Secondary Logo

Journal Logo

The nerve: A fragile balance between physiology and pathophysiology

Estebe, Jean-Pierre; Atchabahian, Arthur

European Journal of Anaesthesiology: March 2017 - Volume 34 - Issue 3 - p 118–126
doi: 10.1097/EJA.0000000000000590
REVIEW ARTICLE
Free

Regarding nerves as simple cables and electrical conduits is a gross error that does not allow us to understand the anomalies and disorders observed postoperatively. Instead, nerves should be seen as a living tissue of which physiological regulation is as complex as that of the blood–brain barrier. This review describes the basic structure and functions of this blood–nerve barrier and highlights the mechanisms of its breakdown and the resultant disorders. For clinical practice, it is important to note that the diffusion of molecules from the perineurium or from the blood is very limited, and so the blood–nerve barrier is a major pharmacologic barrier. Any stress upon neural physiological balance, particularly the terminal vascular blood supply, will induce the classic inflammatory cascade. Due to the complexity of the vascular system, nerve ischaemia will occur more quickly when the terminal blood supply is compromised. This blood supply can adapt in a variety of ways but when these possibilities of adaptation are exceeded, tissue ischaemia may be more extensive. Also, even after the initial injury has subsided, inflammation can cause a secondary insult. This could be particularly important in some patients with subclinical neuropathy.

From the Department of Anaesthesiology, Intensive Care, and Pain Medicine, University of Rennes, CHU of Rennes, Cedex 9, France (J-PE); Department of Anesthesiology, Perioperative Care, and Pain Medicine, NYU School of Medicine, New York, New York, USA (AA)

Correspondence to Jean-Pierre Estebe, Department of Anaesthesiology, Intensive Care, and Pain Medicine, University of Rennes, CHU of Rennes, Rue H Le Guilloux, 35033, Rennes, Cedex 9, France Tel: +33 2 99 28 93 76; e-mail: jean-pierre.estebe@chu-rennes.fr

Published online 18 January 2017

Back to Top | Article Outline

Introduction

Although nerve injury is a well recognised complication of surgical procedures due, for example to surgical trauma, general or regional anaesthesia, patient positioning or use of a pneumatic tourniquet,1–5 other factors may contribute to nerve injury. Among these are genetic factors, immune system activity, age and sex. Pre-existing subclinical neuropathy could also increase the risk of unexpected postoperative nerve injury.6–9 To improve our understanding of the peripheral nerve in clinical practice, we review its anatomy, physiology and pathophysiology.

Back to Top | Article Outline

Anatomy and physiology

Although the structure of peripheral nerves was described nearly 200 years ago, the functional roles of all five main components (epineurium, perineurium, endoneurium, vessels and neural tissue) are still poorly understood and insufficiently studied. Peripheral nerve structure could be compared with spinal cord anatomy but this is an over-simplification. Although the three layers of connective tissue in the peripheral nerves (epineurium, perineurium and endoneurium) (Fig. 1) are a continuation of the dura mater, arachnoid mater and pia mater, respectively, they have a very different anatomy and physiology from the layers enveloping the spinal cord. Axons are surrounded by endoneurium; multiple axons, myelinated and unmyelinated, are grouped into fascicles, which are surrounded by perineurium; and finally, groups of fascicles constituting a nerve are covered by epineurium (Fig. 1a). In specific locations, such as in the sciatic nerve, there is an outermost common layer of connective tissue. This outermost layer forms a ‘paraneurial’ component within a subparaneural compartment and on the outside a subepimyseal perineural compartment (epimysium and muscle). The epimysium is the fibrous tissue envelope that surrounds skeletal muscle. It is a layer of connective tissue that ensheathes the entire muscle and protects muscles from friction against other muscles and bones. This ‘paraneurial’ component facilitates the gliding of the nerves during movement.

