Peripheral neuropathy is a general term that refers to any disorder resulting in deranged function and/or structure of any peripheral nerve (sensory, motor, or autonomic).

Background The overall prevalence of the condition is about 2400 (2.4%) per 100 000 populations, but in people older than 55 years, the prevalence rises to about 8000 (8%) per 100 000 (1). Diabetic polyneuropathy is the most common cause of peripheral neuropathy (34%). Even after thorough investigation, about a quarter of neuropathies are deemed idiopathic. the third place (13%) is shared by inflammatory and toxic (alcohol, chemotherapy induced) neuropathies, followed by hereditary (6%) and other neuropathies (2%).

Presenting features may be altered sensation, pain, muscle weakness, atrophy, and/or autonomic symptoms. Electrodiagnostic studies are helpful in establishing whether the distribution is consistent with mononeuropathy, mononeuropathy multiplex, plexopathy, or polyneuropathy; in differentiating between demyelinating or axonal types. Importantly, they also serve to localise the site of injury and determine the severity of the lesion.

Mononeuropathy implies a focal lesion of a single peripheral nerve. the most common mononeuropathy is carpal tunnel syndrome caused by entrapment of the median nerve in the carpal tunnel, followed closely by ulnar neuropathy due to compression of the nerve at or near the elbow.

Mononeuropathy multiplex describes the involvement of multiple separate non-contiguous peripheral nerves, either simultaneously or serially, with a random, multifocal, and frequently rapidly evolving pattern. Assessment of these patients is an urgent matter as many of them have vasculitis associated with a systemic disease.

Polineuropathy is most commonly distal symmetrical. the prototype is chronic sensory and motor polyneuropathy associated with diabetes mellitus, progressing towards ‘stocking-glove’ sensory loss, distal muscle wasting and weakness, and absent tendon reflexes.

Acute polyneuropathy presenting as rapidly progressive paralysis is usually one of the variants of Guillain-Barre syndrome, which can progress rapidly to respiratory failure.

Chronic polyneuropathies are further classified on the basis of electrophysiological findings into: i) Demyelinating: with uniform symmetrical slowing of nerve conduction usually indicates a genetically determined neuropathy (Charcot-Marie-Tooth disease, hereditary liability to pressure palsies), whereas multifocal slowing and conduction block are indicative of acquired demyelinating neuropathies (chronic inflammatory demyelinating polyradiculopathy, paraproteinaemic etc).

ii) Axonal: the most common variety, with diabetes mellitus being the leading cause. Others include nutritional deficiencies, chronic renal failure, malignant diseases as well as exogenous intoxications from medications, alcohol or chemical agents.

Where lies the problem?Regional anaesthesia is generally considered safe, but several risk factors may contribute to new-onset neurological deficit. Because each of the afore mentioned clinical conditions entail compromise to neural structures, the concern is that further insult from surgical or anaesthetic causes may result in new or worsening postoperative neurologic deficits.

Complicating issues are:

Double-crush phenomenon, whereby a second minor insult may result in impaired axoplasmic flux to the point of denervation; or indeed a single minor insult to the already affected axoplasmic flux (in subclinical neuropathies) may result in denervation. Thus, the performance of peripheral or neuraxial regional techniques in patients with pre-existing neuropathies may theoretically place them at increased risk of a double-crush phenomenon (3).

Postsurgical inflammatory neuropathy (PSIN) is a recently described process clinically manifested in the form of severe postoperative neurologic deficits (focal, multifocal, or diffuse). It is an idiopathic, immune-mediated response to a physiologic stress such as an infectious process, a vaccination, or a surgical procedure. Furthermore, the onset may not be apparent during the immediate postoperative period, and the deficit may be anatomically remote from the surgical site/regional anaesthesia needle entry point. the treatment of choice is suppression of immune response (with prolonged high dose corticosteroids or iv immunoglobulins) in an effort to sufficiently blunt the inflammatory response to allow for axonal regeneration (3).

To block or not to block?The approach to the patient with a pre-existing neuropathy presenting to the operating theatre as a potential candidate for regional anaesthesia entails performing the standard preoperative evaluation and documentation with special emphasis on the neurologic examination. It is imperative to document baseline neurologic function and sensory and motor deficits.

