Share this article on:

The Comparative Cytotoxic Effects of Different Local Anesthetics on a Human Neuroblastoma Cell Line

Malet, Arnaud PhD; Faure, Marie-Odile PhD; Deletage, Nathalie PhD; Pereira, Bruno PhD; Haas, Jerome; Lambert, Gregory PharmD, PhD

doi: 10.1213/ANE.0000000000000562
Anesthetic Pharmacology: Research Report

BACKGROUND: Although local anesthetics (LAs) are generally accepted as being safe, incidental neuronal damage has been reported for all LAs in humans. Therefore, in this study, we compared the dose corresponding to 50% cell lethality (LD50) of articaine, lidocaine, mepivacaine, bupivacaine, prilocaine, and ropivacaine in human neuroblastoma cells.

METHODS: Undifferentiated SH-SY5Y cells were exposed for 20 minutes to different concentrations of each LA. Metabolic activity of viable cells was assessed by a cell viability test with a tetrazolium dye (WST-1) followed by optical density quantification. LD50 was determined by extrapolation of curve response.

RESULTS: As expected, all LAs induced cell death in a concentration-dependent manner. The bupivacaine, lidocaine, prilocaine, mepivacaine, articaine, and ropivacaine LD50 were 0.95 ± 0.08, 3.35 ± 0.33, 4.32 ± 0.39, 4.84 ± 1.28, 8.98 ± 2.07, and 13.43 ± 0.61 mM, respectively, after 20 minutes of incubation on SH-SY5Y cells. Ropivacaine LD50 was significantly different from the bupivacaine, lidocaine, mepivacaine, and prilocaine LD50 (all P ≤ 0.009). No significant difference was obtained between ropivacaine and articaine LD50 and between prilocaine, lidocaine, and mepivacaine LD50. Articaine LD50 was significantly different from lidocaine LD50 (P = 0.03). Bupivacaine LD50 was significantly lower compared with all LAs (all P ≤ 0.003).

CONCLUSIONS: LA neurotoxicity was tested in a validated in vitro model SH-SY5Y, a human neuroblastoma cell line. Three groups of LAs were identified in terms of toxicity: (1) the less (ropivacaine, articaine); (2) medium (mepivacaine, prilocaine, lidocaine); and (3) the high (bupivacaine). Among dental anesthetics, articaine is the least neurotoxic in SH-SY5Y cells.

Published ahead of print December 15, 2014.

From the *Scientific Affairs, Septodont, Saint Maur des Fosses, France; Research and Development, Neuronax, Saint Beauzire, France; and Biostatistics Unit, Clinical Research and Innovation Delegation, University Hospital Clermont-Ferrand, Clermont-Ferrand, France.

Published ahead of print December 15, 2014.

Accepted for publication October 8, 2014.

Funding: This work was entirely funded by Septodont SAS, which contributed to the study and the furnishing of interest products.

Conflict of Interest: See Disclosures at end of article.

Reprints will not be available from the authors.

Drs. Malet and Faure contributed equally to this work.

Address correspondence to Marie-Odile Faure, PhD, Scientific Affairs, Septodont, 105 avenue de Beaurepaire, Saint Maur des Fossés 94100, France. Address e-mail to

Pain management associated with use of local anesthetics (LAs) is an essential aspect of patient safety and comfort in dental procedures. The commonly used dental anesthetics are amide-based anesthetics in which lidocaine is considered the reference. Its onset and duration are particularly adapted to pulpal or soft tissue anesthesia by infiltration or maxillofacial nerve block, which requires a fast onset and a moderate duration.1 Moderate-acting LAs (lidocaine, prilocaine, mepivacaine, and articaine) are less firmly bound in the nerve membrane than longer-acting drugs (ropivacaine and bupivacaine) and therefore are released faster from receptor sites in the sodium channels.2,3 Nevertheless, longer-acting LAs are adapted for several clinical situations such as for dental surgical procedures of long duration, or when inflammation and pain are expected postoperatively.3