Fig. 1

Fig. 1

Back to Top | Article Outline

Epineurium and epineurial space

Like the dura mater, the epineurium (or epineurial membrane) consists of thick, dense layers of connective tissue surrounding the nerves externally and then tethering them to neighbouring structures. Its thickness increases around groups of nerve fascicles.10 This epineurial membrane is not a barrier for therapeutic or toxic molecules located outside the nerve (Fig. 1b). This explains why there is no advantage in injecting local anaesthetics into the perineurial space rather than into the extraneural space for regional anaesthesia. The greater part of the extrafascicular epineurial tissue (or subepineurial space, also, incorrectly called perineurial space) is a cushioning adipose tissue with connective tissue tightly bound to the intermediate perineurial layer.11 Due to an increased amount of connective and fibrotic tissue in diabetic patients compared with non-diabetic patients, the perineurial index is higher (based on measurement of outer and inner diameter of the fascicle).12 The proportion of non-neural tissue (stroma and connective tissue) inside and outside the nerves increases as one moves from proximal to distal.13 As an example, in the brachial plexus, although the amount of neural tissue remains approximately the same between the interscalene/supraclavicular and the mid-infraclavicular/subcoracoid regions, the ratio of neural to non-neural tissue inside the epineurium increases from 1 : 1 to 1 : 2 with nerve fascicles more scattered in a polyfascicular pattern.13 These differences could partially explain why nerve injuries are more commonly associated with more proximal regional anaesthetic techniques.14,15 However, at a given anatomic level, some differences can be explained by histology. As an example, the peroneal nerve is more prone to injury than the tibial nerve. This is usually attributed to its superficial location, but it is probably also due to histological differences between the two nerves: the peroneal nerve has less perineurial extrafascicular tissue (50 to 55% for the tibial nerve vs. 38 to 42% for the peroneal nerve).11 If the epineurium is not a pharmacologic barrier, it is a mechanical barrier (i.e. an injection into the epineurial space increases deleterious pressure) (Fig. 1d). The interfascicular tissue has elastic properties. It is composed of adipocytes and blood vessels with fenestrated capillaries as well as lymphatic vessels. Here, there is little to prevent the spread of molecules arriving with the blood supply. Understanding the difference between macrocirculation (vasa nervorum) and microcirculation (endoneurial vessels) of the peripheral nerve is essential. Due to this complex vascular system (i.e. anastomotic permeable perineurial vessels vs. terminal impervious endoneurial vessels), nerve ischaemia will be quicker when the terminal vascularisation is compromised. The vasa nervorum can adapt but when the possibilities of adaptation are exceeded, tissue ischaemia may be more extensive. This is the reason why the blood perfusion of the nerve does not return to the baseline value after relief of experimental compression.16

Back to Top | Article Outline

Perineurium and endoneurium

Like the arachnoid membrane, the perineurium (or perineurial membrane) surrounds the endoneurium in concentric cellular layers of fibroblastic (or myofibroblastic) origin.17 There is a positive correlation between the number of layers and the diameter of the fascicles (5 to 15 layers).10,18 Each layer is formed by single cells joined together with tight junctions and desmosomes preventing the passive influx of epineurial interstitial fluid into the endoneurium. There is a thick basement membrane. Interposed between perineurial cell layers are collagen fibres aligned in different directions, but pre-dominantly in a longitudinal fashion, with occasional elastic fibres possibly providing the structural basis for the compliance of the perineurium.18 The endoneurium (or endoneurial space) (Fig. 2) consists mainly of collagen fibres produced by Schwann cells and endoneurial fibroblasts (or endoneurial fibroblast-like cells).19 These collagen fibres shape genuine tunnels for each axon and are arranged in circular, longitudinal and oblique bundles. Collagen is a major component of the extracellular matrix; it not only provides structural support, but also affects cell behaviour by triggering intracellular signals (Schwann cell function and myelination).20 The endoneurial microenvironment represents 20 to 25% of the endoneurial area. Unlike other extracellular connective tissue spaces, the endoneurium has a positive pressure gradient (10 mmHg in the spinal cord vs. 2 to 3 mmHg in the peripheral nerve) that counterbalances the low oncotic pressure due to a very low level of albumin, in the non-compressible endoneurial fluid. This endoneurial fluid has a different electrolyte composition than the surrounding tissues and blood, maintained through the so-called endoneurial homeostasis. Non-fenestrated capillaries are present in the endoneurial space. They are vulnerable to compression despite their unusually large diameter (6 to 10 vs. 3 to 6 μm for skeletal muscle capillaries).18 Endoneurial vessels acquire tight junctions when the vessels penetrate the innermost layer of the perineurium despite the high-energy demand of axons. The tight junctions, with membrane proteins such as occludins, cingulins, claudins and zona occludens proteins essentially produced by pericytes (mural contractile cells of blood microvessels surrounding the endothelial cells), control the paracellular transport pathways.18,21,22 The influx transporters of the nerve nutritious microelements require active enzymatic systems. Pericytes also express on their surface efflux transporters like some members of the ATP-binding cassette family such as P-glycoprotein and multi-drug resistance proteins (1 to 6) transporters, which might play a role in regulating xenobiotic transport through the blood–nerve barrier (BNB). These pericytes also share the basal membrane with the vessel endothelial cells.23 It is hypothesised that peripheral nerve pericytes are important for maintaining the specialised BNB characteristic, just like astrocytes and microglia foot processes contribute to the barrier function in the blood–brain barrier (BBB). After mechanical trauma, alterations in endoneurial blood vessel structure (hypertrophy and duplication of endothelial cells, multiplication of layers, basal lamina thickening and fibrosis) contribute to endoneurial hypoxia.16

Fig. 2

Fig. 2

The diffusion of molecules into the endoneurium from the perineurium or from the blood is very limited. The BNB is a selective pharmacological barrier rather than a simple passive restrictive barrier. The absence of lymphatic drainage in the endoneurial space further emphasises the protective nature of the BNB that tightly regulates the homeostasis of the nerve microenvironment.18 Nutritional exchanges depend on the presence of specific pumps, transporter, enzymes and receptors (glucose transporter-1, L-type amino acid transporter, monocarboxylase acid transporter, transferrin receptor and alkaline phosphate),22,24 and there are a large number of cytosolic mitochondria.21

The microcirculation originates from regional extrinsic vessels necessitating radicular terminal vessels that supply the intrinsic plexus circulation. The intrinsic circulation consists of longitudinally oriented vessels that run through the epineurium to the perineurium and ultimately join with the vessels in the endoneurium. Extensive anastomotic connections are present at all levels of the extrinsic circulation.

