Secondary Logo

Journal Logo

Basic science

Sensitivities of rat primary sensory afferent nerves to magnesium: implications for differential nerve blocks

Vastani, Nisha; Seifert, Burkhardt; Spahn, Donat R.; Maurer, Konrad

Author Information
European Journal of Anaesthesiology: January 2013 - Volume 30 - Issue 1 - p 21-28
doi: 10.1097/EJA.0b013e32835949ab
  • Free



Recent studies have investigated the role of magnesium as an adjuvant to local anaesthetics in regional anaesthesia. Clinical trials in humans, as well as in-vivo animal studies, have reported that when magnesium is used as an additive, it can prolong the duration of regional anaesthesia and provide better postoperative analgesia.1–7

However, contrasting findings have also been published. In a study investigating spinal analgesia in rats, intrathecal administration of magnesium sulphate (MgSO4) caused motor paralysis but not complete analgesia.8 In an animal model of acute pain, magnesium failed to display any antinociceptive properties.9 A recent study by Hung et al.10 demonstrated that MgSO4 shortened the duration of rat sciatic nerve block achieved by lidocaine in vivo, although whole-cell patch clamp recordings in the same study revealed contrasting findings.

The present study was, therefore, undertaken to clarify and further investigate the role of magnesium as an adjuvant to a local anaesthetic in the peripheral nervous system. The effects of applying MgSO4 alone or in combination with lidocaine directly on the saphenous nerve were investigated in myelinated Aβ and unmyelinated C fibre compound action potentials (CAPs) in a rat in-vitro model. Threshold tracking techniques were used to provide a sensitive measure of the changes in excitability induced by MgSO4, lidocaine and a combination of both.


Preparation of rat saphenous nerve and experimental setup

All the animal work was performed in accordance with the Swiss Animal Protection Act 2008 and was approved on 14 December 2007 by the Gesundheitsdirektion des Kantons, Zurich, Switzerland (217/2007, Dr R. Vogel).

Experiments were carried out on a total of 30 isolated saphenous nerves. Male Wistar rats between 12 and 16 weeks old were killed with carbon dioxide and the saphenous nerves were excised from the hind limbs with innervated skin. Each preparation was positioned corium-side up in an organ bath superfused with synthetic interstitial fluid (SIF): NaCl 108 mmol l−1, KCl 3.5 mmol l−1, MgSO4 0.7 mmol l−1, NaH2PO4 1.7 mmol l−1, CaCl2 1.5 mmol l−1, sodium gluconate 9.6 mmol l−1, glucose 5.0 mmol l−1 and sucrose 7.6 mmol l−1.11 The SIF was continuously gassed with a mixture of 95% oxygen and 5% carbon dioxide (carbogen) and the temperature was maintained at 32°C. The proximal end of the nerve was placed in a separate recording chamber isolated by paraffin oil in which it was desheathed and positioned on a gold wire electrode under a microscope (Fig. 1a). The distal end was cut, desheathed and drawn into a suction electrode within a self-sealing metal ring to elicit CAPs. Nerves were desheathed to ensure that the surrounding connective tissue did not act as a diffusion barrier which could otherwise prevent the drugs from reaching the axons. Drugs were applied within the metal ring using a circulating system which was independent from the rest of the organ bath, and which was also maintained at 32°C.


All recordings began with a control period lasting at least 10 min, during which only SIF was perfused into the metal ring; SIF containing a drug was then applied for 10 min, followed by washout with SIF alone. Subsequently, further drug solutions were applied, separated by washout periods. Washout periods lasted for 12 to 20 min (Figs 3a and 4a). Recordings started only after the nerve had been in the organ bath for an equilibration period of 1.5 h following the initial dissection.

A stock solution of 1 mmol l−1 lidocaine was made up from lidocaine hydrochloride monohydrate with a molecular weight of 288.81 g mol−1 (Kantonsapotheke Zurich, Zurich, Switzerland). A working solution of 80 μmol l−1 lidocaine was made up in SIF on the day of the experiment. A working solution of 10 mmol l−1 MgSO4 was made up using the molecular weight of the heptahydrate (MgSO4.7H20, Kantonsapotheke Zurich) which is 246.47 g mol−1. For the mixture, double the concentrations of lidocaine and MgSO4 solutions were made and equal parts were then mixed together.