Although available dental LAs are accepted as being safe, incidental neuronal damage has been reported for all LAs in humans. Articaine (4%) and prilocaine (4%) are suspected to be responsible for increased neurotoxic side events compared to lidocaine (2%),4,5 although the incidence of local neurotoxicity events from either prilocaine and articaine were similarly close to those of lidocaine.6 Neurotoxicity among LAs is difficult to compare clinically since the LAs’ capacity to damage neuronal cells remains unclear and numerous factors may be involved in damage to the nerves other than the drugs themselves. Studies in animal and cellular models have proven that all LAs can be neurotoxic when applied to neural tissues in clinical concentrations. Nerve injury subsequent to a variety of LA injections at clinical concentrations has been displayed in both nerve and spinal cord experimental animal studies.7–14 The animal model’s selection for neurotoxicity testing is critical when results have to be extrapolated to humans, suggesting the need for multiple models to fully assess neurotoxicity. Damage to the growth cones and neurites of chicken embryos or snails has been reported in in vitro studies,15,16 but the extrapolation to humans is not obvious. Furthermore, the cytotoxicity of LAs has been evaluated in different in vitro and in vivo models (neurons, Schwann cells, T lymphoma cells, chondrocytes, myocytes, tenocytes) of varying origins.15,17–22 However, the LA toxicity differs depending on the organs or tissues; the cytotoxicity on different cell types could be caused by local factors acting in physiological microenvironments. Werdehausen et al.23 reported that amide and ester types of LAs were cytotoxic in the human neuroblastoma cell line SHEP. Articaine and procaine are the least toxic LAs, whereas bupivacaine and tetracaine presented the highest toxicity. Other anesthetics (mepivacaine, lidocaine, ropivacaine, chloroprocaine, and prilocaine) are associated with moderate cytotoxicity. Perez-Castro et al.24 studied another subclone of neuroblastoma cell line, SH-SY5Y, which is a validated human neuronal cell model in which to perform toxicity tests.25–28 Similarly to Werdehausen et al., Perez-Castro et al. displayed an equivalent range of LA toxicity although LAs such as articaine and prilocaine were not evaluated. Furthermore, this experiment was performed with a shorter LA exposure.

Consequently, few comparative data of local toxicity on human nerve cells are available. In this study, we explored neural cell toxicity after a short exposure to various amide LAs (lidocaine, mepivacaine, prilocaine, articaine, bupivacaine, and ropivacaine) using the human neuroblastoma SH-SY5Y cell line.

Back to Top | Article Outline


Cell Culture

SH-SY5Y (ECACC), an undifferentiated human neuroblastoma cell line at passages P18-P19, was seeded on 96-well tissue culture plates with a plating density of 2.104 cells/well. Cells were maintained in phenol red Dulbecco’s Modified Eagle’s Medium (DMEM; Lonza, Belgium) supplemented with 10% fetal bovine serum (PAA, France), 2 mM of L-glutamine (Lonza, Belgium), penicillin (100 U/mL), and streptomycin (100 µg/mL) (Lonza, Belgium). SH-SY5Y cells were incubated under a humidified atmosphere (95%) containing 5% CO2 at 37°C. Measurements of cell viability were performed 48 hours after seeding.

Back to Top | Article Outline


Mepivacaine, lidocaine, articaine, bupivacaine, and prilocaine were obtained from Septodont. Ropivacaine was purchased from Sigma-Aldrich (France). Stock solutions for mepivacaine (24 mM), lidocaine (15 mM), articaine (30 mM), bupivacaine (12 mM), prilocaine (22 mM), and ropivacaine (36 mM) were made from the powder chemicals and were dissolved in DMEM. Subsequently, pH of each stock solution was stabilized to ensure a physiological pH value (7.10–7.20), except for bupivacaine and ropivacaine where pH was stabilized at pH 6 to preserve the limit of product solubilization.

Back to Top | Article Outline

Exposure to LAs and Measurement of Cell Viability

Cell viability was evaluated using Cell Proliferation Reagent WST-1 (Roche, France). Briefly, 2 days after seeding, culture medium was discarded and cells in 96-well tissue culture plates were gently washed in phosphate-buffered saline (PAA, France). Phenol red DMEM and DMEM-Triton 0.1% were used as positive and negative control baseline cell viability, respectively.