In recent years, the important concept of the neurovascular unit has emerged. Cell–cell and cell–matrix signalling involves pericytes, Schwann cells, axons, macrophages, endoneurial fibroblasts and mast cells. There are complex, fragile immunomodulation systems in place similar to that of the central nervous system microglia.18 Schwann cells support the preservation and function of the peripheral nerve axons. Myelinating Schwann cells wrap multi-lamellar membranes around large-diameter axons to form myelin, whereas non-myelinating Schwann cells form Remak bundles that sheathe several small diameter unmyelinated sensory axons (C fibres). Fibroblast growth factor-2 acts on receptors expressed by Schwann cells but its role in modulating axonal function is unclear due to complex regulation and a wide variety of Schwann cell phenotypes.25

Inside nerve fascicles, a continuous exchange of fibres leads to the formation of an intraneural plexus, and this is true for both longer and shorter nerves.10 This plexus structure could make the nerve more vulnerable to stretching or compression forces. Bands of Fontana, whose function seems to be the accommodation of stretch, are present in the endoneurial bundles.26 Stretching produces an increase in the spacing of spiral bands around the nerve, with a recoiling phenomenon when the distracting force is removed. These bands are not present in regenerated nerve tissue.26

Back to Top | Article Outline

Pathophysiology

Peripheral nerve injury following surgical procedures is usually attributed to stretch, compression, contusion, needle trauma or drug toxicity. In other cases, an inflammatory cause (such as diabetes, alcohol toxicity or chemotherapy), genetic cause (such as sodium channel mutation) or autoimmune mechanisms (such as in Guillain–Barré syndrome, with autoantibodies to the ganglioside components of the axonal membrane; or chronic inflammatory polyneuropathy, with autoantibodies to the cell adhesion molecule) should be considered.6,7,27

Back to Top | Article Outline

Age

Little data are available concerning age-related changes in the peripheral nervous system. Peripheral nerve ultrastructure is significantly affected by ageing with axonal degeneration, demyelination (partly due to oxidative damage accumulated over time28), a change in mitochondrial redox potential29 or a decrease in axonal transport.30 Thickening of the endoneurium and perineurial area is due to an increase in collagen.31

Back to Top | Article Outline

Sex

Some sex differences have been reported in motor nerve conductance and in the nerve regeneration process, which may be due to the protective effects of sex hormones, mainly progesterone.32

Back to Top | Article Outline

Inflammation

Increasing evidence suggests that like microglia in the central nervous system, nerve inflammation that initially has a repair function can quickly create pathological conditions. Growing evidence suggests that macrophages may play a significant role, for example in HIV infection.33 Macrophages produce a large number of different cytokines, chemokines and metabolites that cause neurodegeneration.

Whatever the cause of neural hypoxia, in the initial stages there is activation of resident endoneurial macrophages, which will then release vascular endothelial growth factor. This is a potent inducer of vascular permeability and stimulates the growth of new endoneurial vessels that do not possess tight junctions (Fig. 3a and b).24 This creates another interface for blood-borne immune cells and induces a breach in the BNB (Fig. 3c).34 Hypoxia also impairs nerve autophagy producing homeostatic dysfunction that can be correlated with neurological dysfunction.35

Fig. 3

Fig. 3

An increased number of degranulating mast cells, which release histamine and factors with potent inflammatory properties known to promote vascular permeability such as serotonin and ATP, have been described. Glucocorticoid immunosuppressive treatment reduces neuropathic pain by a reduction of TNFα in mast cells.36

An increase in the number of endoneurial fibroblasts, causing abnormal proliferation of collagen, has been also reported. These cells initiate a cascade of inflammation and immune-related events with production of IL-1α and IL-1β, IL-6, tumour necrosis factor and apolipoprotein.21

A reduction of the Na+/K+ ATPase ion transporter has been observed, probably because of a deficit of ATP, which may result in spontaneous depolarisation with a failure to maintain the resting membrane potential. These electrical changes generate ectopic nerve activity that can also be increased by the activation of the transient receptor potential (TRP) family (as TRP vanilloid 1, TRP malestin 8 and TRP Ankyrin 1).16 An upregulation of voltage-gated sodium channels may also play a role in sensitisation.

Nerve protection seen in the intermediate phase as the clearance of myelin debris is in favour of axonal regeneration, whereas later protective effects may be seen as an increased production of immunosuppressive cytokines (such as TGF-Beta1 and IL-10) from M2 macrophages.37 However, these homeostatic mechanisms can be quickly overwhelmed by the inflammatory process.

After initial local inflammation, a secondary inflammatory injury occurs over several days after the original injury. Infiltration with and activation of blood-borne macrophages occurs (Fig. 3c). The leucocyte trafficking at the BNB is an active rather than passive process.24 Although M1-macrophages typically kill pathogens and clear cellular debris, they are associated with a considerable increase in the production of proinflammatory cytokines. This correlates with the development of neuropathic pain,34 whereas a preventive or curative effect was observed after macrophage depletion (neutrophils and/or monocytes).37 Other chemoattractants such as monocyte chemoattractant protein-1 produced by Schwann cells are involved in these macrophage infiltration phenomena. Various cellular adhesion molecules (such as matrix metalloproteinase, intercellular molecule-1, vascular cell adhesion molecule-1, E-selectin and P-selectin) facilitate extravasation across endothelial membranes.24 Similarly, extravasated blood fibrinogen could induce neural sensitisation.