Axonal excitability was measured using the QTRAC program (QtracP version 3/4/2009, Institute of Neurology, University College London, London, UK) which is suitable for investigating electrophysiological properties of both human nerves in vivo12 and rat nerves in vitro.13 QTRAC was used to record CAPs from peripheral A and C fibres, generate the waveform of the stimulus and display the results. The nerves were stimulated using an isolated linear constant current stimulator (Linear Stimulus Isolator A395, World Precision Instruments, Hertfordshire, UK) with a maximal output of 1 mA. The evoked CAPs were recorded with an isolated bio-amplifier (ISO-80, World Precision Instruments) from the proximal nerve ending using the following settings: low-pass filter 2 kHz; high-pass filter 2 Hz; gain × 1000. Data were then digitised by a data acquisition unit (CED micro1401 MK II, Cambridge Electronic Design Ltd, Cambridge, UK) using a sampling rate of 10 kHz. Fast-conducting Aβ fibres and slowly conducting C fibres were separated by adjusting the stimulus strength and duration. Aβ fibre CAPs were stimulated at a fixed rate of 1.25 Hz with 0.1 ms current pulses, whereas C fibre CAPs were stimulated at a fixed rate of 0.25 Hz with 1 ms current pulses.

Three stimulus conditions were tested in succession using three channels of the QTRAC program (Fig. 1b). First, a stimulus–response curve was recorded by increasing the stimulus current until a maximum sensory nerve action potential was obtained. The peak-to-peak amplitude of the CAP response to supramaximal stimulation was recorded on Channel 1. Second, to investigate the effects of the drugs on the control threshold, the current required to elicit a CAP with amplitude of 40% of the maximal response (Channel 1) was defined as the target response and recorded on Channel 2. The test stimulus current required to maintain this response (referred to as ‘threshold current’) was adjusted by being tracked in a feedback-controlled manner throughout the experiment by the QTRAC program. To investigate the mechanism of the induced excitability changes, a slow polarising ramp current was applied prior to a normal test stimulus. The use of such depolarising subthreshold currents has previously been shown to cause inactivation of sodium channels14 and to be useful in studying the function of axonal voltage-gated sodium channels in vitro.15

The first set of experiments (n = 10) was designed to investigate the role of preconditioning subthreshold polarising currents and whether they alter the drug-induced excitability changes. Channel 3 used the same protocol as Channel 2, but a 300 ms polarising current ramp preceded the test stimulus and was varied in steps from a hyperpolarising current (−10%) to depolarising currents (Aβ 400%; C 120%), at defined fractions of the threshold current on Channel 2. This polarising current was applied first in the presence of normal bath solution, then again with 80 μmol l−1 lidocaine and finally in the presence of 10 mmol l−1 MgSO4.

In the second set of experiments, nerve block induced by MgSO4 alone, lidocaine alone or a combination (all experiments were carried out in this sequence) were investigated in the presence of a test stimulus or using a preconditioning depolarising ramp current of set strength. In both Aβ (Fig. 3a) and C (Fig. 4a) fibres, the preconditioning current which produced the lowest point (nadir) on the ‘U’ shape was chosen as ramp current. The nadir represents the membrane poential (after a given preconditioning subthreshold stimulus) at which the net effect of voltage-gated ion channels results in maximal excitability. In Aβ fibres, this point of maximal excitability was reached with a polarising current which was 110% of the unconditioned threshold current, whereas C fibres only required a current which was 10% of threshold. Using this approach, we expected to see a maximal effect of the drugs on excitability thresholds.

Channel 3 used the same protocol as in the first set, but the 300 ms preconditioning ramp stimulus was only depolarising and fixed to stimulate at 10% of the threshold current for the C fibre recordings (n = 10) and at 110% of the threshold current for Aβ fibre recordings (n = 10), respectively. Experiments were then carried out to investigate the action of 10 mmol l−1 MgSO4, 80 μmol l−1 lidocaine and a mixture of both. Given the stability of the recordings over time, we compared all the threshold changes with the initial control threshold values.