DMEM-LA solutions were previously prepared at the following concentrations: 0.1/3/5/7.5/10/15 mM for lidocaine; 0.1/3/5/10/15/30 mM for articaine; 0.4/1.2/2/4/12/24 mM for mepivacaine; 0.02/0.6/1/2/6/12 mM for bupivacaine; 2/5/6/12/18/22 mM for prilocaine; and 1.2/2/4/12/24/36 mM for ropivacaine. Doses were determined according to preliminary results (data not shown). Tested products were added at cell passages 18–19. Cell viability was determined after 20 minutes’ LA incubation at 37°C. After treatment, DMEM-LA solution and control solutions were immediately discarded, and attached cells were gently rinsed once with phosphate-buffered saline. Then, SH-SY5Y cells were incubated according to WST-1 manufacturer’s instructions. Culture medium with WST-1 without cells was used for background control.

Optic density of the WST-1 incorporated cells was measured against the background control using the spectrophotometer (Thermo, Multiskan Spectrum) at 450 nm.

Absorbance of the positive control was set as 100% of cell viability, and the optic density measured in LA-treated cells was normalized to the corresponding control values and used as an indicator for cell viability (in % versus the untreated control groups). The 50% cell lethality (LD50) was obtained from the equation of concentration-cell viability curves of the different LAs.

The LD50 of each investigated LA was normalized by the lidocaine LD50 to evaluate the toxic equipotency value compared to lidocaine.

Back to Top | Article Outline

Microscopic Observation of SH-SY5Y Morphology Under LA Treatment

Cells in multiwell plates were placed on the stage of a microscope (Nikon, Eclipse TS-100). Images were taken before LA addition and 20 minutes after incubation with the LA at 37°C, using a 10X objective.

Back to Top | Article Outline

Statistical Analysis

Statistical analysis was performed using Stata software, version 13 (StataCorp, College Station, TX). The tests were 2-sided, with a type I error set at α = 0.05. Results (quantitative parameters) were presented as means ± SEM. For each experiment, the 6 concentrations of tested LAs were performed in quintuplets. Considering the potential nonindependence of data due to the experimenter effect, correlated data were analyzed using ANOVA for repeated measures (normality studied by Shapiro-Wilk’s test and homoscedasticity by Bartlett’s test). Pairwise comparisons (multiple comparisons) were considered, if previous analysis (repeated ANOVA) was P < 0.05, by Tukey-Kramer test. A log transformation was improved to access the normality. Differences were considered to be statistically significant with P < 0.05.

Back to Top | Article Outline


The LAs neurotoxicity on neuroblastoma cell line SH-SY5Y was evaluated by cell metabolism activity after a short exposure to different concentrations of each LA using the WST-1 assay, which provides a simple and accurate method to measure cell viability based on the cleavage of the tetrazolium salt WST-1 to formazan by cellular mitochondrial dehydrogenases. After 20 minutes of treatment, all 6 LAs were neurotoxic in a concentration-dependent manner (Fig. 1).

Figure 1

Figure 1

To compare the neurotoxicity of the investigated LAs, the concentrations that induced LD50 were calculated from a concentration-viability equation derived from extrapolation of curve response and are displayed in Table 1 and Figure 2. The LD50s of bupivacaine, lidocaine, prilocaine, mepivacaine, articaine, and ropivacaine was 0.95 ± 0.08, 3.35 ± 0.33, 4.32 ± 0.39, 4.84 ± 1.28, 8.98 ± 2.07, and 13.43 ± 0.61 mM, respectively. Variance analysis (ANOVA) followed by a Tukey-Kramer post hoc test revealed significant differences among the following groups: Group 1 = ropivacaine, articaine; Group 2 = mepivacaine, prilocaine, lidocaine; and Group 3 = bupivacaine (Figs. 2 and 3) (P value associated with Shapiro-Wilk test was 0.21 and Bartlett’s P value was 0.14). Ropivacaine’s LD50 was significantly different from those of mepivacaine, prilocaine, lidocaine, and bupivacaine (all P ≤ 0.009). Articaine’s LD50 was significantly different from lidocaine’s (P = 0.03), and bupivacaine’sLD50 (P < 0.001). Mepivacaine’s and bupivacaine’s LD50s were significantly different (P < 0.001). Prilocaine’s LD50 was significantly different from bupivacaine’s (P < 0.001). Lidocaine’s LD50 was significantly different compared with bupivacaine’s LD50 (P = 0.002). No significant difference was obtained between ropivacaine’s and articaine’s LD50s and between prilocaine’s, lidocaine’s, and mepivacaine’s LD50s (Fig. 3).