The production of endothelin-1 is upregulated by endothelial cells during nerve injury.38 Then endothelin-1 induces vasoconstriction through its action on the endothelin vascular receptors (predominantly ET A Receptor), and induces pain by stimulating the endothelin neural and macrophage receptors (primarily ET-B Receptor).39

An extreme clinical example is the group of chronic inflammatory demyelinating polyneuropathies, particularly the atypical forms (multi-focal acquired demyelinating sensory and motor neuropathy and distal acquired demyelinating symmetric neuropathy). In such cases, cell-mediated immunity (lymphocytes T and macrophages) reacts against the peripheral nerves with an inappropriate secretion of inflammatory mediators (prostaglandins; cytokines: IL-1β, tumour necrosis factor and IL-6; chemokines and matrix metalloproteinases).37,40 Activated macrophages and other components of cell-mediated immunity (like T lymphocytes) strongly enhance macrophage phagocytosis and cause an increase in permeability of the BNB. They then penetrate the basement membrane of Schwann cells and split the myelin lamellae, thereby provoking destruction of the myelin sheath. This effect is at least partially due to alterations in tight junction proteins.41 In immune-mediated peripheral neuropathy (such as Guillain–Barré syndrome), alterations of the myelin (such as swelling, vesicular disorganisation and separation of the myelin lamellae) are due to inflammatory cytokines.42

Occasionally, no obvious cause for a postsurgical neuropathy can be found, and inflammatory or autoimmune causes need to be considered.27 In a subset of perioperative neuropathies, biopsy of superficial sensory nerves could confirm the diagnosis.7,8 The consequences of this neural inflammation are probably underestimated, due to the strong tendency to attribute the cause of nerve injury to surgical or anaesthetic insults.

Back to Top | Article Outline

Vasculitis

Vasculitis can affect the peripheral nervous system. The classification of neuropathy due to vasculitis has become increasingly complex.43 Although the common primary (such as microscopic polyangitis) and secondary (such as rheumatoid vasculitis) systemic vasculitic neuropathies are well recognised, the non-systemic vasculitic neuropathies typically develop slowly, asymmetrically or multi-focally and their frequency is probably underestimated. Some authors have proposed a simpler division into two groups based on the size of involved vessels44: nerve large-arteriole vasculitis (perineurial vessel 75 to 300 μm in diameter) and nerve microvasculitis (<40 μm). The diagnosis can usually only be made by a biopsy of an affected sensory nerve (usually the sural sensory nerve). It includes smaller vessels without internal elastic lamina, fibrinoid necrosis, multi-focal fibre loss, perineurial thickening, haemosiderin-laden macrophages, neovascularisation, active Wallerian degeneration, as well as immune deposits of IgM, complement proteins and fibrinogen in the epineurial vessel walls.44 Thus, the limitations of nerve biopsy reside in the fact that it examines the distal part of the sensory nerve but not the proximal motor portion of nerves, which is especially important for chronic inflammatory demyelinating polyneuropathies.

In diabetic patients, because of the overlap in clinical and electrophysiological characteristics, it can be sometimes difficult to understand the contribution of different pathophysiological mechanisms due to the diabetic sensorimotor polyneuropathy and chronic inflammatory demyelinating polyneuropathy.45 Under chronic hyperglycaemia conditions, reactive oxygen species initiate and amplify the vascular nerve lesions (i.e. directly by accumulation of sorbitol and indirectly through consumption of NADPH increase protein kinase C activity, which effectively inhibits cellular Na+/K+ ATPase).46 Thus, oxidative stress and neuroinflammation are identified to be pivotal pathophysiological triggers in diabetic neuropathy. Microvascular changes precede the development of neuropathy with active epineurial arteriovenous shunts and endothelial dysfunction (exaggerated production of vasoconstrictors and diminished endothelial-derived vasodilatators) that may result in a reduction in the endothelial blood flow and induce a fibrotic response.47 As previously reported, regardless of the cause of endoneurial hypoxia, neural injury is worst with hypoglycaemia or hyperglycaemia.48 In experimental models, it was reported that endoneurial hypoxia induces neuropathic pain, which could be alleviated with hyperbaric oxygen.16

Following nerve trauma, the perineurial permeability increases about two-fold (with decreased expression of intercellular junctional proteins), which allows diffusion of large molecules like albumin.49 This local disturbance of vascular permeability causes an increase in endoneurial fluid pressure, as observed in many different neuropathic pain models (chronic constriction injury, partial sciatic nerve ligation, paclitaxel-induced neuropathy or sciatic inflammatory neuritis).