Statistical analysis

The recordings were computed using the QTRAC software. All results were expressed as mean (SD) except in Figs 2a and 2b, 3a and 4a in which error bars are SEM. Statistical analysis was performed using Stata 11.2 (StataCorp, College Station, Texas, USA). Absolute values were used except for the comparison of different fibre types (Aβ versus C fibre CAP) for which data were normalised before analysis. To compare threshold values of different drugs, individual differences within fibres were computed and compared to 0 or between fibre types. To address clustering within animals, a robust sandwich estimator for standard errors allowing for intragroup correlation [Stata command regress with option vce (cluster animal)] was used. Bonferroni correction was used to address the issue of multiple comparisons between conditions. According to the Bonferroni correction, two-tailed P values of 0.008 or less were considered significant.



We used a total of 19 rats. In every rat, both saphenous nerves were explanted and used for experiments. In eight animals, only one nerve could be used (for technical reasons) and in 11 animals, both nerves could be used. The pH estimates of the working solutions were similar (SIF 7.42; MgSO4 10 mmol l−1 7.42; lidocaine 80 μmol l−1 7.43; mixture 7.43). The preparation was stable over time with regard to the necessary injected control threshold current and the peak size of the CAP after supramaximal stimulation (Fig. 1c). Our results showed that both MgSO4 and lidocaine, whether applied separately or combined, resulted in a reduction in axonal excitability (i.e. an increase in the ‘threshold’ current required to evoke a CAP response which was 40% of maximal).

In Aβ fibres (n = 10), 10 mmol l−1 MgSO4 increased the excitability threshold and it exceeded the control threshold [SIF 0.273 (SD 0.064) mA versus MgSO4 0.413 (SD 0.086) mA, P < 0.001) (Fig. 3b). Lidocaine 80 μmol l−1 not only increased the threshold less than 10 mmol l−1 MgSO4 but also to well above the control SIF threshold [lidocaine 0.358 (SD 0.080) mA, P = 0.001]. The mixture of both did not further increase the threshold current significantly more than MgSO4 alone [mixture 0.470 (SD 0.105) mA, P = 0.012], but the threshold was significantly higher than with lidocaine alone [lidocaine 0.358 (SD 0.080) mA, P < 0.001].

In C fibres (n = 10), both 10 mmol l−1 MgSO4 and 80 μmol l−1 lidocaine significantly increased the excitability thresholds above control threshold [SIF 2.002 (SD 0.562) mA; MgSO4 2.249 (SD 0.629) mA, P = 0.001; lidocaine 2.385 (SD 0.656 mA, P < 0.001] (Fig. 4b). The combination of lidocaine and MgSO4 [mixture 2.531 (SD 0.752)] further enhanced the effects of both MgSO4 (P = 0.001) and lidocaine (P = 0.008).

Differential sensitivities of Aβ and C fibres

Under control conditions (test stimulus only; Channel 2), all the drugs tested reduced the excitability of Aβ fibres (n = 10) more than C fibres (n = 10), as shown by greater percentage increases in electrical thresholds: 10 mmol l−1 MgSO4 [Aβ 150% (SD 12) versus C 112% (SD 4), P < 0.001], 80 μmol l−1 lidocaine [Aβ: 131% (SD 17) versus C 119% (SD 5), P = 0.001] and the mixture [Aβ 172% (SD 25) versus C 126% (SD 5), P < 0.001].

Preconditioning ramp currents enhanced magnesium-induced threshold increases in Aβ fibres but not in C fibres

In the presence of bath solution only, a ‘U’ shaped relationship was observed for the strength of preconditioning ramp applied and the membrane threshold in both Aβ (Fig. 2a; see Figure 11 in the article by Bostock and Grafe16) and C (Fig. 2b) fibres. Weak depolarisation resulted in an increase in membrane excitability. Applying stronger depolarisation had the opposite effect, and the excitability increase was reversed. This indicated that the use of such strong depolarising ramp currents could alone produce a threshold increase in the absence of any drug by inactivation of sodium currents.14 In the presence of 80 μmol l−1 lidocaine, applying weak depolarising ramp currents (Aβ up to 100% and C 5%, as percentages of threshold current) resulted in greater reductions in excitability as compared with the application of a test stimulus alone. This is indicated by the upward (higher threshold currents) and leftward (smaller size of preconditioning current) shift in the ‘U’-shaped profile in both Aβ and C fibres. Applying preconditioning ramp currents in the presence of 10 mmol l−1 MgSO4 resulted in differential effects in Aβ and C fibres. In Aβ fibres, a leftward and upward shift in the ‘U’-shaped profile was observed for strong depolarisation currents (>90% threshold current; Fig. 2a). In C fibres, addition of MgSO4 shifted the ‘U’-shaped profile slightly towards higher threshold currents at small preconditioning ramp currents (Fig. 2b).