Table 1

Table 1

Figure 2

Figure 2

Figure 3

Figure 3

The comparison of LD50 normalized by LD50 lidocaine showed that articaine and ropivacaine present the least neurotoxic effects in SH-SY5Y cells compared with lidocaine (2.7 and 4 ratios, respectively). The toxic profiles of prilocaine and mepivacaine are approximately equivalent to lidocaine ratios (1.3 and 1.4, respectively), whereas bupivacaine has an unfavorable toxic profile compared with that of lidocaine (ratio of 0.3) (Table 1).

Representative morphological changes of SH-SY5Y cells at 20-minute exposure to LAs are presented in Figure 4. The shape of SH-SY5Y cells gradually became rounded with increased LA concentration applied; with 10 mM articaine, 3 mM lidocaine, and 2 mM bupivavaine, almost all cells looked rounded and many of the cells had detached from the bottom of the culture plate. These observations were consistent with the cell viability data from the WST-1 assay (Fig. 1).

Figure 4

Figure 4

Back to Top | Article Outline


All LAs in humans are associated with local reactions, even those widely used and well tolerated. The toxicity of LAs should be studied to improve our knowledge of relative local toxicity of LAs. In the present study, we demonstrated, in an in vitro model using a human neuroblastoma cell line, that all LAs have significant differences in their neurotoxic profile. Our variance analysis statistically distinguished 3 groups of LAs according to their range of intrinsic toxicity. Ropivacaine and articaine have the lowest toxicity; mepivacaine, prilocaine, and lidocaine have medium toxicity; and bupivacaine has the highest toxicity.

In our study, cell viability was determined on SH-SY5Y cells, which are a neural cell line widely used to investigate neurotoxicity of compounds.25,29 Easier to grow and to manipulate for such experiments, this cell line has similar sensitivity to toxicity than primary neuronal cells.30 Moreover, this cell line is also a validated in vitro model for neurotoxicology studies and testing.27

The diffusion of LAs in tissue and the amount distributed over time are affected by the concentration gradient, the affinity for lipids, pKa, and protein-binding ability. As the LAs diffuse into the peripheral nerve that is composed of tightly packed axons and protected by layers of fibrous, elastic tissues, and glial cells, LAs become increasingly diluted by tissue fluids. Hence, Kimi et al.31 demonstrated that the distribution of ropivacaine and lidocaine in the nerve reached a maximum immediately after injection (equivalent to 20%–40% of quantity of anesthetic in contact with palatal tissue), then rapidly decreased and became negligible (<5%) after 20 minutes in rat nerve.31 Consequently, a short and continuous exposure of cultured cells seems representative of a clinical situation to analyze LA neurotoxicity. Notably, contact with test items during 20 to 30 minutes is considered a short time exposure in neuroblastoma cultured cells. On the contrary, 24 hours of contact, which is a long exposure time, was not considered appropriate because of the pharmacokinetics of LAs. Therefore, cell viability was determined after a short exposure time in our model.

Several published studies on neuronal cells and other cell types reported different rankings of LA toxicity depending on the time of contact. Similar to previous studies, bupivacaine was the most toxic of the amide LAs tested in our experiment. Bupivacaine was observed to be more neurotoxic than mepivacaine,24,32 which was more toxic than articaine.23 Prilocaine and mepivacaine were also more toxic than articaine in SH-SY5Y cells than in SHEP cells.23 Furthermore, according to Werdehausen et al., our results confirmed that articaine has low toxicity on both types of neuroblastoma cells. However, there are discrepancies among studies concerning the toxicity of ropivacaine and prilocaine. Toxicity of ropivacaine was found intermediate compared with bupivacaine and lidocaine toxicities in SH-SY5Y cells by Perez-Castro et al. Similarly, Werderhausen et al. found that ropivacaine toxicity value was included between bupivacaine and lidocaine toxicity values in human neuroblastoma SHEP cells.33 However, ropivacaine was the least toxic LA in chondrocytes followed by mepivacaine and finally by bupivacaine.34 In our study, ropivacaine was less neurotoxic than mepivacaine and bupivacaine. Our experiment also found that prilocaine had the same neurotoxic profile as lidocaine in SH-SY5Y cells in contrast to results obtained in SHEP cells from Werderhausen et al. All these observed differences might be explained by the variability in the cell model, cell line subtype, and experimental conditions such as time of contact and culture conditions. Moreover, variability in sensitivity to certain neurotoxic molecules was previously described for SH-SY5Y cells.26

