurotoxicity studies. Reg Anesth, 1993;18:439-447.         [ Links ]



Correspondence to:
Dr. Paulo de Oliveira Vasconcelos Filho
Rua Napoleão de Barros, 420/21 – Vila Clementino
04024-001 São Paulo, SP

Submitted em 18 de maio de 2007
Accepted para publicação em 19 de fevereiro de 2008



* Received from Disciplina de Anestesiologia da Faculdade de Medicina da Universidade de São Paulo (FM/USP), São Paulo, SP

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License