The detrimental role of endothelial nitric oxide synthase on regeneration processes has been reported.50

Back to Top | Article Outline

Toxicity

Antitumour chemotherapy commonly affects sensory nerves more than motor and autonomic nerves in a dose-dependent and time-dependent manner. There are substantial variations in the time to the onset of symptoms, time to peak symptoms, severity and reversibility.51 The exact mechanisms are not certain but evidence is accumulating suggesting mitochondrial dysfunction and oxidative stress as the key targets.52 A neuroinflammatory response is seen comprising activation of Schwann cells and resident macrophages with production of cytokines and chemokines. Various mechanisms have been described, such as disruption of microtubule dynamics (central effect to axonal transport processes), tubulin polymerisation, abnormalities in axonal mitochondria (interfering with energy mechanisms), damage to the nerve vasculature and direct axonal toxicity (axonal degeneration and dysfunction of voltage-gated sodium channels).53 In some cases, infiltration by malignant cells and/or endoneurial deposits of immunoglobulin or amyloid substance may be found on biopsies.54

The mechanism of alcohol-related peripheral neuropathy remains unclear. The theory of a decrease in thiamine absorption is probably not relevant as thiamine supplementation is not clearly effective and neuropathies can occur in the absence of nutritional deficiency.55 However, chronic ethanol exposure induces slowed conduction velocities partially due to demyelination.56 It is believed that inflammatory reactions, oxidative stress and cytokine production mediate the adverse effect of alcohol-induced multi-organ damage, including the nervous system with axonal degeneration.57

It is now widely accepted that neuropathic pain is a neuroimmunological disorder, and the idea that the immune system can regulate higher order behaviour (i.e. fatigue, depression, appetite and sleep–wake cycle) has recently begun to gain traction.58

Little is known about the impact of local anaesthetics on nerves with subclinical neuropathy. There could be potentiation of mechanical trauma, neuronal ischaemia, drug neurotoxicity and genetic susceptibility (as observed in multiple sclerosis and juvenile rheumatoid arthritis). A better understanding of the distinct pathophysiologic mechanisms may allow targeted therapies, such as determination of the minimum effective concentration of local anaesthetics for an optimal duration for individual patients. It is well established that all local anaesthetics induce significant Schwann cell death in a time-dependent and concentration-dependent manner,59–61 but local anaesthetics used in clinical concentrations and doses have been proven to be well tolerated. There is a high risk of inducing additive neurotoxicity with some adjuvants (such as ketamine or midazolam) or solvents either by a specific action or by increasing the toxicity of local anaesthetics.62,63 Similarly, needle-nerve contact,64,65 ultrasound gels,66 blood67,68 or skin antiseptic solutions could induce neurotoxicity or at least increase inflammation and synergistically enhance local anaesthetic neurotoxicity. As an example, in a needle-stick nerve injury model, significant losses of myelinated and unmyelinated axons, intrafascicular oedema, thickened blood vessel walls and doubling of mast cells were reported.69

In an experimental model, a decrease of intraneural blood flow (with congestion in the nerve due to impairment of venous return) has been reported as soon as the pressure on the nerve exceeds 0.15 N, and blood flow completely stops for a pressure over 0.45 N (48.7 mmHg).70 When an injection of local anaesthetic is performed in an enclosed or semi-enclosed space (i.e. epineurial or perineurial space), a risk of reduction of the neural blood flow may occur. The duration of block is extended when the local anaesthetics are applied on pathological nerves, as in diabetic patients.71,72 Similarly, nerve dysfunction increases the electrical nerve stimulation threshold for regional anaesthesia73 or for nerve functional evaluation.74 In theory, nerve stimulation or ultrasound could cause thermal or mechanical damage to nerves, although this has never been noted in humans.75 Toxicity could be modulated in a pharmacokinetic way by a larger mass of fat outside the nerve, which might serve as a reservoir for lipophilic local anaesthetics in more distal areas.13

Ultrasound could be a useful tool for the diagnosis and management of traumatic nerve lesions,76 diabetic neuropathy77 or other peripheral nerve disease.78 High-resolution ultrasound can provide diagnostic information such as visualisation of the inner part of small nerves that cannot be obtained by electrophysiology or even by MRI studies.79

Back to Top | Article Outline

Conclusion

Compared with the BBB, there is only limited knowledge of the function, cell biology and clinical significance of the BNB. This review describes the basic structure and functions of this barrier and highlights the mechanisms of BNB breakdown and the resultant pathophysiology. Postoperative neuropathies are commonly blamed on a mechanical insult, due to the surgical procedure or the anaesthetic block. However, it is important to recognise that inflammatory phenomena and vasculitis play a significant role in the initial insult and can contribute to a secondary insult. Although recommendations can be made to decrease further injury (for instance, by injecting medications whose safety has been demonstrated outside the epineurial space using the lowest effective concentration), our understanding is still insufficient to recommend therapy for established neuropathy.

Back to Top | Article Outline

Acknowledgements relating to this article

Assistance with the review: the author thanks Léna Estebe for assistance in primary full-text acquisition and Nicolas Suraud (www.sunidigital.com) for his outstanding graphics support.

Financial support and sponsorship: none.

Conflicts of interest: none.