Our results were arbitrarily expressed as an LD50 ratio of each LA compared with lidocaine, which is the LA “gold standard,” the drug against which other LAs are compared.35 Each in vitro intrinsic molecule toxicity was relativized with lidocaine thanks to the LD50 ratios of ropivacaine/lidocaine, articaine/lidocaine, mepivacaine/lidocaine, prilocaine/lidocaine, and bupivacaine/lidocaine. Those were respectively equal to 4, 2.7, 1.4, 1.3, and 0.3. Of note, the higher the ratio, the lower the toxicity for the molecule being compared to lidocaine. However, the clinically relevant concentrations (CRC) associated with clinical practice must also be considered when evaluating the clinically equipotent profile of LAs. The CRC ratios indicate that ropivacaine/lidocaine, articaine/lidocaine, mepivacaine/lidocaine, prilocaine/lidocaine, and bupivacaine/lidocaine were equal to 0.2, 1.5, 0.9, 2.1, and 0.2, respectively.35,36 As a consequence, ropivacaine is 5-fold more potent and 4-fold less toxic than lidocaine. Articaine is 1.5-fold less potent than lidocaine and 2.7-fold less toxic than lidocaine. Thus ropivacaine and articaine have the most favorable safety profiles. Similarly, mepivacaine has a favorable safety profile compared with lidocaine (CRC ratio = 0.9; LD50 ratio = 1.4), while prilocaine has an unfavorable safety profile (CRC ratio = 2.1; LD50 ratio = 1.3) (Table 1). Although lidocaine is the LA gold standard, the molecule is classified as an intermediate-acting drug. CRC ratio could also be expressed according to the duration properties. In this situation, ropivacaine has a very favorable safety profile compared with bupivacaine. Distinct molecular and cellular mechanisms could explain the toxic extent of each LA. This includes their direct effects on neurons or glial cells such as Schwann cells and a secondary effect on the neural microenvironment. Inflammation or physical stress can also lead to nerve cell death associated with apoptosis and necrosis. Hence, the neurotoxicity of lidocaine involves mitochondrial injury and activation of apoptotic pathways in vitro37 and in vivo. Lidocaine can also promote neural damage in rats38 by oxidative stress, and bupivacaine can induce apoptosis pathways by reactive oxygen species production in neuronal cells39 and in Schwann cells.40 In contrast, articaine and prilocaine could be associated with a decrease in oxidative stress.41 Overall, LA activity is also related to the dissociation constant. It determines the percentage of LAs’ ionized state at the physiologic pH and LAs’ ability to penetrate into cells and onset. Moreover, LAs interact with membrane lipids’ modifying fluidity and protein conformation and influencing channel functions. Especially, a high lipid:water partition coefficient expressing the lipid solubility of LA and representing the diffusion through cell membranes is correlated to a high LA potency. Thus, bupivacaine and ropivacaine showed different toxicity profiles despite the same pKa (8.1) and a very different lipid:water partition coefficient (27.5 and 2.9, respectively) (Table 1). The toxicity differences between these 2 drugs were observed in other studies.21,42 Specific mechanisms of action in the toxicity potency of all of these molecules may be involved as previously described with lidocaine.43,44 In this study, articaine appeared to be one of the least neurotoxic LAs but also tended to induce an increase of metabolic activity measured when SH-SY5Y cells were exposed to the lowest doses tested. This drug might have a distinct mechanism of activity that could be related to its specific molecular structure and the presence of the thiophene group or its ability to scavenge free radical species at low concentrations.41 This metabolic activity increase has been reported in the presence of low concentrations of richlocaine and bupivacaine whereas these LAs were toxic at high concentrations.45,46

In vitro studies have the advantages of reproducibility and accuracy because they are creating a controlled environment; they provide the possibility to standardize and to compare capacities of different molecules and to evaluate their neurotoxicity mechanisms. Although there are some limitations such as potential genetic instability, human cell lines are often used in these investigations because of their origin and their ease of growing. The absence of different cell type interaction (glial-neuron) and specific environment (extracellular matrix, paracrine effect, etc.) can occasionally limit the homogeneous culture of a cell line. These limitations can be overcome when working with primary cultures even if these cultures cannot mimic important variables such as tissue diffusion, distribution, and absorption, which influence the local pharmacokinetics of the molecule tested. In addition, extrapolation to the clinical situation would also require considering the concomitant use of other drugs such as vasoconstrictors, which modulate contact time of LAs with the neural environment and potentially augment their toxicity. In conclusion, our results suggest that among LAs commonly used in dentistry, articaine and ropivacaine had the least neurotoxic effects in our model of SH-SY5Y cells after an exposure corresponding to the dental practice.