Using a 110% conditioning ramp current in Aβ fibres, 10 mmol l−1 MgSO4 did elevate the excitability threshold, but it never exceeded the control threshold [SIF 0.137 (SD 0.041) mA versus MgSO4 0.204 (SD 0.056) mA, P < 0.001] (Fig. 3c). In contrast, 80 μmol l−1 lidocaine increased the threshold well above the control threshold [lidocaine 0.543 (SD 0.315) mA, P < 0.001] and the mixture of both [mixture 0.620 (SD 0.281) mA] induced even further increases in threshold currents which were higher than the individual drugs alone (MgSO4, P = 0.006; lidocaine, P = 0.005).

After a 10% conditioning ramp current in C fibres, 10 mmol l−1 MgSO4 did not elevate the excitability thresholds above control threshold [SIF 1.697 (SD 0.481) mA versus MgSO4 1.819 (SD 0.489) mA, P < 0.001] (Fig. 4c). The 80 μmol l−1 lidocaine [lidocaine 2.461 (SD 0.693) mA, P < 0.001] and the mixture [mixture 2.412 (SD 0.641) mA] produced increases in thresholds above the control and exceeded that of MgSO4 alone (P < 0.001). However, compared to the effect on Aβ fibres, the effect of the mixture was less than the increase in threshold produced by lidocaine alone, leading to a small decrease in threshold which was not significant (P = 0.17).


Our results confirm earlier evidence that MgSO4 alone can reduce the excitability of peripheral nerve fibres in vitro.17 We also found that the addition of MgSO4 to the local anaesthetic lidocaine enhances its effects in reducing nerve excitability in Aβ fibres. These results are, therefore, in agreement with previous clinical studies which support the use of magnesium in combination with a local anaesthetic to enhance the blocking effect.5,7 However, the study also reveals that the effects of MgSO4 on excitability depend on the type of fibre and on the membrane potential. A possible explanation is that Mg2+ binding is dependent on the different types of voltage-gated sodium channels and their conformational states.

Limitations of the interpretation of the results

Research on peripheral nerves over recent decades has shown that the expression of ion channels is much more complex than that proposed by Hodgkin and Huxley.18 Voltage clamp experiments have identified the complexities of ionic movements, defining membrane potentials and our interpretations are always simplified.19 Today, the most sophisticated computerised membrane models of peripheral nerve consider up to 30 parameters defining membrane potential.20 In our study, we used a technique which provides general information about nerve excitability derived from the size and the threshold current of CAPs. One of the most important factors defining the generation and the size of the action potential is the sodium channel.21 The following discussion therefore mainly focuses on this aspect of the topic.

Myelinated A fibres are more sensitive to magnesium sulphate than unmyelinated C fibres

Voltage-gated ion channels have diverse properties and a varied expression pattern amongst primary sensory neurones. The Aβ fibre CAP recorded in our experiments was produced by stimulating myelinated Aβ fibres. The subtype of sodium channel primarily responsible for this action potential is NaV1.6 which mediates tetrodotoxin (TTX)-sensitive currents.22,23 Our experiments suggest that NaV1.6 is blocked to a considerable degree by MgSO4 at the concentration used. These results are in line with previous patch-clamp experiments which showed that TTX-sensitive sodium channels display a concentration-dependent decrease in sodium current on administration of MgSO4.24