Back to Top | Article Outline

References

1. Dwyer T, Drexler M, Chan VW, et al. Neurological complications related to elective orthopedic surgery. Part 2: Common hip and knee procedures. Reg Anesth Pain Med 2015; 40:443–454.
2. Dwyer T, Henry PDG, Cholvisudhi P, et al. Neurological complications related to elective orthopedic surgery. Part 1: Common shoulder and elbow procedures. Reg Anesth Pain Med 2015; 40:431–442.
3. Agostini J, Goasguen N, Mosnier H. Patient positioning in laparoscopic surgery: tricks and tips. J Visc Surg 2010; 147:e227–e232.
4. Veljkovic A, Dwyer T, Lau JT, et al. Neurological complications related to elective orthopedic surgery. Part 3: Common foot and ankle procedures. Reg Anesth Pain Med 2015; 40:455–466.
5. Estebe JP, Davies JM, Richebe P. The pneumatic tourniquet: mechanical, ischaemia-reperfusion and systemic effects. Eur J Anaesthesiol 2011; 28:404–411.
6. Staff N, Engelstad J, Klein CJ, et al. Postsurgical inflammatory neuropathy. Brain 2010; 133:2866–2880.
7. Staff NP, Dyck PJB, Warner MA. Postsurgical inflammatory neuropathy should be considered in the differential diagnosis of diaphragm paralysis after surgery. Anesthesiology 2014; 120:1057.
8. Rattananan W, Thaisetthawatkul P, Dyck PJ. Postsurgical inflammatory neuropathy: a report of five cases. J Neurol Sci 2014; 337:137–140.
9. Laughlin RS, Dyck PJ, Watson JC, et al. Ipsilateral inflammatory neuropathy after hip surgery. Mayo Clin Proc 2014; 89:454–461.
10. Reina MA, Arriazu R, Collier CB, et al. Electron microscopy of human peripheral nerves of clinical relevance to the practice of nerve blocks. A structural and ultrastructural review based on original experimental and laboratory data. Rev Esp Anestesiol Reanim 2013; 60:552–562.
11. Schraut NB, Walton S, Bou Monsef J, et al. What protects certain nerves from stretch injury? Anat Rec 2016; 299:111–117.
12. Kundalic B, Ugrenovic S, Jovanovic I, et al. Morphometric analysis of connective tissue sheaths of sural nerve in diabetic and nondiabetic patients. Biomed Res Int 2014; 2014:870930.
13. Moayeri N, Bigeleisen PE, Groen GJ. Quantitative architecture of the brachial plexus and surrounding compartments, and their possible significance for plexus block. Anesthesiology 2008; 108:299–304.
14. Welch MB, Brummett CM, Welch TD, et al. Perioperative peripheral nerve injuries: a retrospective study of 380,680 cases during a 10-year period at a single institution. Anesthesiology 2009; 111:490–497.
15. Brull R, McCartney C, Chan V. Neurological complications after regional anesthesia: contemporary estimates of risk. Anesth Analg 2007; 104:965–974.
16. Lim TKY, Shi XQ, Johnson JM, et al. Peripheral nerve injury induces persistent vascular dysfunction and endoneurial hypoxia, contributing to the genesis of neuropathic pain. J Neurosci 2015; 35:3346–3359.
17. Greathouse KM, Palladino SP, Dong C, et al. Modeling leukocyte trafficking at the human blood–nerve barrier in vitro and in vivo geared towards targeted molecular therapies for peripheral neuroinflammation. J Neuroinflammation 2016; 13:3.
18. Mizisin AP, Weerasyuriya A. Homeostatic regulation of the endoneurial microenvironment during development, aging, and in response to trauma, disease and toxic insult. Acta Neuropathol 2011; 121:291–312.
19. Richard L, Topilko P, Magy L, et al. Endoneurial fibroblast-like cells. J Neuropathol Exp Neurol 2012; 71:938–947.
20. Chen P, Cescon M, Bonaldo P. The role of collagens in peripheral nerve myelination and function. Mol Neurobiol 2015; 52:216–225.
21. Radu BM, Bramanti P, Osculati F, et al. Neurovascular unit in chronic pain. Mediators Inflamm 2013; 2013:648268.
22. Ubogu EE. The molecular and biophysical characterization of the human blood–nerve barrier: current concepts. J Vasc Res 2013; 50:289–303.
23. Armulik A, Genove G, Betsholtz C. Pericytes: developmental, physiology, and pathological perspectives, problems, and promises. Dev Cell 2011; 21:193–215.
24. Alit G, Lawrenson JG. The blood–nerve barrier: enzymes, transporter and receptors – a comparison with the blood–brain barrier. Brain Res Bull 2000; 52:1–12.
25. Furusho M, Dupree JL, Bryant M, et al. Disruption of fibroblast growth factor receptor signaling in nonmyelinating Schwann cell causes sensory axonal neuropathy and impairment of thermal pain sensitivity. J Neurosci 2009; 29:1608–1614.
26. Merolli A, Mingarelli L, Rocchi L. A more detailed mechanism to explain the ‘bands of Fontana’ in peripheral nerves. Muscle Nerve 2012; 48:540–547.