Back to Top | Article Outline


Name: Arnaud Malet, PhD.

Contribution: This author helped design the study, analyze data, and prepare the manuscript.

Attestation: Arnaud Malet approved the final manuscript, and attests to the integrity of the original study data and the analysis reported in this manuscript.

Conflicts of Interest: Arnaud Malet is an employee of Septodont.

Name: Marie-Odile Faure, PhD.

Contribution: This author helped design the study, analyze data, and prepare the manuscript.

Attestation: Marie-Odile Faure approved the final manuscript, and attests to the integrity of the original study data and the analysis reported in this manuscript.

Conflicts of Interest: Marie-Odile Faure is an employee of Septodont.

Name: Nathalie Deletage, PhD.

Contribution: This author helped conduct the study, collect data, and revise the manuscript.

Attestation: Nathalie Deletage approved the final manuscript, and attests to the integrity of the original study data and the analysis reported in this manuscript. Nathalie Deletage is the archival author.

Conflicts of Interest: Nathalie Deletage is an employee of Neuronax, which is under contract with Septodont.

Name: Bruno Pereira, PhD.

Contribution: This author performed the statistical analysis.

Attestation: Bruno Pereira approved the final manuscript.

Conflicts of Interest: This author has no conflicts of interest to declare.

Name: Jerome Haas.

Contribution: This author helped design the study, and revise the manuscript.

Attestation: Jerome Haas approved the final manuscript.

Conflicts of Interest: Jerome Haas is an employee of Septodont.

Name: Gregory Lambert, PharmD, PhD.

Contribution: This author helped design the study, and revise the manuscript.

Attestation: Gregory Lambert approved the final manuscript.

Conflicts of Interest: Gregory Lambert is an employee of Septodont.

This manuscript was handled by: Marcel E. Durieux, MD, PhD.