In contrast to Aβ fibres, C fibres are characterised by different subtypes of sodium channels mediating TTX-resistant currents: NaV1.8 and NaV1.9.25 To our knowledge, no study so far has specifically investigated the effect of magnesium on TTX-resistant sodium channels. The results of our C fibre CAP recordings suggest that the subpopulation of voltage-gated sodium channels expressed primarily on these fibres are not as sensitive to MgSO4 as those expressed on Aβ fibres. The reasons for this differential action are as yet unclear. The mechanism of action of MgSO4 on sodium channels is thought to differ radically from that of local anaesthetics. Divalent cations, such as magnesium, do not block sodium currents within the pore but have a strong effect on the gating properties of sodium channels. This effect is usually explained by the ‘surface charge’ theory which hypothesises that divalent cations neutralise the negative charges at the surface of the nerve membrane which affect the voltage-sensitive ion channels and, thereby, alter their electrical threshold.26,27 From patch-clamp studies, we know that gating properties vary considerably among subtypes of sodium channels. Parameters such as rate of inactivation or recovery from inactivation strongly depend on these gating properties and they vary within different subtypes.28,29 Our results suggest, therefore, that the differential blocking effect of MgSO4 is also dependent on these distinct gating properties.

Preconditioning ramp currents enhance blocking effects of lidocaine in Aβ and C fibres

It is well established that the activity of a local anaesthetic depends on the conformational state of the sodium channel. As lidocaine binds more efficiently to open and inactivated sodium channels, the nerve block induced by lidocaine is enhanced when the channels are in one of those states compared with the resting state.30 Accordingly, we found in the present study that lidocaine-induced block was greater in both Aβ and C fibres when preconditioning ramp currents were applied. The ‘U’ shape of the relationship between threshold change and strength of the depolarising preconditioning ramp stimulus shifted in a very characteristic way to higher thresholds and smaller preconditioning ramp currents. This may reflect a combination of two effects. First, opening sodium channels with a small depolarising ramp current facilitates access of the local anaesthetic to the pore and enhances its blocking effect (use-dependent block). Second, stronger depolarising preconditioning ramp currents lead to more sodium channels being in the inactivated state in which they bind more strongly to the lidocaine. Our results are, therefore, in line with the previously described dependency of lidocaine block on conformational state.31

Preconditioning ramp currents with lidocaine enhance differential blocking effects of magnesium sulphate

As MgSO4 acts on sodium channels in a quite different way to lidocaine, its action is not enhanced by the preconditioning ramp currents in the same way. In Fig. 2, it can be seen that MgSO4 had little effect on the ‘U’-shaped relationships between threshold and polarising current, the only significant differences occurring with very depolarised Aβ fibres. In both Aβ and C fibres, the threshold was increased by 10 mmol l−1 MgSO4 alone by about the same proportion whether the fibres were depolarised or not. Interestingly, however, whereas MgSO4 enhanced the effects of lidocaine on depolarised Aβ fibres (Fig. 3c), it reduced the effect of lidocaine on depolarised C fibres (Fig. 4c). This finding implies that, although MgSO4 enhances use-dependent block by lidocaine in peripheral Aβ fibres, this effect should not be relied upon for peripheral C fibres mediating pain.

Clinical implications

The results presented in this study have shown that the application of MgSO4 has a differential nerve effect by reducing the excitability of sensory Aβ fibres more than C fibres. When combined with lidocaine, the excitability reductions in both sensory Aβ and C fibres were stronger than those achieved by lidocaine alone; thus, magnesium enhances the effect. When applying preconditioning ramp currents in the presence of 10 mmol l−1 MgSO4, the differential effects were even more pronounced. So far, there is no clinical evidence for a differential block by MgSO4. The few clinical studies which have investigated analgesia in patients have limitations due to patient selection and defined outcome measures. The study by Karasawa et al.8 showing that magnesium sulphate blocked largely myelinated motor fibres without affecting pain pathways suggests the possibility of differential sensitivities in neurones.8 The results of Hung et al.10 also imply different sensitivities for large myelinated and unmylinated fibres, but again there are limitations due to the use of different models in the same study.


These results suggest that the binding of Mg2+ depends on both the type and conformational state of voltage-gated sodium channels. They may also help to explain the conflicting reports regarding the clinical effects of MgSO4 as an additive in peripheral nerve blocks.


Assistance with the study: The authors thank Professor Hugh Bostock (MRC Centre for Neuromuscular Diseases, National Hospital for Neurology and Neurosurgery, London, UK) for providing the stimulation protocol and for his input to the manuscript.

Financial support and sponsorship: This study was funded by the Swiss Foundation for Anaesthesia Research (SFAR), by the Swiss National Science Foundation (grant number SPUM 33CM30_124117) and the Institute of Anaesthesiology, University Hospital Zurich, Zurich, Switzerland.