27. Martini R, Willison H. Neuroinflammation in the peripheral nerve: cause, modulator, or bystander in peripheral neuropathies? Glia 2016; 64:475–486.
28. Sims-Robinson C, Hur J, Hayes JM, et al. The role of oxidative stress in nervous system aging. PLoS One 2013; 8:e68011.
29. McDonagh B, Scullion SM, Vasilaki A, et al. Ageing-induced changes in the redox status of peripheral motor nerve imply an effect on redox signaling rather than oxidative damages. Free Radic Biol Med 2016; 94:27–35.
30. Milde S, Adalbert R, Elaman MH, et al. Axonal transport declines with age in two distinct phases separated by a period of relative stability. Neurobiology Aging 2015; 36:971–981.
31. Ugrenovic S, Jovanovic I, Vasovic L. Morphometric analysis of human sciatic nerve perineurial collagen type IV content. Micros Res Tech 2011; 74:1127–1133.
32. Stenberg L, Dahlin LB. Gender differences in nerve regeneration after sciatic nerve injury and repair in healthy and in type 2 diabetic Goto–Kakizaki rats. BMC Neurosci 2014; 15:107.
33. Moss PJ, Huang W, Dawes J, et al. Macrophage-sensory neuronal interaction in HIV-1 gp120-induced neurotoxicity. Br J Anaesth 2015; 114:499–508.
34. Lim TKY, Shi XQ, Martin HC, et al. Blood–nerve barrier dysfunction contributes to the generation of neuropathic pain and allows targeting of injured nerves for pain relief. Pain 2014; 155:954–967.
35. Berliocchi L, Russo R, Malaru M, et al. Autophagy impairment in a mouse model of neuropathic pain. Mol Pain 2011; 7:83.
36. Hayasshi R, Xiao W, Kawamoto M, et al. Systemic glucocorticoid therapy reduces pain and the number of endoneurial tumor necrosis factor-alpha (TNFα)-positive mast cell in rats with a painful peripheral neuropathy. J Pharmacol Sci 2008; 106:559–565.
37. Bastien D, Lacroix S. Cytokine pathways regulating glial and leukocyte function after spinal cord and peripheral nerve injury. Exp Neurol 2014; 258:62–77.
38. Klass M, Hord A, Wilcox M, et al. A role for endothelin in neuropathic pain after chronic constriction injury of the sciatic nerve. Anesth Analg 2005; 101:1757–1762.
39. Pomonis JD, Rogers SD, Peters CV, et al. Expression and localization of endothelin receptors: implications for the involvement of peripheral glial in nociception. J Neurosci 2001; 21:999–1006.
40. Svahn J, Antoine JC, Camdessanche JP. Pathophysiology and biomarkers in chronic inflammatory demyelinating polyradiculoneuropathies. Rev Neurol (Paris) 2014; 170:808–817.
41. Shimizu F, Sawai S, Sano Y, et al. Severity and patterns of blood–nerve barrier breakdown in patients with chronic inflammatory demyelinating polyradiculoneuropathy: correlations with clinical subtypes. PLoS One 2014; 9:e104205.
42. Yuan XJ, Wei YJ, Ao Q, et al. Myelin ultrastructure of sciatic nerve in rat experimental autoimmune neuritis model and its correlation with associated protein expression. Int J Clin Exp Pathol 2015; 8:7849–8785.
43. Jennette JC, Falk RJ, Bacon PA, et al. 2012 revised international Chapel Hill conferences nomenclature of vasculitides. Arthritis Rheum 2013; 65:1–11.
44. Gwathmey KG, Burns TM, Collins MP, et al. Vasculitic neuropathies. Lancet Neurol 2014; 13:67–82.
45. Dunnigan SK, Ebadi H, Katzberg HD, et al. Comparison of diabetes patients with ‘demyelinating’ diabetic sensorimotor polyneuropathy to those diagnosed with CIDP. Brain Behav 2013; 3:656–663.
46. Sandireddy R, Yerra VG, Areti A, et al. Neuroinflammation and oxidative stress in diabetic neuropathy: futuristic strategies based on these targets. Int J Endocrinol 2014; 2014:674987.
47. Nukada H. Ischemia and diabetic neuropathy. Hand Clin Neurol 2014; 126:469–487.
48. Punsoni M, Drexler S, Palaja T, et al. Acute anoxic changes in peripheral nerve: anatomic and physiologic correlations. Brain Behav 2015; 7:e00347.
49. Omura K, Ohbayayashi M, Sano M, et al. The recovery of blood–nerve barrier in crush nerve injury: a quantitative analysis utilizing immunohistochemistry. Brain Res 2004; 1001:13–21.
50. Sunico CR, Moreno-Lopez B. Evidence of endothelial nitric oxide as a negative regulator of Schwann cell dedifferentiation after peripheral nerve injury. Neurosci Lett 2010; 47:119–124.
51. Wheeler HE, Wing C, Delaney SM, et al. Modeling chemotherapeutic neurotoxicity with human induced pluripotent stem cell-derived neuronal cells. PLoS One 2015; 10:e0118020.
52. Han Y, Smith MT. Pathobiology of cancer chemotherapy-induced peripheral neuropathy. Front Pharmacol 2013; 4:156.
53. Park SB, Goldstein D, Krishman AV, et al. Chemotherapy-induced peripheral neurotoxicity: a critical analysis. CA Cancer J Clin 2013; 63:419–437.