Back to Top | Article Outline


1. Malamed SF, Falkel M. Advances in local anesthetics: pH buffering and dissolved CO2. Dent Today. 2012;31:88–93
2. Moore PA, Hersh EV. Local anesthetics: pharmacology and toxicity. Dent Clin North Am. 2010;54:587–99
3. Gordon SM, Mischenko AV, Dionne RA. Long-acting local anesthetics and perioperative pain management. Dent Clin North Am. 2010;54:611–20
4. Haas DA. An update on local anesthetics in dentistry. J Can Dent Assoc. 2002;68:546–51
5. Haas DA. Localized complications from local anesthesia. J Calif Dent Assoc. 1998;26:677–82
6. Miller PA, Haas DA. Incidence of local anaesthetics-induced neuropathies in Ontario from 1994–1998. J Dent Res. 2000;79:627
7. Ready LB, Plumer MH, Haschke RH, Austin E, Sumi SM. Neurotoxicity of intrathecal local anesthetics in rabbits. Anesthesiology. 1985;63:364–70
8. Hodgson PS, Neal JM, Pollock JE, Liu SS. The neurotoxicity of drugs given intrathecally (spinal) Anesth Analg. 1999;88:797–809
9. Kanai Y, Katsuki H, Takasaki M. Graded, irreversible changes in crayfish giant axon as manifestations of lidocaine neurotoxicity in vitro. Anesth Analg. 1998;86:569–73
10. Gentili F, Hudson AR, Hunter D, Kline DG. Nerve injection injury with local anesthetic agents: a light and electron microscopic, fluorescent microscopic, and horseradish peroxidase study. Neurosurgery. 1980;6:263–72
11. Myers RR, Kalichman MW, Reisner LS, Powell HC. Neurotoxicity of local anesthetics: altered perineurial permeability, edema, and nerve fiber injury. Anesthesiology. 1986;64:29–35
12. Adams HJ, Mastri AR, Eicholzer AW, Kilpatrick G. Morphologic effects of intrathecal etidocaine and tetracaine on the rabbit spinal cord. Anesth Analg. 1974;53:904–8
13. Li DF, Bahar M, Cole G, Rosen M. Neurological toxicity of the subarachnoid infusion of bupivacaine, lignocaine or 2-chloroprocaine in the rat. Br J Anaesth. 1985;57:424–9
14. Hillerup S, Bakke M, Larsen JO, Thomsen CE, Gerds TA. Concentration-dependent neurotoxicity of articaine: an electrophysiological and stereological study of the rat sciatic nerve. Anesth Analg. 2011;112:1330–8
15. Kasaba T, Onizuka S, Takasaki M. Procaine and mepivacaine have less toxicity in vitro than other clinically used local anesthetics. Anesth Analg. 2003;97:85–90
16. Radwan IA, Saito S, Goto F. The neurotoxicity of local anesthetics on growing neurons: a comparative study of lidocaine, bupivacaine, mepivacaine, and ropivacaine. Anesth Analg. 2002;94:319–24
17. Yang S, Abrahams MS, Hurn PD, Grafe MR, Kirsch JR. Local anesthetic Schwann cell toxicity is time and concentration dependent. Reg Anesth Pain Med. 2011;36:444–51
18. Onizuka S, Yonaha T, Tamura R, Kasiwada M, Shirasaka T, Tsuneyoshi I. Lidocaine depolarizes the mitochondrial membrane potential by intracellular alkalization in rat dorsal root ganglion neurons. J Anesth. 2011;25:229–39
19. Park J, Sutradhar BC, Hong G, Choi SH, Kim G. Comparison of the cytotoxic effects of bupivacaine, lidocaine, and mepivacaine in equine articular chondrocytes. Vet Anaesth Analg. 2011;38:127–33
20. Baker JF, Mulhall KJ. Local anaesthetics and chondrotoxicity: what is the evidence? Knee Surg Sports Traumatol Arthrosc. 2012;20:2294–301
21. Piper SL, Laron D, Manzano G, Pattnaik T, Liu X, Kim HT, Feeley BT. A comparison of lidocaine, ropivacaine and dexamethasone toxicity on bovine tenocytes in culture. J Bone Joint Surg Br. 2012;94:856–62
22. Cai XY, Xiong LM, Yang SH, Shao ZW, Xie M, Gao F, Ding F. Comparison of toxicity effects of ropivacaine, bupivacaine, and lidocaine on rabbit intervertebral disc cells in vitro. Spine J. 2014;14:483–90
23. Werdehausen R, Fazeli S, Braun S, Hermanns H, Essmann F, Hollmann MW, Bauer I, Stevens MF. Apoptosis induction by different local anaesthetics in a neuroblastoma cell line. Br J Anaesth. 2009;103:711–8
24. Perez-Castro R, Patel S, Garavito-Aguilar ZV, Rosenberg A, Recio-Pinto E, Zhang J, Blanck TJ, Xu F. Cytotoxicity of local anesthetics in human neuronal cells. Anesth Analg. 2009;108:997–1007
25. Sanfeliu C, Cristòfol R, Torán N, Rodríguez-Farré E, Kim SU. Use of human central nervous system cell cultures in neurotoxicity testing. Toxicol In Vitro. 1999;13:753–9