If regional anaesthesia is considered a prudent choice for these patients, a comprehensive discussion detailing the relevant risks, benefits, and alternatives should be undertaken and documented. Although the scientific evidence is still inadequate to determine the likelihood of developing new neuropathic pain in these patients, the incidence in the general population may and should be used as the best-case scenario (4). Brull R et al have indicated that the incidence of perioperative neuropathy was generally 100 times higher after peripheral nerve blocks (PNBs) (2.84% after interscalene, 1.48% after axillary, 0.34% after femoral block) than after neuraxial (0.04% and 0.02% after spinal and epidural, respectively) blocks, with the likelihood of complete resolution being much higher after PNB-induced neuropathy (5).

While the most conservative legal approach is to avoid regional anaesthesia in patients with pre-existing neuropathy, the decision to proceed with regional anaesthesia should be made on a case-by-case basis (4).

Nerve blocks may be desirable for their own benefits; or to avoid general anaesthesia and minimise the surgical stress response in vulnerable patients with comorbidities. (6).

In addition, caution should be used to minimise potential risk factors for perioperative nerve injury at each step of patient care.

Hereditary peripheral neuropathy The most common inherited neuropathy is Charcot-Marie-Tooth (CMT) disease. It affects 1:2500 people, caused by mutations in more than 30 genes responsible for manufacturing neurons or the myelin sheath (3). Small case series and anecdotal case reports describe no worsening of the neurological condition after nerve blocks. of note, an analgesic supraclavicular block using 30 ml of bupivacaine 0.5% lasted 30 hours (7).

Hereditary neuropathy with liability to pressure palsy (HNPP) is another rare (2–5 in 100,000) inherited demyelinating peripheral neuropathy linked to a mutation on the PMP-22 gene resulting in reduced myelin production. Patients suffer from repeated motor and sensory neuropathies following brief nerve compression or mild trauma (pressure palsies) (3). the main feature of the condition is sausage-shaped peripheral nerves, called tomaculas, resulting from abnormal myelin thickness. a number of case reports describe uneventful recovery with no worsening of the neurologic condition following labour epidural analgesia (8,9). Another case report describes the development of intraoperative nerve injury and subsequent diagnosis of HNPP in a woman after breast surgery under general anaesthesia (10). Two clinical implications have emerged: i) considering HNPP for differential diagnosis in patients who develop postoperative neuropathy without obvious risk factors ii) ultrasound guidance may be ideally suited for the performance of nerve blocks as it enables visualisation and avoidance of the enlargement during needle advancement toward target nerves (3).

Recommendation: Although definitive recommendations cannot be made, PNB appears safe in patients with stable CMT and HNPP disease states without worsening of their neurologic symptoms. Universal precautions to minimise modifiable risk factors for nerve injury should be taken (3).

Acquired peripheral neuropathy Diabetic polyneuropathy (DPN) is the most common systemic polyneuropathy with the prototype being a distal symmetric sensorimotor pattern. It is present in 4–5% of patients at the time of diagnosis with >50% of long-standing diabetic patients being symptomatic. Nearly all patients have evidence of abnormal nerve conduction (3). the increasing prevalence of diabetes mellitus (DM), increased likelihood of surgery will translate into a larger number of diabetic patients presenting for surgery. In addition, they may present with either undiagnosed DM or known diabetes with uncontrolled hyperglycaemia.

There is an abundance of animal and human data suggesting that diabetic nerves may have an increased risk of neurologic injury after regional anaesthesia compared with non-diabetic nerves. What do we know so far?

• Diabetic nerves have decreased sensitivity to electrical stimulation (11). This may increase the risk of intraneural injection if reliant solely on this technique. US has the added ability to use the cross–sectional area of a peripheral nerve to both identify a neuropathy and intraneural injection.

• Block success rates are higher in diabetic patients (12).

• Block duration is longer thus obviating the need for adding vasoconstrictors (ie adrenaline) (13).

• Minimum effective anaesthetic concentration (ED50) in vivo to achieve a motor sciatic block in diabetic rats is about two thirds of that required in non–diabetic rats (14).

• Neurotoxicity of LA is experimentally proven, clinically inconclusive (15).