54. Duchesne M, Mathis S, Corcia P, et al. Value of nerve biopsy in patients with latent malignant hemopathy and peripheral neuropathy. Medicine (Baltimore) 2015; 94:e394.
55. Mellion ML, Nguyen V, Tong M, et al. Experimental model of alcohol-related peripheral neuropathy. Muscle Nerve 2013; 48:204–211.
56. Koike H, Mori K, Misu K, et al. Painful alcoholic neuropathy with predominant small-fiber loss and normal thiamine status. Neurology 2001; 56:1727–1732.
57. De la Monte SM, Krill JJ. Human alcohol-related neuropathology. Acta Neuropathol 2014; 127:71–90.
58. Austin PJ, Berglund AM, Siu S, et al. Evidence for a distinct neuro-immune signature in rats that develop behavioural disability after nerve injury. J Neuroinflammation 2015; 12:96.
59. Yang S, Abrahams MS, Hurn PD, et al. Local anesthetic Schwann cell toxicity is time and concentration dependent. Reg Anesth Pain Med 2011; 36:444–451.
60. Farber SJ, Saheb-Al-Zamani M, Zieske L, et al. Peripheral nerve injury after local anesthetic injection. Anesth Analg 2013; 117:731–739.
61. Maiet A, Faure MO, Deletage N, et al. The comparative cytotoxic effects of different local anesthetics on a human neuroblastoma cell line. Anesth Analg 2015; 120:589–596.
62. Williams BA, Hough KA, Tsui BY. Neurotoxicity of adjuvants used in perineural anesthesia and analgesia in comparison with ropivacaine. Reg Anesth Pain Med 2011; 36:225–230.
63. Takanami T, Hiruma H, Kaneko H, et al. Effects of sodium bisulfite with or without procaine derivatives on axons of cultured mouse dorsal root ganglion neurons. Reg Anesth Pain Med 2015; 40:62–67.
64. Steinfeldt T, Graf J, Schneider J, et al. Histological consequences of needle-nerve contact following nerve stimulation in a pig model. Anesthesiol Res Pract 2011; 2011:591851.
65. Steinfeldt T, Poeschi S, Nimphius W, et al. Forced needle advancement during needle-nerve contact in a porcine model: histological outcome. Anesth Analg 2011; 113:417–420.
66. Pintaric TS, Cvelko E, Strubenc M, et al. Intraneural and perineural inflammatory changes in piglets after injection of ultrasound gel, endotoxin, 0.9% NaCl, or needle insertion without injection. Anesth Analg 2014; 118:869–873.
67. Steinfeldt T, Wiesmann T, Nimphius W, et al. Perineural hematoma may result in nerve inflammation and myelin damage. Reg Anesth Pain Med 2014; 39:513–519.
68. Scopel GP, Marques Faria JC, Orpheu SC, et al. Intraneural hematoma with extrinsic compression: experimental study in rats and therapeutic options. J Reconst Microsurg 2007; 23:275–281.
69. Klein MM, Lee JW, Siegle SM, et al. Endoneurial pathology of the needlestick-nerve-injury model of complex regional pain syndrome, including rats with and without pain behaviors. Eur J Pain 2012; 16:28–37.
70. Yayama T, Kobayashi S, Nakanishi Y, et al. Effects of graded mechanical compression of rabbit sciatic nerve on nerve blood flow and electrophysiological properties. J Clin Neurosci 2010; 17:501–505.
71. Cuvillon P, Reubrech V, Zoric L, et al. Comparison of subgluteal sciatic nerve block duration in type 2 diabetic and nondiabetic patients. Br J Anaesth 2013; 110:823–830.
72. Lirk P, Verhamme C, Boeckh R, et al. Effects of early and late neuropathy on sciatic nerve block duration and neurotoxicity in Zucker diabetic fatty rats. Br J Anaesth 2015; 114:319–326.
73. Keyl C, Held T, Albiez G, et al. Increased electrical nerve stimulation threshold of the sciatic nerve in patients with diabetic foot gangrene: a prospective parallel cohort study. Eur J Anaesthesiol 2013; 30:435–440.
74. Weisman A, Bril V, Ngo M, et al. Identification and prediction of diabetic sensorimotor polyneuropathy using individual and simple combination of nerve conduction study parameters. PLoS One 2013; 8:e58783.
75. Shankar H, Pagel PS. Potential adverse ultrasound-related biological effects: a critical review. Anesthesiology 2011; 115:1109–1124.
76. Padua L, Di Pasquale A, Liotta G, et al. Ultrasound as a useful tool in the diagnosis and management of traumatic nerve lesions. Clin Neurophysiol 2013; 124:1237–1243.
77. Riazi S, Bril V, Perkins BA, et al. Can ultrasound of the tibial nerve detect diabetic peripheral neuropathy? A cross-sectional study. Diabetes Care 2012; 35:2575–2579.
78. Hobson-Webb LD, Padua L, Martinoli C. Ultrasonography in the diagnosis of peripheral nerve disease. Expert Opin Med Diagn 2012; 6:457–471.
79. Ali ZS, Pisapia JM, Ma TS, et al. Ultrasonographic evaluation of peripheral nerves. World Neurosurg 2016; 85:333–339.
© 2017 European Society of Anaesthesiology