26. Gustafsson H, Runesson J, Lundqvist J, Lindegren H, Axelsson V, Forsby A. Neurofunctional endpoints assessed in human neuroblastoma SH-SY5Y cells for estimation of acute systemic toxicity. Toxicol Appl Pharmacol. 2010;245:191–202
27. Bal-Price AK, Suñol C, Weiss DG, van Vliet E, Westerink RH, Costa LG. Application of in vitro neurotoxicity testing for regulatory purposes: symposium III summary and research needs. Neurotoxicology. 2008;29:520–31
28. Regulska M, Pomierny B, Basta-Kaim A, Starek A, Filip M, Lasoń W, Budziszewska B. Effects of ethylene glycol ethers on cell viability in the human neuroblastoma SH-SY5Y cell line. Pharmacol Rep. 2010;62:1243–9
29. Forsby A, Bal-Price AK, Camins A, Coecke S, Fabre N, Gustafsson H, Honegger P, Kinsner-Ovaskainen A, Pallas M, Rimbau V, Rodríguez-Farré E, Suñol C, Vericat JA, Zurich MG. Neuronal in vitro models for the estimation of acute systemic toxicity. Toxicol In Vitro. 2009;23:1564–9
30. Sanfeliu C, Sebastià J, Ki SU. Methylmercury neurotoxicity in cultures of human neurons, astrocytes, neuroblastoma cells. Neurotoxicology. 2001;22:317–27
31. Kimi H, Yamashiro M, Hashimoto S. The local pharmacokinetics of 3H-ropivacaine and 14C-lidocaine after maxillary infiltration anesthesia in rats. Anesth Prog. 2012;59:75–81
32. Onizuka S, Tamura R, Hosokawa N, Kawasaki Y, Tsuneyoshi I. Local anesthetics depolarize mitochondrial membrane potential by intracellular alkalization in rat dorsal root ganglion neurons. Anesth Analg. 2010;111:775–83
33. Cai XY, Xiong LM, Yang SH, Shao ZW, Xie M, Gao F, Ding F. Comparison of toxicity effects of ropivacaine, bupivacaine, and lidocaine on rabbit intervertebral disc cells in vitro. Spine J. 2014;14:483–90
34. Breu A, Rosenmeier K, Kujat R, Angele P, Zink W. The cytotoxicity of bupivacaine, ropivacaine, and mepivacaine on human chondrocytes and cartilage. Anesth Analg. 2013;117:514–22
35. Malamed SF Handbook of local anesthesia. 20116th ed. St. Louis Mosby
36. Ciechanowicz S, Patil V. Lipid emulsion for local anesthetic systemic toxicity. Anesthesiol Res Pract. 2012;2012:131784
37. Johnson ME, Uhl CB, Spittler KH, Wang H, Gores GJ. Mitochondrial injury and caspase activation by the local anesthetic lidocaine. Anesthesiology. 2004;101:1184–94
38. Cano-Europa E, López-Galindo GE, Hernández-García A, Blas-Valdivia V, Gallardo-Casas CA, Vargas-Lascari M, Ortiz-Butrón R. Lidocaine affects the redox environment and the antioxidant enzymatic system causing oxidative stress in the hippocampus and amygdala of adult rats. Life Sci. 2008;83:681–5
39. Lu J, Xu SY, Zhang QG, Lei HY. Bupivacaine induces reactive oxygen species production via activation of the AMP-activated protein kinase-dependent pathway. Pharmacology. 2011;87:121–9
40. Park CJ, Park SA, Yoon TG, Lee SJ, Yum KW, Kim HJ. Bupivacaine induces apoptosis via ROS in the Schwann cell line. J Dent Res. 2005;84:852–7
41. Günaydin B, Demiryürek AT. Effects of prilocaine and articaine on human leucocytes and reactive oxygen species in vitro. Acta Anaesthesiol Scand. 2001;45:741–5
42. Graf BM, Abraham I, Eberbach N, Kunst G, Stowe DF, Martin E. Differences in cardiotoxicity of bupivacaine and ropivacaine are the result of physicochemical and stereoselective properties. Anesthesiology. 2002;96:1427–34
43. Lirk P, Haller I, Colvin HP, Lang L, Tomaselli B, Klimaschewski L, Gerner P. In vitro, inhibition of mitogen-activated protein kinase pathways protects against bupivacaine- and ropivacaine-induced neurotoxicity. Anesth Analg. 2008;106:1456–64
44. Lirk P, Haller I, Colvin HP, Frauscher S, Kirchmair L, Gerner P, Klimaschewski L. In vitro, lidocaine-induced axonal injury is prevented by peripheral inhibition of the p38 mitogen-activated protein kinase, but not by inhibiting caspase activity. Anesth Analg. 2007;105:1657–64
45. Zadorozhnyi AV, Popkov VL, Galenko-Yaroshevskii PA, Semeikin AV. Effect of richlocaine on proliferative activity of osteoblasts and intracellular calcium content in rats. Bull Exp Biol Med. 2003;136:375–6
46. Fedder C, Beck-Schimmer B, Aguirre J, Hasler M, Roth-Z’graggen B, Urner M, Kalberer S, Schlicker A, Votta-Velis G, Bonvini JM, Graetz K, Borgeat A. In vitro exposure of human fibroblasts to local anaesthetics impairs cell growth. Clin Exp Immunol. 2010;162:280–8
© 2015 International Anesthesia Research Society