Conflicts of interest: DRS was the chairman of the ABC Faculty and is a member of the ABC Trauma Faculty which are both managed by Thomson Physicians World GmbH, Mannheim, Germany and sponsored by an unrestricted educational grant from Novo Nordisk A/S, Bagsvärd, Denmark and CSL Behring GmbH, Hattersheim am Main, Germany. In the past 5 years, DRS has received honoraria or travel support for consulting or lecturing from the following companies: Abbott AG; Baar, Switzerland; AstraZeneca AG, Zug, Switzerland; Bayer (Schweiz) AG, Zurich, Switzerland; Baxter S.p.A., Rome, Italy; B. Braun Melsungen AG, Melsungen, Germany; Boehringer Ingelheim (Schweiz) GmbH, Basel, Switzerland; Bristol-Myers-Squibb, Rueil-Malmaison Cedex, France; CSL Behring GmbH, Hattersheim am Main, Germany and Bern, Switzerland; Curacyte AG, Munich, Germany; Ethicon Biosurgery, Sommerville, New Jersey, USA; Fresenius SE, Bad Homburg v.d.H., Germany; Galenica AG, Bern, Switzerland (including Vifor SA, Villars-sur-Glâne, Switzerland); GlaxoSmithKline GmbH & Co. KG, Hamburg, Germany; Janssen-Cilag AG, Baar, Switzerland; Janssen-Cilag EMEA, Beerse, Belgium; Merck Sharp&Dohme-Chibret AG, Opfikon-Glattbrugg, Switzerland; Novo Nordisk A/S, Bagsvärd, Denmark; Octapharma AG, Lachen, Switzerland; Organon AG, Pfäffikon/SZ, Switzerland; Oxygen Biotherapeutics, Costa Mesa, California, USA; Pentapharm GmbH (now Tem Innovations GmbH), Munich, Germany; Roche Pharma (Schweiz) AG, Reinach, Switzerland; and Schering-Plough International, Inc., Kenilworth, New Jersey, USA.

KM has received travel support for consulting or lecturing from the following companies: Pfizer AG, Zurich, Switzerland; Bristol-Myers Squibb SA, Baar, Switzerland; Mundipharma Medical Company, Basel, Switzerland; Janssen-Cilag AG, Baar, Switzerland; UCB, Bulle, Switzerland; Medtronic, Bern, Switzerland; B. Braun Medical AG, Sempach, Switzerland; Grünenthal Pharma Schweiz, Mitlödi; Switzerland; St Jude Medical AG, Zurich, Switzerland.

NV and BS have no conflicts of interest to declare.