Although the concern for an increased susceptibility to injury appears real, these patients should not be deprived of the benefits of regional anaesthesia. the anaesthetic technique should be instead modified in order to minimise the potential additive risk (US guidance, decreased concentration, reduced total dose, no vasoconstrictors) (3). the diabetic nerve appears swollen on ultrasound due to the increased water content, a by-product from the conversion of glucose to sorbitol. Interestingly, in about 20% of nerves with moderate to advanced diabetic neuropathy intraneural blood flow is detected (16).

Chemotherapy-induced neuropathy (CIPN) occurs in up to 50% of patients on chemotherapy. Patients who recover from CIPN have an increased risk of developing progressive neuropathic symptoms if exposed to additional neurotoxic agents. In addition, it is common for patients to have a subclinical neuropathy that presents only following a second neurologic insult, such as a PNB (3). a retrospective study of 216 PNBs performed in 186 patients previously treated with chemotherapy revealed a 2.2% incidence of peripheral nerve injury (not dissimilar to the baseline incidence in the general population) (17). Although PNB may be safely performed in the presence of CIPN, the possibility of sustaining additional neurologic injury should be considered and caution exercised (3).

Inflammatory neuropathies Guillain-Barré syndrome (GBS) is an acute, inflammatory, demyelinating polyneuropathy. Infection, pregnancy, vaccinations, immunosuppression, systemic illnesses, transfusion have all been incriminated. There are several reports of GBS occurring in the postoperative period after a variety of surgical procedures and types of anaesthetics. Although there are no case reports of PNB in patients with GBS, the general wisdom is that acute neuronal inflammation may be a relative contraindication to regional anaesthesia. the current literature can neither support nor refute this claim. Ultimately, the decision should be made on an individual basis and avoidance of muscle relaxants, opioids, intubation may prevail (3).

Multiple sclerosis (MS) is an inflammatory degenerative disease of poorly defined aetiology, characterised by demyelination lesions in the brain and spinal cord. Neurologic symptoms may worsen postoperatively due to any stressor regardless of the anaesthetic technique. Historically, PNBs have been considered safe until evidence documenting peripheral demyelinating lesions (5–47% of MS patients) emerged (18). PNB has not been definitely shown to be harmful in the setting of MS and thus should not be considered an absolute contraindication. Given that demyelinated fibres may be more prone to LA toxicity, reducing the LA concentration and total dose to the lowest effective level may be prudent in addition to the general precautions. Patients should also be informed about the risk of new or worsening neurologic symptoms during the postoperative period because of exposure to multiple exacerbating factors (3).

In conclusion, no absolute method exists for predicting whether a pre-existing neuropathy will remain static or become exacerbated following PNB. Each case must be evaluated individually, a full appraisal of risks and benefits, and the alternatives to regional anaesthesia must be discussed and documented. Best efforts should be made to reduce the modifiable risk factors to improve the chance for uneventful perioperative events. Ultrasound guidance for both nerve location and dose reduction purposes should be considered.

Further reading 1. Martyn CN, et al. J Neurol Neurosurg Psychiatry 1977;62:310–18.

2. Hanewickel R, et al. Handb Clin Neurol 2016;138:263–82.

3. Kopp SL, et al. Reg Anesth Pain Med 2015;40:467–78.

4. Candido KD. Anesthesiology News Feb 2019.

5. Brull R, et al. Anesth Analg 2007;10:965–74.

6. Lirk P, et al. Br J Anesth 2019;122:16–8.

7. Bui AH, et al. Can J Anaesth 2008;55:718–9.

8. Lepski GR, et al. Int J Obstet Anesth 2001;10:198–201.

9. Berdai S, at al. Ann Fr Anesth Reanim 2004;23:1011–4.

10. Wijayasiri L, at al. Anaesthesia 2006;61:1004–6.

11. Keyl C, et al. Eur J Anaesthesiol 2013;30:435–40.

12. Gebhard RE, et al. Reg Anesth Pain Med 2009;34:404–7.

13. Cuvillon P, et al. Br J Anaesth 2013;110:823–30.

14. Ten Hoope W, et al. Anesthesiology 2018;128:609–19.

15. Kroin JS, at al. Reg Anesth Pain Med 2010;35:343–50.

16. Borire A, et al. Neurology 2018;90(15 Suppl)P1.437.

17. Abcejo AS, et al. Reg Anesth Pain Med 2016;41:685–90.

18. Misawa S, at al. Clin Neurophysiol 2008;119:1829–33.