1. Bilir A, Gulec S, Erkan A, Ozcelik A. Epidural magnesium reduces postoperative analgesic requirement. Br J Anaesth 2007; 98:519–523.
2. Cizmeci P, Ozkose Z. Magnesium sulfate as an adjuvant to total intravenous anesthesia in septorhinoplasty: a randomized controlled study. Aesthetic Plast Surg 2007; 31:167–173.
3. Brill S, Sedgwick PM, Hamann W, Di Vadi PP. Efficacy of intravenous magnesium in neuropathic pain. Br J Anaesth 2002; 89:711–714.
4. Telci L, Esen F, Akcora D, et al. Evaluation of effects of magnesium sulphate in reducing intraoperative anaesthetic requirements. Br J Anaesth 2002; 89:594–598.
5. Tramer MR, Glynn CJ. Magnesium Bier's block for treatment of chronic limb pain: a randomised, double-blind, cross-over study. Pain 2002; 99:235–241.
6. McCarthy RJ, Kroin JS, Tuman KJ, et al. Antinociceptive potentiation and attenuation of tolerance by intrathecal co-infusion of magnesium sulfate and morphine in rats. Anesth Analg 1998; 86:830–836.
7. Turan A, Memis D, Karamanlioglu B, et al. Intravenous regional anesthesia using lidocaine and magnesium. Anesth Analg 2005; 100:1189–1192.
8. Karasawa S, Ishizaki K, Goto F. The effect of intrathecal administration of magnesium sulphate in rats. Anaesthesia 1998; 53:879–886.
9. Ishizaki K, Sasaki M, Karasawa S, et al. The effect of intrathecal magnesium sulphate on nociception in rat acute pain models. Anaesthesia 1999; 54:241–246.
10. Hung YC, Chen CY, Lirk P, et al. Magnesium sulfate diminishes the effects of amide local anesthetics in rat sciatic-nerve block. Reg Anesth Pain Med 2007; 32:288–295.
11. Bretag AH. Synthetic interstitial fluid for isolated mammalian tissue. Life Sci 1969; 8:319–329.
12. Bostock H, Cikurel K, Burke D. Threshold tracking techniques in the study of human peripheral nerve. Muscle Nerve 1998; 21:137–158.
13. Maurer K, Bostock H, Koltzenburg M. A rat in vitro model for the measurement of multiple excitability properties of cutaneous axons. Clin Neurophysiol 2007; 118:2404–2412.
14. Sassen M, Zimmermann M. Differential blocking of myelinated nerve fibres by transient depolarisation. Pflugers Arch 1973; 341:179–195.
15. Lang PM, Hilmer VB, Grafe P. Differential contribution of sodium channel subtypes to action potential generation in unmyelinated human C-type nerve fibers. Anesthesiology 2007; 107:495–501.
16. Bostock H, Grafe P. Activity-dependent excitability changes in normal and demyelinated rat spinal root axons. J Physiol 1985; 365:239–257.
17. Frankenhaeuser B, Meves H. The effect of magnesium and calcium on the frog myelinated nerve fibre. J Physiol 1958; 142:360–365.
18. Hodgkin AL, Huxley AF. The components of membrane conductance in the giant axon of Loligo. J Physiol 1952; 116:473–496.
19. Hille B. Ionic channels in nerve membranes. Prog Biophys Mol Biol 1970; 21:1–32.
20. Bostock H. MEMFIT: a computer program to aid interpretation of multiple excitability measurements on human motor axons. Clin Neurophysiol 2006; 117:S85–S86.
21. Momin A, Wood JN. Sensory neuron voltage-gated sodium channels as analgesic drug targets. Curr Opin Neurobiol 2008; 18:383–388.
22. Caldwell JH, Schaller KL, Lasher RS, et al. Sodium channel Na(v)1.6 is localized at nodes of ranvier, dendrites, and synapses. Proc Natl Acad Sci U S A 2000; 97:5616–5620.
23. van Wart A, Matthews G. Impaired firing and cell-specific compensation in neurons lacking nav1.6 sodium channels. J Neurosci 2006; 26:7172–7180.
24. Sang N, Meng Z. Blockade by magnesium of sodium currents in acutely isolated hippocampal CA1 neurons of rat. Brain Res 2002; 952:218–221.
25. Wood JN, Boorman JP, Okuse K, Baker MD. Voltage-gated sodium channels and pain pathways. J Neurobiol 2004; 61:55–71.
26. Hille B, Woodhull AM, Shapiro BI. Negative surface charge near sodium channels of nerve: divalent ions, monovalent ions, and pH. Philos Trans R Soc Lond B Biol Sci 1975; 270:301–318.
27. Mert T, Gunes Y, Guven M, et al. Effects of calcium and magnesium on peripheral nerve conduction. Pol J Pharmacol 2003; 55:25–30.
28. Chen C, Bharucha V, Chen Y, et al. Reduced sodium channel density, altered voltage dependence of inactivation, and increased susceptibility to seizures in mice lacking sodium channel beta 2-subunits. Proc Natl Acad Sci U S A 2002; 99:17072–17077.
29. Herzog RI, Cummins TR, Ghassemi F, et al. Distinct repriming and closed-state inactivation kinetics of Nav1.6 and Nav1.7 sodium channels in mouse spinal sensory neurons. J Physiol 2003; 551:741–750.
30. Hille B. Local anesthetics: hydrophilic and hydrophobic pathways for the drug-receptor reaction. J Gen Physiol 1977; 69:497–515.
31. Cummins TR. Setting up for the block: the mechanism underlying lidocaine's use-dependent inhibition of sodium channels. J Physiol 2007; 582:11.

magnesium sulphate; nerve excitability; sensory afferents; threshold tracking

© 2013 European Society of Anaesthesiology